Copyright 2014-2018 The Khronos Group Inc.

This Specification is protected by copyright laws and contains material proprietary to Khronos. Except as described by these terms, it or any components may not be reproduced, republished, distributed, transmitted, displayed, broadcast or otherwise exploited in any manner without the express prior written permission of Khronos. Khronos grants a conditional copyright license to use and reproduce the unmodified Specification for any purpose, without fee or royalty, EXCEPT no licenses to any patent, trademark or other intellectual property rights are granted under these terms.

Khronos makes no, and expressly disclaims any, representations or warranties, express or implied, regarding this Specification, including, without limitation: merchantability, fitness for a particular purpose, non-infringement of any intellectual property, correctness, accuracy, completeness, timeliness, and reliability. Under no circumstances will Khronos, or any of its Promoters, Contributors or Members, or their respective partners, officers, directors, employees, agents or representatives be liable for any damages, whether direct, indirect, special or consequential damages for lost revenues, lost profits, or otherwise, arising from or in connection with these materials.

This Specification has been created under the Khronos Intellectual Property Rights Policy, which is Attachment A of the Khronos Group Membership Agreement available at www.khronos.org/files/member_agreement.pdf, and which defines the terms 'Scope', 'Compliant Portion', and 'Necessary Patent Claims'. Parties desiring to implement the Specification and make use of Khronos trademarks in relation to that implementation, and receive reciprocal patent license protection under the Khronos Intellectual Property Rights Policy must become Adopters and confirm the implementation as conformant under the process defined by Khronos for this Specification; see https://www.khronos.org/adopters.

This Specification contains substantially unmodified functionality from, and is a successor to, Khronos specifications including OpenGL, OpenGL ES and OpenCL.

Some parts of this Specification are purely informative and so are EXCLUDED from the Scope of this Specification. The Document Conventions section of the Introduction defines how these parts of the Specification are identified.

Where this Specification uses technical terminology, defined in the Glossary or otherwise, that refer to enabling technologies that are not expressly set forth in this Specification, those enabling technologies are EXCLUDED from the Scope of this Specification. For clarity, enabling technologies not disclosed with particularity in this Specification (e.g. semiconductor manufacturing technology, hardware architecture, processor architecture or microarchitecture, memory architecture, compiler technology, object oriented technology, basic operating system technology, compression technology, algorithms, and so on) are NOT to be considered expressly set forth; only those application program interfaces and data structures disclosed with particularity are included in the Scope of this Specification.

For purposes of the Khronos Intellectual Property Rights Policy as it relates to the definition of Necessary Patent Claims, all recommended or optional features, behaviors and functionality set forth in this Specification, if implemented, are considered to be included as Compliant Portions.

Where this Specification includes normative references to external documents, only the specifically identified sections of those external documents are INCLUDED in the Scope of this Specification. If not created by Khronos, those external documents may contain contributions from non-members of Khronos not covered by the Khronos Intellectual Property Rights Policy.

Vulkan is a registered trademark, and Khronos is a trademark of The Khronos Group Inc. ASTC is a trademark of ARM Holdings PLC; OpenCL is a trademark of Apple Inc.; and OpenGL is a registered trademark of Silicon Graphics International, all used under license by Khronos. All other product names, trademarks, and/or company names are used solely for identification and belong to their respective owners.

1. Introduction

This document, referred to as the “Vulkan Specification” or just the “Specification” hereafter, describes the Vulkan Application Programming Interface (API). Vulkan is a C99 API designed for explicit control of low-level graphics and compute functionality.

The canonical version of the Specification is available in the official Vulkan Registry (http://www.khronos.org/registry/vulkan/). The source files used to generate the Vulkan specification are stored in the Vulkan Documentation Repository (https://github.com/KhronosGroup/Vulkan-Docs). The source repository additionally has a public issue tracker and allows the submission of pull requests that improve the specification.

1.1. Document Conventions

The Vulkan specification is intended for use by both implementors of the API and application developers seeking to make use of the API, forming a contract between these parties. Specification text may address either party; typically the intended audience can be inferred from context, though some sections are defined to address only one of these parties. (For example, Valid Usage sections only address application developers). Any requirements, prohibitions, recommendations or options defined by normative terminology are imposed only on the audience of that text.

Note

Structure and enumerated types defined in extensions that were promoted to core in Vulkan 1.1 are now defined in terms of the equivalent Vulkan 1.1 interfaces. This affects the Vulkan Specification, the Vulkan header files, and the corresponding XML Registry.

1.1.1. Normative Terminology

Within this specification, the key words must, required, should, recommended, may, and optional are to be interpreted as described in RFC 2119 - Key words for use in RFCs to Indicate Requirement Levels (http://www.ietf.org/rfc/rfc2119.txt). These key words are highlighted in the specification for clarity. In text addressing application developers, their use expresses requirements that apply to application behavior. In text addressing implementors, their use expresses requirements that apply to implementations.

In text addressing application developers, the additional key words can and cannot are to be interpreted as describing the capabilities of an application, as follows:

can

This word means that the application is able to perform the action described.

cannot

This word means that the API and/or the execution environment provide no mechanism through which the application can express or accomplish the action described.

These key words are never used in text addressing implementors.

Note

There is an important distinction between cannot and must not, as used in this Specification. Cannot means something the application literally is unable to express or accomplish through the API, while must not means something that the application is capable of expressing through the API, but that the consequences of doing so are undefined and potentially unrecoverable for the implementation.

Unless otherwise noted in the section heading, all sections and appendices in this document are normative.

1.1.2. Technical Terminology

The Vulkan Specification makes use of common engineering and graphics terms such as Pipeline, Shader, and Host to identify and describe Vulkan API constructs and their attributes, states, and behaviors. The Glossary defines the basic meanings of these terms in the context of the Specification. The Specification text provides fuller definitions of the terms and may elaborate, extend, or clarify the Glossary definitions. When a term defined in the Glossary is used in normative language within the Specification, the definitions within the Specification govern and supersede any meanings the terms may have in other technical contexts (i.e. outside the Specification).

1.1.3. Normative References

References to external documents are considered normative references if the Specification uses any of the normative terms defined in Normative Terminology to refer to them or their requirements, either as a whole or in part.

The following documents are referenced by normative sections of the specification:

IEEE Standard for Floating-Point Arithmetic, IEEE Std 754-2008, http://dx.doi.org/10.1109/IEEESTD.2008.4610935, August, 2008.

A. Garrard, Khronos Data Format Specification, version 1.2, https://www.khronos.org/registry/DataFormat/specs/1.2/dataformat.1.2.html, September, 2017.

J. Kessenich, SPIR-V Extended Instructions for GLSL, Version 1.00, https://www.khronos.org/registry/spir-v/, February 10, 2016.

J. Kessenich and B. Ouriel, The Khronos SPIR-V Specification, Version 1.00, https://www.khronos.org/registry/spir-v/, February 10, 2016.

J. Leech and T. Hector, Vulkan Documentation and Extensions: Procedures and Conventions, https://www.khronos.org/registry/vulkan/, July 11, 2016

2. Fundamentals

This chapter introduces fundamental concepts including the Vulkan architecture and execution model, API syntax, queues, pipeline configurations, numeric representation, state and state queries, and the different types of objects and shaders. It provides a framework for interpreting more specific descriptions of commands and behavior in the remainder of the Specification.

2.1. Host and Device Environment

The Vulkan Specification assumes and requires: the following properties of the host environment with respect to Vulkan implementations:

  • The host must have runtime support for 8, 16, 32 and 64-bit signed and unsigned twos-complement integers, all addressable at the granularity of their size in bytes.

  • The host must have runtime support for 32- and 64-bit floating-point types satisfying the range and precision constraints in the Floating Point Computation section.

  • The representation and endianness of these types on the host must match the representation and endianness of the same types on every physical device supported.

Note

Since a variety of data types and structures in Vulkan may be accessible by both host and physical device operations, the implementation should be able to access such data efficiently in both paths in order to facilitate writing portable and performant applications.

2.2. Execution Model

This section outlines the execution model of a Vulkan system.

Vulkan exposes one or more devices, each of which exposes one or more queues which may process work asynchronously to one another. The set of queues supported by a device is partitioned into families. Each family supports one or more types of functionality and may contain multiple queues with similar characteristics. Queues within a single family are considered compatible with one another, and work produced for a family of queues can be executed on any queue within that family. This Specification defines four types of functionality that queues may support: graphics, compute, transfer, and sparse memory management.

Note

A single device may report multiple similar queue families rather than, or as well as, reporting multiple members of one or more of those families. This indicates that while members of those families have similar capabilities, they are not directly compatible with one another.

Device memory is explicitly managed by the application. Each device may advertise one or more heaps, representing different areas of memory. Memory heaps are either device local or host local, but are always visible to the device. Further detail about memory heaps is exposed via memory types available on that heap. Examples of memory areas that may be available on an implementation include:

  • device local is memory that is physically connected to the device.

  • device local, host visible is device local memory that is visible to the host.

  • host local, host visible is memory that is local to the host and visible to the device and host.

On other architectures, there may only be a single heap that can be used for any purpose.

A Vulkan application controls a set of devices through the submission of command buffers which have recorded device commands issued via Vulkan library calls. The content of command buffers is specific to the underlying implementation and is opaque to the application. Once constructed, a command buffer can be submitted once or many times to a queue for execution. Multiple command buffers can be built in parallel by employing multiple threads within the application.

Command buffers submitted to different queues may execute in parallel or even out of order with respect to one another. Command buffers submitted to a single queue respect submission order, as described further in synchronization chapter. Command buffer execution by the device is also asynchronous to host execution. Once a command buffer is submitted to a queue, control may return to the application immediately. Synchronization between the device and host, and between different queues is the responsibility of the application.

2.2.1. Queue Operation

Vulkan queues provide an interface to the execution engines of a device. Commands for these execution engines are recorded into command buffers ahead of execution time. These command buffers are then submitted to queues with a queue submission command for execution in a number of batches. Once submitted to a queue, these commands will begin and complete execution without further application intervention, though the order of this execution is dependent on a number of implicit and explicit ordering constraints.

Work is submitted to queues using queue submission commands that typically take the form vkQueue* (e.g. vkQueueSubmit, vkQueueBindSparse), and optionally take a list of semaphores upon which to wait before work begins and a list of semaphores to signal once work has completed. The work itself, as well as signaling and waiting on the semaphores are all queue operations.

Queue operations on different queues have no implicit ordering constraints, and may execute in any order. Explicit ordering constraints between queues can be expressed with semaphores and fences.

Command buffer submissions to a single queue respect submission order and other implicit ordering guarantees, but otherwise may overlap or execute out of order. Other types of batches and queue submissions against a single queue (e.g. sparse memory binding) have no implicit ordering constraints with any other queue submission or batch. Additional explicit ordering constraints between queue submissions and individual batches can be expressed with semaphores and fences.

Before a fence or semaphore is signaled, it is guaranteed that any previously submitted queue operations have completed execution, and that memory writes from those queue operations are available to future queue operations. Waiting on a signaled semaphore or fence guarantees that previous writes that are available are also visible to subsequent commands.

Command buffer boundaries, both between primary command buffers of the same or different batches or submissions as well as between primary and secondary command buffers, do not introduce any additional ordering constraints. In other words, submitting the set of command buffers (which can include executing secondary command buffers) between any semaphore or fence operations execute the recorded commands as if they had all been recorded into a single primary command buffer, except that the current state is reset on each boundary. Explicit ordering constraints can be expressed with explicit synchronization primitives.

There are a few implicit ordering guarantees between commands within a command buffer, but only covering a subset of execution. Additional explicit ordering constraints can be expressed with the various explicit synchronization primitives.

Note

Implementations have significant freedom to overlap execution of work submitted to a queue, and this is common due to deep pipelining and parallelism in Vulkan devices.

Commands recorded in command buffers either perform actions (draw, dispatch, clear, copy, query/timestamp operations, begin/end subpass operations), set state (bind pipelines, descriptor sets, and buffers, set dynamic state, push constants, set render pass/subpass state), or perform synchronization (set/wait events, pipeline barrier, render pass/subpass dependencies). Some commands perform more than one of these tasks. State setting commands update the current state of the command buffer. Some commands that perform actions (e.g. draw/dispatch) do so based on the current state set cumulatively since the start of the command buffer. The work involved in performing action commands is often allowed to overlap or to be reordered, but doing so must not alter the state to be used by each action command. In general, action commands are those commands that alter framebuffer attachments, read/write buffer or image memory, or write to query pools.

Synchronization commands introduce explicit execution and memory dependencies between two sets of action commands, where the second set of commands depends on the first set of commands. These dependencies enforce that both the execution of certain pipeline stages in the later set occur after the execution of certain stages in the source set, and that the effects of memory accesses performed by certain pipeline stages occur in order and are visible to each other. When not enforced by an explicit dependency or implicit ordering guarantees, action commands may overlap execution or execute out of order, and may not see the side effects of each other’s memory accesses.

The device executes queue operations asynchronously with respect to the host. Control is returned to an application immediately following command buffer submission to a queue. The application must synchronize work between the host and device as needed.

2.3. Object Model

The devices, queues, and other entities in Vulkan are represented by Vulkan objects. At the API level, all objects are referred to by handles. There are two classes of handles, dispatchable and non-dispatchable. Dispatchable handle types are a pointer to an opaque type. This pointer may be used by layers as part of intercepting API commands, and thus each API command takes a dispatchable type as its first parameter. Each object of a dispatchable type must have a unique handle value during its lifetime.

Non-dispatchable handle types are a 64-bit integer type whose meaning is implementation-dependent, and may encode object information directly in the handle rather than acting as a reference to an underlying object. Objects of a non-dispatchable type may not have unique handle values within a type or across types. If handle values are not unique, then destroying one such handle must not cause identical handles of other types to become invalid, and must not cause identical handles of the same type to become invalid if that handle value has been created more times than it has been destroyed.

All objects created or allocated from a VkDevice (i.e. with a VkDevice as the first parameter) are private to that device, and must not be used on other devices.

2.3.1. Object Lifetime

Objects are created or allocated by vkCreate* and vkAllocate* commands, respectively. Once an object is created or allocated, its “structure” is considered to be immutable, though the contents of certain object types is still free to change. Objects are destroyed or freed by vkDestroy* and vkFree* commands, respectively.

Objects that are allocated (rather than created) take resources from an existing pool object or memory heap, and when freed return resources to that pool or heap. While object creation and destruction are generally expected to be low-frequency occurrences during runtime, allocating and freeing objects can occur at high frequency. Pool objects help accommodate improved performance of the allocations and frees.

It is an application’s responsibility to track the lifetime of Vulkan objects, and not to destroy them while they are still in use.

The ownership of application-owned memory is immediately acquired by any Vulkan command it is passed into. Ownership of such memory must be released back to the application at the end of the duration of the command, so that the application can alter or free this memory as soon as all the commands that acquired it have returned.

The following object types are consumed when they are passed into a Vulkan command and not further accessed by the objects they are used to create. They must not be destroyed in the duration of any API command they are passed into:

  • VkShaderModule

  • VkPipelineCache

  • VkValidationCacheEXT

A VkRenderPass object passed as a parameter to create another object is not further accessed by that object after the duration of the command it is passed into. A VkRenderPass used in a command buffer follows the rules described below.

A VkPipelineLayout object must not be destroyed while any command buffer that uses it is in the recording state.

VkDescriptorSetLayout objects may be accessed by commands that operate on descriptor sets allocated using that layout, and those descriptor sets must not be updated with vkUpdateDescriptorSets after the descriptor set layout has been destroyed. Otherwise, a VkDescriptorSetLayout object passed as a parameter to create another object is not further accessed by that object after the duration of the command it is passed into.

The application must not destroy any other type of Vulkan object until all uses of that object by the device (such as via command buffer execution) have completed.

The following Vulkan objects must not be destroyed while any command buffers using the object are in the pending state:

  • VkEvent

  • VkQueryPool

  • VkBuffer

  • VkBufferView

  • VkImage

  • VkImageView

  • VkPipeline

  • VkSampler

  • VkSamplerYcbcrConversion

  • VkDescriptorPool

  • VkFramebuffer

  • VkRenderPass

  • VkCommandBuffer

  • VkCommandPool

  • VkDeviceMemory

  • VkDescriptorSet

  • VkObjectTableNVX

  • VkIndirectCommandsLayout

Destroying these objects will move any command buffers that are in the recording or executable state, and are using those objects, to the invalid state.

The following Vulkan objects must not be destroyed while any queue is executing commands that use the object:

  • VkFence

  • VkSemaphore

  • VkCommandBuffer

  • VkCommandPool

In general, objects can be destroyed or freed in any order, even if the object being freed is involved in the use of another object (e.g. use of a resource in a view, use of a view in a descriptor set, use of an object in a command buffer, binding of a memory allocation to a resource), as long as any object that uses the freed object is not further used in any way except to be destroyed or to be reset in such a way that it no longer uses the other object (such as resetting a command buffer). If the object has been reset, then it can be used as if it never used the freed object. An exception to this is when there is a parent/child relationship between objects. In this case, the application must not destroy a parent object before its children, except when the parent is explicitly defined to free its children when it is destroyed (e.g. for pool objects, as defined below).

VkCommandPool objects are parents of VkCommandBuffer objects. VkDescriptorPool objects are parents of VkDescriptorSet objects. VkDevice objects are parents of many object types (all that take a VkDevice as a parameter to their creation).

The following Vulkan objects have specific restrictions for when they can be destroyed:

  • VkQueue objects cannot be explicitly destroyed. Instead, they are implicitly destroyed when the VkDevice object they are retrieved from is destroyed.

  • Destroying a pool object implicitly frees all objects allocated from that pool. Specifically, destroying VkCommandPool frees all VkCommandBuffer objects that were allocated from it, and destroying VkDescriptorPool frees all VkDescriptorSet objects that were allocated from it.

  • VkDevice objects can be destroyed when all VkQueue objects retrieved from them are idle, and all objects created from them have been destroyed. This includes the following objects:

    • VkFence

    • VkSemaphore

    • VkEvent

    • VkQueryPool

    • VkBuffer

    • VkBufferView

    • VkImage

    • VkImageView

    • VkShaderModule

    • VkPipelineCache

    • VkPipeline

    • VkPipelineLayout

    • VkSampler

    • VkSamplerYcbcrConversion

    • VkDescriptorSetLayout

    • VkDescriptorPool

    • VkFramebuffer

    • VkRenderPass

    • VkCommandPool

    • VkCommandBuffer

    • VkDeviceMemory

  • VkValidationCacheEXT

  • VkPhysicalDevice objects cannot be explicitly destroyed. Instead, they are implicitly destroyed when the VkInstance object they are retrieved from is destroyed.

  • VkInstance objects can be destroyed once all VkDevice objects created from any of its VkPhysicalDevice objects have been destroyed.

2.3.2. External Object Handles

As defined above, the scope of object handles created or allocated from a VkDevice is limited to that logical device. Objects which are not in scope are said to be external. To bring an external object into scope, an external handle must be exported from the object in the source scope and imported into the destination scope.

Note

The scope of external handles and their associated resources may vary according to their type, but they can generally be shared across process and API boundaries.

2.4. Application Binary Interface

The mechanism by which Vulkan is made available to applications is platform- or implementation- defined. On many platforms the C interface described in this Specification is provided by a shared library. Since shared libraries can be changed independently of the applications that use them, they present particular compatibility challenges, and this Specification places some requirements on them.

Shared library implementations must use the default Application Binary Interface (ABI) of the standard C compiler for the platform, or provide customized API headers that cause application code to use the implementation’s non-default ABI. An ABI in this context means the size, alignment, and layout of C data types; the procedure calling convention; and the naming convention for shared library symbols corresponding to C functions. Customizing the calling convention for a platform is usually accomplished by defining calling convention macros appropriately in vk_platform.h.

On platforms where Vulkan is provided as a shared library, library symbols beginning with “vk” and followed by a digit or uppercase letter are reserved for use by the implementation. Applications which use Vulkan must not provide definitions of these symbols. This allows the Vulkan shared library to be updated with additional symbols for new API versions or extensions without causing symbol conflicts with existing applications.

Shared library implementations should provide library symbols for commands in the highest version of this Specification they support, and for Window System Integration extensions relevant to the platform. They may also provide library symbols for commands defined by additional extensions.

Note

These requirements and recommendations are intended to allow implementors to take advantage of platform-specific conventions for SDKs, ABIs, library versioning mechanisms, etc. while still minimizing the code changes necessary to port applications or libraries between platforms. Platform vendors, or providers of the de facto standard Vulkan shared library for a platform, are encouraged to document what symbols the shared library provides and how it will be versioned when new symbols are added.

Applications should only rely on shared library symbols for commands in the minimum core version required by the application. vkGetInstanceProcAddr and vkGetDeviceProcAddr should be used to obtain function pointers for commands in core versions beyond the application’s minimum required version.

2.5. Command Syntax and Duration

The Specification describes Vulkan commands as functions or procedures using C99 syntax. Language bindings for other languages such as C++ and JavaScript may allow for stricter parameter passing, or object-oriented interfaces.

Vulkan uses the standard C types for the base type of scalar parameters (e.g. types from <stdint.h>), with exceptions described below, or elsewhere in the text when appropriate:

VkBool32 represents boolean True and False values, since C does not have a sufficiently portable built-in boolean type:

typedef uint32_t VkBool32;

VK_TRUE represents a boolean True (integer 1) value, and VK_FALSE a boolean False (integer 0) value.

All values returned from a Vulkan implementation in a VkBool32 will be either VK_TRUE or VK_FALSE.

Applications must not pass any other values than VK_TRUE or VK_FALSE into a Vulkan implementation where a VkBool32 is expected.

VkDeviceSize represents device memory size and offset values:

typedef uint64_t VkDeviceSize;

Commands that create Vulkan objects are of the form vkCreate* and take Vk*CreateInfo structures with the parameters needed to create the object. These Vulkan objects are destroyed with commands of the form vkDestroy*.

The last in-parameter to each command that creates or destroys a Vulkan object is pAllocator. The pAllocator parameter can be set to a non-NULL value such that allocations for the given object are delegated to an application provided callback; refer to the Memory Allocation chapter for further details.

Commands that allocate Vulkan objects owned by pool objects are of the form vkAllocate*, and take Vk*AllocateInfo structures. These Vulkan objects are freed with commands of the form vkFree*. These objects do not take allocators; if host memory is needed, they will use the allocator that was specified when their parent pool was created.

Commands are recorded into a command buffer by calling API commands of the form vkCmd*. Each such command may have different restrictions on where it can be used: in a primary and/or secondary command buffer, inside and/or outside a render pass, and in one or more of the supported queue types. These restrictions are documented together with the definition of each such command.

The duration of a Vulkan command refers to the interval between calling the command and its return to the caller.

2.5.1. Lifetime of Retrieved Results

Information is retrieved from the implementation with commands of the form vkGet* and vkEnumerate*.

Unless otherwise specified for an individual command, the results are invariant; that is, they will remain unchanged when retrieved again by calling the same command with the same parameters, so long as those parameters themselves all remain valid.

2.6. Threading Behavior

Vulkan is intended to provide scalable performance when used on multiple host threads. All commands support being called concurrently from multiple threads, but certain parameters, or components of parameters are defined to be externally synchronized. This means that the caller must guarantee that no more than one thread is using such a parameter at a given time.

More precisely, Vulkan commands use simple stores to update the state of Vulkan objects. A parameter declared as externally synchronized may have its contents updated at any time during the host execution of the command. If two commands operate on the same object and at least one of the commands declares the object to be externally synchronized, then the caller must guarantee not only that the commands do not execute simultaneously, but also that the two commands are separated by an appropriate memory barrier (if needed).

Note

Memory barriers are particularly relevant for hosts based on the ARM CPU architecture, which is more weakly ordered than many developers are accustomed to from x86/x64 programming. Fortunately, most higher-level synchronization primitives (like the pthread library) perform memory barriers as a part of mutual exclusion, so mutexing Vulkan objects via these primitives will have the desired effect.

Similarly the application must avoid any potential data hazard of application-owned memory that has its ownership temporarily acquired by a Vulkan command. While the ownership of application-owned memory remains acquired by a command the implementation may read the memory at any point, and it may write non-const qualified memory at any point. Parameters referring to non-const qualified application-owned memory are not marked explicitly as externally synchronized in the Specification.

Many object types are immutable, meaning the objects cannot change once they have been created. These types of objects never need external synchronization, except that they must not be destroyed while they are in use on another thread. In certain special cases mutable object parameters are internally synchronized, making external synchronization unnecessary. One example of this is the use of a VkPipelineCache in vkCreateGraphicsPipelines and vkCreateComputePipelines, where external synchronization around such a heavyweight command would be impractical. The implementation must internally synchronize the cache in this example, and may be able to do so in the form of a much finer-grained mutex around the command. Any command parameters that are not labeled as externally synchronized are either not mutated by the command or are internally synchronized. Additionally, certain objects related to a command’s parameters (e.g. command pools and descriptor pools) may be affected by a command, and must also be externally synchronized. These implicit parameters are documented as described below.

Parameters of commands that are externally synchronized are listed below.

Externally Synchronized Parameters

There are also a few instances where a command can take in a user allocated list whose contents are externally synchronized parameters. In these cases, the caller must guarantee that at most one thread is using a given element within the list at a given time. These parameters are listed below.

Externally Synchronized Parameter Lists
  • Each element of the pWaitSemaphores member of each element of the pSubmits parameter in vkQueueSubmit

  • Each element of the pSignalSemaphores member of each element of the pSubmits parameter in vkQueueSubmit

  • Each element of the pWaitSemaphores member of each element of the pBindInfo parameter in vkQueueBindSparse

  • Each element of the pSignalSemaphores member of each element of the pBindInfo parameter in vkQueueBindSparse

  • The buffer member of each element of the pBufferBinds member of each element of the pBindInfo parameter in vkQueueBindSparse

  • The image member of each element of the pImageOpaqueBinds member of each element of the pBindInfo parameter in vkQueueBindSparse

  • The image member of each element of the pImageBinds member of each element of the pBindInfo parameter in vkQueueBindSparse

  • Each element of the pFences parameter in vkResetFences

  • Each element of the pDescriptorSets parameter in vkFreeDescriptorSets

  • The dstSet member of each element of the pDescriptorWrites parameter in vkUpdateDescriptorSets

  • The dstSet member of each element of the pDescriptorCopies parameter in vkUpdateDescriptorSets

  • Each element of the pCommandBuffers parameter in vkFreeCommandBuffers

  • Each element of the pWaitSemaphores member of the pPresentInfo parameter in vkQueuePresentKHR

  • Each element of the pSwapchains member of the pPresentInfo parameter in vkQueuePresentKHR

  • The surface member of each element of the pCreateInfos parameter in vkCreateSharedSwapchainsKHR

  • The oldSwapchain member of each element of the pCreateInfos parameter in vkCreateSharedSwapchainsKHR

In addition, there are some implicit parameters that need to be externally synchronized. For example, all commandBuffer parameters that need to be externally synchronized imply that the commandPool that was passed in when creating that command buffer also needs to be externally synchronized. The implicit parameters and their associated object are listed below.

Implicit Externally Synchronized Parameters

2.7. Errors

Vulkan is a layered API. The lowest layer is the core Vulkan layer, as defined by this Specification. The application can use additional layers above the core for debugging, validation, and other purposes.

One of the core principles of Vulkan is that building and submitting command buffers should be highly efficient. Thus error checking and validation of state in the core layer is minimal, although more rigorous validation can be enabled through the use of layers.

The core layer assumes applications are using the API correctly. Except as documented elsewhere in the Specification, the behavior of the core layer to an application using the API incorrectly is undefined, and may include program termination. However, implementations must ensure that incorrect usage by an application does not affect the integrity of the operating system, the Vulkan implementation, or other Vulkan client applications in the system. In particular, any guarantees made by an operating system about whether memory from one process can be visible to another process or not must not be violated by a Vulkan implementation for any memory allocation. Vulkan implementations are not required to make additional security or integrity guarantees beyond those provided by the OS unless explicitly directed by the application’s use of a particular feature or extension (e.g. via robust buffer access).

Note

For instance, if an operating system guarantees that data in all its memory allocations are set to zero when newly allocated, the Vulkan implementation must make the same guarantees for any allocations it controls (e.g. VkDeviceMemory).

Applications can request stronger robustness guarantees by enabling the robustBufferAccess feature as described in Features, Limits, and Formats.

Validation of correct API usage is left to validation layers. Applications should be developed with validation layers enabled, to help catch and eliminate errors. Once validated, released applications should not enable validation layers by default.

2.7.1. Valid Usage

Valid usage defines a set of conditions which must be met in order to achieve well-defined run-time behavior in an application. These conditions depend only on Vulkan state, and the parameters or objects whose usage is constrained by the condition.

Some valid usage conditions have dependencies on run-time limits or feature availability. It is possible to validate these conditions against Vulkan’s minimum supported values for these limits and features, or some subset of other known values.

Valid usage conditions do not cover conditions where well-defined behavior (including returning an error code) exists.

Valid usage conditions should apply to the command or structure where complete information about the condition would be known during execution of an application. This is such that a validation layer or linter can be written directly against these statements at the point they are specified.

Note

This does lead to some non-obvious places for valid usage statements. For instance, the valid values for a structure might depend on a separate value in the calling command. In this case, the structure itself will not reference this valid usage as it is impossible to determine validity from the structure that it is invalid - instead this valid usage would be attached to the calling command.

Another example is draw state - the state setters are independent, and can cause a legitimately invalid state configuration between draw calls; so the valid usage statements are attached to the place where all state needs to be valid - at the draw command.

Valid usage conditions are described in a block labelled “Valid Usage” following each command or structure they apply to.

2.7.2. Implicit Valid Usage

Some valid usage conditions apply to all commands and structures in the API, unless explicitly denoted otherwise for a specific command or structure. These conditions are considered implicit, and are described in a block labelled “Valid Usage (Implicit)” following each command or structure they apply to. Implicit valid usage conditions are described in detail below.

Valid Usage for Object Handles

Any input parameter to a command that is an object handle must be a valid object handle, unless otherwise specified. An object handle is valid if:

  • It has been created or allocated by a previous, successful call to the API. Such calls are noted in the Specification.

  • It has not been deleted or freed by a previous call to the API. Such calls are noted in the Specification.

  • Any objects used by that object, either as part of creation or execution, must also be valid.

The reserved values VK_NULL_HANDLE and NULL can be used in place of valid non-dispatchable handles and dispatchable handles, respectively, when explicitly called out in the Specification. Any command that creates an object successfully must not return these values. It is valid to pass these values to vkDestroy* or vkFree* commands, which will silently ignore these values.

Valid Usage for Pointers

Any parameter that is a pointer must be a valid pointer only if it is explicitly called out by a Valid Usage statement.

A pointer is “valid” if it points at memory containing values of the number and type(s) expected by the command, and all fundamental types accessed through the pointer (e.g. as elements of an array or as members of a structure) satisfy the alignment requirements of the host processor.

Valid Usage for Strings

Any parameter that is a pointer to char must be a finite sequence of values terminated by a null character, or if explicitly called out in the Specification, can be NULL.

Valid Usage for Enumerated Types

Any parameter of an enumerated type must be a valid enumerant for that type. A enumerant is valid if:

  • The enumerant is defined as part of the enumerated type.

  • The enumerant is not one of the special values defined for the enumerated type, which are suffixed with _BEGIN_RANGE, _END_RANGE, _RANGE_SIZE or _MAX_ENUM1.

    1

    The meaning of these special tokens is not exposed in the Vulkan Specification. They are not part of the API, and they should not be used by applications. Their original intended use was for internal consumption by Vulkan implementations. Even that use will no longer be supported in the future, but they will be retained for backwards compatibility reasons.

Any enumerated type returned from a query command or otherwise output from Vulkan to the application must not have a reserved value. Reserved values are values not defined by any extension for that enumerated type.

Note

This language is intended to accomodate cases such as “hidden” extensions known only to driver internals, or layers enabling extensions without knowledge of the application, without allowing return of values not defined by any extension.

Valid Usage for Flags

A collection of flags is represented by a bitmask using the type VkFlags:

typedef uint32_t VkFlags;

Bitmasks are passed to many commands and structures to compactly represent options, but VkFlags is not used directly in the API. Instead, a Vk*Flags type which is an alias of VkFlags, and whose name matches the corresponding Vk*FlagBits that are valid for that type, is used.

Any Vk*Flags member or parameter used in the API as an input must be a valid combination of bit flags. A valid combination is either zero or the bitwise OR of valid bit flags. A bit flag is valid if:

  • The bit flag is defined as part of the Vk*FlagBits type, where the bits type is obtained by taking the flag type and replacing the trailing Flags with FlagBits. For example, a flag value of type VkColorComponentFlags must contain only bit flags defined by VkColorComponentFlagBits.

  • The flag is allowed in the context in which it is being used. For example, in some cases, certain bit flags or combinations of bit flags are mutually exclusive.

Any Vk*Flags member or parameter returned from a query command or otherwise output from Vulkan to the application may contain bit flags undefined in its corresponding Vk*FlagBits type. An application cannot rely on the state of these unspecified bits.

Valid Usage for Structure Types

Any parameter that is a structure containing a sType member must have a value of sType which is a valid VkStructureType value matching the type of the structure.

Structure types supported by the Vulkan API include:

typedef enum VkStructureType {
    VK_STRUCTURE_TYPE_APPLICATION_INFO = 0,
    VK_STRUCTURE_TYPE_INSTANCE_CREATE_INFO = 1,
    VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO = 2,
    VK_STRUCTURE_TYPE_DEVICE_CREATE_INFO = 3,
    VK_STRUCTURE_TYPE_SUBMIT_INFO = 4,
    VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO = 5,
    VK_STRUCTURE_TYPE_MAPPED_MEMORY_RANGE = 6,
    VK_STRUCTURE_TYPE_BIND_SPARSE_INFO = 7,
    VK_STRUCTURE_TYPE_FENCE_CREATE_INFO = 8,
    VK_STRUCTURE_TYPE_SEMAPHORE_CREATE_INFO = 9,
    VK_STRUCTURE_TYPE_EVENT_CREATE_INFO = 10,
    VK_STRUCTURE_TYPE_QUERY_POOL_CREATE_INFO = 11,
    VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO = 12,
    VK_STRUCTURE_TYPE_BUFFER_VIEW_CREATE_INFO = 13,
    VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO = 14,
    VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO = 15,
    VK_STRUCTURE_TYPE_SHADER_MODULE_CREATE_INFO = 16,
    VK_STRUCTURE_TYPE_PIPELINE_CACHE_CREATE_INFO = 17,
    VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO = 18,
    VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO = 19,
    VK_STRUCTURE_TYPE_PIPELINE_INPUT_ASSEMBLY_STATE_CREATE_INFO = 20,
    VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_STATE_CREATE_INFO = 21,
    VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO = 22,
    VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_CREATE_INFO = 23,
    VK_STRUCTURE_TYPE_PIPELINE_MULTISAMPLE_STATE_CREATE_INFO = 24,
    VK_STRUCTURE_TYPE_PIPELINE_DEPTH_STENCIL_STATE_CREATE_INFO = 25,
    VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_STATE_CREATE_INFO = 26,
    VK_STRUCTURE_TYPE_PIPELINE_DYNAMIC_STATE_CREATE_INFO = 27,
    VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO = 28,
    VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_CREATE_INFO = 29,
    VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO = 30,
    VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO = 31,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO = 32,
    VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO = 33,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO = 34,
    VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET = 35,
    VK_STRUCTURE_TYPE_COPY_DESCRIPTOR_SET = 36,
    VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO = 37,
    VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO = 38,
    VK_STRUCTURE_TYPE_COMMAND_POOL_CREATE_INFO = 39,
    VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO = 40,
    VK_STRUCTURE_TYPE_COMMAND_BUFFER_INHERITANCE_INFO = 41,
    VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO = 42,
    VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO = 43,
    VK_STRUCTURE_TYPE_BUFFER_MEMORY_BARRIER = 44,
    VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER = 45,
    VK_STRUCTURE_TYPE_MEMORY_BARRIER = 46,
    VK_STRUCTURE_TYPE_LOADER_INSTANCE_CREATE_INFO = 47,
    VK_STRUCTURE_TYPE_LOADER_DEVICE_CREATE_INFO = 48,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SUBGROUP_PROPERTIES = 1000094000,
    VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO = 1000157000,
    VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO = 1000157001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES = 1000083000,
    VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS = 1000127000,
    VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO = 1000127001,
    VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO = 1000060000,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO = 1000060003,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO = 1000060004,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO = 1000060005,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO = 1000060006,
    VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO = 1000060013,
    VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO = 1000060014,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES = 1000070000,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO = 1000070001,
    VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2 = 1000146000,
    VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2 = 1000146001,
    VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2 = 1000146002,
    VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2 = 1000146003,
    VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2 = 1000146004,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2 = 1000059000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2 = 1000059001,
    VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2 = 1000059002,
    VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2 = 1000059003,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2 = 1000059004,
    VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2 = 1000059005,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2 = 1000059006,
    VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2 = 1000059007,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SPARSE_IMAGE_FORMAT_INFO_2 = 1000059008,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES = 1000117000,
    VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO = 1000117001,
    VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO = 1000117002,
    VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO = 1000117003,
    VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO = 1000053000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES = 1000053001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES = 1000053002,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES = 1000120000,
    VK_STRUCTURE_TYPE_PROTECTED_SUBMIT_INFO = 1000145000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_FEATURES = 1000145001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_PROPERTIES = 1000145002,
    VK_STRUCTURE_TYPE_DEVICE_QUEUE_INFO_2 = 1000145003,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO = 1000156000,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO = 1000156001,
    VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO = 1000156002,
    VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO = 1000156003,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES = 1000156004,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_IMAGE_FORMAT_PROPERTIES = 1000156005,
    VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO = 1000085000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_IMAGE_FORMAT_INFO = 1000071000,
    VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES = 1000071001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_BUFFER_INFO = 1000071002,
    VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES = 1000071003,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES = 1000071004,
    VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_BUFFER_CREATE_INFO = 1000072000,
    VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO = 1000072001,
    VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO = 1000072002,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_FENCE_INFO = 1000112000,
    VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES = 1000112001,
    VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO = 1000113000,
    VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO = 1000077000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_SEMAPHORE_INFO = 1000076000,
    VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES = 1000076001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES = 1000168000,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT = 1000168001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_DRAW_PARAMETER_FEATURES = 1000063000,
    VK_STRUCTURE_TYPE_SWAPCHAIN_CREATE_INFO_KHR = 1000001000,
    VK_STRUCTURE_TYPE_PRESENT_INFO_KHR = 1000001001,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_CAPABILITIES_KHR = 1000060007,
    VK_STRUCTURE_TYPE_IMAGE_SWAPCHAIN_CREATE_INFO_KHR = 1000060008,
    VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_SWAPCHAIN_INFO_KHR = 1000060009,
    VK_STRUCTURE_TYPE_ACQUIRE_NEXT_IMAGE_INFO_KHR = 1000060010,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_INFO_KHR = 1000060011,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_SWAPCHAIN_CREATE_INFO_KHR = 1000060012,
    VK_STRUCTURE_TYPE_DISPLAY_MODE_CREATE_INFO_KHR = 1000002000,
    VK_STRUCTURE_TYPE_DISPLAY_SURFACE_CREATE_INFO_KHR = 1000002001,
    VK_STRUCTURE_TYPE_DISPLAY_PRESENT_INFO_KHR = 1000003000,
    VK_STRUCTURE_TYPE_XLIB_SURFACE_CREATE_INFO_KHR = 1000004000,
    VK_STRUCTURE_TYPE_XCB_SURFACE_CREATE_INFO_KHR = 1000005000,
    VK_STRUCTURE_TYPE_WAYLAND_SURFACE_CREATE_INFO_KHR = 1000006000,
    VK_STRUCTURE_TYPE_MIR_SURFACE_CREATE_INFO_KHR = 1000007000,
    VK_STRUCTURE_TYPE_ANDROID_SURFACE_CREATE_INFO_KHR = 1000008000,
    VK_STRUCTURE_TYPE_WIN32_SURFACE_CREATE_INFO_KHR = 1000009000,
    VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT = 1000011000,
    VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_RASTERIZATION_ORDER_AMD = 1000018000,
    VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_NAME_INFO_EXT = 1000022000,
    VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_TAG_INFO_EXT = 1000022001,
    VK_STRUCTURE_TYPE_DEBUG_MARKER_MARKER_INFO_EXT = 1000022002,
    VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_IMAGE_CREATE_INFO_NV = 1000026000,
    VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_BUFFER_CREATE_INFO_NV = 1000026001,
    VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_MEMORY_ALLOCATE_INFO_NV = 1000026002,
    VK_STRUCTURE_TYPE_TEXTURE_LOD_GATHER_FORMAT_PROPERTIES_AMD = 1000041000,
    VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_NV = 1000056000,
    VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_NV = 1000056001,
    VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_NV = 1000057000,
    VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_NV = 1000057001,
    VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_NV = 1000058000,
    VK_STRUCTURE_TYPE_VALIDATION_FLAGS_EXT = 1000061000,
    VK_STRUCTURE_TYPE_VI_SURFACE_CREATE_INFO_NN = 1000062000,
    VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_KHR = 1000073000,
    VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_KHR = 1000073001,
    VK_STRUCTURE_TYPE_MEMORY_WIN32_HANDLE_PROPERTIES_KHR = 1000073002,
    VK_STRUCTURE_TYPE_MEMORY_GET_WIN32_HANDLE_INFO_KHR = 1000073003,
    VK_STRUCTURE_TYPE_IMPORT_MEMORY_FD_INFO_KHR = 1000074000,
    VK_STRUCTURE_TYPE_MEMORY_FD_PROPERTIES_KHR = 1000074001,
    VK_STRUCTURE_TYPE_MEMORY_GET_FD_INFO_KHR = 1000074002,
    VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_KHR = 1000075000,
    VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR = 1000078000,
    VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR = 1000078001,
    VK_STRUCTURE_TYPE_D3D12_FENCE_SUBMIT_INFO_KHR = 1000078002,
    VK_STRUCTURE_TYPE_SEMAPHORE_GET_WIN32_HANDLE_INFO_KHR = 1000078003,
    VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_FD_INFO_KHR = 1000079000,
    VK_STRUCTURE_TYPE_SEMAPHORE_GET_FD_INFO_KHR = 1000079001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PUSH_DESCRIPTOR_PROPERTIES_KHR = 1000080000,
    VK_STRUCTURE_TYPE_PRESENT_REGIONS_KHR = 1000084000,
    VK_STRUCTURE_TYPE_OBJECT_TABLE_CREATE_INFO_NVX = 1000086000,
    VK_STRUCTURE_TYPE_INDIRECT_COMMANDS_LAYOUT_CREATE_INFO_NVX = 1000086001,
    VK_STRUCTURE_TYPE_CMD_PROCESS_COMMANDS_INFO_NVX = 1000086002,
    VK_STRUCTURE_TYPE_CMD_RESERVE_SPACE_FOR_COMMANDS_INFO_NVX = 1000086003,
    VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_LIMITS_NVX = 1000086004,
    VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_FEATURES_NVX = 1000086005,
    VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_W_SCALING_STATE_CREATE_INFO_NV = 1000087000,
    VK_STRUCTURE_TYPE_SURFACE_CAPABILITIES_2_EXT = 1000090000,
    VK_STRUCTURE_TYPE_DISPLAY_POWER_INFO_EXT = 1000091000,
    VK_STRUCTURE_TYPE_DEVICE_EVENT_INFO_EXT = 1000091001,
    VK_STRUCTURE_TYPE_DISPLAY_EVENT_INFO_EXT = 1000091002,
    VK_STRUCTURE_TYPE_SWAPCHAIN_COUNTER_CREATE_INFO_EXT = 1000091003,
    VK_STRUCTURE_TYPE_PRESENT_TIMES_INFO_GOOGLE = 1000092000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PER_VIEW_ATTRIBUTES_PROPERTIES_NVX = 1000097000,
    VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_SWIZZLE_STATE_CREATE_INFO_NV = 1000098000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DISCARD_RECTANGLE_PROPERTIES_EXT = 1000099000,
    VK_STRUCTURE_TYPE_PIPELINE_DISCARD_RECTANGLE_STATE_CREATE_INFO_EXT = 1000099001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_CONSERVATIVE_RASTERIZATION_PROPERTIES_EXT = 1000101000,
    VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_CONSERVATIVE_STATE_CREATE_INFO_EXT = 1000101001,
    VK_STRUCTURE_TYPE_HDR_METADATA_EXT = 1000105000,
    VK_STRUCTURE_TYPE_SHARED_PRESENT_SURFACE_CAPABILITIES_KHR = 1000111000,
    VK_STRUCTURE_TYPE_IMPORT_FENCE_WIN32_HANDLE_INFO_KHR = 1000114000,
    VK_STRUCTURE_TYPE_EXPORT_FENCE_WIN32_HANDLE_INFO_KHR = 1000114001,
    VK_STRUCTURE_TYPE_FENCE_GET_WIN32_HANDLE_INFO_KHR = 1000114002,
    VK_STRUCTURE_TYPE_IMPORT_FENCE_FD_INFO_KHR = 1000115000,
    VK_STRUCTURE_TYPE_FENCE_GET_FD_INFO_KHR = 1000115001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SURFACE_INFO_2_KHR = 1000119000,
    VK_STRUCTURE_TYPE_SURFACE_CAPABILITIES_2_KHR = 1000119001,
    VK_STRUCTURE_TYPE_SURFACE_FORMAT_2_KHR = 1000119002,
    VK_STRUCTURE_TYPE_IOS_SURFACE_CREATE_INFO_MVK = 1000122000,
    VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK = 1000123000,
    VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_NAME_INFO_EXT = 1000128000,
    VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_TAG_INFO_EXT = 1000128001,
    VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT = 1000128002,
    VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CALLBACK_DATA_EXT = 1000128003,
    VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT = 1000128004,
    VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_USAGE_ANDROID = 1000129000,
    VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_PROPERTIES_ANDROID = 1000129001,
    VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_FORMAT_PROPERTIES_ANDROID = 1000129002,
    VK_STRUCTURE_TYPE_IMPORT_ANDROID_HARDWARE_BUFFER_INFO_ANDROID = 1000129003,
    VK_STRUCTURE_TYPE_MEMORY_GET_ANDROID_HARDWARE_BUFFER_INFO_ANDROID = 1000129004,
    VK_STRUCTURE_TYPE_EXTERNAL_FORMAT_ANDROID = 1000129005,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_FILTER_MINMAX_PROPERTIES_EXT = 1000130000,
    VK_STRUCTURE_TYPE_SAMPLER_REDUCTION_MODE_CREATE_INFO_EXT = 1000130001,
    VK_STRUCTURE_TYPE_SAMPLE_LOCATIONS_INFO_EXT = 1000143000,
    VK_STRUCTURE_TYPE_RENDER_PASS_SAMPLE_LOCATIONS_BEGIN_INFO_EXT = 1000143001,
    VK_STRUCTURE_TYPE_PIPELINE_SAMPLE_LOCATIONS_STATE_CREATE_INFO_EXT = 1000143002,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLE_LOCATIONS_PROPERTIES_EXT = 1000143003,
    VK_STRUCTURE_TYPE_MULTISAMPLE_PROPERTIES_EXT = 1000143004,
    VK_STRUCTURE_TYPE_IMAGE_FORMAT_LIST_CREATE_INFO_KHR = 1000147000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_FEATURES_EXT = 1000148000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_PROPERTIES_EXT = 1000148001,
    VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_ADVANCED_STATE_CREATE_INFO_EXT = 1000148002,
    VK_STRUCTURE_TYPE_PIPELINE_COVERAGE_TO_COLOR_STATE_CREATE_INFO_NV = 1000149000,
    VK_STRUCTURE_TYPE_PIPELINE_COVERAGE_MODULATION_STATE_CREATE_INFO_NV = 1000152000,
    VK_STRUCTURE_TYPE_VALIDATION_CACHE_CREATE_INFO_EXT = 1000160000,
    VK_STRUCTURE_TYPE_SHADER_MODULE_VALIDATION_CACHE_CREATE_INFO_EXT = 1000160001,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_BINDING_FLAGS_CREATE_INFO_EXT = 1000161000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_FEATURES_EXT = 1000161001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_PROPERTIES_EXT = 1000161002,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_ALLOCATE_INFO_EXT = 1000161003,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_LAYOUT_SUPPORT_EXT = 1000161004,
    VK_STRUCTURE_TYPE_DEVICE_QUEUE_GLOBAL_PRIORITY_CREATE_INFO_EXT = 1000174000,
    VK_STRUCTURE_TYPE_IMPORT_MEMORY_HOST_POINTER_INFO_EXT = 1000178000,
    VK_STRUCTURE_TYPE_MEMORY_HOST_POINTER_PROPERTIES_EXT = 1000178001,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_MEMORY_HOST_PROPERTIES_EXT = 1000178002,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_CORE_PROPERTIES_AMD = 1000185000,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VERTEX_ATTRIBUTE_DIVISOR_PROPERTIES_EXT = 1000190000,
    VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_DIVISOR_STATE_CREATE_INFO_EXT = 1000190001,
    VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2,
    VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2,
    VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2,
    VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2,
    VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2_KHR = VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SPARSE_IMAGE_FORMAT_INFO_2_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SPARSE_IMAGE_FORMAT_INFO_2,
    VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO_KHR = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO_KHR = VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO_KHR = VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO_KHR = VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO_KHR = VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO,
    VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO_KHR = VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO,
    VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO_KHR = VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES,
    VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_IMAGE_FORMAT_INFO_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_IMAGE_FORMAT_INFO,
    VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES_KHR = VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_BUFFER_INFO_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_BUFFER_INFO,
    VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES_KHR = VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES,
    VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_BUFFER_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_BUFFER_CREATE_INFO,
    VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO,
    VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_KHR = VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_SEMAPHORE_INFO_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_SEMAPHORE_INFO,
    VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES_KHR = VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES,
    VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES,
    VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_FENCE_INFO_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_FENCE_INFO,
    VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES_KHR = VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES,
    VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES,
    VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO,
    VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES,
    VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS_KHR = VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS,
    VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO_KHR = VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO,
    VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2_KHR = VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2,
    VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2_KHR = VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2,
    VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2_KHR = VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2,
    VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2_KHR = VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2,
    VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2_KHR = VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO_KHR = VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO_KHR = VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO,
    VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO_KHR = VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO,
    VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO_KHR = VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES,
    VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_IMAGE_FORMAT_PROPERTIES_KHR = VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_IMAGE_FORMAT_PROPERTIES,
    VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO_KHR = VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO,
    VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO_KHR = VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO,
    VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES_KHR = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES,
    VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT_KHR = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT,
} VkStructureType;

Each value corresponds to a particular structure with a sType member with a matching name. As a general rule, the name of each VkStructureType value is obtained by taking the name of the structure, stripping the leading Vk, prefixing each capital letter with _, converting the entire resulting string to upper case, and prefixing it with VK_STRUCTURE_TYPE_. For example, structures of type VkImageCreateInfo correspond to a VkStructureType of VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO, and thus its sType member must equal that when it is passed to the API.

The values VK_STRUCTURE_TYPE_LOADER_INSTANCE_CREATE_INFO and VK_STRUCTURE_TYPE_LOADER_DEVICE_CREATE_INFO are reserved for internal use by the loader, and do not have corresponding Vulkan structures in this Specification.

Valid Usage for Structure Pointer Chains

Any parameter that is a structure containing a void* pNext member must have a value of pNext that is either NULL, or points to a valid structure defined by an extension, containing sType and pNext members as described in the Vulkan Documentation and Extensions document in the section “Extension Interactions”. The set of structures connected by pNext pointers is referred to as a pNext chain. If that extension is supported by the implementation, then it must be enabled.

Each type of valid structure must not appear more than once in a pNext chain.

Any component of the implementation (the loader, any enabled layers, and drivers) must skip over, without processing (other than reading the sType and pNext members) any structures in the chain with sType values not defined by extensions supported by that component.

Extension structures are not described in the base Vulkan Specification, but either in layered Specifications incorporating those extensions, or in separate vendor-provided documents.

As a convenience to implementations and layers needing to iterate through a structure pointer chain, the Vulkan API provides the following base structures:

typedef struct VkBaseInStructure {
    VkStructureType                    sType;
    const struct VkBaseInStructure*    pNext;
} VkBaseInStructure;
typedef struct VkBaseOutStructure {
    VkStructureType               sType;
    struct VkBaseOutStructure*    pNext;
} VkBaseOutStructure;

VkBaseInStructure can be used to facilitate iterating through a read-only structure pointer chain. VkBaseOutStructure can be used to facilitate iterating through a structure pointer chain that returns data back to the application. These structures allow for some type safety and can be used by Vulkan API functions that operate on generic inputs and outputs.

Valid Usage for Nested Structures

The above conditions also apply recursively to members of structures provided as input to a command, either as a direct argument to the command, or themselves a member of another structure.

Specifics on valid usage of each command are covered in their individual sections.

Valid Usage for Extensions

Instance-level functionality or behavior added by an instance extension to the API must not be used unless that extension is supported by the instance as determined by vkEnumerateInstanceExtensionProperties, and that extension is enabled in VkInstanceCreateInfo.

Physical-device-level functionality or behavior added by an instance extension to the API must not be used unless that extension is supported by the instance as determined by vkEnumerateInstanceExtensionProperties, and that extension is enabled in VkInstanceCreateInfo.

Physical-device-level functionality or behavior added by a device extension to the API must not be used unless the conditions described in Extending Physical Device Core Functionality are met.

Device functionality or behavior added by a device extension to the API must not be used unless that extension is supported by the device as determined by vkEnumerateDeviceExtensionProperties, and that extension is enabled in VkDeviceCreateInfo.

Valid Usage for Newer Core Versions

Instance-level functionality or behavior added by a new core version of the API must not be used unless it is supported by the instance as determined by vkEnumerateInstanceVersion.

Physical-device-level functionality or behavior added by a new core version of the API must not be used unless it is supported by the physical device as determined by vkGetPhysicalDeviceProperties.

Device-level functionality or behavior added by a new core version of the API must not be used unless it is supported by the device as determined by vkGetPhysicalDeviceProperties.

2.7.3. Return Codes

While the core Vulkan API is not designed to capture incorrect usage, some circumstances still require return codes. Commands in Vulkan return their status via return codes that are in one of two categories:

  • Successful completion codes are returned when a command needs to communicate success or status information. All successful completion codes are non-negative values.

  • Run time error codes are returned when a command needs to communicate a failure that could only be detected at run time. All run time error codes are negative values.

All return codes in Vulkan are reported via VkResult return values. The possible codes are:

typedef enum VkResult {
    VK_SUCCESS = 0,
    VK_NOT_READY = 1,
    VK_TIMEOUT = 2,
    VK_EVENT_SET = 3,
    VK_EVENT_RESET = 4,
    VK_INCOMPLETE = 5,
    VK_ERROR_OUT_OF_HOST_MEMORY = -1,
    VK_ERROR_OUT_OF_DEVICE_MEMORY = -2,
    VK_ERROR_INITIALIZATION_FAILED = -3,
    VK_ERROR_DEVICE_LOST = -4,
    VK_ERROR_MEMORY_MAP_FAILED = -5,
    VK_ERROR_LAYER_NOT_PRESENT = -6,
    VK_ERROR_EXTENSION_NOT_PRESENT = -7,
    VK_ERROR_FEATURE_NOT_PRESENT = -8,
    VK_ERROR_INCOMPATIBLE_DRIVER = -9,
    VK_ERROR_TOO_MANY_OBJECTS = -10,
    VK_ERROR_FORMAT_NOT_SUPPORTED = -11,
    VK_ERROR_FRAGMENTED_POOL = -12,
    VK_ERROR_OUT_OF_POOL_MEMORY = -1000069000,
    VK_ERROR_INVALID_EXTERNAL_HANDLE = -1000072003,
    VK_ERROR_SURFACE_LOST_KHR = -1000000000,
    VK_ERROR_NATIVE_WINDOW_IN_USE_KHR = -1000000001,
    VK_SUBOPTIMAL_KHR = 1000001003,
    VK_ERROR_OUT_OF_DATE_KHR = -1000001004,
    VK_ERROR_INCOMPATIBLE_DISPLAY_KHR = -1000003001,
    VK_ERROR_VALIDATION_FAILED_EXT = -1000011001,
    VK_ERROR_INVALID_SHADER_NV = -1000012000,
    VK_ERROR_FRAGMENTATION_EXT = -1000161000,
    VK_ERROR_NOT_PERMITTED_EXT = -1000174001,
    VK_ERROR_OUT_OF_POOL_MEMORY_KHR = VK_ERROR_OUT_OF_POOL_MEMORY,
    VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR = VK_ERROR_INVALID_EXTERNAL_HANDLE,
} VkResult;
Success Codes
  • VK_SUCCESS Command successfully completed

  • VK_NOT_READY A fence or query has not yet completed

  • VK_TIMEOUT A wait operation has not completed in the specified time

  • VK_EVENT_SET An event is signaled

  • VK_EVENT_RESET An event is unsignaled

  • VK_INCOMPLETE A return array was too small for the result

  • VK_SUBOPTIMAL_KHR A swapchain no longer matches the surface properties exactly, but can still be used to present to the surface successfully.

Error codes
  • VK_ERROR_OUT_OF_HOST_MEMORY A host memory allocation has failed.

  • VK_ERROR_OUT_OF_DEVICE_MEMORY A device memory allocation has failed.

  • VK_ERROR_INITIALIZATION_FAILED Initialization of an object could not be completed for implementation-specific reasons.

  • VK_ERROR_DEVICE_LOST The logical or physical device has been lost. See Lost Device

  • VK_ERROR_MEMORY_MAP_FAILED Mapping of a memory object has failed.

  • VK_ERROR_LAYER_NOT_PRESENT A requested layer is not present or could not be loaded.

  • VK_ERROR_EXTENSION_NOT_PRESENT A requested extension is not supported.

  • VK_ERROR_FEATURE_NOT_PRESENT A requested feature is not supported.

  • VK_ERROR_INCOMPATIBLE_DRIVER The requested version of Vulkan is not supported by the driver or is otherwise incompatible for implementation-specific reasons.

  • VK_ERROR_TOO_MANY_OBJECTS Too many objects of the type have already been created.

  • VK_ERROR_FORMAT_NOT_SUPPORTED A requested format is not supported on this device.

  • VK_ERROR_FRAGMENTED_POOL A pool allocation has failed due to fragmentation of the pool’s memory. This must only be returned if no attempt to allocate host or device memory was made to accomodate the new allocation. This should be returned in preference to VK_ERROR_OUT_OF_POOL_MEMORY, but only if the implementation is certain that the pool allocation failure was due to fragmentation.

  • VK_ERROR_SURFACE_LOST_KHR A surface is no longer available.

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR The requested window is already in use by Vulkan or another API in a manner which prevents it from being used again.

  • VK_ERROR_OUT_OF_DATE_KHR A surface has changed in such a way that it is no longer compatible with the swapchain, and further presentation requests using the swapchain will fail. Applications must query the new surface properties and recreate their swapchain if they wish to continue presenting to the surface.

  • VK_ERROR_INCOMPATIBLE_DISPLAY_KHR The display used by a swapchain does not use the same presentable image layout, or is incompatible in a way that prevents sharing an image.

  • VK_ERROR_INVALID_SHADER_NV One or more shaders failed to compile or link. More details are reported back to the application via VK_EXT_debug_report if enabled.

  • VK_ERROR_OUT_OF_POOL_MEMORY A pool memory allocation has failed. This must only be returned if no attempt to allocate host or device memory was made to accomodate the new allocation. If the failure was definitely due to fragmentation of the pool, VK_ERROR_FRAGMENTED_POOL should be returned instead.

  • VK_ERROR_INVALID_EXTERNAL_HANDLE An external handle is not a valid handle of the specified type.

  • VK_ERROR_FRAGMENTATION_EXT A descriptor pool creation has failed due to fragmentation.

If a command returns a run time error, unless otherwise specified any output parameters will have undefined contents, except that if the output parameter is a structure with sType and pNext fields, those fields will be unmodified. Any structures chained from pNext will also have undefined contents, except that sType and pNext will be unmodified.

Out of memory errors do not damage any currently existing Vulkan objects. Objects that have already been successfully created can still be used by the application.

Performance-critical commands generally do not have return codes. If a run time error occurs in such commands, the implementation will defer reporting the error until a specified point. For commands that record into command buffers (vkCmd*) run time errors are reported by vkEndCommandBuffer.

2.8. Numeric Representation and Computation

Implementations normally perform computations in floating-point, and must meet the range and precision requirements defined under “Floating-Point Computation” below.

These requirements only apply to computations performed in Vulkan operations outside of shader execution, such as texture image specification and sampling, and per-fragment operations. Range and precision requirements during shader execution differ and are specified by the Precision and Operation of SPIR-V Instructions section.

In some cases, the representation and/or precision of operations is implicitly limited by the specified format of vertex or texel data consumed by Vulkan. Specific floating-point formats are described later in this section.

2.8.1. Floating-Point Computation

Most floating-point computation is performed in SPIR-V shader modules. The properties of computation within shaders are constrained as defined by the Precision and Operation of SPIR-V Instructions section.

Some floating-point computation is performed outside of shaders, such as viewport and depth range calculations. For these computations, we do not specify how floating-point numbers are to be represented, or the details of how operations on them are performed, but only place minimal requirements on representation and precision as described in the remainder of this section.

editing-note

(Jon, Bug 14966) This is a rat’s nest of complexity, both in terms of describing/enumerating places such computation may take place (other than “not shader code”) and in how implementations may do it. We have consciously deferred the resolution of this issue to post-1.0, and in the meantime, the following language inherited from the OpenGL Specification is inserted as a placeholder. Hopefully it can be tightened up considerably.

We require simply that numbers’ floating-point parts contain enough bits and that their exponent fields are large enough so that individual results of floating-point operations are accurate to about 1 part in 105. The maximum representable magnitude for all floating-point values must be at least 232.

x × 0 = 0 × x = 0 for any non-infinite and non-NaN x.

1 × x = x × 1 = x.

x + 0 = 0 + x = x.

00 = 1.

Occasionally, further requirements will be specified. Most single-precision floating-point formats meet these requirements.

The special values Inf and -Inf encode values with magnitudes too large to be represented; the special value NaN encodes “Not A Number” values resulting from undefined arithmetic operations such as 0 / 0. Implementations may support Inf and NaN in their floating-point computations.

Any representable floating-point value is legal as input to a Vulkan command that requires floating-point data. The result of providing a value that is not a floating-point number to such a command is unspecified, but must not lead to Vulkan interruption or termination. In IEEE 754 arithmetic, for example, providing a negative zero or a denormalized number to an Vulkan command must yield deterministic results, while providing a NaN or Inf yields unspecified results.

2.8.2. 16-Bit Floating-Point Numbers

16-bit floating point numbers are defined in the “16-bit floating point numbers” section of the Khronos Data Format Specification.

Any representable 16-bit floating-point value is legal as input to a Vulkan command that accepts 16-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN) to such a command is unspecified, but must not lead to Vulkan interruption or termination. Providing a denormalized number or negative zero to Vulkan must yield deterministic results.

2.8.3. Unsigned 11-Bit Floating-Point Numbers

Unsigned 11-bit floating point numbers are defined in the “Unsigned 11-bit floating point numbers” section of the Khronos Data Format Specification.

When a floating-point value is converted to an unsigned 11-bit floating-point representation, finite values are rounded to the closest representable finite value.

While less accurate, implementations are allowed to always round in the direction of zero. This means negative values are converted to zero. Likewise, finite positive values greater than 65024 (the maximum finite representable unsigned 11-bit floating-point value) are converted to 65024. Additionally: negative infinity is converted to zero; positive infinity is converted to positive infinity; and both positive and negative NaN are converted to positive NaN.

Any representable unsigned 11-bit floating-point value is legal as input to a Vulkan command that accepts 11-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN) to such a command is unspecified, but must not lead to Vulkan interruption or termination. Providing a denormalized number to Vulkan must yield deterministic results.

2.8.4. Unsigned 10-Bit Floating-Point Numbers

Unsigned 10-bit floating point numbers are defined in the “Unsigned 10-bit floating point numbers” section of the Khronos Data Format Specification.

When a floating-point value is converted to an unsigned 10-bit floating-point representation, finite values are rounded to the closest representable finite value.

While less accurate, implementations are allowed to always round in the direction of zero. This means negative values are converted to zero. Likewise, finite positive values greater than 64512 (the maximum finite representable unsigned 10-bit floating-point value) are converted to 64512. Additionally: negative infinity is converted to zero; positive infinity is converted to positive infinity; and both positive and negative NaN are converted to positive NaN.

Any representable unsigned 10-bit floating-point value is legal as input to a Vulkan command that accepts 10-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN) to such a command is unspecified, but must not lead to Vulkan interruption or termination. Providing a denormalized number to Vulkan must yield deterministic results.

2.8.5. General Requirements

Some calculations require division. In such cases (including implied divisions performed by vector normalization), division by zero produces an unspecified result but must not lead to Vulkan interruption or termination.

2.9. Fixed-Point Data Conversions

When generic vertex attributes and pixel color or depth components are represented as integers, they are often (but not always) considered to be normalized. Normalized integer values are treated specially when being converted to and from floating-point values, and are usually referred to as normalized fixed-point.

In the remainder of this section, b denotes the bit width of the fixed-point integer representation. When the integer is one of the types defined by the API, b is the bit width of that type. When the integer comes from an image containing color or depth component texels, b is the number of bits allocated to that component in its specified image format.

The signed and unsigned fixed-point representations are assumed to be b-bit binary two’s-complement integers and binary unsigned integers, respectively.

2.9.1. Conversion from Normalized Fixed-Point to Floating-Point

Unsigned normalized fixed-point integers represent numbers in the range [0,1]. The conversion from an unsigned normalized fixed-point value c to the corresponding floating-point value f is defined as

\[f = { c \over { 2^b - 1 } }\]

Signed normalized fixed-point integers represent numbers in the range [-1,1]. The conversion from a signed normalized fixed-point value c to the corresponding floating-point value f is performed using

\[f = \max\left( {c \over {2^{b-1} - 1}}, -1.0 \right)\]

Only the range [-2b-1 + 1, 2b-1 - 1] is used to represent signed fixed-point values in the range [-1,1]. For example, if b = 8, then the integer value -127 corresponds to -1.0 and the value 127 corresponds to 1.0. Note that while zero is exactly expressible in this representation, one value (-128 in the example) is outside the representable range, and must be clamped before use. This equation is used everywhere that signed normalized fixed-point values are converted to floating-point.

2.9.2. Conversion from Floating-Point to Normalized Fixed-Point

The conversion from a floating-point value f to the corresponding unsigned normalized fixed-point value c is defined by first clamping f to the range [0,1], then computing

c = convertFloatToUint(f × (2b - 1), b)

where convertFloatToUint}(r,b) returns one of the two unsigned binary integer values with exactly b bits which are closest to the floating-point value r. Implementations should round to nearest. If r is equal to an integer, then that integer value must be returned. In particular, if f is equal to 0.0 or 1.0, then c must be assigned 0 or 2b - 1, respectively.

The conversion from a floating-point value f to the corresponding signed normalized fixed-point value c is performed by clamping f to the range [-1,1], then computing

c = convertFloatToInt(f × (2b-1 - 1), b)

where convertFloatToInt(r,b) returns one of the two signed two’s-complement binary integer values with exactly b bits which are closest to the floating-point value r. Implementations should round to nearest. If r is equal to an integer, then that integer value must be returned. In particular, if f is equal to -1.0, 0.0, or 1.0, then c must be assigned -(2b-1 - 1), 0, or 2b-1 - 1, respectively.

This equation is used everywhere that floating-point values are converted to signed normalized fixed-point.

2.10. API Version Numbers and Semantics

The Vulkan version number is used in several places in the API. In each such use, the API major version number, minor version number, and patch version number are packed into a 32-bit integer as follows:

  • The major version number is a 10-bit integer packed into bits 31-22.

  • The minor version number is a 10-bit integer packed into bits 21-12.

  • The patch version number is a 12-bit integer packed into bits 11-0.

Differences in any of the Vulkan version numbers indicates a change to the API in some way, with each part of the version number indicating a different scope of changes.

A difference in patch version numbers indicates that some usually small part of the Specification or header has been modified, typically to fix a bug, and may have an impact on the behavior of existing functionality. Differences in this version number should not affect either full compatibility or backwards compatibility between two versions, or add additional interfaces to the API.

A difference in minor version numbers indicates that some amount of new functionality has been added. This will usually include new interfaces in the header, and may also include behavior changes and bug fixes. Functionality may be deprecated in a minor revision, but will not be removed. The patch version will continue to increment through minor version number changes since all minor versions are generated from the same source files, and changes to the source files may affect all minor versions within a major version. Differences in the patch version should not affect backwards compatibility, but will affect full compatibility. The patch version of the Specification is taken from VK_HEADER_VERSION.

A difference in major version numbers indicates a large set of changes to the API, potentially including new functionality and header interfaces, behavioral changes, removal of deprecated features, modification or outright replacement of any feature, and is thus very likely to break any and all compatibility. Differences in this version will typically require significant modification to an application in order for it to function.

C language macros for manipulating version numbers are defined in the Version Number Macros appendix.

2.11. Common Object Types

Some types of Vulkan objects are used in many different structures and command parameters, and are described here. These types include offsets, extents, and rectangles.

2.11.1. Offsets

Offsets are used to describe a pixel location within an image or framebuffer, as an (x,y) location for two-dimensional images, or an (x,y,z) location for three-dimensional images.

A two-dimensional offsets is defined by the structure:

typedef struct VkOffset2D {
    int32_t    x;
    int32_t    y;
} VkOffset2D;
  • x is the x offset.

  • y is the y offset.

A three-dimensional offset is defined by the structure:

typedef struct VkOffset3D {
    int32_t    x;
    int32_t    y;
    int32_t    z;
} VkOffset3D;
  • x is the x offset.

  • y is the y offset.

  • z is the z offset.

2.11.2. Extents

Extents are used to describe the size of a rectangular region of pixels within an image or framebuffer, as (width,height) for two-dimensional images, or as (width,height,depth) for three-dimensional images.

A two-dimensional extent is defined by the structure:

typedef struct VkExtent2D {
    uint32_t    width;
    uint32_t    height;
} VkExtent2D;
  • width is the width of the extent.

  • height is the height of the extent.

A three-dimensional extent is defined by the structure:

typedef struct VkExtent3D {
    uint32_t    width;
    uint32_t    height;
    uint32_t    depth;
} VkExtent3D;
  • width is the width of the extent.

  • height is the height of the extent.

  • depth is the depth of the extent.

2.11.3. Rectangles

Rectangles are used to describe a specified rectangular region of pixels within an image or framebuffer. Rectangles include both an offset and an extent of the same dimensionality, as described above. Two-dimensional rectangles are defined by the structure

typedef struct VkRect2D {
    VkOffset2D    offset;
    VkExtent2D    extent;
} VkRect2D;
  • offset is a VkOffset2D specifying the rectangle offset.

  • extent is a VkExtent2D specifying the rectangle extent.

3. Initialization

Before using Vulkan, an application must initialize it by loading the Vulkan commands, and creating a VkInstance object.

3.1. Command Function Pointers

Vulkan commands are not necessarily exposed statically on a platform. Function pointers for all Vulkan commands can be obtained with the command:

PFN_vkVoidFunction vkGetInstanceProcAddr(
    VkInstance                                  instance,
    const char*                                 pName);
  • instance is the instance that the function pointer will be compatible with, or NULL for commands not dependent on any instance.

  • pName is the name of the command to obtain.

vkGetInstanceProcAddr itself is obtained in a platform- and loader- specific manner. Typically, the loader library will export this command as a function symbol, so applications can link against the loader library, or load it dynamically and look up the symbol using platform-specific APIs.

The table below defines the various use cases for vkGetInstanceProcAddr and expected return value (“fp” is “function pointer”) for each case.

The returned function pointer is of type PFN_vkVoidFunction, and must be cast to the type of the command being queried.

Table 1. vkGetInstanceProcAddr behavior
instance pName return value

*

NULL

undefined

invalid instance

*

undefined

NULL

vkEnumerateInstanceVersion

fp

NULL

vkEnumerateInstanceExtensionProperties

fp

NULL

vkEnumerateInstanceLayerProperties

fp

NULL

vkCreateInstance

fp

NULL

* (any pName not covered above)

NULL

instance

core Vulkan command

fp1

instance

enabled instance extension commands for instance

fp1

instance

available device extension2 commands for instance

fp1

instance

* (any pName not covered above)

NULL

1

The returned function pointer must only be called with a dispatchable object (the first parameter) that is instance or a child of instance, e.g. VkInstance, VkPhysicalDevice, VkDevice, VkQueue, or VkCommandBuffer.

2

An “available device extension” is a device extension supported by any physical device enumerated by instance.

Valid Usage (Implicit)
  • If instance is not NULL, instance must be a valid VkInstance handle

  • pName must be a null-terminated UTF-8 string

In order to support systems with multiple Vulkan implementations, the function pointers returned by vkGetInstanceProcAddr may point to dispatch code that calls a different real implementation for different VkDevice objects or their child objects. The overhead of the internal dispatch for VkDevice objects can be avoided by obtaining device-specific function pointers for any commands that use a device or device-child object as their dispatchable object. Such function pointers can be obtained with the command:

PFN_vkVoidFunction vkGetDeviceProcAddr(
    VkDevice                                    device,
    const char*                                 pName);

The table below defines the various use cases for vkGetDeviceProcAddr and expected return value for each case.

The returned function pointer is of type PFN_vkVoidFunction, and must be cast to the type of the command being queried. The function pointer must only be called with a dispatchable object (the first parameter) that is device or a child of device.

Table 2. vkGetDeviceProcAddr behavior
device pName return value

NULL

*

undefined

invalid device

*

undefined

device

NULL

undefined

device

core device-level Vulkan command

fp

device

enabled device extension commands

fp

device

* (any pName not covered above)

NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pName must be a null-terminated UTF-8 string

The definition of PFN_vkVoidFunction is:

typedef void (VKAPI_PTR *PFN_vkVoidFunction)(void);

3.1.1. Extending Physical Device Core Functionality

New core physical-device-level functionality can be used when the physical-device version is greater than or equal to the version of Vulkan that added the new functionality. The Vulkan version supported by a physical device can be obtained by calling vkGetPhysicalDeviceProperties.

3.1.2. Extending Physical Device From Device Extensions

When the VK_KHR_get_physical_device_properties2 extension is enabled, or when both the instance and the physical-device versions are at least 1.1, physical-device-level functionality of a device extension can be used with a physical device if the corresponding extension is enumerated by vkEnumerateDeviceExtensionProperties for that physical device, even before a logical device has been created.

To obtain a function pointer for a physical-device-level command from a device extension, an application can use vkGetInstanceProcAddr. This function pointer may point to dispatch code, which calls a different real implementation for different VkPhysicalDevice objects. Behavior is undefined if an extension physical-device command is called on a physical device that does not support the extension.

Device extensions may define structures that can be added to the pNext chain of physical-device-level commands. Behavior is undefined if such an extension structure is passed to a physical-device-level command for a physical device that does not support the extension.

3.2. Instances

There is no global state in Vulkan and all per-application state is stored in a VkInstance object. Creating a VkInstance object initializes the Vulkan library and allows the application to pass information about itself to the implementation.

Instances are represented by VkInstance handles:

VK_DEFINE_HANDLE(VkInstance)

The version of Vulkan that is supported by an instance may be different than the version of Vulkan supported by a device or physical device. To query properties that can be used in creating an instance, call:

VkResult vkEnumerateInstanceVersion(
    uint32_t*                                   pApiVersion);
  • pApiVersion points to a uint32_t, which is the version of Vulkan supported by instance-level functionality, encoded as described in the API Version Numbers and Semantics section.

Valid Usage (Implicit)
  • pApiVersion must be a valid pointer to a uint32_t value

Return Codes
Success
  • VK_SUCCESS

To create an instance object, call:

VkResult vkCreateInstance(
    const VkInstanceCreateInfo*                 pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkInstance*                                 pInstance);
  • pCreateInfo points to an instance of VkInstanceCreateInfo controlling creation of the instance.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pInstance points a VkInstance handle in which the resulting instance is returned.

vkCreateInstance verifies that the requested layers exist. If not, vkCreateInstance will return VK_ERROR_LAYER_NOT_PRESENT. Next vkCreateInstance verifies that the requested extensions are supported (e.g. in the implementation or in any enabled instance layer) and if any requested extension is not supported, vkCreateInstance must return VK_ERROR_EXTENSION_NOT_PRESENT. After verifying and enabling the instance layers and extensions the VkInstance object is created and returned to the application. If a requested extension is only supported by a layer, both the layer and the extension need to be specified at vkCreateInstance time for the creation to succeed.

Valid Usage
Valid Usage (Implicit)
  • pCreateInfo must be a valid pointer to a valid VkInstanceCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pInstance must be a valid pointer to a VkInstance handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INITIALIZATION_FAILED

  • VK_ERROR_LAYER_NOT_PRESENT

  • VK_ERROR_EXTENSION_NOT_PRESENT

  • VK_ERROR_INCOMPATIBLE_DRIVER

The VkInstanceCreateInfo structure is defined as:

typedef struct VkInstanceCreateInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkInstanceCreateFlags       flags;
    const VkApplicationInfo*    pApplicationInfo;
    uint32_t                    enabledLayerCount;
    const char* const*          ppEnabledLayerNames;
    uint32_t                    enabledExtensionCount;
    const char* const*          ppEnabledExtensionNames;
} VkInstanceCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • pApplicationInfo is NULL or a pointer to an instance of VkApplicationInfo. If not NULL, this information helps implementations recognize behavior inherent to classes of applications. VkApplicationInfo is defined in detail below.

  • enabledLayerCount is the number of global layers to enable.

  • ppEnabledLayerNames is a pointer to an array of enabledLayerCount null-terminated UTF-8 strings containing the names of layers to enable for the created instance. See the Layers section for further details.

  • enabledExtensionCount is the number of global extensions to enable.

  • ppEnabledExtensionNames is a pointer to an array of enabledExtensionCount null-terminated UTF-8 strings containing the names of extensions to enable.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_INSTANCE_CREATE_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkDebugReportCallbackCreateInfoEXT, VkDebugUtilsMessengerCreateInfoEXT, or VkValidationFlagsEXT

  • Each sType member in the pNext chain must be unique

  • flags must be 0

  • If pApplicationInfo is not NULL, pApplicationInfo must be a valid pointer to a valid VkApplicationInfo structure

  • If enabledLayerCount is not 0, ppEnabledLayerNames must be a valid pointer to an array of enabledLayerCount null-terminated UTF-8 strings

  • If enabledExtensionCount is not 0, ppEnabledExtensionNames must be a valid pointer to an array of enabledExtensionCount null-terminated UTF-8 strings

typedef VkFlags VkInstanceCreateFlags;

VkInstanceCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

When creating a Vulkan instance for which you wish to disable validation checks, add a VkValidationFlagsEXT structure to the pNext chain of the VkInstanceCreateInfo structure, specifying the checks to be disabled.

typedef struct VkValidationFlagsEXT {
    VkStructureType          sType;
    const void*              pNext;
    uint32_t                 disabledValidationCheckCount;
    VkValidationCheckEXT*    pDisabledValidationChecks;
} VkValidationFlagsEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • disabledValidationCheckCount is the number of checks to disable.

  • pDisabledValidationChecks is a pointer to an array of VkValidationCheckEXT values specifying the validation checks to be disabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_VALIDATION_FLAGS_EXT

  • pDisabledValidationChecks must be a valid pointer to an array of disabledValidationCheckCount VkValidationCheckEXT values

  • disabledValidationCheckCount must be greater than 0

Possible values of elements of the VkValidationFlagsEXT::pDisabledValidationChecks array, specifying validation checks to be disabled, are:

typedef enum VkValidationCheckEXT {
    VK_VALIDATION_CHECK_ALL_EXT = 0,
    VK_VALIDATION_CHECK_SHADERS_EXT = 1,
} VkValidationCheckEXT;
  • VK_VALIDATION_CHECK_ALL_EXT specifies that all validation checks are disabled.

  • VK_VALIDATION_CHECK_SHADERS_EXT specifies that shader validation is disabled.

The VkApplicationInfo structure is defined as:

typedef struct VkApplicationInfo {
    VkStructureType    sType;
    const void*        pNext;
    const char*        pApplicationName;
    uint32_t           applicationVersion;
    const char*        pEngineName;
    uint32_t           engineVersion;
    uint32_t           apiVersion;
} VkApplicationInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pApplicationName is NULL or is a pointer to a null-terminated UTF-8 string containing the name of the application.

  • applicationVersion is an unsigned integer variable containing the developer-supplied version number of the application.

  • pEngineName is NULL or is a pointer to a null-terminated UTF-8 string containing the name of the engine (if any) used to create the application.

  • engineVersion is an unsigned integer variable containing the developer-supplied version number of the engine used to create the application.

  • apiVersion must be the highest version of Vulkan that the application is designed to use, encoded as described in the API Version Numbers and Semantics section. The patch version number specified in apiVersion is ignored when creating an instance object. Only the major and minor versions of the instance must match those requested in apiVersion.

Vulkan 1.0 implementations were required to return VK_ERROR_INCOMPATIBLE_DRIVER if apiVersion was larger than 1.0. Implementations that support Vulkan 1.1 or later must not return VK_ERROR_INCOMPATIBLE_DRIVER for any value of apiVersion.

Note

Because Vulkan 1.0 implementations may fail with VK_ERROR_INCOMPATIBLE_DRIVER, applications should determine the version of Vulkan available before calling vkCreateInstance. If the vkGetInstanceProcAddr returns NULL for vkEnumerateInstanceVersion, it is a Vulkan 1.0 implementation. Otherwise, the application can call vkEnumerateInstanceVersion to determine the version of Vulkan.

Implicit layers must be disabled if they do not support a version at least as high as apiVersion. See the "Vulkan Loader Specification and Architecture Overview" document for additional information.

Note

Providing a NULL VkInstanceCreateInfo::pApplicationInfo or providing an apiVersion of 0 is equivalent to providing an apiVersion of VK_MAKE_VERSION(1,0,0).

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_APPLICATION_INFO

  • pNext must be NULL

  • If pApplicationName is not NULL, pApplicationName must be a null-terminated UTF-8 string

  • If pEngineName is not NULL, pEngineName must be a null-terminated UTF-8 string

To destroy an instance, call:

void vkDestroyInstance(
    VkInstance                                  instance,
    const VkAllocationCallbacks*                pAllocator);
  • instance is the handle of the instance to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All child objects created using instance must have been destroyed prior to destroying instance

  • If VkAllocationCallbacks were provided when instance was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when instance was created, pAllocator must be NULL

Valid Usage (Implicit)
  • If instance is not NULL, instance must be a valid VkInstance handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

Host Synchronization
  • Host access to instance must be externally synchronized

4. Devices and Queues

Once Vulkan is initialized, devices and queues are the primary objects used to interact with a Vulkan implementation.

Vulkan separates the concept of physical and logical devices. A physical device usually represents a single complete implementation of Vulkan (excluding instance-level functionality) available to the host, of which there are a finite number. A logical device represents an instance of that implementation with its own state and resources independent of other logical devices.

Physical devices are represented by VkPhysicalDevice handles:

VK_DEFINE_HANDLE(VkPhysicalDevice)

4.1. Physical Devices

To retrieve a list of physical device objects representing the physical devices installed in the system, call:

VkResult vkEnumeratePhysicalDevices(
    VkInstance                                  instance,
    uint32_t*                                   pPhysicalDeviceCount,
    VkPhysicalDevice*                           pPhysicalDevices);
  • instance is a handle to a Vulkan instance previously created with vkCreateInstance.

  • pPhysicalDeviceCount is a pointer to an integer related to the number of physical devices available or queried, as described below.

  • pPhysicalDevices is either NULL or a pointer to an array of VkPhysicalDevice handles.

If pPhysicalDevices is NULL, then the number of physical devices available is returned in pPhysicalDeviceCount. Otherwise, pPhysicalDeviceCount must point to a variable set by the user to the number of elements in the pPhysicalDevices array, and on return the variable is overwritten with the number of handles actually written to pPhysicalDevices. If pPhysicalDeviceCount is less than the number of physical devices available, at most pPhysicalDeviceCount structures will be written. If pPhysicalDeviceCount is smaller than the number of physical devices available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available physical devices were returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pPhysicalDeviceCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPhysicalDeviceCount is not 0, and pPhysicalDevices is not NULL, pPhysicalDevices must be a valid pointer to an array of pPhysicalDeviceCount VkPhysicalDevice handles

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INITIALIZATION_FAILED

To query general properties of physical devices once enumerated, call:

void vkGetPhysicalDeviceProperties(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties*                 pProperties);
  • physicalDevice is the handle to the physical device whose properties will be queried.

  • pProperties points to an instance of the VkPhysicalDeviceProperties structure, that will be filled with returned information.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pProperties must be a valid pointer to a VkPhysicalDeviceProperties structure

The VkPhysicalDeviceProperties structure is defined as:

typedef struct VkPhysicalDeviceProperties {
    uint32_t                            apiVersion;
    uint32_t                            driverVersion;
    uint32_t                            vendorID;
    uint32_t                            deviceID;
    VkPhysicalDeviceType                deviceType;
    char                                deviceName[VK_MAX_PHYSICAL_DEVICE_NAME_SIZE];
    uint8_t                             pipelineCacheUUID[VK_UUID_SIZE];
    VkPhysicalDeviceLimits              limits;
    VkPhysicalDeviceSparseProperties    sparseProperties;
} VkPhysicalDeviceProperties;
  • apiVersion is the version of Vulkan supported by the device, encoded as described in the API Version Numbers and Semantics section.

  • driverVersion is the vendor-specified version of the driver.

  • vendorID is a unique identifier for the vendor (see below) of the physical device.

  • deviceID is a unique identifier for the physical device among devices available from the vendor.

  • deviceType is a VkPhysicalDeviceType specifying the type of device.

  • deviceName is a null-terminated UTF-8 string containing the name of the device.

  • pipelineCacheUUID is an array of size VK_UUID_SIZE, containing 8-bit values that represent a universally unique identifier for the device.

  • limits is the VkPhysicalDeviceLimits structure which specifies device-specific limits of the physical device. See Limits for details.

  • sparseProperties is the VkPhysicalDeviceSparseProperties structure which specifies various sparse related properties of the physical device. See Sparse Properties for details.

Note

The value of apiVersion may be different than the version returned by vkEnumerateInstanceVersion; either higher or lower. In such cases, the application must not use functionality that exceeds the version of Vulkan associated with a given object. The pApiVersion parameter returned by vkEnumerateInstanceVersion is the version associated with a VkInstance and its children, except for a VkPhysicalDevice and its children. VkPhysicalDeviceProperties::apiVersion is the version associated with a VkPhysicalDevice and its children.

The vendorID and deviceID fields are provided to allow applications to adapt to device characteristics that are not adequately exposed by other Vulkan queries.

Note

These may include performance profiles, hardware errata, or other characteristics.

The vendor identified by vendorID is the entity responsible for the most salient characteristics of the underlying implementation of the VkPhysicalDevice being queried.

Note

For example, in the case of a discrete GPU implementation, this should be the GPU chipset vendor. In the case of a hardware accelerator integrated into a system-on-chip (SoC), this should be the supplier of the silicon IP used to create the accelerator.

If the vendor has a PCI vendor ID, the low 16 bits of vendorID must contain that PCI vendor ID, and the remaining bits must be set to zero. Otherwise, the value returned must be a valid Khronos vendor ID, obtained as described in the Vulkan Documentation and Extensions document in the section “Registering a Vendor ID with Khronos”. Khronos vendor IDs are allocated starting at 0x10000, to distinguish them from the PCI vendor ID namespace.

The vendor is also responsible for the value returned in deviceID. If the implementation is driven primarily by a PCI device with a PCI device ID, the low 16 bits of deviceID must contain that PCI device ID, and the remaining bits must be set to zero. Otherwise, the choice of what values to return may be dictated by operating system or platform policies - but should uniquely identify both the device version and any major configuration options (for example, core count in the case of multicore devices).

Note

The same device ID should be used for all physical implementations of that device version and configuration. For example, all uses of a specific silicon IP GPU version and configuration should use the same device ID, even if those uses occur in different SoCs.

The physical device types which may be returned in VkPhysicalDeviceProperties::deviceType are:

typedef enum VkPhysicalDeviceType {
    VK_PHYSICAL_DEVICE_TYPE_OTHER = 0,
    VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU = 1,
    VK_PHYSICAL_DEVICE_TYPE_DISCRETE_GPU = 2,
    VK_PHYSICAL_DEVICE_TYPE_VIRTUAL_GPU = 3,
    VK_PHYSICAL_DEVICE_TYPE_CPU = 4,
} VkPhysicalDeviceType;
  • VK_PHYSICAL_DEVICE_TYPE_OTHER - the device does not match any other available types.

  • VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU - the device is typically one embedded in or tightly coupled with the host.

  • VK_PHYSICAL_DEVICE_TYPE_DISCRETE_GPU - the device is typically a separate processor connected to the host via an interlink.

  • VK_PHYSICAL_DEVICE_TYPE_VIRTUAL_GPU - the device is typically a virtual node in a virtualization environment.

  • VK_PHYSICAL_DEVICE_TYPE_CPU - the device is typically running on the same processors as the host.

The physical device type is advertised for informational purposes only, and does not directly affect the operation of the system. However, the device type may correlate with other advertised properties or capabilities of the system, such as how many memory heaps there are.

To query general properties of physical devices once enumerated, call:

void vkGetPhysicalDeviceProperties2(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties2*                pProperties);

or the equivalent command

void vkGetPhysicalDeviceProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties2*                pProperties);
  • physicalDevice is the handle to the physical device whose properties will be queried.

  • pProperties points to an instance of the VkPhysicalDeviceProperties2 structure, that will be filled with returned information.

Each structure in pProperties and its pNext chain contain members corresponding to properties or implementation-dependent limits. vkGetPhysicalDeviceProperties2 writes each member to a value indicating the value of that property or limit.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pProperties must be a valid pointer to a VkPhysicalDeviceProperties2 structure

The VkPhysicalDeviceProperties2 structure is defined as:

typedef struct VkPhysicalDeviceProperties2 {
    VkStructureType               sType;
    void*                         pNext;
    VkPhysicalDeviceProperties    properties;
} VkPhysicalDeviceProperties2;

or the equivalent

typedef VkPhysicalDeviceProperties2 VkPhysicalDeviceProperties2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • properties is a structure of type VkPhysicalDeviceProperties describing the properties of the physical device. This structure is written with the same values as if it were written by vkGetPhysicalDeviceProperties.

The pNext chain of this structure is used to extend the structure with properties defined by extensions.

To query the UUID and LUID of a device, add VkPhysicalDeviceIDProperties to the pNext chain of the VkPhysicalDeviceProperties2 structure. The VkPhysicalDeviceIDProperties structure is defined as:

typedef struct VkPhysicalDeviceIDProperties {
    VkStructureType    sType;
    void*              pNext;
    uint8_t            deviceUUID[VK_UUID_SIZE];
    uint8_t            driverUUID[VK_UUID_SIZE];
    uint8_t            deviceLUID[VK_LUID_SIZE];
    uint32_t           deviceNodeMask;
    VkBool32           deviceLUIDValid;
} VkPhysicalDeviceIDProperties;

or the equivalent

typedef VkPhysicalDeviceIDProperties VkPhysicalDeviceIDPropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • deviceUUID is an array of size VK_UUID_SIZE, containing 8-bit values that represent a universally unique identifier for the device.

  • driverUUID is an array of size VK_UUID_SIZE, containing 8-bit values that represent a universally unique identifier for the driver build in use by the device.

  • deviceLUID is an array of size VK_LUID_SIZE, containing 8-bit values that represent a locally unique identifier for the device.

  • deviceNodeMask is a bitfield identifying the node within a linked device adapter corresponding to the device.

  • deviceLUIDValid is a boolean value that will be VK_TRUE if deviceLUID contains a valid LUID and deviceNodeMask contains a valid node mask, and VK_FALSE if they do not.

deviceUUID must be immutable for a given device across instances, processes, driver APIs, driver versions, and system reboots.

Applications can compare the driverUUID value across instance and process boundaries, and can make similar queries in external APIs to determine whether they are capable of sharing memory objects and resources using them with the device.

deviceUUID and/or driverUUID must be used to determine whether a particular external object can be shared between driver components, where such a restriction exists as defined in the compatibility table for the particular object type:

If deviceLUIDValid is VK_FALSE, the contents of deviceLUID and deviceNodeMask are undefined. If deviceLUIDValid is VK_TRUE and Vulkan is running on the Windows operating system, the contents of deviceLUID can be cast to an LUID object and must be equal to the locally unique identifier of a IDXGIAdapter1 object that corresponds to physicalDevice. If deviceLUIDValid is VK_TRUE, deviceNodeMask must contain exactly one bit. If Vulkan is running on an operating system that supports the Direct3D 12 API and physicalDevice corresponds to an individual device in a linked device adapter, deviceNodeMask identifies the Direct3D 12 node corresponding to physicalDevice. Otherwise, deviceNodeMask must be 1.

Note

Although they have identical descriptions, VkPhysicalDeviceIDProperties::deviceUUID may differ from VkPhysicalDeviceProperties2::pipelineCacheUUID. The former is intended to identify and correlate devices across API and driver boundaries, while the latter is used to identify a compatible device and driver combination to use when serializing and de-serializing pipeline state.

Note

While VkPhysicalDeviceIDProperties::deviceUUID is specified to remain consistent across driver versions and system reboots, it is not intended to be usable as a serializable persistent identifier for a device. It may change when a device is physically added to, removed from, or moved to a different connector in a system while that system is powered down. Further, there is no reasonable way to verify with conformance testing that a given device retains the same UUID in a given system across all driver versions supported in that system. While implementations should make every effort to report consistent device UUIDs across driver versions, applications should avoid relying on the persistence of this value for uses other than identifying compatible devices for external object sharing purposes.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES

To query properties of queues available on a physical device, call:

void vkGetPhysicalDeviceQueueFamilyProperties(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pQueueFamilyPropertyCount,
    VkQueueFamilyProperties*                    pQueueFamilyProperties);
  • physicalDevice is the handle to the physical device whose properties will be queried.

  • pQueueFamilyPropertyCount is a pointer to an integer related to the number of queue families available or queried, as described below.

  • pQueueFamilyProperties is either NULL or a pointer to an array of VkQueueFamilyProperties structures.

If pQueueFamilyProperties is NULL, then the number of queue families available is returned in pQueueFamilyPropertyCount. Otherwise, pQueueFamilyPropertyCount must point to a variable set by the user to the number of elements in the pQueueFamilyProperties array, and on return the variable is overwritten with the number of structures actually written to pQueueFamilyProperties. If pQueueFamilyPropertyCount is less than the number of queue families available, at most pQueueFamilyPropertyCount structures will be written.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pQueueFamilyPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pQueueFamilyPropertyCount is not 0, and pQueueFamilyProperties is not NULL, pQueueFamilyProperties must be a valid pointer to an array of pQueueFamilyPropertyCount VkQueueFamilyProperties structures

The VkQueueFamilyProperties structure is defined as:

typedef struct VkQueueFamilyProperties {
    VkQueueFlags    queueFlags;
    uint32_t        queueCount;
    uint32_t        timestampValidBits;
    VkExtent3D      minImageTransferGranularity;
} VkQueueFamilyProperties;
  • queueFlags is a bitmask of VkQueueFlagBits indicating capabilities of the queues in this queue family.

  • queueCount is the unsigned integer count of queues in this queue family.

  • timestampValidBits is the unsigned integer count of meaningful bits in the timestamps written via vkCmdWriteTimestamp. The valid range for the count is 36..64 bits, or a value of 0, indicating no support for timestamps. Bits outside the valid range are guaranteed to be zeros.

  • minImageTransferGranularity is the minimum granularity supported for image transfer operations on the queues in this queue family.

The value returned in minImageTransferGranularity has a unit of compressed texel blocks for images having a block-compressed format, and a unit of texels otherwise.

Possible values of minImageTransferGranularity are:

  • (0,0,0) which indicates that only whole mip levels must be transferred using the image transfer operations on the corresponding queues. In this case, the following restrictions apply to all offset and extent parameters of image transfer operations:

    • The x, y, and z members of a VkOffset3D parameter must always be zero.

    • The width, height, and depth members of a VkExtent3D parameter must always match the width, height, and depth of the image subresource corresponding to the parameter, respectively.

  • (Ax, Ay, Az) where Ax, Ay, and Az are all integer powers of two. In this case the following restrictions apply to all image transfer operations:

    • x, y, and z of a VkOffset3D parameter must be integer multiples of Ax, Ay, and Az, respectively.

    • width of a VkExtent3D parameter must be an integer multiple of Ax, or else x + width must equal the width of the image subresource corresponding to the parameter.

    • height of a VkExtent3D parameter must be an integer multiple of Ay, or else y + height must equal the height of the image subresource corresponding to the parameter.

    • depth of a VkExtent3D parameter must be an integer multiple of Az, or else z + depth must equal the depth of the image subresource corresponding to the parameter.

    • If the format of the image corresponding to the parameters is one of the block-compressed formats then for the purposes of the above calculations the granularity must be scaled up by the compressed texel block dimensions.

Queues supporting graphics and/or compute operations must report (1,1,1) in minImageTransferGranularity, meaning that there are no additional restrictions on the granularity of image transfer operations for these queues. Other queues supporting image transfer operations are only required to support whole mip level transfers, thus minImageTransferGranularity for queues belonging to such queue families may be (0,0,0).

The Device Memory section describes memory properties queried from the physical device.

For physical device feature queries see the Features chapter.

Bits which may be set in VkQueueFamilyProperties::queueFlags indicating capabilities of queues in a queue family are:

typedef enum VkQueueFlagBits {
    VK_QUEUE_GRAPHICS_BIT = 0x00000001,
    VK_QUEUE_COMPUTE_BIT = 0x00000002,
    VK_QUEUE_TRANSFER_BIT = 0x00000004,
    VK_QUEUE_SPARSE_BINDING_BIT = 0x00000008,
    VK_QUEUE_PROTECTED_BIT = 0x00000010,
} VkQueueFlagBits;
  • VK_QUEUE_GRAPHICS_BIT specifies that queues in this queue family support graphics operations.

  • VK_QUEUE_COMPUTE_BIT specifies that queues in this queue family support compute operations.

  • VK_QUEUE_TRANSFER_BIT specifies that queues in this queue family support transfer operations.

  • VK_QUEUE_SPARSE_BINDING_BIT specifies that queues in this queue family support sparse memory management operations (see Sparse Resources). If any of the sparse resource features are enabled, then at least one queue family must support this bit.

  • if VK_QUEUE_PROTECTED_BIT is set, then the queues in this queue family support the VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT bit. (see Protected Memory). If the protected memory physical device feature is supported, then at least one queue family of at least one physical device exposed by the implementation must support this bit.

If an implementation exposes any queue family that supports graphics operations, at least one queue family of at least one physical device exposed by the implementation must support both graphics and compute operations.

Furthermore, if the protected memory physical device feature is supported, then at least one queue family of at least one physical device exposed by the implementation must support graphics operations, compute operations, and protected memory operations.

Note

All commands that are allowed on a queue that supports transfer operations are also allowed on a queue that supports either graphics or compute operations. Thus, if the capabilities of a queue family include VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT, then reporting the VK_QUEUE_TRANSFER_BIT capability separately for that queue family is optional.

For further details see Queues.

typedef VkFlags VkQueueFlags;

VkQueueFlags is a bitmask type for setting a mask of zero or more VkQueueFlagBits.

To query properties of queues available on a physical device, call:

void vkGetPhysicalDeviceQueueFamilyProperties2(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pQueueFamilyPropertyCount,
    VkQueueFamilyProperties2*                   pQueueFamilyProperties);

or the equivalent command

void vkGetPhysicalDeviceQueueFamilyProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pQueueFamilyPropertyCount,
    VkQueueFamilyProperties2*                   pQueueFamilyProperties);
  • physicalDevice is the handle to the physical device whose properties will be queried.

  • pQueueFamilyPropertyCount is a pointer to an integer related to the number of queue families available or queried, as described in vkGetPhysicalDeviceQueueFamilyProperties.

  • pQueueFamilyProperties is either NULL or a pointer to an array of VkQueueFamilyProperties2 structures.

vkGetPhysicalDeviceQueueFamilyProperties2 behaves similarly to vkGetPhysicalDeviceQueueFamilyProperties, with the ability to return extended information in a pNext chain of output structures.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pQueueFamilyPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pQueueFamilyPropertyCount is not 0, and pQueueFamilyProperties is not NULL, pQueueFamilyProperties must be a valid pointer to an array of pQueueFamilyPropertyCount VkQueueFamilyProperties2 structures

The VkQueueFamilyProperties2 structure is defined as:

typedef struct VkQueueFamilyProperties2 {
    VkStructureType            sType;
    void*                      pNext;
    VkQueueFamilyProperties    queueFamilyProperties;
} VkQueueFamilyProperties2;

or the equivalent

typedef VkQueueFamilyProperties2 VkQueueFamilyProperties2KHR;
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2

  • pNext must be NULL

4.2. Devices

Device objects represent logical connections to physical devices. Each device exposes a number of queue families each having one or more queues. All queues in a queue family support the same operations.

As described in Physical Devices, a Vulkan application will first query for all physical devices in a system. Each physical device can then be queried for its capabilities, including its queue and queue family properties. Once an acceptable physical device is identified, an application will create a corresponding logical device. An application must create a separate logical device for each physical device it will use. The created logical device is then the primary interface to the physical device.

How to enumerate the physical devices in a system and query those physical devices for their queue family properties is described in the Physical Device Enumeration section above.

A single logical device can also be created from multiple physical devices, if those physical devices belong to the same device group. A device group is a set of physical devices that support accessing each other’s memory and recording a single command buffer that can be executed on all the physical devices. Device groups are enumerated by calling vkEnumeratePhysicalDeviceGroups, and a logical device is created from a subset of the physical devices in a device group by passing the physical devices through VkDeviceGroupDeviceCreateInfo.

To retrieve a list of the device groups present in the system, call:

VkResult vkEnumeratePhysicalDeviceGroups(
    VkInstance                                  instance,
    uint32_t*                                   pPhysicalDeviceGroupCount,
    VkPhysicalDeviceGroupProperties*            pPhysicalDeviceGroupProperties);

or the equivalent command

VkResult vkEnumeratePhysicalDeviceGroupsKHR(
    VkInstance                                  instance,
    uint32_t*                                   pPhysicalDeviceGroupCount,
    VkPhysicalDeviceGroupProperties*            pPhysicalDeviceGroupProperties);
  • instance is a handle to a Vulkan instance previously created with vkCreateInstance.

  • pPhysicalDeviceGroupCount is a pointer to an integer related to the number of device groups available or queried, as described below.

  • pPhysicalDeviceGroupProperties is either NULL or a pointer to an array of VkPhysicalDeviceGroupProperties structures.

If pPhysicalDeviceGroupProperties is NULL, then the number of device groups available is returned in pPhysicalDeviceGroupCount. Otherwise, pPhysicalDeviceGroupCount must point to a variable set by the user to the number of elements in the pPhysicalDeviceGroupProperties array, and on return the variable is overwritten with the number of structures actually written to pPhysicalDeviceGroupProperties. If pPhysicalDeviceGroupCount is less than the number of device groups available, at most pPhysicalDeviceGroupCount structures will be written. If pPhysicalDeviceGroupCount is smaller than the number of device groups available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available device groups were returned.

Every physical device must be in exactly one device group.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pPhysicalDeviceGroupCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPhysicalDeviceGroupCount is not 0, and pPhysicalDeviceGroupProperties is not NULL, pPhysicalDeviceGroupProperties must be a valid pointer to an array of pPhysicalDeviceGroupCount VkPhysicalDeviceGroupProperties structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INITIALIZATION_FAILED

The VkPhysicalDeviceGroupProperties structure is defined as:

typedef struct VkPhysicalDeviceGroupProperties {
    VkStructureType     sType;
    void*               pNext;
    uint32_t            physicalDeviceCount;
    VkPhysicalDevice    physicalDevices[VK_MAX_DEVICE_GROUP_SIZE];
    VkBool32            subsetAllocation;
} VkPhysicalDeviceGroupProperties;

or the equivalent

typedef VkPhysicalDeviceGroupProperties VkPhysicalDeviceGroupPropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • physicalDeviceCount is the number of physical devices in the group.

  • physicalDevices is an array of physical device handles representing all physical devices in the group. The first physicalDeviceCount elements of the array will be valid.

  • subsetAllocation specifies whether logical devices created from the group support allocating device memory on a subset of devices, via the deviceMask member of the VkMemoryAllocateFlagsInfo. If this is VK_FALSE, then all device memory allocations are made across all physical devices in the group. If physicalDeviceCount is 1, then subsetAllocation must be VK_FALSE.

4.2.1. Device Creation

Logical devices are represented by VkDevice handles:

VK_DEFINE_HANDLE(VkDevice)

A logical device is created as a connection to a physical device. To create a logical device, call:

VkResult vkCreateDevice(
    VkPhysicalDevice                            physicalDevice,
    const VkDeviceCreateInfo*                   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDevice*                                   pDevice);
  • physicalDevice must be one of the device handles returned from a call to vkEnumeratePhysicalDevices (see Physical Device Enumeration).

  • pCreateInfo is a pointer to a VkDeviceCreateInfo structure containing information about how to create the device.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pDevice points to a handle in which the created VkDevice is returned.

vkCreateDevice verifies that extensions and features requested in the ppEnabledExtensionNames and pEnabledFeatures members of pCreateInfo, respectively, are supported by the implementation. If any requested extension is not supported, vkCreateDevice must return VK_ERROR_EXTENSION_NOT_PRESENT. If any requested feature is not supported, vkCreateDevice must return VK_ERROR_FEATURE_NOT_PRESENT. Support for extensions can be checked before creating a device by querying vkEnumerateDeviceExtensionProperties. Support for features can similarly be checked by querying vkGetPhysicalDeviceFeatures.

After verifying and enabling the extensions the VkDevice object is created and returned to the application. If a requested extension is only supported by a layer, both the layer and the extension need to be specified at vkCreateInstance time for the creation to succeed.

Multiple logical devices can be created from the same physical device. Logical device creation may fail due to lack of device-specific resources (in addition to the other errors). If that occurs, vkCreateDevice will return VK_ERROR_TOO_MANY_OBJECTS.

Valid Usage
Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pCreateInfo must be a valid pointer to a valid VkDeviceCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pDevice must be a valid pointer to a VkDevice handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INITIALIZATION_FAILED

  • VK_ERROR_EXTENSION_NOT_PRESENT

  • VK_ERROR_FEATURE_NOT_PRESENT

  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_DEVICE_LOST

The VkDeviceCreateInfo structure is defined as:

typedef struct VkDeviceCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkDeviceCreateFlags                flags;
    uint32_t                           queueCreateInfoCount;
    const VkDeviceQueueCreateInfo*     pQueueCreateInfos;
    uint32_t                           enabledLayerCount;
    const char* const*                 ppEnabledLayerNames;
    uint32_t                           enabledExtensionCount;
    const char* const*                 ppEnabledExtensionNames;
    const VkPhysicalDeviceFeatures*    pEnabledFeatures;
} VkDeviceCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • queueCreateInfoCount is the unsigned integer size of the pQueueCreateInfos array. Refer to the Queue Creation section below for further details.

  • pQueueCreateInfos is a pointer to an array of VkDeviceQueueCreateInfo structures describing the queues that are requested to be created along with the logical device. Refer to the Queue Creation section below for further details.

  • enabledLayerCount is deprecated and ignored.

  • ppEnabledLayerNames is deprecated and ignored. See Device Layer Deprecation.

  • enabledExtensionCount is the number of device extensions to enable.

  • ppEnabledExtensionNames is a pointer to an array of enabledExtensionCount null-terminated UTF-8 strings containing the names of extensions to enable for the created device. See the Extensions section for further details.

  • pEnabledFeatures is NULL or a pointer to a VkPhysicalDeviceFeatures structure that contains boolean indicators of all the features to be enabled. Refer to the Features section for further details.

Valid Usage
  • The queueFamilyIndex member of each element of pQueueCreateInfos must be unique within pQueueCreateInfos, except that two members can share the same queueFamilyIndex if one is a protected-capable queue and one is not a protected-capable queue.

  • If the pNext chain includes a VkPhysicalDeviceFeatures2 structure, then pEnabledFeatures must be NULL

  • ppEnabledExtensionNames must not contain VK_AMD_negative_viewport_height

Valid Usage (Implicit)
typedef VkFlags VkDeviceCreateFlags;

VkDeviceCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

A logical device can be created that connects to one or more physical devices by including a VkDeviceGroupDeviceCreateInfo structure in the pNext chain of VkDeviceCreateInfo. The VkDeviceGroupDeviceCreateInfo structure is defined as:

typedef struct VkDeviceGroupDeviceCreateInfo {
    VkStructureType            sType;
    const void*                pNext;
    uint32_t                   physicalDeviceCount;
    const VkPhysicalDevice*    pPhysicalDevices;
} VkDeviceGroupDeviceCreateInfo;

or the equivalent

typedef VkDeviceGroupDeviceCreateInfo VkDeviceGroupDeviceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • physicalDeviceCount is the number of elements in the pPhysicalDevices array.

  • pPhysicalDevices is an array of physical device handles belonging to the same device group.

The elements of the pPhysicalDevices array are an ordered list of the physical devices that the logical device represents. These must be a subset of a single device group, and need not be in the same order as they were enumerated. The order of the physical devices in the pPhysicalDevices array determines the device index of each physical device, with element i being assigned a device index of i. Certain commands and structures refer to one or more physical devices by using device indices or device masks formed using device indices.

A logical device created without using VkDeviceGroupDeviceCreateInfo, or with physicalDeviceCount equal to zero, is equivalent to a physicalDeviceCount of one and pPhysicalDevices pointing to the physicalDevice parameter to vkCreateDevice. In particular, the device index of that physical device is zero.

Valid Usage
  • Each element of pPhysicalDevices must be unique

  • All elements of pPhysicalDevices must be in the same device group as enumerated by vkEnumeratePhysicalDeviceGroups

  • If physicalDeviceCount is not 0, the physicalDevice parameter of vkCreateDevice must be an element of pPhysicalDevices.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO

  • If physicalDeviceCount is not 0, pPhysicalDevices must be a valid pointer to an array of physicalDeviceCount valid VkPhysicalDevice handles

4.2.2. Device Use

The following is a high-level list of VkDevice uses along with references on where to find more information:

4.2.3. Lost Device

A logical device may become lost for a number of implementation-specific reasons, indicating that pending and future command execution may fail and cause resources and backing memory to become undefined.

Note

Typical reasons for device loss will include things like execution timing out (to prevent denial of service), power management events, platform resource management, or implementation errors.

When this happens, certain commands will return VK_ERROR_DEVICE_LOST (see Error Codes for a list of such commands). After any such event, the logical device is considered lost. It is not possible to reset the logical device to a non-lost state, however the lost state is specific to a logical device (VkDevice), and the corresponding physical device (VkPhysicalDevice) may be otherwise unaffected.

In some cases, the physical device may also be lost, and attempting to create a new logical device will fail, returning VK_ERROR_DEVICE_LOST. This is usually indicative of a problem with the underlying implementation, or its connection to the host. If the physical device has not been lost, and a new logical device is successfully created from that physical device, it must be in the non-lost state.

Note

Whilst logical device loss may be recoverable, in the case of physical device loss, it is unlikely that an application will be able to recover unless additional, unaffected physical devices exist on the system. The error is largely informational and intended only to inform the user that a platform issue has occurred, and should be investigated further. For example, underlying hardware may have developed a fault or become physically disconnected from the rest of the system. In many cases, physical device loss may cause other more serious issues such as the operating system crashing; in which case it may not be reported via the Vulkan API.

Note

Undefined behavior caused by an application error may cause a device to become lost. However, such undefined behavior may also cause unrecoverable damage to the process, and it is then not guaranteed that the API objects, including the VkPhysicalDevice or the VkInstance are still valid or that the error is recoverable.

When a device is lost, its child objects are not implicitly destroyed and their handles are still valid. Those objects must still be destroyed before their parents or the device can be destroyed (see the Object Lifetime section). The host address space corresponding to device memory mapped using vkMapMemory is still valid, and host memory accesses to these mapped regions are still valid, but the contents are undefined. It is still legal to call any API command on the device and child objects.

Once a device is lost, command execution may fail, and commands that return a VkResult may return VK_ERROR_DEVICE_LOST. Commands that do not allow run-time errors must still operate correctly for valid usage and, if applicable, return valid data.

Commands that wait indefinitely for device execution (namely vkDeviceWaitIdle, vkQueueWaitIdle, vkWaitForFences or vkAcquireNextImageKHR with a maximum timeout, and vkGetQueryPoolResults with the VK_QUERY_RESULT_WAIT_BIT bit set in flags) must return in finite time even in the case of a lost device, and return either VK_SUCCESS or VK_ERROR_DEVICE_LOST. For any command that may return VK_ERROR_DEVICE_LOST, for the purpose of determining whether a command buffer is in the pending state, or whether resources are considered in-use by the device, a return value of VK_ERROR_DEVICE_LOST is equivalent to VK_SUCCESS.

The content of any external memory objects that have been exported from or imported to a lost device become undefined. Objects on other logical devices or in other APIs which are associated with the same underlying memory resource as the external memory objects on the lost device are unaffected other than their content becoming undefined. The layout of subresources of images on other logical devices that are bound to VkDeviceMemory objects associated with the same underlying memory resources as external memory objects on the lost device becomes VK_IMAGE_LAYOUT_UNDEFINED.

The state of VkSemaphore objects on other logical devices created by importing a semaphore payload with temporary permanence which was exported from the lost device is undefined. The state of VkSemaphore objects on other logical devices that permanently share a semaphore payload with a VkSemaphore object on the lost device is undefined, and remains undefined following any subsequent signal operations. Implementations must ensure pending and subsequently submitted wait operations on such semaphores behave as defined in Semaphore State Requirements For Wait Operations for external semaphores not in a valid state for a wait operation.

editing-note

TODO (piman) - I do not think we are very clear about what “in-use by the device” means.

4.2.4. Device Destruction

To destroy a device, call:

void vkDestroyDevice(
    VkDevice                                    device,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

To ensure that no work is active on the device, vkDeviceWaitIdle can be used to gate the destruction of the device. Prior to destroying a device, an application is responsible for destroying/freeing any Vulkan objects that were created using that device as the first parameter of the corresponding vkCreate* or vkAllocate* command.

Note

The lifetime of each of these objects is bound by the lifetime of the VkDevice object. Therefore, to avoid resource leaks, it is critical that an application explicitly free all of these resources prior to calling vkDestroyDevice.

Valid Usage
  • All child objects created on device must have been destroyed prior to destroying device

  • If VkAllocationCallbacks were provided when device was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when device was created, pAllocator must be NULL

Valid Usage (Implicit)
  • If device is not NULL, device must be a valid VkDevice handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

Host Synchronization
  • Host access to device must be externally synchronized

4.3. Queues

4.3.1. Queue Family Properties

As discussed in the Physical Device Enumeration section above, the vkGetPhysicalDeviceQueueFamilyProperties command is used to retrieve details about the queue families and queues supported by a device.

Each index in the pQueueFamilyProperties array returned by vkGetPhysicalDeviceQueueFamilyProperties describes a unique queue family on that physical device. These indices are used when creating queues, and they correspond directly with the queueFamilyIndex that is passed to the vkCreateDevice command via the VkDeviceQueueCreateInfo structure as described in the Queue Creation section below.

Grouping of queue families within a physical device is implementation-dependent.

Note

The general expectation is that a physical device groups all queues of matching capabilities into a single family. However, while implementations should do this, it is possible that a physical device may return two separate queue families with the same capabilities.

Once an application has identified a physical device with the queue(s) that it desires to use, it will create those queues in conjunction with a logical device. This is described in the following section.

4.3.2. Queue Creation

Creating a logical device also creates the queues associated with that device. The queues to create are described by a set of VkDeviceQueueCreateInfo structures that are passed to vkCreateDevice in pQueueCreateInfos.

Queues are represented by VkQueue handles:

VK_DEFINE_HANDLE(VkQueue)

The VkDeviceQueueCreateInfo structure is defined as:

typedef struct VkDeviceQueueCreateInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkDeviceQueueCreateFlags    flags;
    uint32_t                    queueFamilyIndex;
    uint32_t                    queueCount;
    const float*                pQueuePriorities;
} VkDeviceQueueCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask indicating behavior of the queue.

  • queueFamilyIndex is an unsigned integer indicating the index of the queue family to create on this device. This index corresponds to the index of an element of the pQueueFamilyProperties array that was returned by vkGetPhysicalDeviceQueueFamilyProperties.

  • queueCount is an unsigned integer specifying the number of queues to create in the queue family indicated by queueFamilyIndex.

  • pQueuePriorities is an array of queueCount normalized floating point values, specifying priorities of work that will be submitted to each created queue. See Queue Priority for more information.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties

  • queueCount must be less than or equal to the queueCount member of the VkQueueFamilyProperties structure, as returned by vkGetPhysicalDeviceQueueFamilyProperties in the pQueueFamilyProperties[queueFamilyIndex]

  • Each element of pQueuePriorities must be between 0.0 and 1.0 inclusive

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkDeviceQueueGlobalPriorityCreateInfoEXT

  • flags must be a valid combination of VkDeviceQueueCreateFlagBits values

  • pQueuePriorities must be a valid pointer to an array of queueCount float values

  • queueCount must be greater than 0

Bits which can be set in VkDeviceQueueCreateInfo::flags to specify usage behavior of the queue are:

typedef enum VkDeviceQueueCreateFlagBits {
    VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT = 0x00000001,
} VkDeviceQueueCreateFlagBits;
  • VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT specifies that the device queue is a protected-capable queue. If the protected memory feature is not enabled, the VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT bit of flags must not be set.

typedef VkFlags VkDeviceQueueCreateFlags;

VkDeviceQueueCreateFlags is a bitmask type for setting a mask of zero or more VkDeviceQueueCreateFlagBits.

A queue can be created with a system-wide priority by including a VkDeviceQueueGlobalPriorityCreateInfoEXT structure in the pNext chain of VkDeviceQueueCreateInfo.

The VkDeviceQueueGlobalPriorityCreateInfoEXT structure is defined as:

typedef struct VkDeviceQueueGlobalPriorityCreateInfoEXT {
    VkStructureType             sType;
    const void*                 pNext;
    VkQueueGlobalPriorityEXT    globalPriority;
} VkDeviceQueueGlobalPriorityCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • globalPriority is the system-wide priority associated to this queue as specified by VkQueueGlobalPriorityEXT

A queue created without specifying VkDeviceQueueGlobalPriorityCreateInfoEXT will default to VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_QUEUE_GLOBAL_PRIORITY_CREATE_INFO_EXT

  • globalPriority must be a valid VkQueueGlobalPriorityEXT value

Possible values of VkDeviceQueueGlobalPriorityCreateInfoEXT::globalPriority, specifying a system-wide priority level are:

typedef enum VkQueueGlobalPriorityEXT {
    VK_QUEUE_GLOBAL_PRIORITY_LOW_EXT = 128,
    VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT = 256,
    VK_QUEUE_GLOBAL_PRIORITY_HIGH_EXT = 512,
    VK_QUEUE_GLOBAL_PRIORITY_REALTIME_EXT = 1024,
} VkQueueGlobalPriorityEXT;

Priority values are sorted in ascending order. A comparison operation on the enum values can be used to determine the priority order.

  • VK_QUEUE_GLOBAL_PRIORITY_LOW_EXT is below the system default. Useful for non-interactive tasks.

  • VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT is the system default priority.

  • VK_QUEUE_GLOBAL_PRIORITY_HIGH_EXT is above the system default.

  • VK_QUEUE_GLOBAL_PRIORITY_REALTIME_EXT is the highest priority. Useful for critical tasks.

Queues with higher system priority may be allotted more processing time than queues with lower priority. An implementation may allow a higher-priority queue to starve a lower-priority queue until the higher-priority queue has no further commands to execute.

Priorities imply no ordering or scheduling constraints.

No specific guarantees are made about higher priority queues receiving more processing time or better quality of service than lower priority queues.

The global priority level of a queue takes precedence over the per-process queue priority (VkDeviceQueueCreateInfo::pQueuePriorities).

Abuse of this feature may result in starving the rest of the system of implementation resources. Therefore, the driver implementation may deny requests to acquire a priority above the default priority (VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT) if the caller does not have sufficient privileges. In this scenario VK_ERROR_NOT_PERMITTED_EXT is returned.

The driver implementation may fail the queue allocation request if resources required to complete the operation have been exhausted (either by the same process or a different process). In this scenario VK_ERROR_INITIALIZATION_FAILED is returned.

To retrieve a handle to a VkQueue object, call:

void vkGetDeviceQueue(
    VkDevice                                    device,
    uint32_t                                    queueFamilyIndex,
    uint32_t                                    queueIndex,
    VkQueue*                                    pQueue);
  • device is the logical device that owns the queue.

  • queueFamilyIndex is the index of the queue family to which the queue belongs.

  • queueIndex is the index within this queue family of the queue to retrieve.

  • pQueue is a pointer to a VkQueue object that will be filled with the handle for the requested queue.

vkGetDeviceQueue must only be used to get queues that were created with the flags parameter of VkDeviceQueueCreateInfo set to zero. To get queues that were created with a non-zero flags parameter use vkGetDeviceQueue2.

Valid Usage
  • queueFamilyIndex must be one of the queue family indices specified when device was created, via the VkDeviceQueueCreateInfo structure

  • queueIndex must be less than the number of queues created for the specified queue family index when device was created, via the queueCount member of the VkDeviceQueueCreateInfo structure

  • VkDeviceQueueCreateInfo::flags must have been set to zero when device was created

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pQueue must be a valid pointer to a VkQueue handle

To retrieve a handle to a VkQueue object with specific VkDeviceQueueCreateFlags creation flags, call:

void vkGetDeviceQueue2(
    VkDevice                                    device,
    const VkDeviceQueueInfo2*                   pQueueInfo,
    VkQueue*                                    pQueue);
  • device is the logical device that owns the queue.

  • pQueueInfo points to an instance of the VkDeviceQueueInfo2 structure, describing the parameters used to create the device queue.

  • pQueue is a pointer to a VkQueue object that will be filled with the handle for the requested queue.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pQueueInfo must be a valid pointer to a valid VkDeviceQueueInfo2 structure

  • pQueue must be a valid pointer to a VkQueue handle

The VkDeviceQueueInfo2 structure is defined as:

typedef struct VkDeviceQueueInfo2 {
    VkStructureType             sType;
    const void*                 pNext;
    VkDeviceQueueCreateFlags    flags;
    uint32_t                    queueFamilyIndex;
    uint32_t                    queueIndex;
} VkDeviceQueueInfo2;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure. The pNext chain of VkDeviceQueueInfo2 is used to provide additional image parameters to vkGetDeviceQueue2.

  • flags is a VkDeviceQueueCreateFlags value indicating the flags used to create the device queue.

  • queueFamilyIndex is the index of the queue family to which the queue belongs.

  • queueIndex is the index within this queue family of the queue to retrieve.

The queue returned by vkGetDeviceQueue2 must have the same flags value from this structure as that used at device creation time in a VkDeviceQueueCreateInfo instance. If no matching flags were specified at device creation time then pQueue will return VK_NULL_HANDLE.

Valid Usage
  • queueFamilyIndex must be one of the queue family indices specified when device was created, via the VkDeviceQueueCreateInfo structure

  • queueIndex must be less than the number of queues created for the specified queue family index and VkDeviceQueueCreateFlags member flags equal to this flags value when device was created, via the queueCount member of the VkDeviceQueueCreateInfo structure

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_QUEUE_INFO_2

  • pNext must be NULL

  • flags must be a valid combination of VkDeviceQueueCreateFlagBits values

  • flags must not be 0

4.3.3. Queue Family Index

The queue family index is used in multiple places in Vulkan in order to tie operations to a specific family of queues.

When retrieving a handle to the queue via vkGetDeviceQueue, the queue family index is used to select which queue family to retrieve the VkQueue handle from as described in the previous section.

When creating a VkCommandPool object (see Command Pools), a queue family index is specified in the VkCommandPoolCreateInfo structure. Command buffers from this pool can only be submitted on queues corresponding to this queue family.

When creating VkImage (see Images) and VkBuffer (see Buffers) resources, a set of queue families is included in the VkImageCreateInfo and VkBufferCreateInfo structures to specify the queue families that can access the resource.

When inserting a VkBufferMemoryBarrier or VkImageMemoryBarrier (see Events) a source and destination queue family index is specified to allow the ownership of a buffer or image to be transferred from one queue family to another. See the Resource Sharing section for details.

4.3.4. Queue Priority

Each queue is assigned a priority, as set in the VkDeviceQueueCreateInfo structures when creating the device. The priority of each queue is a normalized floating point value between 0.0 and 1.0, which is then translated to a discrete priority level by the implementation. Higher values indicate a higher priority, with 0.0 being the lowest priority and 1.0 being the highest.

Within the same device, queues with higher priority may be allotted more processing time than queues with lower priority. The implementation makes no guarantees with regards to ordering or scheduling among queues with the same priority, other than the constraints defined by any explicit synchronization primitives. The implementation make no guarantees with regards to queues across different devices.

An implementation may allow a higher-priority queue to starve a lower-priority queue on the same VkDevice until the higher-priority queue has no further commands to execute. The relationship of queue priorities must not cause queues on one VkDevice to starve queues on another VkDevice.

No specific guarantees are made about higher priority queues receiving more processing time or better quality of service than lower priority queues.

4.3.5. Queue Submission

Work is submitted to a queue via queue submission commands such as vkQueueSubmit. Queue submission commands define a set of queue operations to be executed by the underlying physical device, including synchronization with semaphores and fences.

Submission commands take as parameters a target queue, zero or more batches of work, and an optional fence to signal upon completion. Each batch consists of three distinct parts:

  1. Zero or more semaphores to wait on before execution of the rest of the batch.

  2. Zero or more work items to execute.

    • If present, these describe a queue operation matching the work described.

  3. Zero or more semaphores to signal upon completion of the work items.

If a fence is present in a queue submission, it describes a fence signal operation.

All work described by a queue submission command must be submitted to the queue before the command returns.

Sparse Memory Binding

In Vulkan it is possible to sparsely bind memory to buffers and images as described in the Sparse Resource chapter. Sparse memory binding is a queue operation. A queue whose flags include the VK_QUEUE_SPARSE_BINDING_BIT must be able to support the mapping of a virtual address to a physical address on the device. This causes an update to the page table mappings on the device. This update must be synchronized on a queue to avoid corrupting page table mappings during execution of graphics commands. By binding the sparse memory resources on queues, all commands that are dependent on the updated bindings are synchronized to only execute after the binding is updated. See the Synchronization and Cache Control chapter for how this synchronization is accomplished.

4.3.6. Queue Destruction

Queues are created along with a logical device during vkCreateDevice. All queues associated with a logical device are destroyed when vkDestroyDevice is called on that device.

5. Command Buffers

Command buffers are objects used to record commands which can be subsequently submitted to a device queue for execution. There are two levels of command buffers - primary command buffers, which can execute secondary command buffers, and which are submitted to queues, and secondary command buffers, which can be executed by primary command buffers, and which are not directly submitted to queues.

Command buffers are represented by VkCommandBuffer handles:

VK_DEFINE_HANDLE(VkCommandBuffer)

Recorded commands include commands to bind pipelines and descriptor sets to the command buffer, commands to modify dynamic state, commands to draw (for graphics rendering), commands to dispatch (for compute), commands to execute secondary command buffers (for primary command buffers only), commands to copy buffers and images, and other commands.

Each command buffer manages state independently of other command buffers. There is no inheritance of state across primary and secondary command buffers, or between secondary command buffers. When a command buffer begins recording, all state in that command buffer is undefined. When secondary command buffer(s) are recorded to execute on a primary command buffer, the secondary command buffer inherits no state from the primary command buffer, and all state of the primary command buffer is undefined after an execute secondary command buffer command is recorded. There is one exception to this rule - if the primary command buffer is inside a render pass instance, then the render pass and subpass state is not disturbed by executing secondary command buffers. Whenever the state of a command buffer is undefined, the application must set all relevant state on the command buffer before any state dependent commands such as draws and dispatches are recorded, otherwise the behavior of executing that command buffer is undefined.

Unless otherwise specified, and without explicit synchronization, the various commands submitted to a queue via command buffers may execute in arbitrary order relative to each other, and/or concurrently. Also, the memory side-effects of those commands may not be directly visible to other commands without explicit memory dependencies. This is true within a command buffer, and across command buffers submitted to a given queue. See the synchronization chapter for information on implicit and explicit synchronization between commands.

5.1. Command Buffer Lifecycle

Each command buffer is always in one of the following states:

Initial

When a command buffer is allocated, it is in the initial state. Some commands are able to reset a command buffer, or a set of command buffers, back to this state from any of the executable, recording or invalid state. Command buffers in the initial state can only be moved to the recording state, or freed.

Recording

vkBeginCommandBuffer changes the state of a command buffer from the initial state to the recording state. Once a command buffer is in the recording state, vkCmd* commands can be used to record to the command buffer.

Executable

vkEndCommandBuffer ends the recording of a command buffer, and moves it from the recording state to the executable state. Executable command buffers can be submitted, reset, or recorded to another command buffer.

Pending

Queue submission of a command buffer changes the state of a command buffer from the executable state to the pending state. Whilst in the pending state, applications must not attempt to modify the command buffer in any way - as the device may be processing the commands recorded to it. Once execution of a command buffer completes, the command buffer reverts back to either the executable state, or the invalid state if it was recorded with VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT. A synchronization command should be used to detect when this occurs.

Invalid

Some operations, such as modifying or deleting a resource that was used in a command recorded to a command buffer, will transition the state of that command buffer into the invalid state. Command buffers in the invalid state can only be reset or freed.

commandbuffer lifecycle
Figure 1. Lifecycle of a command buffer

Any given command that operates on a command buffer has its own requirements on what state a command buffer must be in, which are detailed in the valid usage constraints for that command.

Resetting a command buffer is an operation that discards any previously recorded commands and puts a command buffer in the initial state. Resetting occurs as a result of vkResetCommandBuffer or vkResetCommandPool, or as part of vkBeginCommandBuffer (which additionally puts the command buffer in the recording state).

Secondary command buffers can be recorded to a primary command buffer via vkCmdExecuteCommands. This partially ties the lifecycle of the two command buffers together - if the primary is submitted to a queue, both the primary and any secondaries recorded to it move to the pending state. Once execution of the primary completes, so does any secondary recorded within it, and once all executions of each command buffer complete, they move to the executable state. If a secondary moves to any other state whilst it is recorded to another command buffer, the primary moves to the invalid state. A primary moving to any other state does not affect the state of the secondary. Resetting or freeing a primary command buffer removes the linkage to any secondary command buffers that were recorded to it.

5.2. Command Pools

Command pools are opaque objects that command buffer memory is allocated from, and which allow the implementation to amortize the cost of resource creation across multiple command buffers. Command pools are externally synchronized, meaning that a command pool must not be used concurrently in multiple threads. That includes use via recording commands on any command buffers allocated from the pool, as well as operations that allocate, free, and reset command buffers or the pool itself.

Command pools are represented by VkCommandPool handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkCommandPool)

To create a command pool, call:

VkResult vkCreateCommandPool(
    VkDevice                                    device,
    const VkCommandPoolCreateInfo*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkCommandPool*                              pCommandPool);
  • device is the logical device that creates the command pool.

  • pCreateInfo is a pointer to an instance of the VkCommandPoolCreateInfo structure specifying the state of the command pool object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pCommandPool points to a VkCommandPool handle in which the created pool is returned.

Valid Usage
  • pCreateInfo::queueFamilyIndex must be the index of a queue family available in the logical device device.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkCommandPoolCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pCommandPool must be a valid pointer to a VkCommandPool handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandPoolCreateInfo structure is defined as:

typedef struct VkCommandPoolCreateInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkCommandPoolCreateFlags    flags;
    uint32_t                    queueFamilyIndex;
} VkCommandPoolCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkCommandPoolCreateFlagBits indicating usage behavior for the pool and command buffers allocated from it.

  • queueFamilyIndex designates a queue family as described in section Queue Family Properties. All command buffers allocated from this command pool must be submitted on queues from the same queue family.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_COMMAND_POOL_CREATE_INFO

  • pNext must be NULL

  • flags must be a valid combination of VkCommandPoolCreateFlagBits values

Bits which can be set in VkCommandPoolCreateInfo::flags to specify usage behavior for a command pool are:

typedef enum VkCommandPoolCreateFlagBits {
    VK_COMMAND_POOL_CREATE_TRANSIENT_BIT = 0x00000001,
    VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT = 0x00000002,
    VK_COMMAND_POOL_CREATE_PROTECTED_BIT = 0x00000004,
} VkCommandPoolCreateFlagBits;
  • VK_COMMAND_POOL_CREATE_TRANSIENT_BIT specifies that command buffers allocated from the pool will be short-lived, meaning that they will be reset or freed in a relatively short timeframe. This flag may be used by the implementation to control memory allocation behavior within the pool.

  • VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT allows any command buffer allocated from a pool to be individually reset to the initial state; either by calling vkResetCommandBuffer, or via the implicit reset when calling vkBeginCommandBuffer. If this flag is not set on a pool, then vkResetCommandBuffer must not be called for any command buffer allocated from that pool.

  • VK_COMMAND_POOL_CREATE_PROTECTED_BIT specifies that command buffers allocated from the pool are protected command buffers. If the protected memory feature is not enabled, the VK_COMMAND_POOL_CREATE_PROTECTED_BIT bit of flags must not be set.

typedef VkFlags VkCommandPoolCreateFlags;

VkCommandPoolCreateFlags is a bitmask type for setting a mask of zero or more VkCommandPoolCreateFlagBits.

To trim a command pool, call:

void vkTrimCommandPool(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    VkCommandPoolTrimFlags                      flags);

or the equivalent command

void vkTrimCommandPoolKHR(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    VkCommandPoolTrimFlags                      flags);
  • device is the logical device that owns the command pool.

  • commandPool is the command pool to trim.

  • flags is reserved for future use.

Trimming a command pool recycles unused memory from the command pool back to the system. Command buffers allocated from the pool are not affected by the command.

Note

This command provides applications with some control over the internal memory allocations used by command pools.

Unused memory normally arises from command buffers that have been recorded and later reset, such that they are no longer using the memory. On reset, a command buffer can return memory to its command pool, but the only way to release memory from a command pool to the system requires calling vkResetCommandPool, which cannot be executed while any command buffers from that pool are still in use. Subsequent recording operations into command buffers will re-use this memory but since total memory requirements fluctuate over time, unused memory can accumulate.

In this situation, trimming a command pool may be useful to return unused memory back to the system, returning the total outstanding memory allocated by the pool back to a more “average” value.

Implementations utilize many internal allocation strategies that make it impossible to guarantee that all unused memory is released back to the system. For instance, an implementation of a command pool may involve allocating memory in bulk from the system and sub-allocating from that memory. In such an implementation any live command buffer that holds a reference to a bulk allocation would prevent that allocation from being freed, even if only a small proportion of the bulk allocation is in use.

In most cases trimming will result in a reduction in allocated but unused memory, but it does not guarantee the “ideal” behaviour.

Trimming may be an expensive operation, and should not be called frequently. Trimming should be treated as a way to relieve memory pressure after application-known points when there exists enough unused memory that the cost of trimming is “worth” it.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • commandPool must be a valid VkCommandPool handle

  • flags must be 0

  • commandPool must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to commandPool must be externally synchronized

typedef VkFlags VkCommandPoolTrimFlags;

or the equivalent

typedef VkCommandPoolTrimFlags VkCommandPoolTrimFlagsKHR;

VkCommandPoolTrimFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To reset a command pool, call:

VkResult vkResetCommandPool(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    VkCommandPoolResetFlags                     flags);
  • device is the logical device that owns the command pool.

  • commandPool is the command pool to reset.

  • flags is a bitmask of VkCommandPoolResetFlagBits controlling the reset operation.

Resetting a command pool recycles all of the resources from all of the command buffers allocated from the command pool back to the command pool. All command buffers that have been allocated from the command pool are put in the initial state.

Any primary command buffer allocated from another VkCommandPool that is in the recording or executable state and has a secondary command buffer allocated from commandPool recorded into it, becomes invalid.

Valid Usage
  • All VkCommandBuffer objects allocated from commandPool must not be in the pending state

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • commandPool must be a valid VkCommandPool handle

  • flags must be a valid combination of VkCommandPoolResetFlagBits values

  • commandPool must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to commandPool must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Bits which can be set in vkResetCommandPool::flags to control the reset operation are:

typedef enum VkCommandPoolResetFlagBits {
    VK_COMMAND_POOL_RESET_RELEASE_RESOURCES_BIT = 0x00000001,
} VkCommandPoolResetFlagBits;
  • VK_COMMAND_POOL_RESET_RELEASE_RESOURCES_BIT specifies that resetting a command pool recycles all of the resources from the command pool back to the system.

typedef VkFlags VkCommandPoolResetFlags;

VkCommandPoolResetFlags is a bitmask type for setting a mask of zero or more VkCommandPoolResetFlagBits.

To destroy a command pool, call:

void vkDestroyCommandPool(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the command pool.

  • commandPool is the handle of the command pool to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

When a pool is destroyed, all command buffers allocated from the pool are freed.

Any primary command buffer allocated from another VkCommandPool that is in the recording or executable state and has a secondary command buffer allocated from commandPool recorded into it, becomes invalid.

Valid Usage
  • All VkCommandBuffer objects allocated from commandPool must not be in the pending state.

  • If VkAllocationCallbacks were provided when commandPool was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when commandPool was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If commandPool is not VK_NULL_HANDLE, commandPool must be a valid VkCommandPool handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If commandPool is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to commandPool must be externally synchronized

5.3. Command Buffer Allocation and Management

To allocate command buffers, call:

VkResult vkAllocateCommandBuffers(
    VkDevice                                    device,
    const VkCommandBufferAllocateInfo*          pAllocateInfo,
    VkCommandBuffer*                            pCommandBuffers);
  • device is the logical device that owns the command pool.

  • pAllocateInfo is a pointer to an instance of the VkCommandBufferAllocateInfo structure describing parameters of the allocation.

  • pCommandBuffers is a pointer to an array of VkCommandBuffer handles in which the resulting command buffer objects are returned. The array must be at least the length specified by the commandBufferCount member of pAllocateInfo. Each allocated command buffer begins in the initial state.

vkAllocateCommandBuffers can be used to create multiple command buffers. If the creation of any of those command buffers fails, the implementation must destroy all successfully created command buffer objects from this command, set all entries of the pCommandBuffers array to NULL and return the error.

When command buffers are first allocated, they are in the initial state.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pAllocateInfo must be a valid pointer to a valid VkCommandBufferAllocateInfo structure

  • pCommandBuffers must be a valid pointer to an array of pAllocateInfo::commandBufferCount VkCommandBuffer handles

Host Synchronization
  • Host access to pAllocateInfo::commandPool must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandBufferAllocateInfo structure is defined as:

typedef struct VkCommandBufferAllocateInfo {
    VkStructureType         sType;
    const void*             pNext;
    VkCommandPool           commandPool;
    VkCommandBufferLevel    level;
    uint32_t                commandBufferCount;
} VkCommandBufferAllocateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • commandPool is the command pool from which the command buffers are allocated.

  • level is an VkCommandBufferLevel value specifying the command buffer level.

  • commandBufferCount is the number of command buffers to allocate from the pool.

Valid Usage
  • commandBufferCount must be greater than 0

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO

  • pNext must be NULL

  • commandPool must be a valid VkCommandPool handle

  • level must be a valid VkCommandBufferLevel value

Possible values of VkCommandBufferAllocateInfo::level, specifying the command buffer level, are:

typedef enum VkCommandBufferLevel {
    VK_COMMAND_BUFFER_LEVEL_PRIMARY = 0,
    VK_COMMAND_BUFFER_LEVEL_SECONDARY = 1,
} VkCommandBufferLevel;
  • VK_COMMAND_BUFFER_LEVEL_PRIMARY specifies a primary command buffer.

  • VK_COMMAND_BUFFER_LEVEL_SECONDARY specifies a secondary command buffer.

To reset command buffers, call:

VkResult vkResetCommandBuffer(
    VkCommandBuffer                             commandBuffer,
    VkCommandBufferResetFlags                   flags);

Any primary command buffer that is in the recording or executable state and has commandBuffer recorded into it, becomes invalid.

Valid Usage
  • commandBuffer must not be in the pending state

  • commandBuffer must have been allocated from a pool that was created with the VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT

Valid Usage (Implicit)
Host Synchronization
  • Host access to commandBuffer must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Bits which can be set in vkResetCommandBuffer::flags to control the reset operation are:

typedef enum VkCommandBufferResetFlagBits {
    VK_COMMAND_BUFFER_RESET_RELEASE_RESOURCES_BIT = 0x00000001,
} VkCommandBufferResetFlagBits;
  • VK_COMMAND_BUFFER_RESET_RELEASE_RESOURCES_BIT specifies that most or all memory resources currently owned by the command buffer should be returned to the parent command pool. If this flag is not set, then the command buffer may hold onto memory resources and reuse them when recording commands. commandBuffer is moved to the initial state.

typedef VkFlags VkCommandBufferResetFlags;

VkCommandBufferResetFlags is a bitmask type for setting a mask of zero or more VkCommandBufferResetFlagBits.

To free command buffers, call:

void vkFreeCommandBuffers(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    uint32_t                                    commandBufferCount,
    const VkCommandBuffer*                      pCommandBuffers);
  • device is the logical device that owns the command pool.

  • commandPool is the command pool from which the command buffers were allocated.

  • commandBufferCount is the length of the pCommandBuffers array.

  • pCommandBuffers is an array of handles of command buffers to free.

Any primary command buffer that is in the recording or executable state and has any element of pCommandBuffers recorded into it, becomes invalid.

Valid Usage
  • All elements of pCommandBuffers must not be in the pending state

  • pCommandBuffers must be a valid pointer to an array of commandBufferCount VkCommandBuffer handles, each element of which must either be a valid handle or NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • commandPool must be a valid VkCommandPool handle

  • commandBufferCount must be greater than 0

  • commandPool must have been created, allocated, or retrieved from device

  • Each element of pCommandBuffers that is a valid handle must have been created, allocated, or retrieved from commandPool

Host Synchronization
  • Host access to commandPool must be externally synchronized

  • Host access to each member of pCommandBuffers must be externally synchronized

5.4. Command Buffer Recording

To begin recording a command buffer, call:

VkResult vkBeginCommandBuffer(
    VkCommandBuffer                             commandBuffer,
    const VkCommandBufferBeginInfo*             pBeginInfo);
  • commandBuffer is the handle of the command buffer which is to be put in the recording state.

  • pBeginInfo is an instance of the VkCommandBufferBeginInfo structure, which defines additional information about how the command buffer begins recording.

Valid Usage
  • commandBuffer must not be in the recording or pending state.

  • If commandBuffer was allocated from a VkCommandPool which did not have the VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT flag set, commandBuffer must be in the initial state.

  • If commandBuffer is a secondary command buffer, the pInheritanceInfo member of pBeginInfo must be a valid VkCommandBufferInheritanceInfo structure

  • If commandBuffer is a secondary command buffer and either the occlusionQueryEnable member of the pInheritanceInfo member of pBeginInfo is VK_FALSE, or the precise occlusion queries feature is not enabled, the queryFlags member of the pInheritanceInfo member pBeginInfo must not contain VK_QUERY_CONTROL_PRECISE_BIT

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pBeginInfo must be a valid pointer to a valid VkCommandBufferBeginInfo structure

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandBufferBeginInfo structure is defined as:

typedef struct VkCommandBufferBeginInfo {
    VkStructureType                          sType;
    const void*                              pNext;
    VkCommandBufferUsageFlags                flags;
    const VkCommandBufferInheritanceInfo*    pInheritanceInfo;
} VkCommandBufferBeginInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkCommandBufferUsageFlagBits specifying usage behavior for the command buffer.

  • pInheritanceInfo is a pointer to a VkCommandBufferInheritanceInfo structure, which is used if commandBuffer is a secondary command buffer. If this is a primary command buffer, then this value is ignored.

Valid Usage
  • If flags contains VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT, the renderPass member of pInheritanceInfo must be a valid VkRenderPass

  • If flags contains VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT, the subpass member of pInheritanceInfo must be a valid subpass index within the renderPass member of pInheritanceInfo

  • If flags contains VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT, the framebuffer member of pInheritanceInfo must be either VK_NULL_HANDLE, or a valid VkFramebuffer that is compatible with the renderPass member of pInheritanceInfo

Valid Usage (Implicit)

Bits which can be set in VkCommandBufferBeginInfo::flags to specify usage behavior for a command buffer are:

typedef enum VkCommandBufferUsageFlagBits {
    VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT = 0x00000001,
    VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT = 0x00000002,
    VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT = 0x00000004,
} VkCommandBufferUsageFlagBits;
  • VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT specifies that each recording of the command buffer will only be submitted once, and the command buffer will be reset and recorded again between each submission.

  • VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT specifies that a secondary command buffer is considered to be entirely inside a render pass. If this is a primary command buffer, then this bit is ignored.

  • VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT specifies that a command buffer can be resubmitted to a queue while it is in the pending state, and recorded into multiple primary command buffers.

typedef VkFlags VkCommandBufferUsageFlags;

VkCommandBufferUsageFlags is a bitmask type for setting a mask of zero or more VkCommandBufferUsageFlagBits.

If the command buffer is a secondary command buffer, then the VkCommandBufferInheritanceInfo structure defines any state that will be inherited from the primary command buffer:

typedef struct VkCommandBufferInheritanceInfo {
    VkStructureType                  sType;
    const void*                      pNext;
    VkRenderPass                     renderPass;
    uint32_t                         subpass;
    VkFramebuffer                    framebuffer;
    VkBool32                         occlusionQueryEnable;
    VkQueryControlFlags              queryFlags;
    VkQueryPipelineStatisticFlags    pipelineStatistics;
} VkCommandBufferInheritanceInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • renderPass is a VkRenderPass object defining which render passes the VkCommandBuffer will be compatible with and can be executed within. If the VkCommandBuffer will not be executed within a render pass instance, renderPass is ignored.

  • subpass is the index of the subpass within the render pass instance that the VkCommandBuffer will be executed within. If the VkCommandBuffer will not be executed within a render pass instance, subpass is ignored.

  • framebuffer optionally refers to the VkFramebuffer object that the VkCommandBuffer will be rendering to if it is executed within a render pass instance. It can be VK_NULL_HANDLE if the framebuffer is not known, or if the VkCommandBuffer will not be executed within a render pass instance.

    Note

    Specifying the exact framebuffer that the secondary command buffer will be executed with may result in better performance at command buffer execution time.

  • occlusionQueryEnable specifies whether the command buffer can be executed while an occlusion query is active in the primary command buffer. If this is VK_TRUE, then this command buffer can be executed whether the primary command buffer has an occlusion query active or not. If this is VK_FALSE, then the primary command buffer must not have an occlusion query active.

  • queryFlags specifies the query flags that can be used by an active occlusion query in the primary command buffer when this secondary command buffer is executed. If this value includes the VK_QUERY_CONTROL_PRECISE_BIT bit, then the active query can return boolean results or actual sample counts. If this bit is not set, then the active query must not use the VK_QUERY_CONTROL_PRECISE_BIT bit.

  • pipelineStatistics is a bitmask of VkQueryPipelineStatisticFlagBits specifying the set of pipeline statistics that can be counted by an active query in the primary command buffer when this secondary command buffer is executed. If this value includes a given bit, then this command buffer can be executed whether the primary command buffer has a pipeline statistics query active that includes this bit or not. If this value excludes a given bit, then the active pipeline statistics query must not be from a query pool that counts that statistic.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_INHERITANCE_INFO

  • pNext must be NULL

  • Both of framebuffer, and renderPass that are valid handles must have been created, allocated, or retrieved from the same VkDevice

If VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT was not set when creating a command buffer, that command buffer must not be submitted to a queue whilst it is already in the pending state. If VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT is not set on a secondary command buffer, that command buffer must not be used more than once in a given primary command buffer.

Note

On some implementations, not using the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT bit enables command buffers to be patched in-place if needed, rather than creating a copy of the command buffer.

If a command buffer is in the invalid, or executable state, and the command buffer was allocated from a command pool with the VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT flag set, then vkBeginCommandBuffer implicitly resets the command buffer, behaving as if vkResetCommandBuffer had been called with VK_COMMAND_BUFFER_RESET_RELEASE_RESOURCES_BIT not set. After the implicit reset, commandBuffer is moved to the recording state.

Once recording starts, an application records a sequence of commands (vkCmd*) to set state in the command buffer, draw, dispatch, and other commands.

Several commands can also be recorded indirectly from VkBuffer content, see Device-Generated Commands.

To complete recording of a command buffer, call:

VkResult vkEndCommandBuffer(
    VkCommandBuffer                             commandBuffer);
  • commandBuffer is the command buffer to complete recording.

If there was an error during recording, the application will be notified by an unsuccessful return code returned by vkEndCommandBuffer. If the application wishes to further use the command buffer, the command buffer must be reset. The command buffer must have been in the recording state, and is moved to the executable state.

Valid Usage
  • commandBuffer must be in the recording state.

  • If commandBuffer is a primary command buffer, there must not be an active render pass instance

  • All queries made active during the recording of commandBuffer must have been made inactive

  • If commandBuffer is a secondary command buffer, there must not be an outstanding vkCmdBeginDebugUtilsLabelEXT command recorded to commandBuffer that has not previously been ended by a call to vkCmdEndDebugUtilsLabelEXT.

  • If commandBuffer is a secondary command buffer, there must not be an outstanding vkCmdDebugMarkerBeginEXT command recorded to commandBuffer that has not previously been ended by a call to vkCmdDebugMarkerEndEXT.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

When a command buffer is in the executable state, it can be submitted to a queue for execution.

5.5. Command Buffer Submission

To submit command buffers to a queue, call:

VkResult vkQueueSubmit(
    VkQueue                                     queue,
    uint32_t                                    submitCount,
    const VkSubmitInfo*                         pSubmits,
    VkFence                                     fence);
  • queue is the queue that the command buffers will be submitted to.

  • submitCount is the number of elements in the pSubmits array.

  • pSubmits is a pointer to an array of VkSubmitInfo structures, each specifying a command buffer submission batch.

  • fence is an optional handle to a fence to be signaled once all submitted command buffers have completed execution. If fence is not VK_NULL_HANDLE, it defines a fence signal operation.

Note

Submission can be a high overhead operation, and applications should attempt to batch work together into as few calls to vkQueueSubmit as possible.

vkQueueSubmit is a queue submission command, with each batch defined by an element of pSubmits as an instance of the VkSubmitInfo structure. Batches begin execution in the order they appear in pSubmits, but may complete out of order.

Fence and semaphore operations submitted with vkQueueSubmit have additional ordering constraints compared to other submission commands, with dependencies involving previous and subsequent queue operations. Information about these additional constraints can be found in the semaphore and fence sections of the synchronization chapter.

Details on the interaction of pWaitDstStageMask with synchronization are described in the semaphore wait operation section of the synchronization chapter.

The order that batches appear in pSubmits is used to determine submission order, and thus all the implicit ordering guarantees that respect it. Other than these implicit ordering guarantees and any explicit synchronization primitives, these batches may overlap or otherwise execute out of order.

If any command buffer submitted to this queue is in the executable state, it is moved to the pending state. Once execution of all submissions of a command buffer complete, it moves from the pending state, back to the executable state. If a command buffer was recorded with the VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT flag, it instead moves back to the invalid state.

If vkQueueSubmit fails, it may return VK_ERROR_OUT_OF_HOST_MEMORY or VK_ERROR_OUT_OF_DEVICE_MEMORY. If it does, the implementation must ensure that the state and contents of any resources or synchronization primitives referenced by the submitted command buffers and any semaphores referenced by pSubmits is unaffected by the call or its failure. If vkQueueSubmit fails in such a way that the implementation is unable to make that guarantee, the implementation must return VK_ERROR_DEVICE_LOST. See Lost Device.

Valid Usage
  • If fence is not VK_NULL_HANDLE, fence must be unsignaled

  • If fence is not VK_NULL_HANDLE, fence must not be associated with any other queue command that has not yet completed execution on that queue

  • Any calls to vkCmdSetEvent, vkCmdResetEvent or vkCmdWaitEvents that have been recorded into any of the command buffer elements of the pCommandBuffers member of any element of pSubmits, must not reference any VkEvent that is referenced by any of those commands in a command buffer that has been submitted to another queue and is still in the pending state.

  • Any stage flag included in any element of the pWaitDstStageMask member of any element of pSubmits must be a pipeline stage supported by one of the capabilities of queue, as specified in the table of supported pipeline stages.

  • Each element of the pSignalSemaphores member of any element of pSubmits must be unsignaled when the semaphore signal operation it defines is executed on the device

  • When a semaphore unsignal operation defined by any element of the pWaitSemaphores member of any element of pSubmits executes on queue, no other queue must be waiting on the same semaphore.

  • All elements of the pWaitSemaphores member of all elements of pSubmits must be semaphores that are signaled, or have semaphore signal operations previously submitted for execution.

  • Each element of the pCommandBuffers member of each element of pSubmits must be in the pending or executable state.

  • If any element of the pCommandBuffers member of any element of pSubmits was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT, it must not be in the pending state.

  • Any secondary command buffers recorded into any element of the pCommandBuffers member of any element of pSubmits must be in the pending or executable state.

  • If any secondary command buffers recorded into any element of the pCommandBuffers member of any element of pSubmits was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT, it must not be in the pending state.

  • Each element of the pCommandBuffers member of each element of pSubmits must have been allocated from a VkCommandPool that was created for the same queue family queue belongs to.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

  • If submitCount is not 0, pSubmits must be a valid pointer to an array of submitCount valid VkSubmitInfo structures

  • If fence is not VK_NULL_HANDLE, fence must be a valid VkFence handle

  • Both of fence, and queue that are valid handles must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to queue must be externally synchronized

  • Host access to pSubmits[].pWaitSemaphores[] must be externally synchronized

  • Host access to pSubmits[].pSignalSemaphores[] must be externally synchronized

  • Host access to fence must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

The VkSubmitInfo structure is defined as:

typedef struct VkSubmitInfo {
    VkStructureType                sType;
    const void*                    pNext;
    uint32_t                       waitSemaphoreCount;
    const VkSemaphore*             pWaitSemaphores;
    const VkPipelineStageFlags*    pWaitDstStageMask;
    uint32_t                       commandBufferCount;
    const VkCommandBuffer*         pCommandBuffers;
    uint32_t                       signalSemaphoreCount;
    const VkSemaphore*             pSignalSemaphores;
} VkSubmitInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • waitSemaphoreCount is the number of semaphores upon which to wait before executing the command buffers for the batch.

  • pWaitSemaphores is a pointer to an array of semaphores upon which to wait before the command buffers for this batch begin execution. If semaphores to wait on are provided, they define a semaphore wait operation.

  • pWaitDstStageMask is a pointer to an array of pipeline stages at which each corresponding semaphore wait will occur.

  • commandBufferCount is the number of command buffers to execute in the batch.

  • pCommandBuffers is a pointer to an array of command buffers to execute in the batch.

  • signalSemaphoreCount is the number of semaphores to be signaled once the commands specified in pCommandBuffers have completed execution.

  • pSignalSemaphores is a pointer to an array of semaphores which will be signaled when the command buffers for this batch have completed execution. If semaphores to be signaled are provided, they define a semaphore signal operation.

The order that command buffers appear in pCommandBuffers is used to determine submission order, and thus all the implicit ordering guarantees that respect it. Other than these implicit ordering guarantees and any explicit synchronization primitives, these command buffers may overlap or otherwise execute out of order.

Valid Usage
  • Each element of pCommandBuffers must not have been allocated with VK_COMMAND_BUFFER_LEVEL_SECONDARY

  • If the geometry shaders feature is not enabled, each element of pWaitDstStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, each element of pWaitDstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • Each element of pWaitDstStageMask must not include VK_PIPELINE_STAGE_HOST_BIT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SUBMIT_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkD3D12FenceSubmitInfoKHR, VkDeviceGroupSubmitInfo, VkProtectedSubmitInfo, VkWin32KeyedMutexAcquireReleaseInfoKHR, or VkWin32KeyedMutexAcquireReleaseInfoNV

  • Each sType member in the pNext chain must be unique

  • If waitSemaphoreCount is not 0, pWaitSemaphores must be a valid pointer to an array of waitSemaphoreCount valid VkSemaphore handles

  • If waitSemaphoreCount is not 0, pWaitDstStageMask must be a valid pointer to an array of waitSemaphoreCount valid combinations of VkPipelineStageFlagBits values

  • Each element of pWaitDstStageMask must not be 0

  • If commandBufferCount is not 0, pCommandBuffers must be a valid pointer to an array of commandBufferCount valid VkCommandBuffer handles

  • If signalSemaphoreCount is not 0, pSignalSemaphores must be a valid pointer to an array of signalSemaphoreCount valid VkSemaphore handles

  • Each of the elements of pCommandBuffers, the elements of pSignalSemaphores, and the elements of pWaitSemaphores that are valid handles must have been created, allocated, or retrieved from the same VkDevice

To specify the values to use when waiting for and signaling semaphores whose current payload refers to a Direct3D 12 fence, add the VkD3D12FenceSubmitInfoKHR structure to the pNext chain of the VkSubmitInfo structure. The VkD3D12FenceSubmitInfoKHR structure is defined as:

typedef struct VkD3D12FenceSubmitInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           waitSemaphoreValuesCount;
    const uint64_t*    pWaitSemaphoreValues;
    uint32_t           signalSemaphoreValuesCount;
    const uint64_t*    pSignalSemaphoreValues;
} VkD3D12FenceSubmitInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • waitSemaphoreValuesCount is the number of semaphore wait values specified in pWaitSemaphoreValues.

  • pWaitSemaphoreValues is an array of length waitSemaphoreValuesCount containing values for the corresponding semaphores in VkSubmitInfo::pWaitSemaphores to wait for.

  • signalSemaphoreValuesCount is the number of semaphore signal values specified in pSignalSemaphoreValues.

  • pSignalSemaphoreValues is an array of length signalSemaphoreValuesCount containing values for the corresponding semaphores in VkSubmitInfo::pSignalSemaphores to set when signaled.

If the semaphore in VkSubmitInfo::pWaitSemaphores or VkSubmitInfo::pSignalSemaphores corresponding to an entry in pWaitSemaphoreValues or pSignalSemaphoreValues respectively does not currently have a payload referring to a Direct3D 12 fence, the implementation must ignore the value in the pWaitSemaphoreValues or pSignalSemaphoreValues entry.

Valid Usage
  • waitSemaphoreValuesCount must be the same value as VkSubmitInfo::waitSemaphoreCount, where VkSubmitInfo is in the pNext chain of this VkD3D12FenceSubmitInfoKHR structure.

  • signalSemaphoreValuesCount must be the same value as VkSubmitInfo::signalSemaphoreCount, where VkSubmitInfo is in the pNext chain of this VkD3D12FenceSubmitInfoKHR structure.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_D3D12_FENCE_SUBMIT_INFO_KHR

  • If waitSemaphoreValuesCount is not 0, and pWaitSemaphoreValues is not NULL, pWaitSemaphoreValues must be a valid pointer to an array of waitSemaphoreValuesCount uint64_t values

  • If signalSemaphoreValuesCount is not 0, and pSignalSemaphoreValues is not NULL, pSignalSemaphoreValues must be a valid pointer to an array of signalSemaphoreValuesCount uint64_t values

When submitting work that operates on memory imported from a Direct3D 11 resource to a queue, the keyed mutex mechanism may be used in addition to Vulkan semaphores to synchronize the work. Keyed mutexes are a property of a properly created shareable Direct3D 11 resource. They can only be used if the imported resource was created with the D3D11_RESOURCE_MISC_SHARED_KEYEDMUTEX flag.

To acquire keyed mutexes before submitted work and/or release them after, add a VkWin32KeyedMutexAcquireReleaseInfoKHR structure to the pNext chain of the VkSubmitInfo structure.

The VkWin32KeyedMutexAcquireReleaseInfoKHR structure is defined as:

typedef struct VkWin32KeyedMutexAcquireReleaseInfoKHR {
    VkStructureType          sType;
    const void*              pNext;
    uint32_t                 acquireCount;
    const VkDeviceMemory*    pAcquireSyncs;
    const uint64_t*          pAcquireKeys;
    const uint32_t*          pAcquireTimeouts;
    uint32_t                 releaseCount;
    const VkDeviceMemory*    pReleaseSyncs;
    const uint64_t*          pReleaseKeys;
} VkWin32KeyedMutexAcquireReleaseInfoKHR;
  • acquireCount is the number of entries in the pAcquireSyncs, pAcquireKeys, and pAcquireTimeoutMilliseconds arrays.

  • pAcquireSyncs is a pointer to an array of VkDeviceMemory objects which were imported from Direct3D 11 resources.

  • pAcquireKeys is a pointer to an array of mutex key values to wait for prior to beginning the submitted work. Entries refer to the keyed mutex associated with the corresponding entries in pAcquireSyncs.

  • pAcquireTimeoutMilliseconds is an array of timeout values, in millisecond units, for each acquire specified in pAcquireKeys.

  • releaseCount is the number of entries in the pReleaseSyncs and pReleaseKeys arrays.

  • pReleaseSyncs is a pointer to an array of VkDeviceMemory objects which were imported from Direct3D 11 resources.

  • pReleaseKeys is a pointer to an array of mutex key values to set when the submitted work has completed. Entries refer to the keyed mutex associated with the corresponding entries in pReleaseSyncs.

Valid Usage
  • Each member of pAcquireSyncs and pReleaseSyncs must be a device memory object imported by setting VkImportMemoryWin32HandleInfoKHR::handleType to VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_KHR

  • If acquireCount is not 0, pAcquireSyncs must be a valid pointer to an array of acquireCount valid VkDeviceMemory handles

  • If acquireCount is not 0, pAcquireKeys must be a valid pointer to an array of acquireCount uint64_t values

  • If acquireCount is not 0, pAcquireTimeouts must be a valid pointer to an array of acquireCount uint32_t values

  • If releaseCount is not 0, pReleaseSyncs must be a valid pointer to an array of releaseCount valid VkDeviceMemory handles

  • If releaseCount is not 0, pReleaseKeys must be a valid pointer to an array of releaseCount uint64_t values

  • Both of the elements of pAcquireSyncs, and the elements of pReleaseSyncs that are valid handles must have been created, allocated, or retrieved from the same VkDevice

When submitting work that operates on memory imported from a Direct3D 11 resource to a queue, the keyed mutex mechanism may be used in addition to Vulkan semaphores to synchronize the work. Keyed mutexes are a property of a properly created shareable Direct3D 11 resource. They can only be used if the imported resource was created with the D3D11_RESOURCE_MISC_SHARED_KEYEDMUTEX flag.

To acquire keyed mutexes before submitted work and/or release them after, add a VkWin32KeyedMutexAcquireReleaseInfoNV structure to the pNext chain of the VkSubmitInfo structure.

The VkWin32KeyedMutexAcquireReleaseInfoNV structure is defined as:

typedef struct VkWin32KeyedMutexAcquireReleaseInfoNV {
    VkStructureType          sType;
    const void*              pNext;
    uint32_t                 acquireCount;
    const VkDeviceMemory*    pAcquireSyncs;
    const uint64_t*          pAcquireKeys;
    const uint32_t*          pAcquireTimeoutMilliseconds;
    uint32_t                 releaseCount;
    const VkDeviceMemory*    pReleaseSyncs;
    const uint64_t*          pReleaseKeys;
} VkWin32KeyedMutexAcquireReleaseInfoNV;
  • acquireCount is the number of entries in the pAcquireSyncs, pAcquireKeys, and pAcquireTimeoutMilliseconds arrays.

  • pAcquireSyncs is a pointer to an array of VkDeviceMemory objects which were imported from Direct3D 11 resources.

  • pAcquireKeys is a pointer to an array of mutex key values to wait for prior to beginning the submitted work. Entries refer to the keyed mutex associated with the corresponding entries in pAcquireSyncs.

  • pAcquireTimeoutMilliseconds is an array of timeout values, in millisecond units, for each acquire specified in pAcquireKeys.

  • releaseCount is the number of entries in the pReleaseSyncs and pReleaseKeys arrays.

  • pReleaseSyncs is a pointer to an array of VkDeviceMemory objects which were imported from Direct3D 11 resources.

  • pReleaseKeys is a pointer to an array of mutex key values to set when the submitted work has completed. Entries refer to the keyed mutex associated with the corresponding entries in pReleaseSyncs.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_NV

  • If acquireCount is not 0, pAcquireSyncs must be a valid pointer to an array of acquireCount valid VkDeviceMemory handles

  • If acquireCount is not 0, pAcquireKeys must be a valid pointer to an array of acquireCount uint64_t values

  • If acquireCount is not 0, pAcquireTimeoutMilliseconds must be a valid pointer to an array of acquireCount uint32_t values

  • If releaseCount is not 0, pReleaseSyncs must be a valid pointer to an array of releaseCount valid VkDeviceMemory handles

  • If releaseCount is not 0, pReleaseKeys must be a valid pointer to an array of releaseCount uint64_t values

  • Both of the elements of pAcquireSyncs, and the elements of pReleaseSyncs that are valid handles must have been created, allocated, or retrieved from the same VkDevice

If the pNext chain of VkSubmitInfo includes a VkProtectedSubmitInfo structure, then the structure indicates whether the batch is protected. The VkProtectedSubmitInfo structure is defined as:

typedef struct VkProtectedSubmitInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkBool32           protectedSubmit;
} VkProtectedSubmitInfo;
  • protectedSubmit specifies whether the batch is protected. If protectedSubmit is VK_TRUE, the batch is protected. If protectedSubmit is VK_FALSE, the batch is unprotected. If the VkSubmitInfo::pNext chain does not contain this structure, the batch is unprotected.

Valid Usage
  • If the protected memory feature is not enabled, protectedSubmit must not be VK_TRUE.

  • If protectedSubmit is VK_TRUE, then each element of the pCommandBuffers array must be a protected command buffer.

  • If protectedSubmit is VK_FALSE, then each element of the pCommandBuffers array must be an unprotected command buffer.

  • If the VkSubmitInfo::pNext chain does not include a VkProtectedSubmitInfo structure, then each element of the command buffer of the pCommandBuffers array must be an unprotected command buffer.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PROTECTED_SUBMIT_INFO

If the pNext chain of VkSubmitInfo includes a VkDeviceGroupSubmitInfo structure, then that structure includes device indices and masks specifying which physical devices execute semaphore operations and command buffers.

The VkDeviceGroupSubmitInfo structure is defined as:

typedef struct VkDeviceGroupSubmitInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           waitSemaphoreCount;
    const uint32_t*    pWaitSemaphoreDeviceIndices;
    uint32_t           commandBufferCount;
    const uint32_t*    pCommandBufferDeviceMasks;
    uint32_t           signalSemaphoreCount;
    const uint32_t*    pSignalSemaphoreDeviceIndices;
} VkDeviceGroupSubmitInfo;

or the equivalent

typedef VkDeviceGroupSubmitInfo VkDeviceGroupSubmitInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • waitSemaphoreCount is the number of elements in the pWaitSemaphoreDeviceIndices array.

  • pWaitSemaphoreDeviceIndices is an array of device indices indicating which physical device executes the semaphore wait operation in the corresponding element of VkSubmitInfo::pWaitSemaphores.

  • commandBufferCount is the number of elements in the pCommandBufferDeviceMasks array.

  • pCommandBufferDeviceMasks is an array of device masks indicating which physical devices execute the command buffer in the corresponding element of VkSubmitInfo::pCommandBuffers. A physical device executes the command buffer if the corresponding bit is set in the mask.

  • signalSemaphoreCount is the number of elements in the pSignalSemaphoreDeviceIndices array.

  • pSignalSemaphoreDeviceIndices is an array of device indices indicating which physical device executes the semaphore signal operation in the corresponding element of VkSubmitInfo::pSignalSemaphores.

If this structure is not present, semaphore operations and command buffers execute on device index zero.

Valid Usage
  • waitSemaphoreCount must equal VkSubmitInfo::waitSemaphoreCount

  • commandBufferCount must equal VkSubmitInfo::commandBufferCount

  • signalSemaphoreCount must equal VkSubmitInfo::signalSemaphoreCount

  • All elements of pWaitSemaphoreDeviceIndices and pSignalSemaphoreDeviceIndices must be valid device indices

  • All elements of pCommandBufferDeviceMasks must be valid device masks

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO

  • If waitSemaphoreCount is not 0, pWaitSemaphoreDeviceIndices must be a valid pointer to an array of waitSemaphoreCount uint32_t values

  • If commandBufferCount is not 0, pCommandBufferDeviceMasks must be a valid pointer to an array of commandBufferCount uint32_t values

  • If signalSemaphoreCount is not 0, pSignalSemaphoreDeviceIndices must be a valid pointer to an array of signalSemaphoreCount uint32_t values

5.6. Queue Forward Progress

The application must ensure that command buffer submissions will be able to complete without any subsequent operations by the application on any queue. After any call to vkQueueSubmit, for every queued wait on a semaphore there must be a prior signal of that semaphore that will not be consumed by a different wait on the semaphore.

Command buffers in the submission can include vkCmdWaitEvents commands that wait on events that will not be signaled by earlier commands in the queue. Such events must be signaled by the application using vkSetEvent, and the vkCmdWaitEvents commands that wait upon them must not be inside a render pass instance. Implementations may have limits on how long the command buffer will wait, in order to avoid interfering with progress of other clients of the device. If the event is not signaled within these limits, results are undefined and may include device loss.

5.7. Secondary Command Buffer Execution

A secondary command buffer must not be directly submitted to a queue. Instead, secondary command buffers are recorded to execute as part of a primary command buffer with the command:

void vkCmdExecuteCommands(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    commandBufferCount,
    const VkCommandBuffer*                      pCommandBuffers);
  • commandBuffer is a handle to a primary command buffer that the secondary command buffers are executed in.

  • commandBufferCount is the length of the pCommandBuffers array.

  • pCommandBuffers is an array of secondary command buffer handles, which are recorded to execute in the primary command buffer in the order they are listed in the array.

If any element of pCommandBuffers was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flag, and it was recorded into any other primary command buffer which is currently in the executable or recording state, that primary command buffer becomes invalid.

Valid Usage
  • commandBuffer must have been allocated with a level of VK_COMMAND_BUFFER_LEVEL_PRIMARY

  • Each element of pCommandBuffers must have been allocated with a level of VK_COMMAND_BUFFER_LEVEL_SECONDARY

  • Each element of pCommandBuffers must be in the pending or executable state.

  • If any element of pCommandBuffers was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flag, and it was recorded into any other primary command buffer, that primary command buffer must not be in the pending state

  • If any element of pCommandBuffers was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flag, it must not be in the pending state.

  • If any element of pCommandBuffers was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flag, it must not have already been recorded to commandBuffer.

  • If any element of pCommandBuffers was not recorded with the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flag, it must not appear more than once in pCommandBuffers.

  • Each element of pCommandBuffers must have been allocated from a VkCommandPool that was created for the same queue family as the VkCommandPool from which commandBuffer was allocated

  • If vkCmdExecuteCommands is being called within a render pass instance, that render pass instance must have been begun with the contents parameter of vkCmdBeginRenderPass set to VK_SUBPASS_CONTENTS_SECONDARY_COMMAND_BUFFERS

  • If vkCmdExecuteCommands is being called within a render pass instance, each element of pCommandBuffers must have been recorded with the VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT

  • If vkCmdExecuteCommands is being called within a render pass instance, each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceInfo::subpass set to the index of the subpass which the given command buffer will be executed in

  • If vkCmdExecuteCommands is being called within a render pass instance, the render passes specified in the pname::pBeginInfo::pInheritanceInfo::renderPass members of the vkBeginCommandBuffer commands used to begin recording each element of pCommandBuffers must be compatible with the current render pass.

  • If vkCmdExecuteCommands is being called within a render pass instance, and any element of pCommandBuffers was recorded with VkCommandBufferInheritanceInfo::framebuffer not equal to VK_NULL_HANDLE, that VkFramebuffer must match the VkFramebuffer used in the current render pass instance

  • If vkCmdExecuteCommands is not being called within a render pass instance, each element of pCommandBuffers must not have been recorded with the VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT

  • If the inherited queries feature is not enabled, commandBuffer must not have any queries active

  • If commandBuffer has a VK_QUERY_TYPE_OCCLUSION query active, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceInfo::occlusionQueryEnable set to VK_TRUE

  • If commandBuffer has a VK_QUERY_TYPE_OCCLUSION query active, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceInfo::queryFlags having all bits set that are set for the query

  • If commandBuffer has a VK_QUERY_TYPE_PIPELINE_STATISTICS query active, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceInfo::pipelineStatistics having all bits set that are set in the VkQueryPool the query uses

  • Each element of pCommandBuffers must not begin any query types that are active in commandBuffer

  • If commandBuffer is a protected command buffer, then each element of pCommandBuffers must be a protected command buffer.

  • If commandBuffer is an unprotected command buffer, then each element of pCommandBuffers must be an unprotected command buffer.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pCommandBuffers must be a valid pointer to an array of commandBufferCount valid VkCommandBuffer handles

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • commandBuffer must be a primary VkCommandBuffer

  • commandBufferCount must be greater than 0

  • Both of commandBuffer, and the elements of pCommandBuffers must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary

Both

Transfer
Graphics
Compute

5.8. Command Buffer Device Mask

Each command buffer has a piece of state storing the current device mask of the command buffer. This mask controls which physical devices within the logical device all subsequent commands will execute on, including state-setting commands, action commands, and synchronization commands.

Scissor and viewport state can be set to different values on each physical device (only when set as dynamic state), and each physical device will render using its local copy of the state. Other state is shared between physical devices, such that all physical devices use the most recently set values for the state. However, when recording an action command that uses a piece of state, the most recent command that set that state must have included all physical devices that execute the action command in its current device mask.

The command buffer’s device mask is orthogonal to the pCommandBufferDeviceMasks member of VkDeviceGroupSubmitInfo. Commands only execute on a physical device if the device index is set in both device masks.

If the pNext chain of VkCommandBufferBeginInfo includes a VkDeviceGroupCommandBufferBeginInfo structure, then that structure includes an initial device mask for the command buffer.

The VkDeviceGroupCommandBufferBeginInfo structure is defined as:

typedef struct VkDeviceGroupCommandBufferBeginInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceMask;
} VkDeviceGroupCommandBufferBeginInfo;

or the equivalent

typedef VkDeviceGroupCommandBufferBeginInfo VkDeviceGroupCommandBufferBeginInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • deviceMask is the initial value of the command buffer’s device mask.

The initial device mask also acts as an upper bound on the set of devices that can ever be in the device mask in the command buffer.

If this structure is not present, the initial value of a command buffer’s device mask is set to include all physical devices in the logical device when the command buffer begins recording.

Valid Usage
  • deviceMask must be a valid device mask value

  • deviceMask must not be zero

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO

To update the current device mask of a command buffer, call:

void vkCmdSetDeviceMask(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    deviceMask);

or the equivalent command

void vkCmdSetDeviceMaskKHR(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    deviceMask);
  • commandBuffer is command buffer whose current device mask is modified.

  • deviceMask is the new value of the current device mask.

deviceMask is used to filter out subsequent commands from executing on all physical devices whose bit indices are not set in the mask.

Valid Usage
  • deviceMask must be a valid device mask value

  • deviceMask must not be zero

  • deviceMask must not include any set bits that were not in the VkDeviceGroupCommandBufferBeginInfo::deviceMask value when the command buffer began recording.

  • If vkCmdSetDeviceMask is called inside a render pass instance, deviceMask must not include any set bits that were not in the VkDeviceGroupRenderPassBeginInfo::deviceMask value when the render pass instance began recording.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, compute, or transfer operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute
Transfer

6. Synchronization and Cache Control

Synchronization of access to resources is primarily the responsibility of the application in Vulkan. The order of execution of commands with respect to the host and other commands on the device has few implicit guarantees, and needs to be explicitly specified. Memory caches and other optimizations are also explicitly managed, requiring that the flow of data through the system is largely under application control.

Whilst some implicit guarantees exist between commands, five explicit synchronization mechanisms are exposed by Vulkan:

Fences

Fences can be used to communicate to the host that execution of some task on the device has completed.

Semaphores

Semaphores can be used to control resource access across multiple queues.

Events

Events provide a fine-grained synchronization primitive which can be signaled either within a command buffer or by the host, and can be waited upon within a command buffer or queried on the host.

Pipeline Barriers

Pipeline barriers also provide synchronization control within a command buffer, but at a single point, rather than with separate signal and wait operations.

Render Passes

Render passes provide a useful synchronization framework for most rendering tasks, built upon the concepts in this chapter. Many cases that would otherwise need an application to use other synchronization primitives can be expressed more efficiently as part of a render pass.

6.1. Execution and Memory Dependencies

An operation is an arbitrary amount of work to be executed on the host, a device, or an external entity such as a presentation engine. Synchronization commands introduce explicit execution dependencies, and memory dependencies between two sets of operations defined by the command’s two synchronization scopes.

The synchronization scopes define which other operations a synchronization command is able to create execution dependencies with. Any type of operation that is not in a synchronization command’s synchronization scopes will not be included in the resulting dependency. For example, for many synchronization commands, the synchronization scopes can be limited to just operations executing in specific pipeline stages, which allows other pipeline stages to be excluded from a dependency. Other scoping options are possible, depending on the particular command.

An execution dependency is a guarantee that for two sets of operations, the first set must happen-before the second set. If an operation happens-before another operation, then the first operation must complete before the second operation is initiated. More precisely:

  • Let A and B be separate sets of operations.

  • Let S be a synchronization command.

  • Let AS and BS be the synchronization scopes of S.

  • Let A' be the intersection of sets A and AS.

  • Let B' be the intersection of sets B and BS.

  • Submitting A, S and B for execution, in that order, will result in execution dependency E between A' and B'.

  • Execution dependency E guarantees that A' happens-before B'.

An execution dependency chain is a sequence of execution dependencies that form a happens-before relation between the first dependency’s A' and the final dependency’s B'. For each consecutive pair of execution dependencies, a chain exists if the intersection of BS in the first dependency and AS in the second dependency is not an empty set. The formation of a single execution dependency from an execution dependency chain can be described by substituting the following in the description of execution dependencies:

  • Let S be a set of synchronization commands that generate an execution dependency chain.

  • Let AS be the first synchronization scope of the first command in S.

  • Let BS be the second synchronization scope of the last command in S.

Note

An execution dependency is inherently also multiple execution dependencies - a dependency exists between each subset of A' and each subset of B', and the same is true for execution dependency chains. For example, a synchronization command with multiple pipeline stages in its stage masks effectively generates one dependency between each source stage and each destination stage. This can be useful to think about when considering how execution chains are formed if they do not involve all parts of a synchronization command’s dependency. Similarly, any set of adjacent dependencies in an execution dependency chain can be considered an execution dependency chain in its own right.

Execution dependencies alone are not sufficient to guarantee that values resulting from writes in one set of operations can be read from another set of operations.

Two additional types of operation are used to control memory access. Availability operations cause the values generated by specified memory write accesses to become available for future access. Any available value remains available until a subsequent write to the same memory location occurs (whether it is made available or not) or the memory is freed. Visibility operations cause any available values to become visible to specified memory accesses.

A memory dependency is an execution dependency which includes availability and visibility operations such that:

  • The first set of operations happens-before the availability operation.

  • The availability operation happens-before the visibility operation.

  • The visibility operation happens-before the second set of operations.

Once written values are made visible to a particular type of memory access, they can be read or written by that type of memory access. Most synchronization commands in Vulkan define a memory dependency.

The specific memory accesses that are made available and visible are defined by the access scopes of a memory dependency. Any type of access that is in a memory dependency’s first access scope and occurs in A' is made available. Any type of access that is in a memory dependency’s second access scope and occurs in B' has any available writes made visible to it. Any type of operation that is not in a synchronization command’s access scopes will not be included in the resulting dependency.

A memory dependency enforces availability and visibility of memory accesses and execution order between two sets of operations. Adding to the description of execution dependency chains:

  • Let a be the set of memory accesses performed by A'.

  • Let b be the set of memory accesses performed by B'.

  • Let aS be the first access scope of the first command in S.

  • Let bS be the second access scope of the last command in S.

  • Let a' be the intersection of sets a and aS.

  • Let b' be the intersection of sets b and bS.

  • Submitting A, S and B for execution, in that order, will result in a memory dependency m between A' and B'.

  • Memory dependency m guarantees that:

    • Memory writes in a' are made available.

    • Available memory writes, including those from a', are made visible to b'.

Note

Execution and memory dependencies are used to solve data hazards, i.e. to ensure that read and write operations occur in a well-defined order. Write-after-read hazards can be solved with just an execution dependency, but read-after-write and write-after-write hazards need appropriate memory dependencies to be included between them. If an application does not include dependencies to solve these hazards, the results and execution orders of memory accesses are undefined.

6.1.1. Image Layout Transitions

Image subresources can be transitioned from one layout to another as part of a memory dependency (e.g. by using an image memory barrier). When a layout transition is specified in a memory dependency, it happens-after the availability operations in the memory dependency, and happens-before the visibility operations. Image layout transitions may perform read and write accesses on all memory bound to the image subresource range, so applications must ensure that all memory writes have been made available before a layout transition is executed. Available memory is automatically made visible to a layout transition, and writes performed by a layout transition are automatically made available.

Layout transitions always apply to a particular image subresource range, and specify both an old layout and new layout. If the old layout does not match the new layout, a transition occurs. The old layout must match the current layout of the image subresource range, with one exception. The old layout can always be specified as VK_IMAGE_LAYOUT_UNDEFINED, though doing so invalidates the contents of the image subresource range.

As image layout transitions may perform read and write accesses on the memory bound to the image, if the image subresource affected by the layout transition is bound to peer memory for any device in the current device mask then the memory heap the bound memory comes from must support the VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT and VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT capabilities as returned by vkGetDeviceGroupPeerMemoryFeatures.

Note

Setting the old layout to VK_IMAGE_LAYOUT_UNDEFINED implies that the contents of the image subresource need not be preserved. Implementations may use this information to avoid performing expensive data transition operations.

Note

Applications must ensure that layout transitions happen-after all operations accessing the image with the old layout, and happen-before any operations that will access the image with the new layout. Layout transitions are potentially read/write operations, so not defining appropriate memory dependencies to guarantee this will result in a data race.

Image layout transitions interact with memory aliasing.

6.1.2. Pipeline Stages

The work performed by an action or synchronization command consists of multiple operations, which are performed as a sequence of logically independent steps known as pipeline stages. The exact pipeline stages executed depend on the particular command that is used, and current command buffer state when the command was recorded. Drawing commands, dispatching commands, copy commands, clear commands, and synchronization commands all execute in different sets of pipeline stages. Synchronization commands don’t execute in a defined pipeline, but do execute VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT and VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT.

Note

Operations performed by synchronization commands (e.g. availability and visibility operations) are not executed by a defined pipeline stage. However other commands can still synchronize with them via the VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT and VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT pipeline stages.

Execution of operations across pipeline stages must adhere to implicit ordering guarantees, particularly including pipeline stage order. Otherwise, execution across pipeline stages may overlap or execute out of order with regards to other stages, unless otherwise enforced by an execution dependency.

Several of the synchronization commands include pipeline stage parameters, restricting the synchronization scopes for that command to just those stages. This allows fine grained control over the exact execution dependencies and accesses performed by action commands. Implementations should use these pipeline stages to avoid unnecessary stalls or cache flushing.

Bits which can be set, specifying pipeline stages, are:

typedef enum VkPipelineStageFlagBits {
    VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT = 0x00000001,
    VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT = 0x00000002,
    VK_PIPELINE_STAGE_VERTEX_INPUT_BIT = 0x00000004,
    VK_PIPELINE_STAGE_VERTEX_SHADER_BIT = 0x00000008,
    VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT = 0x00000010,
    VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT = 0x00000020,
    VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT = 0x00000040,
    VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT = 0x00000080,
    VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT = 0x00000100,
    VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT = 0x00000200,
    VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT = 0x00000400,
    VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT = 0x00000800,
    VK_PIPELINE_STAGE_TRANSFER_BIT = 0x00001000,
    VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT = 0x00002000,
    VK_PIPELINE_STAGE_HOST_BIT = 0x00004000,
    VK_PIPELINE_STAGE_ALL_GRAPHICS_BIT = 0x00008000,
    VK_PIPELINE_STAGE_ALL_COMMANDS_BIT = 0x00010000,
    VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX = 0x00020000,
} VkPipelineStageFlagBits;
  • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT specifies the stage of the pipeline where any commands are initially received by the queue.

  • VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX specifies the stage of the pipeline where device-side generation of commands via vkCmdProcessCommandsNVX is handled.

  • VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT specifies the stage of the pipeline where Draw/DispatchIndirect data structures are consumed. This stage also includes reading commands written by vkCmdProcessCommandsNVX.

  • VK_PIPELINE_STAGE_VERTEX_INPUT_BIT specifies the stage of the pipeline where vertex and index buffers are consumed.

  • VK_PIPELINE_STAGE_VERTEX_SHADER_BIT specifies the vertex shader stage.

  • VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT specifies the tessellation control shader stage.

  • VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT specifies the tessellation evaluation shader stage.

  • VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT specifies the geometry shader stage.

  • VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT specifies the fragment shader stage.

  • VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT specifies the stage of the pipeline where early fragment tests (depth and stencil tests before fragment shading) are performed. This stage also includes subpass load operations for framebuffer attachments with a depth/stencil format.

  • VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT specifies the stage of the pipeline where late fragment tests (depth and stencil tests after fragment shading) are performed. This stage also includes subpass store operations for framebuffer attachments with a depth/stencil format.

  • VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT specifies the stage of the pipeline after blending where the final color values are output from the pipeline. This stage also includes subpass load and store operations and multisample resolve operations for framebuffer attachments with a color format.

  • VK_PIPELINE_STAGE_TRANSFER_BIT specifies the execution of copy commands. This includes the operations resulting from all copy commands, clear commands (with the exception of vkCmdClearAttachments), and vkCmdCopyQueryPoolResults.

  • VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT specifies the execution of a compute shader.

  • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT specifies the final stage in the pipeline where operations generated by all commands complete execution.

  • VK_PIPELINE_STAGE_HOST_BIT specifies a pseudo-stage indicating execution on the host of reads/writes of device memory. This stage is not invoked by any commands recorded in a command buffer.

  • VK_PIPELINE_STAGE_ALL_GRAPHICS_BIT specifies the execution of all graphics pipeline stages, and is equivalent to the logical OR of:

    • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

    • VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

    • VK_PIPELINE_STAGE_VERTEX_INPUT_BIT

    • VK_PIPELINE_STAGE_VERTEX_SHADER_BIT

    • VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT

    • VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

    • VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

    • VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT

    • VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT

    • VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

    • VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

    • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

  • VK_PIPELINE_STAGE_ALL_COMMANDS_BIT is equivalent to the logical OR of every other pipeline stage flag that is supported on the queue it is used with.

Note

An execution dependency with only VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT in the destination stage mask will only prevent that stage from executing in subsequently submitted commands. As this stage does not perform any actual execution, this is not observable - in effect, it does not delay processing of subsequent commands. Similarly an execution dependency with only VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT in the source stage mask will effectively not wait for any prior commands to complete.

When defining a memory dependency, using only VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT or VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT would never make any accesses available and/or visible because these stages do not access memory.

VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT and VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT are useful for accomplishing layout transitions and queue ownership operations when the required execution dependency is satisfied by other means - for example, semaphore operations between queues.

typedef VkFlags VkPipelineStageFlags;

VkPipelineStageFlags is a bitmask type for setting a mask of zero or more VkPipelineStageFlagBits.

If a synchronization command includes a source stage mask, its first synchronization scope only includes execution of the pipeline stages specified in that mask, and its first access scope only includes memory access performed by pipeline stages specified in that mask. If a synchronization command includes a destination stage mask, its second synchronization scope only includes execution of the pipeline stages specified in that mask, and its second access scope only includes memory access performed by pipeline stages specified in that mask.

Note

Including a particular pipeline stage in the first synchronization scope of a command implicitly includes logically earlier pipeline stages in the synchronization scope. Similarly, the second synchronization scope includes logically later pipeline stages.

However, note that access scopes are not affected in this way - only the precise stages specified are considered part of each access scope.

Certain pipeline stages are only available on queues that support a particular set of operations. The following table lists, for each pipeline stage flag, which queue capability flag must be supported by the queue. When multiple flags are enumerated in the second column of the table, it means that the pipeline stage is supported on the queue if it supports any of the listed capability flags. For further details on queue capabilities see Physical Device Enumeration and Queues.

Table 3. Supported pipeline stage flags
Pipeline stage flag Required queue capability flag

VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

None required

VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT

VK_PIPELINE_STAGE_VERTEX_INPUT_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_VERTEX_SHADER_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

VK_QUEUE_COMPUTE_BIT

VK_PIPELINE_STAGE_TRANSFER_BIT

VK_QUEUE_GRAPHICS_BIT, VK_QUEUE_COMPUTE_BIT or VK_QUEUE_TRANSFER_BIT

VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

None required

VK_PIPELINE_STAGE_HOST_BIT

None required

VK_PIPELINE_STAGE_ALL_GRAPHICS_BIT

VK_QUEUE_GRAPHICS_BIT

VK_PIPELINE_STAGE_ALL_COMMANDS_BIT

None required

VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT

Pipeline stages that execute as a result of a command logically complete execution in a specific order, such that completion of a logically later pipeline stage must not happen-before completion of a logically earlier stage. This means that including any stage in the source stage mask for a particular synchronization command also implies that any logically earlier stages are included in AS for that command.

Similarly, initiation of a logically earlier pipeline stage must not happen-after initiation of a logically later pipeline stage. Including any given stage in the destination stage mask for a particular synchronization command also implies that any logically later stages are included in BS for that command.

Note

Implementations may not support synchronization at every pipeline stage for every synchronization operation. If a pipeline stage that an implementation does not support synchronization for appears in a source stage mask, it may substitute any logically later stage in its place for the first synchronization scope. If a pipeline stage that an implementation does not support synchronization for appears in a destination stage mask, it may substitute any logically earlier stage in its place for the second synchronization scope.

For example, if an implementation is unable to signal an event immediately after vertex shader execution is complete, it may instead signal the event after color attachment output has completed.

If an implementation makes such a substitution, it must not affect the semantics of execution or memory dependencies or image and buffer memory barriers.

The order and set of pipeline stages executed by a given command is determined by the command’s pipeline type, as described below:

For the graphics pipeline, the following stages occur in this order:

  • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

  • VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

  • VK_PIPELINE_STAGE_VERTEX_INPUT_BIT

  • VK_PIPELINE_STAGE_VERTEX_SHADER_BIT

  • VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT

  • VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT

  • VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT

  • VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

  • VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

  • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

For the compute pipeline, the following stages occur in this order:

  • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

  • VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

  • VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

  • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

For the transfer pipeline, the following stages occur in this order:

  • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

  • VK_PIPELINE_STAGE_TRANSFER_BIT

  • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

For host operations, only one pipeline stage occurs, so no order is guaranteed:

  • VK_PIPELINE_STAGE_HOST_BIT

For the command processing pipeline, the following stages occur in this order:

  • VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT

  • VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

  • VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT

6.1.3. Access Types

Memory in Vulkan can be accessed from within shader invocations and via some fixed-function stages of the pipeline. The access type is a function of the descriptor type used, or how a fixed-function stage accesses memory. Each access type corresponds to a bit flag in VkAccessFlagBits.

Some synchronization commands take sets of access types as parameters to define the access scopes of a memory dependency. If a synchronization command includes a source access mask, its first access scope only includes accesses via the access types specified in that mask. Similarly, if a synchronization command includes a destination access mask, its second access scope only includes accesses via the access types specified in that mask.

Access types that can be set in an access mask include:

typedef enum VkAccessFlagBits {
    VK_ACCESS_INDIRECT_COMMAND_READ_BIT = 0x00000001,
    VK_ACCESS_INDEX_READ_BIT = 0x00000002,
    VK_ACCESS_VERTEX_ATTRIBUTE_READ_BIT = 0x00000004,
    VK_ACCESS_UNIFORM_READ_BIT = 0x00000008,
    VK_ACCESS_INPUT_ATTACHMENT_READ_BIT = 0x00000010,
    VK_ACCESS_SHADER_READ_BIT = 0x00000020,
    VK_ACCESS_SHADER_WRITE_BIT = 0x00000040,
    VK_ACCESS_COLOR_ATTACHMENT_READ_BIT = 0x00000080,
    VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT = 0x00000100,
    VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT = 0x00000200,
    VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT = 0x00000400,
    VK_ACCESS_TRANSFER_READ_BIT = 0x00000800,
    VK_ACCESS_TRANSFER_WRITE_BIT = 0x00001000,
    VK_ACCESS_HOST_READ_BIT = 0x00002000,
    VK_ACCESS_HOST_WRITE_BIT = 0x00004000,
    VK_ACCESS_MEMORY_READ_BIT = 0x00008000,
    VK_ACCESS_MEMORY_WRITE_BIT = 0x00010000,
    VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX = 0x00020000,
    VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX = 0x00040000,
    VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT = 0x00080000,
} VkAccessFlagBits;
  • VK_ACCESS_INDIRECT_COMMAND_READ_BIT specifies read access to an indirect command structure read as part of an indirect drawing or dispatch command.

  • VK_ACCESS_INDEX_READ_BIT specifies read access to an index buffer as part of an indexed drawing command, bound by vkCmdBindIndexBuffer.

  • VK_ACCESS_VERTEX_ATTRIBUTE_READ_BIT specifies read access to a vertex buffer as part of a drawing command, bound by vkCmdBindVertexBuffers.

  • VK_ACCESS_UNIFORM_READ_BIT specifies read access to a uniform buffer.

  • VK_ACCESS_INPUT_ATTACHMENT_READ_BIT specifies read access to an input attachment within a render pass during fragment shading.

  • VK_ACCESS_SHADER_READ_BIT specifies read access to a storage buffer, uniform texel buffer, storage texel buffer, sampled image, or storage image.

  • VK_ACCESS_SHADER_WRITE_BIT specifies write access to a storage buffer, storage texel buffer, or storage image.

  • VK_ACCESS_COLOR_ATTACHMENT_READ_BIT specifies read access to a color attachment, such as via blending, logic operations, or via certain subpass load operations. It does not include advanced blend operations.

  • VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT is similar to VK_ACCESS_COLOR_ATTACHMENT_READ_BIT, but also includes advanced blend operations.

  • VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT specifies write access to a color or resolve attachment during a render pass or via certain subpass load and store operations.

  • VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT specifies read access to a depth/stencil attachment, via depth or stencil operations or via certain subpass load operations.

  • VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT specifies write access to a depth/stencil attachment, via depth or stencil operations or via certain subpass load and store operations.

  • VK_ACCESS_TRANSFER_READ_BIT specifies read access to an image or buffer in a copy operation.

  • VK_ACCESS_TRANSFER_WRITE_BIT specifies write access to an image or buffer in a clear or copy operation.

  • VK_ACCESS_HOST_READ_BIT specifies read access by a host operation. Accesses of this type are not performed through a resource, but directly on memory.

  • VK_ACCESS_HOST_WRITE_BIT specifies write access by a host operation. Accesses of this type are not performed through a resource, but directly on memory.

  • VK_ACCESS_MEMORY_READ_BIT specifies read access via non-specific entities. These entities include the Vulkan device and host, but may also include entities external to the Vulkan device or otherwise not part of the core Vulkan pipeline. When included in a destination access mask, makes all available writes visible to all future read accesses on entities known to the Vulkan device.

  • VK_ACCESS_MEMORY_WRITE_BIT specifies write access via non-specific entities. These entities include the Vulkan device and host, but may also include entities external to the Vulkan device or otherwise not part of the core Vulkan pipeline. When included in a source access mask, all writes that are performed by entities known to the Vulkan device are made available. When included in a destination access mask, makes all available writes visible to all future write accesses on entities known to the Vulkan device.

  • VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX specifies reads from VkBuffer inputs to vkCmdProcessCommandsNVX.

  • VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX specifies writes to the target command buffer in vkCmdProcessCommandsNVX.

Certain access types are only performed by a subset of pipeline stages. Any synchronization command that takes both stage masks and access masks uses both to define the access scopes - only the specified access types performed by the specified stages are included in the access scope. An application must not specify an access flag in a synchronization command if it does not include a pipeline stage in the corresponding stage mask that is able to perform accesses of that type. The following table lists, for each access flag, which pipeline stages can perform that type of access.

Table 4. Supported access types
Access flag Supported pipeline stages

VK_ACCESS_INDIRECT_COMMAND_READ_BIT

VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

VK_ACCESS_INDEX_READ_BIT

VK_PIPELINE_STAGE_VERTEX_INPUT_BIT

VK_ACCESS_VERTEX_ATTRIBUTE_READ_BIT

VK_PIPELINE_STAGE_VERTEX_INPUT_BIT

VK_ACCESS_UNIFORM_READ_BIT

VK_PIPELINE_STAGE_VERTEX_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT, VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, or VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

VK_ACCESS_INPUT_ATTACHMENT_READ_BIT

VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT

VK_ACCESS_SHADER_READ_BIT

VK_PIPELINE_STAGE_VERTEX_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT, VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, or VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

VK_ACCESS_SHADER_WRITE_BIT

VK_PIPELINE_STAGE_VERTEX_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT, VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, or VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

VK_ACCESS_COLOR_ATTACHMENT_READ_BIT

VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT

VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT

VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT

VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT, or VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT

VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT, or VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

VK_ACCESS_TRANSFER_READ_BIT

VK_PIPELINE_STAGE_TRANSFER_BIT

VK_ACCESS_TRANSFER_WRITE_BIT

VK_PIPELINE_STAGE_TRANSFER_BIT

VK_ACCESS_HOST_READ_BIT

VK_PIPELINE_STAGE_HOST_BIT

VK_ACCESS_HOST_WRITE_BIT

VK_PIPELINE_STAGE_HOST_BIT

VK_ACCESS_MEMORY_READ_BIT

N/A

VK_ACCESS_MEMORY_WRITE_BIT

N/A

VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX

VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX

VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

If a memory object does not have the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT property, then vkFlushMappedMemoryRanges must be called in order to guarantee that writes to the memory object from the host are made visible to the VK_ACCESS_HOST_WRITE_BIT access type, where it can be further made available to the device by synchronization commands. Similarly, vkInvalidateMappedMemoryRanges must be called to guarantee that writes which are visible to the VK_ACCESS_HOST_READ_BIT access type are made visible to host operations.

If the memory object does have the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT property flag, writes to the memory object from the host are automatically made visible to the VK_ACCESS_HOST_WRITE_BIT access type. Similarly, writes made visible to the VK_ACCESS_HOST_READ_BIT access type are automatically made visible to the host.

Note

The vkQueueSubmit command automatically guarantees that host writes flushed to VK_ACCESS_HOST_WRITE_BIT are made available if they were flushed before the command executed, so in most cases an explicit memory barrier is not needed for this case. In the few circumstances where a submit does not occur between the host write and the device read access, writes can be made available by using an explicit memory barrier.

typedef VkFlags VkAccessFlags;

VkAccessFlags is a bitmask type for setting a mask of zero or more VkAccessFlagBits.

6.1.4. Framebuffer Region Dependencies

Pipeline stages that operate on, or with respect to, the framebuffer are collectively the framebuffer-space pipeline stages. These stages are:

  • VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT

  • VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT

  • VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT

  • VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

For these pipeline stages, an execution or memory dependency from the first set of operations to the second set can either be a single framebuffer-global dependency, or split into multiple framebuffer-local dependencies. A dependency with non-framebuffer-space pipeline stages is neither framebuffer-global nor framebuffer-local.

A framebuffer region is a set of sample (x, y, layer, sample) coordinates that is a subset of the entire framebuffer.

Both synchronization scopes of a framebuffer-local dependency include only the operations performed within corresponding framebuffer regions (as defined below). No ordering guarantees are made between different framebuffer regions for a framebuffer-local dependency.

Both synchronization scopes of a framebuffer-global dependency include operations on all framebuffer-regions.

If the first synchronization scope includes operations on pixels/fragments with N samples and the second synchronization scope includes operations on pixels/fragments with M samples, where N does not equal M, then a framebuffer region containing all samples at a given (x, y, layer) coordinate in the first synchronization scope corresponds to a region containing all samples at the same coordinate in the second synchronization scope. In other words, it is a pixel granularity dependency. If N equals M, then a framebuffer region containing a single (x, y, layer, sample) coordinate in the first synchronization scope corresponds to a region containing the same sample at the same coordinate in the second synchronization scope. In other words, it is a sample granularity dependency.

Note

Since fragment invocations are not specified to run in any particular groupings, the size of a framebuffer region is implementation-dependent, not known to the application, and must be assumed to be no larger than specified above.

Note

Practically, the pixel vs sample granularity dependency means that if an input attachment has a different number of samples than the pipeline’s rasterizationSamples, then a fragment can access any sample in the input attachment’s pixel even if it only uses framebuffer-local dependencies. If the input attachment has the same number of samples, then the fragment can only access the covered samples in its input SampleMask (i.e. the fragment operations happen-after a framebuffer-local dependency for each sample the fragment covers). To access samples that are not covered, a framebuffer-global dependency is required.

If a synchronization command includes a dependencyFlags parameter, and specifies the VK_DEPENDENCY_BY_REGION_BIT flag, then it defines framebuffer-local dependencies for the framebuffer-space pipeline stages in that synchronization command, for all framebuffer regions. If no dependencyFlags parameter is included, or the VK_DEPENDENCY_BY_REGION_BIT flag is not specified, then a framebuffer-global dependency is specified for those stages. The VK_DEPENDENCY_BY_REGION_BIT flag does not affect the dependencies between non-framebuffer-space pipeline stages, nor does it affect the dependencies between framebuffer-space and non-framebuffer-space pipeline stages.

Note

Framebuffer-local dependencies are more optimal for most architectures; particularly tile-based architectures - which can keep framebuffer-regions entirely in on-chip registers and thus avoid external bandwidth across such a dependency. Including a framebuffer-global dependency in your rendering will usually force all implementations to flush data to memory, or to a higher level cache, breaking any potential locality optimizations.

6.1.5. View-Local Dependencies

In a render pass instance that has multiview enabled, dependencies can be either view-local or view-global.

A view-local dependency only includes operations from a single source view from the source subpass in the first synchronization scope, and only includes operations from a single destination view from the destination subpass in the second synchronization scope. A view-global dependency includes all views in the view mask of the source and destination subpasses in the corresponding synchronization scopes.

If a synchronization command includes a dependencyFlags parameter and specifies the VK_DEPENDENCY_VIEW_LOCAL_BIT flag, then it defines view-local dependencies for that synchronization command, for all views. If no dependencyFlags parameter is included or the VK_DEPENDENCY_VIEW_LOCAL_BIT flag is not specified, then a view-global dependency is specified.

6.1.6. Device-Local Dependencies

Dependencies can be either device-local or non-device-local. A device-local dependency acts as multiple separate dependencies, one for each physical device that executes the synchronization command, where each dependency only includes operations from that physical device in both synchronization scopes. A non-device-local dependency is a single dependency where both synchronization scopes include operations from all physical devices that participate in the synchronization command. For subpass dependencies, all physical devices in the VkDeviceGroupRenderPassBeginInfo::deviceMask participate in the dependency, and for pipeline barriers all physical devices that are set in the command buffer’s current device mask participate in the dependency.

If a synchronization command includes a dependencyFlags parameter and specifies the VK_DEPENDENCY_DEVICE_GROUP_BIT flag, then it defines a non-device-local dependency for that synchronization command. If no dependencyFlags parameter is included or the VK_DEPENDENCY_DEVICE_GROUP_BIT flag is not specified, then it defines device-local dependencies for that synchronization command, for all participating physical devices.

Semaphore and event dependencies are device-local and only execute on the one physical device that performs the dependency.

6.2. Implicit Synchronization Guarantees

A small number of implicit ordering guarantees are provided by Vulkan, ensuring that the order in which commands are submitted is meaningful, and avoiding unnecessary complexity in common operations.

Submission order is a fundamental ordering in Vulkan, giving meaning to the order in which action and synchronization commands are recorded and submitted to a single queue. Explicit and implicit ordering guarantees between commands in Vulkan all work on the premise that this ordering is meaningful. This order does not itself define any execution or memory dependencies; synchronization commands and other orderings within the API use this ordering to define their scopes.

Submission order for any given set of commands is based on the order in which they were recorded to command buffers and then submitted. This order is determined as follows:

  1. The initial order is determined by the order in which vkQueueSubmit commands are executed on the host, for a single queue, from first to last.

  2. The order in which VkSubmitInfo structures are specified in the pSubmits parameter of vkQueueSubmit, from lowest index to highest.

  3. The order in which command buffers are specified in the pCommandBuffers member of VkSubmitInfo, from lowest index to highest.

  4. The order in which commands were recorded to a command buffer on the host, from first to last:

    • For commands recorded outside a render pass, this includes all other commands recorded outside a render pass, including vkCmdBeginRenderPass and vkCmdEndRenderPass commands; it does not directly include commands inside a render pass.

    • For commands recorded inside a render pass, this includes all other commands recorded inside the same subpass, including the vkCmdBeginRenderPass and vkCmdEndRenderPass commands that delimit the same render pass instance; it does not include commands recorded to other subpasses.

Action and synchronization commands recorded to a command buffer execute the VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT pipeline stage in submission order - forming an implicit execution dependency between this stage in each command.

State commands do not execute any operations on the device, instead they set the state of the command buffer when they execute on the host, in the order that they are recorded. Action commands consume the current state of the command buffer when they are recorded, and will execute state changes on the device as required to match the recorded state.

Execution of pipeline stages within a given command also has a loose ordering, dependent only on a single command.

6.3. Fences

Fences are a synchronization primitive that can be used to insert a dependency from a queue to the host. Fences have two states - signaled and unsignaled. A fence can be signaled as part of the execution of a queue submission command. Fences can be unsignaled on the host with vkResetFences. Fences can be waited on by the host with the vkWaitForFences command, and the current state can be queried with vkGetFenceStatus.

As with most objects in Vulkan, fences are an interface to internal data which is typically opaque to applications. This internal data is referred to as a fence’s payload.

However, in order to enable communication with agents outside of the current device, it is necessary to be able to export that payload to a commonly understood format, and subsequently import from that format as well.

The internal data of a fence may include a reference to any resources and pending work associated with signal or unsignal operations performed on that fence object. Mechanisms to import and export that internal data to and from fences are provided below. These mechanisms indirectly enable applications to share fence state between two or more fences and other synchronization primitives across process and API boundaries.

Fences are represented by VkFence handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkFence)

To create a fence, call:

VkResult vkCreateFence(
    VkDevice                                    device,
    const VkFenceCreateInfo*                    pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkFence*                                    pFence);
  • device is the logical device that creates the fence.

  • pCreateInfo is a pointer to an instance of the VkFenceCreateInfo structure which contains information about how the fence is to be created.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pFence points to a handle in which the resulting fence object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkFenceCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pFence must be a valid pointer to a VkFence handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkFenceCreateInfo structure is defined as:

typedef struct VkFenceCreateInfo {
    VkStructureType       sType;
    const void*           pNext;
    VkFenceCreateFlags    flags;
} VkFenceCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkFenceCreateFlagBits specifying the initial state and behavior of the fence.

Valid Usage (Implicit)
typedef enum VkFenceCreateFlagBits {
    VK_FENCE_CREATE_SIGNALED_BIT = 0x00000001,
} VkFenceCreateFlagBits;
  • VK_FENCE_CREATE_SIGNALED_BIT specifies that the fence object is created in the signaled state. Otherwise, it is created in the unsignaled state.

typedef VkFlags VkFenceCreateFlags;

VkFenceCreateFlags is a bitmask type for setting a mask of zero or more VkFenceCreateFlagBits.

To create a fence whose payload can be exported to external handles, add the VkExportFenceCreateInfo structure to the pNext chain of the VkFenceCreateInfo structure. The VkExportFenceCreateInfo structure is defined as:

typedef struct VkExportFenceCreateInfo {
    VkStructureType                   sType;
    const void*                       pNext;
    VkExternalFenceHandleTypeFlags    handleTypes;
} VkExportFenceCreateInfo;

or the equivalent

typedef VkExportFenceCreateInfo VkExportFenceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalFenceHandleTypeFlagBits specifying one or more fence handle types the application can export from the resulting fence. The application can request multiple handle types for the same fence.

Valid Usage
Valid Usage (Implicit)

To specify additional attributes of NT handles exported from a fence, add the VkExportFenceWin32HandleInfoKHR structure to the pNext chain of the VkFenceCreateInfo structure. The VkExportFenceWin32HandleInfoKHR structure is defined as:

typedef struct VkExportFenceWin32HandleInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    const SECURITY_ATTRIBUTES*    pAttributes;
    DWORD                         dwAccess;
    LPCWSTR                       name;
} VkExportFenceWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pAttributes is a pointer to a Windows SECURITY_ATTRIBUTES structure specifying security attributes of the handle.

  • dwAccess is a DWORD specifying access rights of the handle.

  • name is a NULL-terminated UTF-16 string to associate with the underlying synchronization primitive referenced by NT handles exported from the created fence.

If this structure is not present, or if pAttributes is set to NULL, default security descriptor values will be used, and child processes created by the application will not inherit the handle, as described in the MSDN documentation for “Synchronization Object Security and Access Rights”1. Further, if the structure is not present, the access rights will be

DXGI_SHARED_RESOURCE_READ | DXGI_SHARED_RESOURCE_WRITE

for handles of the following types:

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXPORT_FENCE_WIN32_HANDLE_INFO_KHR

  • If pAttributes is not NULL, pAttributes must be a valid pointer to a valid SECURITY_ATTRIBUTES value

To export a Windows handle representing the state of a fence, call:

VkResult vkGetFenceWin32HandleKHR(
    VkDevice                                    device,
    const VkFenceGetWin32HandleInfoKHR*         pGetWin32HandleInfo,
    HANDLE*                                     pHandle);
  • device is the logical device that created the fence being exported.

  • pGetWin32HandleInfo is a pointer to an instance of the VkFenceGetWin32HandleInfoKHR structure containing parameters of the export operation.

  • pHandle will return the Windows handle representing the fence state.

For handle types defined as NT handles, the handles returned by vkGetFenceWin32HandleKHR are owned by the application. To avoid leaking resources, the application must release ownership of them using the CloseHandle system call when they are no longer needed.

Exporting a Windows handle from a fence may have side effects depending on the transference of the specified handle type, as described in Importing Fence Payloads.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetWin32HandleInfo must be a valid pointer to a valid VkFenceGetWin32HandleInfoKHR structure

  • pHandle must be a valid pointer to a HANDLE value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkFenceGetWin32HandleInfoKHR structure is defined as:

typedef struct VkFenceGetWin32HandleInfoKHR {
    VkStructureType                      sType;
    const void*                          pNext;
    VkFence                              fence;
    VkExternalFenceHandleTypeFlagBits    handleType;
} VkFenceGetWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • fence is the fence from which state will be exported.

  • handleType is the type of handle requested.

The properties of the handle returned depend on the value of handleType. See VkExternalFenceHandleTypeFlagBits for a description of the properties of the defined external fence handle types.

Valid Usage
  • handleType must have been included in VkExportFenceCreateInfo::handleTypes when the fence’s current payload was created.

  • If handleType is defined as an NT handle, vkGetFenceWin32HandleKHR must be called no more than once for each valid unique combination of fence and handleType.

  • fence must not currently have its payload replaced by an imported payload as described below in Importing Fence Payloads unless that imported payload’s handle type was included in VkExternalFenceProperties::exportFromImportedHandleTypes for handleType.

  • If handleType refers to a handle type with copy payload transference semantics, fence must be signaled, or have an associated fence signal operation pending execution.

  • handleType must be defined as an NT handle or a global share handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_FENCE_GET_WIN32_HANDLE_INFO_KHR

  • pNext must be NULL

  • fence must be a valid VkFence handle

  • handleType must be a valid VkExternalFenceHandleTypeFlagBits value

To export a POSIX file descriptor representing the payload of a fence, call:

VkResult vkGetFenceFdKHR(
    VkDevice                                    device,
    const VkFenceGetFdInfoKHR*                  pGetFdInfo,
    int*                                        pFd);
  • device is the logical device that created the fence being exported.

  • pGetFdInfo is a pointer to an instance of the VkFenceGetFdInfoKHR structure containing parameters of the export operation.

  • pFd will return the file descriptor representing the fence payload.

Each call to vkGetFenceFdKHR must create a new file descriptor and transfer ownership of it to the application. To avoid leaking resources, the application must release ownership of the file descriptor when it is no longer needed.

Note

Ownership can be released in many ways. For example, the application can call close() on the file descriptor, or transfer ownership back to Vulkan by using the file descriptor to import a fence payload.

If pGetFdInfo::handleType is VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT and the fence is signaled at the time vkGetFenceFdKHR is called, pFd may return the value -1 instead of a valid file descriptor.

Where supported by the operating system, the implementation must set the file descriptor to be closed automatically when an execve system call is made.

Exporting a file descriptor from a fence may have side effects depending on the transference of the specified handle type, as described in Importing Fence State.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetFdInfo must be a valid pointer to a valid VkFenceGetFdInfoKHR structure

  • pFd must be a valid pointer to a int value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkFenceGetFdInfoKHR structure is defined as:

typedef struct VkFenceGetFdInfoKHR {
    VkStructureType                      sType;
    const void*                          pNext;
    VkFence                              fence;
    VkExternalFenceHandleTypeFlagBits    handleType;
} VkFenceGetFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • fence is the fence from which state will be exported.

  • handleType is the type of handle requested.

The properties of the file descriptor returned depend on the value of handleType. See VkExternalFenceHandleTypeFlagBits for a description of the properties of the defined external fence handle types.

Valid Usage
  • handleType must have been included in VkExportFenceCreateInfo::handleTypes when fence’s current payload was created.

  • If handleType refers to a handle type with copy payload transference semantics, fence must be signaled, or have an associated fence signal operation pending execution.

  • fence must not currently have its payload replaced by an imported payload as described below in Importing Fence Payloads unless that imported payload’s handle type was included in VkExternalFenceProperties::exportFromImportedHandleTypes for handleType.

  • handleType must be defined as a POSIX file descriptor handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_FENCE_GET_FD_INFO_KHR

  • pNext must be NULL

  • fence must be a valid VkFence handle

  • handleType must be a valid VkExternalFenceHandleTypeFlagBits value

To destroy a fence, call:

void vkDestroyFence(
    VkDevice                                    device,
    VkFence                                     fence,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the fence.

  • fence is the handle of the fence to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All queue submission commands that refer to fence must have completed execution

  • If VkAllocationCallbacks were provided when fence was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when fence was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If fence is not VK_NULL_HANDLE, fence must be a valid VkFence handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If fence is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to fence must be externally synchronized

To query the status of a fence from the host, call:

VkResult vkGetFenceStatus(
    VkDevice                                    device,
    VkFence                                     fence);
  • device is the logical device that owns the fence.

  • fence is the handle of the fence to query.

Upon success, vkGetFenceStatus returns the status of the fence object, with the following return codes:

Table 5. Fence Object Status Codes
Status Meaning

VK_SUCCESS

The fence specified by fence is signaled.

VK_NOT_READY

The fence specified by fence is unsignaled.

VK_ERROR_DEVICE_LOST

The device has been lost. See Lost Device.

If a queue submission command is pending execution, then the value returned by this command may immediately be out of date.

If the device has been lost (see Lost Device), vkGetFenceStatus may return any of the above status codes. If the device has been lost and vkGetFenceStatus is called repeatedly, it will eventually return either VK_SUCCESS or VK_ERROR_DEVICE_LOST.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • fence must be a valid VkFence handle

  • fence must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_NOT_READY

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

To set the state of fences to unsignaled from the host, call:

VkResult vkResetFences(
    VkDevice                                    device,
    uint32_t                                    fenceCount,
    const VkFence*                              pFences);
  • device is the logical device that owns the fences.

  • fenceCount is the number of fences to reset.

  • pFences is a pointer to an array of fence handles to reset.

If any member of pFences currently has its payload imported with temporary permanence, that fence’s prior permanent payload is first restored. The remaining operations described therefore operate on the restored payload.

When vkResetFences is executed on the host, it defines a fence unsignal operation for each fence, which resets the fence to the unsignaled state.

If any member of pFences is already in the unsignaled state when vkResetFences is executed, then vkResetFences has no effect on that fence.

Valid Usage
  • Each element of pFences must not be currently associated with any queue command that has not yet completed execution on that queue

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pFences must be a valid pointer to an array of fenceCount valid VkFence handles

  • fenceCount must be greater than 0

  • Each element of pFences must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to each member of pFences must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

When a fence is submitted to a queue as part of a queue submission command, it defines a memory dependency on the batches that were submitted as part of that command, and defines a fence signal operation which sets the fence to the signaled state.

The first synchronization scope includes every batch submitted in the same queue submission command. Fence signal operations that are defined by vkQueueSubmit additionally include in the first synchronization scope all commands that occur earlier in submission order.

The second synchronization scope only includes the fence signal operation.

The first access scope includes all memory access performed by the device.

The second access scope is empty.

To wait for one or more fences to enter the signaled state on the host, call:

VkResult vkWaitForFences(
    VkDevice                                    device,
    uint32_t                                    fenceCount,
    const VkFence*                              pFences,
    VkBool32                                    waitAll,
    uint64_t                                    timeout);
  • device is the logical device that owns the fences.

  • fenceCount is the number of fences to wait on.

  • pFences is a pointer to an array of fenceCount fence handles.

  • waitAll is the condition that must be satisfied to successfully unblock the wait. If waitAll is VK_TRUE, then the condition is that all fences in pFences are signaled. Otherwise, the condition is that at least one fence in pFences is signaled.

  • timeout is the timeout period in units of nanoseconds. timeout is adjusted to the closest value allowed by the implementation-dependent timeout accuracy, which may be substantially longer than one nanosecond, and may be longer than the requested period.

If the condition is satisfied when vkWaitForFences is called, then vkWaitForFences returns immediately. If the condition is not satisfied at the time vkWaitForFences is called, then vkWaitForFences will block and wait up to timeout nanoseconds for the condition to become satisfied.

If timeout is zero, then vkWaitForFences does not wait, but simply returns the current state of the fences. VK_TIMEOUT will be returned in this case if the condition is not satisfied, even though no actual wait was performed.

If the specified timeout period expires before the condition is satisfied, vkWaitForFences returns VK_TIMEOUT. If the condition is satisfied before timeout nanoseconds has expired, vkWaitForFences returns VK_SUCCESS.

If device loss occurs (see Lost Device) before the timeout has expired, vkWaitForFences must return in finite time with either VK_SUCCESS or VK_ERROR_DEVICE_LOST.

Note

While we guarantee that vkWaitForFences must return in finite time, no guarantees are made that it returns immediately upon device loss. However, the client can reasonably expect that the delay will be on the order of seconds and that calling vkWaitForFences will not result in a permanently (or seemingly permanently) dead process.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pFences must be a valid pointer to an array of fenceCount valid VkFence handles

  • fenceCount must be greater than 0

  • Each element of pFences must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_TIMEOUT

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

An execution dependency is defined by waiting for a fence to become signaled, either via vkWaitForFences or by polling on vkGetFenceStatus.

The first synchronization scope includes only the fence signal operation.

The second synchronization scope includes the host operations of vkWaitForFences or vkGetFenceStatus indicating that the fence has become signaled.

Note

Signaling a fence and waiting on the host does not guarantee that the results of memory accesses will be visible to the host, as the access scope of a memory dependency defined by a fence only includes device access. A memory barrier or other memory dependency must be used to guarantee this. See the description of host access types for more information.

6.3.1. Alternate Methods to Signal Fences

Besides submitting a fence to a queue as part of a queue submission command, a fence may also be signaled when a particular event occurs on a device or display.

To create a fence that will be signaled when an event occurs on a device, call:

VkResult vkRegisterDeviceEventEXT(
    VkDevice                                    device,
    const VkDeviceEventInfoEXT*                 pDeviceEventInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkFence*                                    pFence);
  • device is a logical device on which the event may occur.

  • pDeviceEventInfo is a pointer to an instance of the VkDeviceEventInfoEXT structure describing the event of interest to the application.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pFence points to a handle in which the resulting fence object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pDeviceEventInfo must be a valid pointer to a valid VkDeviceEventInfoEXT structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pFence must be a valid pointer to a VkFence handle

Return Codes
Success
  • VK_SUCCESS

The VkDeviceEventInfoEXT structure is defined as:

typedef struct VkDeviceEventInfoEXT {
    VkStructureType         sType;
    const void*             pNext;
    VkDeviceEventTypeEXT    deviceEvent;
} VkDeviceEventInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • device is a VkDeviceEventTypeEXT value specifying when the fence will be signaled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_EVENT_INFO_EXT

  • pNext must be NULL

  • deviceEvent must be a valid VkDeviceEventTypeEXT value

Possible values of VkDeviceEventInfoEXT::device, specifying when a fence will be signaled, are:

typedef enum VkDeviceEventTypeEXT {
    VK_DEVICE_EVENT_TYPE_DISPLAY_HOTPLUG_EXT = 0,
} VkDeviceEventTypeEXT;
  • VK_DEVICE_EVENT_TYPE_DISPLAY_HOTPLUG_EXT specifies that the fence is signaled when a display is plugged into or unplugged from the specified device. Applications can use this notification to determine when they need to re-enumerate the available displays on a device.

To create a fence that will be signaled when an event occurs on a VkDisplayKHR object, call:

VkResult vkRegisterDisplayEventEXT(
    VkDevice                                    device,
    VkDisplayKHR                                display,
    const VkDisplayEventInfoEXT*                pDisplayEventInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkFence*                                    pFence);
  • device is a logical device associated with display

  • display is the display on which the event may occur.

  • pDisplayEventInfo is a pointer to an instance of the VkDisplayEventInfoEXT structure describing the event of interest to the application.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pFence points to a handle in which the resulting fence object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • display must be a valid VkDisplayKHR handle

  • pDisplayEventInfo must be a valid pointer to a valid VkDisplayEventInfoEXT structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pFence must be a valid pointer to a VkFence handle

Return Codes
Success
  • VK_SUCCESS

The VkDisplayEventInfoEXT structure is defined as:

typedef struct VkDisplayEventInfoEXT {
    VkStructureType          sType;
    const void*              pNext;
    VkDisplayEventTypeEXT    displayEvent;
} VkDisplayEventInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • displayEvent is a VkDisplayEventTypeEXT specifying when the fence will be signaled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DISPLAY_EVENT_INFO_EXT

  • pNext must be NULL

  • displayEvent must be a valid VkDisplayEventTypeEXT value

Possible values of VkDisplayEventInfoEXT::displayEvent, specifying when a fence will be signaled, are:

typedef enum VkDisplayEventTypeEXT {
    VK_DISPLAY_EVENT_TYPE_FIRST_PIXEL_OUT_EXT = 0,
} VkDisplayEventTypeEXT;
  • VK_DISPLAY_EVENT_TYPE_FIRST_PIXEL_OUT_EXT specifies that the fence is signaled when the first pixel of the next display refresh cycle leaves the display engine for the display.

6.3.2. Importing Fence Payloads

Applications can import a fence payload into an existing fence using an external fence handle. The effects of the import operation will be either temporary or permanent, as specified by the application. If the import is temporary, the fence will be restored to its permanent state the next time that fence is passed to vkResetFences.

Note

Restoring a fence to its prior permanent payload is a distinct operation from resetting a fence payload. See vkResetFences for more detail.

Performing a subsequent temporary import on a fence before resetting it has no effect on this requirement; the next unsignal of the fence must still restore its last permanent state. A permanent payload import behaves as if the target fence was destroyed, and a new fence was created with the same handle but the imported payload. Because importing a fence payload temporarily or permanently detaches the existing payload from a fence, similar usage restrictions to those applied to vkDestroyFence are applied to any command that imports a fence payload. Which of these import types is used is referred to as the import operation’s permanence. Each handle type supports either one or both types of permanence.

The implementation must perform the import operation by either referencing or copying the payload referred to by the specified external fence handle, depending on the handle’s type. The import method used is referred to as the handle type’s transference. When using handle types with reference transference, importing a payload to a fence adds the fence to the set of all fences sharing that payload. This set includes the fence from which the payload was exported. Fence signaling, waiting, and resetting operations performed on any fence in the set must behave as if the set were a single fence. Importing a payload using handle types with copy transference creates a duplicate copy of the payload at the time of import, but makes no further reference to it. Fence signaling, waiting, and resetting operations performed on the target of copy imports must not affect any other fence or payload.

Export operations have the same transference as the specified handle type’s import operations. Additionally, exporting a fence payload to a handle with copy transference has the same side effects on the source fence’s payload as executing a fence reset operation. If the fence was using a temporarily imported payload, the fence’s prior permanent payload will be restored.

Note

The tables Handle Types Supported by VkImportFenceWin32HandleInfoKHR and Handle Types Supported by VkImportFenceFdInfoKHR define the permanence and transference of each handle type.

External synchronization allows implementations to modify an object’s internal state, i.e. payload, without internal synchronization. However, for fences sharing a payload across processes, satisfying the external synchronization requirements of VkFence parameters as if all fences in the set were the same object is sometimes infeasible. Satisfying valid usage constraints on the state of a fence would similarly require impractical coordination or levels of trust between processes. Therefore, these constraints only apply to a specific fence handle, not to its payload. For distinct fence objects which share a payload:

  • If multiple commands which queue a signal operation, or which unsignal a fence, are called concurrently, behavior will be as if the commands were called in an arbitrary sequential order.

  • If a queue submission command is called with a fence that is sharing a payload, and the payload is already associated with another queue command that has not yet completed execution, either one or both of the commands will cause the fence to become signaled when they complete execution.

  • If a fence payload is reset while it is associated with a queue command that has not yet completed execution, the payload will become unsignaled, but may become signaled again when the command completes execution.

  • In the preceding cases, any of the devices associated with the fences sharing the payload may be lost, or any of the queue submission or fence reset commands may return VK_ERROR_INITIALIZATION_FAILED.

Other than these non-deterministic results, behavior is well defined. In particular:

  • The implementation must not crash or enter an internally inconsistent state where future valid Vulkan commands might cause undefined results,

  • Timeouts on future wait commands on fences sharing the payload must be effective.

Note

These rules allow processes to synchronize access to shared memory without trusting each other. However, such processes must still be cautious not to use the shared fence for more than synchronizing access to the shared memory. For example, a process should not use a fence with shared payload to tell when commands it submitted to a queue have completed and objects used by those commands may be destroyed, since the other process can accidentally or maliciously cause the fence to signal before the commands actually complete.

When a fence is using an imported payload, its VkExportFenceCreateInfo::handleTypes value is that specified when creating the fence from which the payload was exported, rather than that specified when creating the fence. Additionally, VkExternalFenceProperties::exportFromImportedHandleTypes restricts which handle types can be exported from such a fence based on the specific handle type used to import the current payload. Passing a fence to vkAcquireNextImageKHR is equivalent to temporarily importing a fence payload to that fence.

Note

Because the exportable handle types of an imported fence correspond to its current imported payload, and vkAcquireNextImageKHR behaves the same as a temporary import operation for which the source fence is opaque to the application, applications have no way of determining whether any external handle types can be exported from a fence in this state. Therefore, applications must not attempt to export handles from fences using a temporarily imported payload from vkAcquireNextImageKHR.

When importing a fence payload, it is the responsibility of the application to ensure the external handles meet all valid usage requirements. However, implementations must perform sufficient validation of external handles to ensure that the operation results in a valid fence which will not cause program termination, device loss, queue stalls, host thread stalls, or corruption of other resources when used as allowed according to its import parameters. If the external handle provided does not meet these requirements, the implementation must fail the fence payload import operation with the error code VK_ERROR_INVALID_EXTERNAL_HANDLE.

To import a fence payload from a Windows handle, call:

VkResult vkImportFenceWin32HandleKHR(
    VkDevice                                    device,
    const VkImportFenceWin32HandleInfoKHR*      pImportFenceWin32HandleInfo);
  • device is the logical device that created the fence.

  • pImportFenceWin32HandleInfo points to a VkImportFenceWin32HandleInfoKHR structure specifying the fence and import parameters.

Importing a fence payload from Windows handles does not transfer ownership of the handle to the Vulkan implementation. For handle types defined as NT handles, the application must release ownership using the CloseHandle system call when the handle is no longer needed.

Applications can import the same fence payload into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pImportFenceWin32HandleInfo must be a valid pointer to a valid VkImportFenceWin32HandleInfoKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportFenceWin32HandleInfoKHR structure is defined as:

typedef struct VkImportFenceWin32HandleInfoKHR {
    VkStructureType                      sType;
    const void*                          pNext;
    VkFence                              fence;
    VkFenceImportFlags                   flags;
    VkExternalFenceHandleTypeFlagBits    handleType;
    HANDLE                               handle;
    LPCWSTR                              name;
} VkImportFenceWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • fence is the fence into which the state will be imported.

  • flags is a bitmask of VkFenceImportFlagBits specifying additional parameters for the fence payload import operation.

  • handleType specifies the type of handle.

  • handle is the external handle to import, or NULL.

  • name is the NULL-terminated UTF-16 string naming the underlying synchronization primitive to import, or NULL.

The handle types supported by handleType are:

Table 6. Handle Types Supported by VkImportFenceWin32HandleInfoKHR
Handle Type Transference Permanence Supported

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT

Reference

Temporary,Permanent

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT

Reference

Temporary,Permanent

Valid Usage
  • handleType must be a value included in the Handle Types Supported by VkImportFenceWin32HandleInfoKHR table.

  • If handleType is not VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT, name must be NULL.

  • If handleType is not 0 and handle is NULL, name must name a valid synchronization primitive of the type specified by handleType.

  • If handleType is not 0 and name is NULL, handle must be a valid handle of the type specified by handleType.

  • If handle is not NULL, name must be NULL.

  • If handle is not NULL, it must obey any requirements listed for handleType in external fence handle types compatibility.

  • If name is not NULL, it must obey any requirements listed for handleType in external fence handle types compatibility.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMPORT_FENCE_WIN32_HANDLE_INFO_KHR

  • pNext must be NULL

  • fence must be a valid VkFence handle

  • flags must be a valid combination of VkFenceImportFlagBits values

  • If handleType is not 0, handleType must be a valid VkExternalFenceHandleTypeFlagBits value

Host Synchronization
  • Host access to fence must be externally synchronized

To import a fence payload from a POSIX file descriptor, call:

VkResult vkImportFenceFdKHR(
    VkDevice                                    device,
    const VkImportFenceFdInfoKHR*               pImportFenceFdInfo);
  • device is the logical device that created the fence.

  • pImportFenceFdInfo points to a VkImportFenceFdInfoKHR structure specifying the fence and import parameters.

Importing a fence payload from a file descriptor transfers ownership of the file descriptor from the application to the Vulkan implementation. The application must not perform any operations on the file descriptor after a successful import.

Applications can import the same fence payload into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance.

Valid Usage
  • fence must not be associated with any queue command that has not yet completed execution on that queue

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pImportFenceFdInfo must be a valid pointer to a valid VkImportFenceFdInfoKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportFenceFdInfoKHR structure is defined as:

typedef struct VkImportFenceFdInfoKHR {
    VkStructureType                      sType;
    const void*                          pNext;
    VkFence                              fence;
    VkFenceImportFlags                   flags;
    VkExternalFenceHandleTypeFlagBits    handleType;
    int                                  fd;
} VkImportFenceFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • fence is the fence into which the payload will be imported.

  • flags is a bitmask of VkFenceImportFlagBits specifying additional parameters for the fence payload import operation.

  • handleType specifies the type of fd.

  • fd is the external handle to import.

The handle types supported by handleType are:

Table 7. Handle Types Supported by VkImportFenceFdInfoKHR
Handle Type Transference Permanence Supported

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT

Reference

Temporary,Permanent

VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT

Copy

Temporary

Valid Usage

If handleType is VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT, the special value -1 for fd is treated like a valid sync file descriptor referring to an object that has already signaled. The import operation will succeed and the VkFence will have a temporarily imported payload as if a valid file descriptor had been provided.

Note

This special behavior for importing an invalid sync file descriptor allows easier interoperability with other system APIs which use the convention that an invalid sync file descriptor represents work that has already completed and doesn’t need to be waited for. It is consistent with the option for implementations to return a -1 file descriptor when exporting a VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT from a VkFence which is signaled.

Valid Usage (Implicit)
Host Synchronization
  • Host access to fence must be externally synchronized

Bits which can be set in VkImportFenceWin32HandleInfoKHR::flags and VkImportFenceFdInfoKHR::flags specifying additional parameters of a fence import operation are:

typedef enum VkFenceImportFlagBits {
    VK_FENCE_IMPORT_TEMPORARY_BIT = 0x00000001,
    VK_FENCE_IMPORT_TEMPORARY_BIT_KHR = VK_FENCE_IMPORT_TEMPORARY_BIT,
} VkFenceImportFlagBits;

or the equivalent

typedef VkFenceImportFlagBits VkFenceImportFlagBitsKHR;
  • VK_FENCE_IMPORT_TEMPORARY_BIT specifies that the fence payload will be imported only temporarily, as described in Importing Fence Payloads, regardless of the permanence of handleType.

typedef VkFlags VkFenceImportFlags;

or the equivalent

typedef VkFenceImportFlags VkFenceImportFlagsKHR;

VkFenceImportFlags is a bitmask type for setting a mask of zero or more VkFenceImportFlagBits.

6.4. Semaphores

Semaphores are a synchronization primitive that can be used to insert a dependency between batches submitted to queues. Semaphores have two states - signaled and unsignaled. The state of a semaphore can be signaled after execution of a batch of commands is completed. A batch can wait for a semaphore to become signaled before it begins execution, and the semaphore is also unsignaled before the batch begins execution.

As with most objects in Vulkan, semaphores are an interface to internal data which is typically opaque to applications. This internal data is referred to as a semaphore’s payload.

However, in order to enable communication with agents outside of the current device, it is necessary to be able to export that payload to a commonly understood format, and subsequently import from that format as well.

The internal data of a semaphore may include a reference to any resources and pending work associated with signal or unsignal operations performed on that semaphore object. Mechanisms to import and export that internal data to and from semaphores are provided below. These mechanisms indirectly enable applications to share semaphore state between two or more semaphores and other synchronization primitives across process and API boundaries.

Semaphores are represented by VkSemaphore handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSemaphore)

To create a semaphore, call:

VkResult vkCreateSemaphore(
    VkDevice                                    device,
    const VkSemaphoreCreateInfo*                pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSemaphore*                                pSemaphore);
  • device is the logical device that creates the semaphore.

  • pCreateInfo is a pointer to an instance of the VkSemaphoreCreateInfo structure which contains information about how the semaphore is to be created.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pSemaphore points to a handle in which the resulting semaphore object is returned.

When created, the semaphore is in the unsignaled state.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkSemaphoreCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSemaphore must be a valid pointer to a VkSemaphore handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkSemaphoreCreateInfo structure is defined as:

typedef struct VkSemaphoreCreateInfo {
    VkStructureType           sType;
    const void*               pNext;
    VkSemaphoreCreateFlags    flags;
} VkSemaphoreCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SEMAPHORE_CREATE_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkExportSemaphoreCreateInfo or VkExportSemaphoreWin32HandleInfoKHR

  • Each sType member in the pNext chain must be unique

  • flags must be 0

typedef VkFlags VkSemaphoreCreateFlags;

VkSemaphoreCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To create a semaphore whose payload can be exported to external handles, add the VkExportSemaphoreCreateInfo structure to the pNext chain of the VkSemaphoreCreateInfo structure. The VkExportSemaphoreCreateInfo structure is defined as:

typedef struct VkExportSemaphoreCreateInfo {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalSemaphoreHandleTypeFlags    handleTypes;
} VkExportSemaphoreCreateInfo;

or the equivalent

typedef VkExportSemaphoreCreateInfo VkExportSemaphoreCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalSemaphoreHandleTypeFlagBits specifying one or more semaphore handle types the application can export from the resulting semaphore. The application can request multiple handle types for the same semaphore.

Valid Usage
Valid Usage (Implicit)

To specify additional attributes of NT handles exported from a semaphore, add the VkExportSemaphoreWin32HandleInfoKHR structure to the pNext chain of the VkSemaphoreCreateInfo structure. The VkExportSemaphoreWin32HandleInfoKHR structure is defined as:

typedef struct VkExportSemaphoreWin32HandleInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    const SECURITY_ATTRIBUTES*    pAttributes;
    DWORD                         dwAccess;
    LPCWSTR                       name;
} VkExportSemaphoreWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pAttributes is a pointer to a Windows SECURITY_ATTRIBUTES structure specifying security attributes of the handle.

  • dwAccess is a DWORD specifying access rights of the handle.

  • name is a NULL-terminated UTF-16 string to associate with the underlying synchronization primitive referenced by NT handles exported from the created semaphore.

If this structure is not present, or if pAttributes is set to NULL, default security descriptor values will be used, and child processes created by the application will not inherit the handle, as described in the MSDN documentation for “Synchronization Object Security and Access Rights”1. Further, if the structure is not present, the access rights will be

DXGI_SHARED_RESOURCE_READ | DXGI_SHARED_RESOURCE_WRITE

for handles of the following types:

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT

And

GENERIC_ALL

for handles of the following types:

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT

Valid Usage
  • If VkExportSemaphoreCreateInfo::handleTypes does not include VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT or VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT, VkExportSemaphoreWin32HandleInfoKHR must not be in the pNext chain of VkSemaphoreCreateInfo.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR

  • If pAttributes is not NULL, pAttributes must be a valid pointer to a valid SECURITY_ATTRIBUTES value

To export a Windows handle representing the payload of a semaphore, call:

VkResult vkGetSemaphoreWin32HandleKHR(
    VkDevice                                    device,
    const VkSemaphoreGetWin32HandleInfoKHR*     pGetWin32HandleInfo,
    HANDLE*                                     pHandle);
  • device is the logical device that created the semaphore being exported.

  • pGetWin32HandleInfo is a pointer to an instance of the VkSemaphoreGetWin32HandleInfoKHR structure containing parameters of the export operation.

  • pHandle will return the Windows handle representing the semaphore state.

For handle types defined as NT handles, the handles returned by vkGetSemaphoreWin32HandleKHR are owned by the application. To avoid leaking resources, the application must release ownership of them using the CloseHandle system call when they are no longer needed.

Exporting a Windows handle from a semaphore may have side effects depending on the transference of the specified handle type, as described in Importing Semaphore Payloads.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetWin32HandleInfo must be a valid pointer to a valid VkSemaphoreGetWin32HandleInfoKHR structure

  • pHandle must be a valid pointer to a HANDLE value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkSemaphoreGetWin32HandleInfoKHR structure is defined as:

typedef struct VkSemaphoreGetWin32HandleInfoKHR {
    VkStructureType                          sType;
    const void*                              pNext;
    VkSemaphore                              semaphore;
    VkExternalSemaphoreHandleTypeFlagBits    handleType;
} VkSemaphoreGetWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • semaphore is the semaphore from which state will be exported.

  • handleType is the type of handle requested.

The properties of the handle returned depend on the value of handleType. See VkExternalSemaphoreHandleTypeFlagBits for a description of the properties of the defined external semaphore handle types.

Valid Usage
  • handleType must have been included in VkExportSemaphoreCreateInfo::handleTypes when the semaphore’s current payload was created.

  • If handleType is defined as an NT handle, vkGetSemaphoreWin32HandleKHR must be called no more than once for each valid unique combination of semaphore and handleType.

  • semaphore must not currently have its payload replaced by an imported payload as described below in Importing Semaphore Payloads unless that imported payload’s handle type was included in VkExternalSemaphoreProperties::exportFromImportedHandleTypes for handleType.

  • If handleType refers to a handle type with copy payload transference semantics, as defined below in Importing Semaphore Payloads, there must be no queue waiting on semaphore.

  • If handleType refers to a handle type with copy payload transference semantics, semaphore must be signaled, or have an associated semaphore signal operation pending execution.

  • handleType must be defined as an NT handle or a global share handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SEMAPHORE_GET_WIN32_HANDLE_INFO_KHR

  • pNext must be NULL

  • semaphore must be a valid VkSemaphore handle

  • handleType must be a valid VkExternalSemaphoreHandleTypeFlagBits value

To export a POSIX file descriptor representing the payload of a semaphore, call:

VkResult vkGetSemaphoreFdKHR(
    VkDevice                                    device,
    const VkSemaphoreGetFdInfoKHR*              pGetFdInfo,
    int*                                        pFd);
  • device is the logical device that created the semaphore being exported.

  • pGetFdInfo is a pointer to an instance of the VkSemaphoreGetFdInfoKHR structure containing parameters of the export operation.

  • pFd will return the file descriptor representing the semaphore payload.

Each call to vkGetSemaphoreFdKHR must create a new file descriptor and transfer ownership of it to the application. To avoid leaking resources, the application must release ownership of the file descriptor when it is no longer needed.

Note

Ownership can be released in many ways. For example, the application can call close() on the file descriptor, or transfer ownership back to Vulkan by using the file descriptor to import a semaphore payload.

Where supported by the operating system, the implementation must set the file descriptor to be closed automatically when an execve system call is made.

Exporting a file descriptor from a semaphore may have side effects depending on the transference of the specified handle type, as described in Importing Semaphore State.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetFdInfo must be a valid pointer to a valid VkSemaphoreGetFdInfoKHR structure

  • pFd must be a valid pointer to a int value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkSemaphoreGetFdInfoKHR structure is defined as:

typedef struct VkSemaphoreGetFdInfoKHR {
    VkStructureType                          sType;
    const void*                              pNext;
    VkSemaphore                              semaphore;
    VkExternalSemaphoreHandleTypeFlagBits    handleType;
} VkSemaphoreGetFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • semaphore is the semaphore from which state will be exported.

  • handleType is the type of handle requested.

The properties of the file descriptor returned depend on the value of handleType. See VkExternalSemaphoreHandleTypeFlagBits for a description of the properties of the defined external semaphore handle types.

Valid Usage
  • handleType must have been included in VkExportSemaphoreCreateInfo::handleTypes when semaphore’s current payload was created.

  • semaphore must not currently have its payload replaced by an imported payload as described below in Importing Semaphore Payloads unless that imported payload’s handle type was included in VkExternalSemaphoreProperties::exportFromImportedHandleTypes for handleType.

  • If handleType refers to a handle type with copy payload transference semantics, as defined below in Importing Semaphore Payloads, there must be no queue waiting on semaphore.

  • If handleType refers to a handle type with copy payload transference semantics, semaphore must be signaled, or have an associated semaphore signal operation pending execution.

  • handleType must be defined as a POSIX file descriptor handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SEMAPHORE_GET_FD_INFO_KHR

  • pNext must be NULL

  • semaphore must be a valid VkSemaphore handle

  • handleType must be a valid VkExternalSemaphoreHandleTypeFlagBits value

To destroy a semaphore, call:

void vkDestroySemaphore(
    VkDevice                                    device,
    VkSemaphore                                 semaphore,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the semaphore.

  • semaphore is the handle of the semaphore to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted batches that refer to semaphore must have completed execution

  • If VkAllocationCallbacks were provided when semaphore was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when semaphore was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If semaphore is not VK_NULL_HANDLE, semaphore must be a valid VkSemaphore handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If semaphore is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to semaphore must be externally synchronized

6.4.1. Semaphore Signaling

When a batch is submitted to a queue via a queue submission, and it includes semaphores to be signaled, it defines a memory dependency on the batch, and defines semaphore signal operations which set the semaphores to the signaled state.

The first synchronization scope includes every command submitted in the same batch. Semaphore signal operations that are defined by vkQueueSubmit additionally include all commands that occur earlier in submission order.

The second synchronization scope includes only the semaphore signal operation.

The first access scope includes all memory access performed by the device.

The second access scope is empty.

6.4.2. Semaphore Waiting & Unsignaling

When a batch is submitted to a queue via a queue submission, and it includes semaphores to be waited on, it defines a memory dependency between prior semaphore signal operations and the batch, and defines semaphore unsignal operations which set the semaphores to the unsignaled state.

The first synchronization scope includes all semaphore signal operations that operate on semaphores waited on in the same batch, and that happen-before the wait completes.

The second synchronization scope includes every command submitted in the same batch. In the case of vkQueueSubmit, the second synchronization scope is limited to operations on the pipeline stages determined by the destination stage mask specified by the corresponding element of pWaitDstStageMask. Also, in the case of vkQueueSubmit, the second synchronization scope additionally includes all commands that occur later in submission order.

The first access scope is empty.

The second access scope includes all memory access performed by the device.

The semaphore unsignal operation happens-after the first set of operations in the execution dependency, and happens-before the second set of operations in the execution dependency.

Note

Unlike fences or events, the act of waiting for a semaphore also unsignals that semaphore. If two operations are separately specified to wait for the same semaphore, and there are no other execution dependencies between those operations, behaviour is undefined. An execution dependency must be present that guarantees that the semaphore unsignal operation for the first of those waits, happens-before the semaphore is signalled again, and before the second unsignal operation. Semaphore waits and signals should thus occur in discrete 1:1 pairs.

Note

A common scenario for using pWaitDstStageMask with values other than VK_PIPELINE_STAGE_ALL_COMMANDS_BIT is when synchronizing a window system presentation operation against subsequent command buffers which render the next frame. In this case, a presentation image must not be overwritten until the presentation operation completes, but other pipeline stages can execute without waiting. A mask of VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT prevents subsequent color attachment writes from executing until the semaphore signals. Some implementations may be able to execute transfer operations and/or vertex processing work before the semaphore is signaled.

If an image layout transition needs to be performed on a presentable image before it is used in a framebuffer, that can be performed as the first operation submitted to the queue after acquiring the image, and should not prevent other work from overlapping with the presentation operation. For example, a VkImageMemoryBarrier could use:

  • srcStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

  • srcAccessMask = 0

  • dstStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT

  • dstAccessMask = VK_ACCESS_COLOR_ATTACHMENT_READ_BIT | VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT.

  • oldLayout = VK_IMAGE_LAYOUT_PRESENT_SRC_KHR

  • newLayout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL

Alternatively, oldLayout can be VK_IMAGE_LAYOUT_UNDEFINED, if the image’s contents need not be preserved.

This barrier accomplishes a dependency chain between previous presentation operations and subsequent color attachment output operations, with the layout transition performed in between, and does not introduce a dependency between previous work and any vertex processing stages. More precisely, the semaphore signals after the presentation operation completes, the semaphore wait stalls the VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT stage, and there is a dependency from that same stage to itself with the layout transition performed in between.

6.4.3. Semaphore State Requirements For Wait Operations

Before waiting on a semaphore, the application must ensure the semaphore is in a valid state for a wait operation. Specifically, when a semaphore wait and unsignal operation is submitted to a queue:

  • The semaphore must be signaled, or have an associated semaphore signal operation that is pending execution.

  • There must be no other queue waiting on the same semaphore when the operation executes.

6.4.4. Importing Semaphore Payloads

Applications can import a semaphore payload into an existing semaphore using an external semaphore handle. The effects of the import operation will be either temporary or permanent, as specified by the application. If the import is temporary, the implementation must restore the semaphore to its prior permanent state after submitting the next semaphore wait operation. Performing a subsequent temporary import on a semaphore before performing a semaphore wait has no effect on this requirement; the next wait submitted on the semaphore must still restore its last permanent state. A permanent payload import behaves as if the target semaphore was destroyed, and a new semaphore was created with the same handle but the imported payload. Because importing a semaphore payload temporarily or permanently detaches the existing payload from a semaphore, similar usage restrictions to those applied to vkDestroySemaphore are applied to any command that imports a semaphore payload. Which of these import types is used is referred to as the import operation’s permanence. Each handle type supports either one or both types of permanence.

The implementation must perform the import operation by either referencing or copying the payload referred to by the specified external semaphore handle, depending on the handle’s type. The import method used is referred to as the handle type’s transference. When using handle types with reference transference, importing a payload to a semaphore adds the semaphore to the set of all semaphores sharing that payload. This set includes the semaphore from which the payload was exported. Semaphore signaling and waiting operations performed on any semaphore in the set must behave as if the set were a single semaphore. Importing a payload using handle types with copy transference creates a duplicate copy of the payload at the time of import, but makes no further reference to it. Semaphore signaling and waiting operations performed on the target of copy imports must not affect any other semaphore or payload.

Export operations have the same transference as the specified handle type’s import operations. Additionally, exporting a semaphore payload to a handle with copy transference has the same side effects on the source semaphore’s payload as executing a semaphore wait operation. If the semaphore was using a temporarily imported payload, the semaphore’s prior permanent payload will be restored.

Note

The tables Handle Types Supported by VkImportSemaphoreWin32HandleInfoKHR and Handle Types Supported by VkImportSemaphoreFdInfoKHR define the permanence and transference of each handle type.

External synchronization allows implementations to modify an object’s internal state, i.e. payload, without internal synchronization. However, for semaphores sharing a payload across processes, satisfying the external synchronization requirements of VkSemaphore parameters as if all semaphores in the set were the same object is sometimes infeasible. Satisfying the wait operation state requirements would similarly require impractical coordination or levels of trust between processes. Therefore, these constraints only apply to a specific semaphore handle, not to its payload. For distinct semaphore objects which share a payload, if the semaphores are passed to separate queue submission commands concurrently, behavior will be as if the commands were called in an arbitrary sequential order. If the wait operation state requirements are violated for the shared payload by a queue submission command, or if a signal operation is queued for a shared payload that is already signaled or has a pending signal operation, effects must be limited to one or more of the following:

  • Returning VK_ERROR_INITIALIZATION_FAILED from the command which resulted in the violation.

  • Losing the logical device on which the violation occured immediately or at a future time, resulting in a VK_ERROR_DEVICE_LOST error from subsequent commands, including the one causing the violation.

  • Continuing execution of the violating command or operation as if the semaphore wait completed successfully after an implementation-dependent timeout. In this case, the state of the payload becomes undefined, and future operations on semaphores sharing the payload will be subject to these same rules. The semaphore must be destroyed or have its payload replaced by an import operation to again have a well-defined state.

Note

These rules allow processes to synchronize access to shared memory without trusting each other. However, such processes must still be cautious not to use the shared semaphore for more than synchronizing access to the shared memory. For example, a process should not use a shared semaphore as part of an execution dependency chain that, when complete, leads to objects being destroyed, if it does not trust other processes sharing the semaphore payload.

When a semaphore is using an imported payload, its VkExportSemaphoreCreateInfo::handleTypes value is that specified when creating the semaphore from which the payload was exported, rather than that specified when creating the semaphore. Additionally, VkExternalSemaphoreProperties::exportFromImportedHandleTypes restricts which handle types can be exported from such a semaphore based on the specific handle type used to import the current payload. Passing a semaphore to vkAcquireNextImageKHR is equivalent to temporarily importing a semaphore payload to that semaphore.

Note

Because the exportable handle types of an imported semaphore correspond to its current imported payload, and vkAcquireNextImageKHR behaves the same as a temporary import operation for which the source semaphore is opaque to the application, applications have no way of determining whether any external handle types can be exported from a semaphore in this state. Therefore, applications must not attempt to export external handles from semaphores using a temporarily imported payload from vkAcquireNextImageKHR.

When importing a semaphore payload, it is the responsibility of the application to ensure the external handles meet all valid usage requirements. However, implementations must perform sufficient validation of external handles to ensure that the operation results in a valid semaphore which will not cause program termination, device loss, queue stalls, or corruption of other resources when used as allowed according to its import parameters, and excepting those side effects allowed for violations of the valid semaphore state for wait operations rules. If the external handle provided does not meet these requirements, the implementation must fail the semaphore payload import operation with the error code VK_ERROR_INVALID_EXTERNAL_HANDLE.

To import a semaphore payload from a Windows handle, call:

VkResult vkImportSemaphoreWin32HandleKHR(
    VkDevice                                    device,
    const VkImportSemaphoreWin32HandleInfoKHR*  pImportSemaphoreWin32HandleInfo);
  • device is the logical device that created the semaphore.

  • pImportSemaphoreWin32HandleInfo points to a VkImportSemaphoreWin32HandleInfoKHR structure specifying the semaphore and import parameters.

Importing a semaphore payload from Windows handles does not transfer ownership of the handle to the Vulkan implementation. For handle types defined as NT handles, the application must release ownership using the CloseHandle system call when the handle is no longer needed.

Applications can import the same semaphore payload into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pImportSemaphoreWin32HandleInfo must be a valid pointer to a valid VkImportSemaphoreWin32HandleInfoKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportSemaphoreWin32HandleInfoKHR structure is defined as:

typedef struct VkImportSemaphoreWin32HandleInfoKHR {
    VkStructureType                          sType;
    const void*                              pNext;
    VkSemaphore                              semaphore;
    VkSemaphoreImportFlags                   flags;
    VkExternalSemaphoreHandleTypeFlagBits    handleType;
    HANDLE                                   handle;
    LPCWSTR                                  name;
} VkImportSemaphoreWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • semaphore is the semaphore into which the payload will be imported.

  • flags is a bitmask of VkSemaphoreImportFlagBits specifying additional parameters for the semaphore payload import operation.

  • handleType specifies the type of handle.

  • handle is the external handle to import, or NULL.

  • name is a NULL-terminated UTF-16 string naming the underlying synchronization primitive to import, or NULL.

The handle types supported by handleType are:

Table 8. Handle Types Supported by VkImportSemaphoreWin32HandleInfoKHR
Handle Type Transference Permanence Supported

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT

Reference

Temporary,Permanent

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT

Reference

Temporary,Permanent

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT

Reference

Temporary,Permanent

Valid Usage
  • handleType must be a value included in the Handle Types Supported by VkImportSemaphoreWin32HandleInfoKHR table.

  • If handleType is not VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT or VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT, name must be NULL.

  • If handleType is not 0 and handle is NULL, name must name a valid synchronization primitive of the type specified by handleType.

  • If handleType is not 0 and name is NULL, handle must be a valid handle of the type specified by handleType.

  • If handle is not NULL, name must be NULL.

  • If handle is not NULL, it must obey any requirements listed for handleType in external semaphore handle types compatibility.

  • If name is not NULL, it must obey any requirements listed for handleType in external semaphore handle types compatibility.

Valid Usage (Implicit)
Host Synchronization
  • Host access to semaphore must be externally synchronized

To import a semaphore payload from a POSIX file descriptor, call:

VkResult vkImportSemaphoreFdKHR(
    VkDevice                                    device,
    const VkImportSemaphoreFdInfoKHR*           pImportSemaphoreFdInfo);
  • device is the logical device that created the semaphore.

  • pImportSemaphoreFdInfo points to a VkImportSemaphoreFdInfoKHR structure specifying the semaphore and import parameters.

Importing a semaphore payload from a file descriptor transfers ownership of the file descriptor from the application to the Vulkan implementation. The application must not perform any operations on the file descriptor after a successful import.

Applications can import the same semaphore payload into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance.

Valid Usage
  • semaphore must not be associated with any queue command that has not yet completed execution on that queue

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pImportSemaphoreFdInfo must be a valid pointer to a valid VkImportSemaphoreFdInfoKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportSemaphoreFdInfoKHR structure is defined as:

typedef struct VkImportSemaphoreFdInfoKHR {
    VkStructureType                          sType;
    const void*                              pNext;
    VkSemaphore                              semaphore;
    VkSemaphoreImportFlags                   flags;
    VkExternalSemaphoreHandleTypeFlagBits    handleType;
    int                                      fd;
} VkImportSemaphoreFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • semaphore is the semaphore into which the payload will be imported.

  • flags is a bitmask of VkSemaphoreImportFlagBits specifying additional parameters for the semaphore payload import operation.

  • handleType specifies the type of fd.

  • fd is the external handle to import.

The handle types supported by handleType are:

Table 9. Handle Types Supported by VkImportSemaphoreFdInfoKHR
Handle Type Transference Permanence Supported

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT

Reference

Temporary,Permanent

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT

Copy

Temporary

Valid Usage
Valid Usage (Implicit)
Host Synchronization
  • Host access to semaphore must be externally synchronized

Additional parameters of a semaphore import operation are specified by VkImportSemaphoreWin32HandleInfoKHR::flags or VkImportSemaphoreFdInfoKHR::flags . Bits which can be set include:

typedef enum VkSemaphoreImportFlagBits {
    VK_SEMAPHORE_IMPORT_TEMPORARY_BIT = 0x00000001,
    VK_SEMAPHORE_IMPORT_TEMPORARY_BIT_KHR = VK_SEMAPHORE_IMPORT_TEMPORARY_BIT,
} VkSemaphoreImportFlagBits;

or the equivalent

typedef VkSemaphoreImportFlagBits VkSemaphoreImportFlagBitsKHR;

These bits have the following meanings:

  • VK_SEMAPHORE_IMPORT_TEMPORARY_BIT specifies that the semaphore payload will be imported only temporarily, as described in Importing Semaphore Payloads, regardless of the permanence of handleType.

typedef VkFlags VkSemaphoreImportFlags;

or the equivalent

typedef VkSemaphoreImportFlags VkSemaphoreImportFlagsKHR;

VkSemaphoreImportFlags is a bitmask type for setting a mask of zero or more VkSemaphoreImportFlagBits.

6.5. Events

Events are a synchronization primitive that can be used to insert a fine-grained dependency between commands submitted to the same queue, or between the host and a queue. Events must not be used to insert a dependency between commands submitted to different queues. Events have two states - signaled and unsignaled. An application can signal an event, or unsignal it, on either the host or the device. A device can wait for an event to become signaled before executing further operations. No command exists to wait for an event to become signaled on the host, but the current state of an event can be queried.

Events are represented by VkEvent handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkEvent)

To create an event, call:

VkResult vkCreateEvent(
    VkDevice                                    device,
    const VkEventCreateInfo*                    pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkEvent*                                    pEvent);
  • device is the logical device that creates the event.

  • pCreateInfo is a pointer to an instance of the VkEventCreateInfo structure which contains information about how the event is to be created.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pEvent points to a handle in which the resulting event object is returned.

When created, the event object is in the unsignaled state.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkEventCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pEvent must be a valid pointer to a VkEvent handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkEventCreateInfo structure is defined as:

typedef struct VkEventCreateInfo {
    VkStructureType       sType;
    const void*           pNext;
    VkEventCreateFlags    flags;
} VkEventCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EVENT_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

typedef VkFlags VkEventCreateFlags;

VkEventCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To destroy an event, call:

void vkDestroyEvent(
    VkDevice                                    device,
    VkEvent                                     event,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the event.

  • event is the handle of the event to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to event must have completed execution

  • If VkAllocationCallbacks were provided when event was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when event was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If event is not VK_NULL_HANDLE, event must be a valid VkEvent handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If event is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to event must be externally synchronized

To query the state of an event from the host, call:

VkResult vkGetEventStatus(
    VkDevice                                    device,
    VkEvent                                     event);
  • device is the logical device that owns the event.

  • event is the handle of the event to query.

Upon success, vkGetEventStatus returns the state of the event object with the following return codes:

Table 10. Event Object Status Codes
Status Meaning

VK_EVENT_SET

The event specified by event is signaled.

VK_EVENT_RESET

The event specified by event is unsignaled.

If a vkCmdSetEvent or vkCmdResetEvent command is in a command buffer that is in the pending state, then the value returned by this command may immediately be out of date.

The state of an event can be updated by the host. The state of the event is immediately changed, and subsequent calls to vkGetEventStatus will return the new state. If an event is already in the requested state, then updating it to the same state has no effect.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • event must be a valid VkEvent handle

  • event must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_EVENT_SET

  • VK_EVENT_RESET

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

To set the state of an event to signaled from the host, call:

VkResult vkSetEvent(
    VkDevice                                    device,
    VkEvent                                     event);
  • device is the logical device that owns the event.

  • event is the event to set.

When vkSetEvent is executed on the host, it defines an event signal operation which sets the event to the signaled state.

If event is already in the signaled state when vkSetEvent is executed, then vkSetEvent has no effect, and no event signal operation occurs.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • event must be a valid VkEvent handle

  • event must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to event must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

To set the state of an event to unsignaled from the host, call:

VkResult vkResetEvent(
    VkDevice                                    device,
    VkEvent                                     event);
  • device is the logical device that owns the event.

  • event is the event to reset.

When vkResetEvent is executed on the host, it defines an event unsignal operation which resets the event to the unsignaled state.

If event is already in the unsignaled state when vkResetEvent is executed, then vkResetEvent has no effect, and no event unsignal operation occurs.

Valid Usage
  • event must not be waited on by a vkCmdWaitEvents command that is currently executing

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • event must be a valid VkEvent handle

  • event must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to event must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The state of an event can also be updated on the device by commands inserted in command buffers.

To set the state of an event to signaled from a device, call:

void vkCmdSetEvent(
    VkCommandBuffer                             commandBuffer,
    VkEvent                                     event,
    VkPipelineStageFlags                        stageMask);
  • commandBuffer is the command buffer into which the command is recorded.

  • event is the event that will be signaled.

  • stageMask specifies the source stage mask used to determine when the event is signaled.

When vkCmdSetEvent is submitted to a queue, it defines an execution dependency on commands that were submitted before it, and defines an event signal operation which sets the event to the signaled state.

The first synchronization scope includes all commands that occur earlier in submission order. The synchronization scope is limited to operations on the pipeline stages determined by the source stage mask specified by stageMask.

The second synchronization scope includes only the event signal operation.

If event is already in the signaled state when vkCmdSetEvent is executed on the device, then vkCmdSetEvent has no effect, no event signal operation occurs, and no execution dependency is generated.

Valid Usage
  • stageMask must not include VK_PIPELINE_STAGE_HOST_BIT

  • If the geometry shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • commandBuffer’s current device mask must include exactly one physical device.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • event must be a valid VkEvent handle

  • stageMask must be a valid combination of VkPipelineStageFlagBits values

  • stageMask must not be 0

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • Both of commandBuffer, and event must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics
Compute

To set the state of an event to unsignaled from a device, call:

void vkCmdResetEvent(
    VkCommandBuffer                             commandBuffer,
    VkEvent                                     event,
    VkPipelineStageFlags                        stageMask);
  • commandBuffer is the command buffer into which the command is recorded.

  • event is the event that will be unsignaled.

  • stageMask is a bitmask of VkPipelineStageFlagBits specifying the source stage mask used to determine when the event is unsignaled.

When vkCmdResetEvent is submitted to a queue, it defines an execution dependency on commands that were submitted before it, and defines an event unsignal operation which resets the event to the unsignaled state.

The first synchronization scope includes all commands that occur earlier in submission order. The synchronization scope is limited to operations on the pipeline stages determined by the source stage mask specified by stageMask.

The second synchronization scope includes only the event unsignal operation.

If event is already in the unsignaled state when vkCmdResetEvent is executed on the device, then vkCmdResetEvent has no effect, no event unsignal operation occurs, and no execution dependency is generated.

Valid Usage
  • stageMask must not include VK_PIPELINE_STAGE_HOST_BIT

  • If the geometry shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • When this command executes, event must not be waited on by a vkCmdWaitEvents command that is currently executing

  • commandBuffer’s current device mask must include exactly one physical device.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • event must be a valid VkEvent handle

  • stageMask must be a valid combination of VkPipelineStageFlagBits values

  • stageMask must not be 0

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • Both of commandBuffer, and event must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics
Compute

To wait for one or more events to enter the signaled state on a device, call:

void vkCmdWaitEvents(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    eventCount,
    const VkEvent*                              pEvents,
    VkPipelineStageFlags                        srcStageMask,
    VkPipelineStageFlags                        dstStageMask,
    uint32_t                                    memoryBarrierCount,
    const VkMemoryBarrier*                      pMemoryBarriers,
    uint32_t                                    bufferMemoryBarrierCount,
    const VkBufferMemoryBarrier*                pBufferMemoryBarriers,
    uint32_t                                    imageMemoryBarrierCount,
    const VkImageMemoryBarrier*                 pImageMemoryBarriers);
  • commandBuffer is the command buffer into which the command is recorded.

  • eventCount is the length of the pEvents array.

  • pEvents is an array of event object handles to wait on.

  • srcStageMask is a bitmask of VkPipelineStageFlagBits specifying the source stage mask.

  • dstStageMask is a bitmask of VkPipelineStageFlagBits specifying the destination stage mask.

  • memoryBarrierCount is the length of the pMemoryBarriers array.

  • pMemoryBarriers is a pointer to an array of VkMemoryBarrier structures.

  • bufferMemoryBarrierCount is the length of the pBufferMemoryBarriers array.

  • pBufferMemoryBarriers is a pointer to an array of VkBufferMemoryBarrier structures.

  • imageMemoryBarrierCount is the length of the pImageMemoryBarriers array.

  • pImageMemoryBarriers is a pointer to an array of VkImageMemoryBarrier structures.

When vkCmdWaitEvents is submitted to a queue, it defines a memory dependency between prior event signal operations on the same queue or the host, and subsequent commands. vkCmdWaitEvents must not be used to wait on event signal operations occuring on other queues.

The first synchronization scope only includes event signal operations that operate on members of pEvents, and the operations that happened-before the event signal operations. Event signal operations performed by vkCmdSetEvent that occur earlier in submission order are included in the first synchronization scope, if the logically latest pipeline stage in their stageMask parameter is logically earlier than or equal to the logically latest pipeline stage in srcStageMask. Event signal operations performed by vkSetEvent are only included in the first synchronization scope if VK_PIPELINE_STAGE_HOST_BIT is included in srcStageMask.

The second synchronization scope includes all commands that occur later in submission order. The second synchronization scope is limited to operations on the pipeline stages determined by the destination stage mask specified by dstStageMask.

The first access scope is limited to access in the pipeline stages determined by the source stage mask specified by srcStageMask. Within that, the first access scope only includes the first access scopes defined by elements of the pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers arrays, which each define a set of memory barriers. If no memory barriers are specified, then the first access scope includes no accesses.

The second access scope is limited to access in the pipeline stages determined by the destination stage mask specified by dstStageMask. Within that, the second access scope only includes the second access scopes defined by elements of the pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers arrays, which each define a set of memory barriers. If no memory barriers are specified, then the second access scope includes no accesses.

Note

vkCmdWaitEvents is used with vkCmdSetEvent to define a memory dependency between two sets of action commands, roughly in the same way as pipeline barriers, but split into two commands such that work between the two may execute unhindered.

Note

Applications should be careful to avoid race conditions when using events. There is no direct ordering guarantee between a vkCmdResetEvent command and a vkCmdWaitEvents command submitted after it, so some other execution dependency must be included between these commands (e.g. a semaphore).

Valid Usage
  • srcStageMask must be the bitwise OR of the stageMask parameter used in previous calls to vkCmdSetEvent with any of the members of pEvents and VK_PIPELINE_STAGE_HOST_BIT if any of the members of pEvents was set using vkSetEvent

  • If the geometry shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the geometry shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • If the tessellation shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • If pEvents includes one or more events that will be signaled by vkSetEvent after commandBuffer has been submitted to a queue, then vkCmdWaitEvents must not be called inside a render pass instance

  • Any pipeline stage included in srcStageMask or dstStageMask must be supported by the capabilities of the queue family specified by the queueFamilyIndex member of the VkCommandPoolCreateInfo structure that was used to create the VkCommandPool that commandBuffer was allocated from, as specified in the table of supported pipeline stages.

  • Each element of pMemoryBarriers, pBufferMemoryBarriers or pImageMemoryBarriers must not have any access flag included in its srcAccessMask member if that bit is not supported by any of the pipeline stages in srcStageMask, as specified in the table of supported access types.

  • Each element of pMemoryBarriers, pBufferMemoryBarriers or pImageMemoryBarriers must not have any access flag included in its dstAccessMask member if that bit is not supported by any of the pipeline stages in dstStageMask, as specified in the table of supported access types.

  • commandBuffer’s current device mask must include exactly one physical device.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pEvents must be a valid pointer to an array of eventCount valid VkEvent handles

  • srcStageMask must be a valid combination of VkPipelineStageFlagBits values

  • srcStageMask must not be 0

  • dstStageMask must be a valid combination of VkPipelineStageFlagBits values

  • dstStageMask must not be 0

  • If memoryBarrierCount is not 0, pMemoryBarriers must be a valid pointer to an array of memoryBarrierCount valid VkMemoryBarrier structures

  • If bufferMemoryBarrierCount is not 0, pBufferMemoryBarriers must be a valid pointer to an array of bufferMemoryBarrierCount valid VkBufferMemoryBarrier structures

  • If imageMemoryBarrierCount is not 0, pImageMemoryBarriers must be a valid pointer to an array of imageMemoryBarrierCount valid VkImageMemoryBarrier structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • eventCount must be greater than 0

  • Both of commandBuffer, and the elements of pEvents must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

6.6. Pipeline Barriers

vkCmdPipelineBarrier is a synchronization command that inserts a dependency between commands submitted to the same queue, or between commands in the same subpass.

To record a pipeline barrier, call:

void vkCmdPipelineBarrier(
    VkCommandBuffer                             commandBuffer,
    VkPipelineStageFlags                        srcStageMask,
    VkPipelineStageFlags                        dstStageMask,
    VkDependencyFlags                           dependencyFlags,
    uint32_t                                    memoryBarrierCount,
    const VkMemoryBarrier*                      pMemoryBarriers,
    uint32_t                                    bufferMemoryBarrierCount,
    const VkBufferMemoryBarrier*                pBufferMemoryBarriers,
    uint32_t                                    imageMemoryBarrierCount,
    const VkImageMemoryBarrier*                 pImageMemoryBarriers);
  • commandBuffer is the command buffer into which the command is recorded.

  • srcStageMask is a bitmask of VkPipelineStageFlagBits specifying the source stage mask.

  • dstStageMask is a bitmask of VkPipelineStageFlagBits specifying the destination stage mask.

  • dependencyFlags is a bitmask of VkDependencyFlagBits specifying how execution and memory dependencies are formed.

  • memoryBarrierCount is the length of the pMemoryBarriers array.

  • pMemoryBarriers is a pointer to an array of VkMemoryBarrier structures.

  • bufferMemoryBarrierCount is the length of the pBufferMemoryBarriers array.

  • pBufferMemoryBarriers is a pointer to an array of VkBufferMemoryBarrier structures.

  • imageMemoryBarrierCount is the length of the pImageMemoryBarriers array.

  • pImageMemoryBarriers is a pointer to an array of VkImageMemoryBarrier structures.

When vkCmdPipelineBarrier is submitted to a queue, it defines a memory dependency between commands that were submitted before it, and those submitted after it.

If vkCmdPipelineBarrier was recorded outside a render pass instance, the first synchronization scope includes all commands that occur earlier in submission order. If vkCmdPipelineBarrier was recorded inside a render pass instance, the first synchronization scope includes only commands that occur earlier in submission order within the same subpass. In either case, the first synchronization scope is limited to operations on the pipeline stages determined by the source stage mask specified by srcStageMask.

If vkCmdPipelineBarrier was recorded outside a render pass instance, the second synchronization scope includes all commands that occur later in submission order. If vkCmdPipelineBarrier was recorded inside a render pass instance, the second synchronization scope includes only commands that occur later in submission order within the same subpass. In either case, the second synchronization scope is limited to operations on the pipeline stages determined by the destination stage mask specified by dstStageMask.

The first access scope is limited to access in the pipeline stages determined by the source stage mask specified by srcStageMask. Within that, the first access scope only includes the first access scopes defined by elements of the pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers arrays, which each define a set of memory barriers. If no memory barriers are specified, then the first access scope includes no accesses.

The second access scope is limited to access in the pipeline stages determined by the destination stage mask specified by dstStageMask. Within that, the second access scope only includes the second access scopes defined by elements of the pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers arrays, which each define a set of memory barriers. If no memory barriers are specified, then the second access scope includes no accesses.

If dependencyFlags includes VK_DEPENDENCY_BY_REGION_BIT, then any dependency between framebuffer-space pipeline stages is framebuffer-local - otherwise it is framebuffer-global.

Valid Usage
  • If the geometry shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the geometry shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • If the tessellation shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • If vkCmdPipelineBarrier is called within a render pass instance, the render pass must have been created with a VkSubpassDependency instance in pDependencies that expresses a dependency from the current subpass to itself.

  • If vkCmdPipelineBarrier is called within a render pass instance, srcStageMask must contain a subset of the bit values in the srcStageMask member of that instance of VkSubpassDependency

  • If vkCmdPipelineBarrier is called within a render pass instance, dstStageMask must contain a subset of the bit values in the dstStageMask member of that instance of VkSubpassDependency

  • If vkCmdPipelineBarrier is called within a render pass instance, the srcAccessMask of any element of pMemoryBarriers or pImageMemoryBarriers must contain a subset of the bit values the srcAccessMask member of that instance of VkSubpassDependency

  • If vkCmdPipelineBarrier is called within a render pass instance, the dstAccessMask of any element of pMemoryBarriers or pImageMemoryBarriers must contain a subset of the bit values the dstAccessMask member of that instance of VkSubpassDependency

  • If vkCmdPipelineBarrier is called within a render pass instance, dependencyFlags must be equal to the dependencyFlags member of that instance of VkSubpassDependency

  • If vkCmdPipelineBarrier is called within a render pass instance, bufferMemoryBarrierCount must be 0

  • If vkCmdPipelineBarrier is called within a render pass instance, the image member of any element of pImageMemoryBarriers must be equal to one of the elements of pAttachments that the current framebuffer was created with, that is also referred to by one of the elements of the pColorAttachments, pResolveAttachments or pDepthStencilAttachment members of the VkSubpassDescription instance that the current subpass was created with

  • If vkCmdPipelineBarrier is called within a render pass instance, the oldLayout and newLayout members of any element of pImageMemoryBarriers must be equal to the layout member of an element of the pColorAttachments, pResolveAttachments or pDepthStencilAttachment members of the VkSubpassDescription instance that the current subpass was created with, that refers to the same image

  • If vkCmdPipelineBarrier is called within a render pass instance, the oldLayout and newLayout members of an element of pImageMemoryBarriers must be equal

  • If vkCmdPipelineBarrier is called within a render pass instance, the srcQueueFamilyIndex and dstQueueFamilyIndex members of any element of pImageMemoryBarriers must be VK_QUEUE_FAMILY_IGNORED

  • Any pipeline stage included in srcStageMask or dstStageMask must be supported by the capabilities of the queue family specified by the queueFamilyIndex member of the VkCommandPoolCreateInfo structure that was used to create the VkCommandPool that commandBuffer was allocated from, as specified in the table of supported pipeline stages.

  • Each element of pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers must not have any access flag included in its srcAccessMask member if that bit is not supported by any of the pipeline stages in srcStageMask, as specified in the table of supported access types.

  • Each element of pMemoryBarriers, pBufferMemoryBarriers and pImageMemoryBarriers must not have any access flag included in its dstAccessMask member if that bit is not supported by any of the pipeline stages in dstStageMask, as specified in the table of supported access types.

  • If vkCmdPipelineBarrier is called outside of a render pass instance, dependencyFlags must not include VK_DEPENDENCY_VIEW_LOCAL_BIT

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcStageMask must be a valid combination of VkPipelineStageFlagBits values

  • srcStageMask must not be 0

  • dstStageMask must be a valid combination of VkPipelineStageFlagBits values

  • dstStageMask must not be 0

  • dependencyFlags must be a valid combination of VkDependencyFlagBits values

  • If memoryBarrierCount is not 0, pMemoryBarriers must be a valid pointer to an array of memoryBarrierCount valid VkMemoryBarrier structures

  • If bufferMemoryBarrierCount is not 0, pBufferMemoryBarriers must be a valid pointer to an array of bufferMemoryBarrierCount valid VkBufferMemoryBarrier structures

  • If imageMemoryBarrierCount is not 0, pImageMemoryBarriers must be a valid pointer to an array of imageMemoryBarrierCount valid VkImageMemoryBarrier structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Transfer
Graphics
Compute

Bits which can be set in vkCmdPipelineBarrier::dependencyFlags, specifying how execution and memory dependencies are formed, are:

typedef enum VkDependencyFlagBits {
    VK_DEPENDENCY_BY_REGION_BIT = 0x00000001,
    VK_DEPENDENCY_DEVICE_GROUP_BIT = 0x00000004,
    VK_DEPENDENCY_VIEW_LOCAL_BIT = 0x00000002,
    VK_DEPENDENCY_VIEW_LOCAL_BIT_KHR = VK_DEPENDENCY_VIEW_LOCAL_BIT,
    VK_DEPENDENCY_DEVICE_GROUP_BIT_KHR = VK_DEPENDENCY_DEVICE_GROUP_BIT,
} VkDependencyFlagBits;
typedef VkFlags VkDependencyFlags;

VkDependencyFlags is a bitmask type for setting a mask of zero or more VkDependencyFlagBits.

6.6.1. Subpass Self-dependency

If vkCmdPipelineBarrier is called inside a render pass instance, the following restrictions apply. For a given subpass to allow a pipeline barrier, the render pass must declare a self-dependency from that subpass to itself. That is, there must exist a VkSubpassDependency in the subpass dependency list for the render pass with srcSubpass and dstSubpass equal to that subpass index. More than one self-dependency can be declared for each subpass. Self-dependencies must only include pipeline stage bits that are graphics stages. Self-dependencies must not have any earlier pipeline stages depend on any later pipeline stages (according to the order of graphics pipeline stages), unless all of the stages are framebuffer-space stages. If the source and destination stage masks both include framebuffer-space stages, then dependencyFlags must include VK_DEPENDENCY_BY_REGION_BIT. If the subpass has more than one view, then dependencyFlags must include VK_DEPENDENCY_VIEW_LOCAL_BIT.

A vkCmdPipelineBarrier command inside a render pass instance must be a subset of one of the self-dependencies of the subpass it is used in, meaning that the stage masks and access masks must each include only a subset of the bits of the corresponding mask in that self-dependency. If the self-dependency has VK_DEPENDENCY_BY_REGION_BIT or VK_DEPENDENCY_VIEW_LOCAL_BIT set, then so must the pipeline barrier. Pipeline barriers within a render pass instance can only be types VkMemoryBarrier or VkImageMemoryBarrier. If a VkImageMemoryBarrier is used, the image and image subresource range specified in the barrier must be a subset of one of the image views used by the framebuffer in the current subpass. Additionally, oldLayout must be equal to newLayout, and both the srcQueueFamilyIndex and dstQueueFamilyIndex must be VK_QUEUE_FAMILY_IGNORED.

6.7. Memory Barriers

Memory barriers are used to explicitly control access to buffer and image subresource ranges. Memory barriers are used to transfer ownership between queue families, change image layouts, and define availability and visibility operations. They explicitly define the access types and buffer and image subresource ranges that are included in the access scopes of a memory dependency that is created by a synchronization command that includes them.

6.7.1. Global Memory Barriers

Global memory barriers apply to memory accesses involving all memory objects that exist at the time of its execution.

The VkMemoryBarrier structure is defined as:

typedef struct VkMemoryBarrier {
    VkStructureType    sType;
    const void*        pNext;
    VkAccessFlags      srcAccessMask;
    VkAccessFlags      dstAccessMask;
} VkMemoryBarrier;

The first access scope is limited to access types in the source access mask specified by srcAccessMask.

The second access scope is limited to access types in the destination access mask specified by dstAccessMask.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_BARRIER

  • pNext must be NULL

  • srcAccessMask must be a valid combination of VkAccessFlagBits values

  • dstAccessMask must be a valid combination of VkAccessFlagBits values

6.7.2. Buffer Memory Barriers

Buffer memory barriers only apply to memory accesses involving a specific buffer range. That is, a memory dependency formed from an buffer memory barrier is scoped to access via the specified buffer range. Buffer memory barriers can also be used to define a queue family ownership transfer for the specified buffer range.

The VkBufferMemoryBarrier structure is defined as:

typedef struct VkBufferMemoryBarrier {
    VkStructureType    sType;
    const void*        pNext;
    VkAccessFlags      srcAccessMask;
    VkAccessFlags      dstAccessMask;
    uint32_t           srcQueueFamilyIndex;
    uint32_t           dstQueueFamilyIndex;
    VkBuffer           buffer;
    VkDeviceSize       offset;
    VkDeviceSize       size;
} VkBufferMemoryBarrier;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • srcAccessMask is a bitmask of VkAccessFlagBits specifying a source access mask.

  • dstAccessMask is a bitmask of VkAccessFlagBits specifying a destination access mask.

  • srcQueueFamilyIndex is the source queue family for a queue family ownership transfer.

  • dstQueueFamilyIndex is the destination queue family for a queue family ownership transfer.

  • buffer is a handle to the buffer whose backing memory is affected by the barrier.

  • offset is an offset in bytes into the backing memory for buffer; this is relative to the base offset as bound to the buffer (see vkBindBufferMemory).

  • size is a size in bytes of the affected area of backing memory for buffer, or VK_WHOLE_SIZE to use the range from offset to the end of the buffer.

The first access scope is limited to access to memory through the specified buffer range, via access types in the source access mask specified by srcAccessMask. If srcAccessMask includes VK_ACCESS_HOST_WRITE_BIT, memory writes performed by that access type are also made visible, as that access type is not performed through a resource.

The second access scope is limited to access to memory through the specified buffer range, via access types in the destination access mask. specified by dstAccessMask. If dstAccessMask includes VK_ACCESS_HOST_WRITE_BIT or VK_ACCESS_HOST_READ_BIT, available memory writes are also made visible to accesses of those types, as those access types are not performed through a resource.

If srcQueueFamilyIndex is not equal to dstQueueFamilyIndex, and srcQueueFamilyIndex is equal to the current queue family, then the memory barrier defines a queue family release operation for the specified buffer range, and the second access scope includes no access, as if dstAccessMask was 0.

If dstQueueFamilyIndex is not equal to srcQueueFamilyIndex, and dstQueueFamilyIndex is equal to the current queue family, then the memory barrier defines a queue family acquire operation for the specified buffer range, and the first access scope includes no access, as if srcAccessMask was 0.

Valid Usage
  • offset must be less than the size of buffer

  • If size is not equal to VK_WHOLE_SIZE, size must be greater than 0

  • If size is not equal to VK_WHOLE_SIZE, size must be less than or equal to than the size of buffer minus offset

  • If buffer was created with a sharing mode of VK_SHARING_MODE_CONCURRENT, at least one of srcQueueFamilyIndex and dstQueueFamilyIndex must be VK_QUEUE_FAMILY_IGNORED

  • If buffer was created with a sharing mode of VK_SHARING_MODE_CONCURRENT, and one of srcQueueFamilyIndex and dstQueueFamilyIndex is VK_QUEUE_FAMILY_IGNORED, the other must be VK_QUEUE_FAMILY_IGNORED or a special queue family reserved for external memory ownership transfers, as described in Queue Family Ownership Transfer.

  • If buffer was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and srcQueueFamilyIndex is VK_QUEUE_FAMILY_IGNORED, dstQueueFamilyIndex must also be VK_QUEUE_FAMILY_IGNORED

  • If buffer was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and srcQueueFamilyIndex is not VK_QUEUE_FAMILY_IGNORED, it must be a valid queue family or a special queue family reserved for external memory transfers, as described in Queue Family Ownership Transfer.

  • If buffer was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and dstQueueFamilyIndex is not VK_QUEUE_FAMILY_IGNORED, it must be a valid queue family or a special queue family reserved for external memory transfers, as described in Queue Family Ownership Transfer.

  • If buffer was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex and dstQueueFamilyIndex are not VK_QUEUE_FAMILY_IGNORED, at least one of them must be the same as the family of the queue that will execute this barrier

  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BUFFER_MEMORY_BARRIER

  • pNext must be NULL

  • srcAccessMask must be a valid combination of VkAccessFlagBits values

  • dstAccessMask must be a valid combination of VkAccessFlagBits values

  • buffer must be a valid VkBuffer handle

6.7.3. Image Memory Barriers

Image memory barriers only apply to memory accesses involving a specific image subresource range. That is, a memory dependency formed from an image memory barrier is scoped to access via the specified image subresource range. Image memory barriers can also be used to define image layout transitions or a queue family ownership transfer for the specified image subresource range.

The VkImageMemoryBarrier structure is defined as:

typedef struct VkImageMemoryBarrier {
    VkStructureType            sType;
    const void*                pNext;
    VkAccessFlags              srcAccessMask;
    VkAccessFlags              dstAccessMask;
    VkImageLayout              oldLayout;
    VkImageLayout              newLayout;
    uint32_t                   srcQueueFamilyIndex;
    uint32_t                   dstQueueFamilyIndex;
    VkImage                    image;
    VkImageSubresourceRange    subresourceRange;
} VkImageMemoryBarrier;

The first access scope is limited to access to memory through the specified image subresource range, via access types in the source access mask specified by srcAccessMask. If srcAccessMask includes VK_ACCESS_HOST_WRITE_BIT, memory writes performed by that access type are also made visible, as that access type is not performed through a resource.

The second access scope is limited to access to memory through the specified image subresource range, via access types in the destination access mask specified by dstAccessMask. If dstAccessMask includes VK_ACCESS_HOST_WRITE_BIT or VK_ACCESS_HOST_READ_BIT, available memory writes are also made visible to accesses of those types, as those access types are not performed through a resource.

If srcQueueFamilyIndex is not equal to dstQueueFamilyIndex, and srcQueueFamilyIndex is equal to the current queue family, then the memory barrier defines a queue family release operation for the specified image subresource range, and the second access scope includes no access, as if dstAccessMask was 0.

If dstQueueFamilyIndex is not equal to srcQueueFamilyIndex, and dstQueueFamilyIndex is equal to the current queue family, then the memory barrier defines a queue family acquire operation for the specified image subresource range, and the first access scope includes no access, as if srcAccessMask was 0.

If oldLayout is not equal to newLayout, then the memory barrier defines an image layout transition for the specified image subresource range.

Layout transitions that are performed via image memory barriers execute in their entirety in submission order, relative to other image layout transitions submitted to the same queue, including those performed by render passes. In effect there is an implicit execution dependency from each such layout transition to all layout transitions previously submitted to the same queue.

The image layout of each image subresource of a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT is dependent on the last sample locations used to render to the image subresource as a depth/stencil attachment, thus when the image member of an VkImageMemoryBarrier is an image created with this flag the application can chain a VkSampleLocationsInfoEXT structure to the pNext chain of VkImageMemoryBarrier to specify the sample locations to use during the image layout transition.

If the VkSampleLocationsInfoEXT structure in the pNext chain of VkImageMemoryBarrier does not match the sample location state last used to render to the image subresource range specified by subresourceRange or if no VkSampleLocationsInfoEXT structure is in the pNext chain of VkImageMemoryBarrier then the contents of the given image subresource range becomes undefined as if oldLayout would equal VK_IMAGE_LAYOUT_UNDEFINED.

If image has a multi-planar format and the image is disjoint, then including VK_IMAGE_ASPECT_COLOR_BIT in the aspectMask member of subresourceRange is equivalent to including VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, and (for three-plane formats only) VK_IMAGE_ASPECT_PLANE_2_BIT.

Valid Usage
  • oldLayout must be VK_IMAGE_LAYOUT_UNDEFINED or the current layout of the image subresources affected by the barrier

  • newLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED

  • If image was created with a sharing mode of VK_SHARING_MODE_CONCURRENT, at least one of srcQueueFamilyIndex and dstQueueFamilyIndex must be VK_QUEUE_FAMILY_IGNORED

  • If image was created with a sharing mode of VK_SHARING_MODE_CONCURRENT, and one of srcQueueFamilyIndex and dstQueueFamilyIndex is VK_QUEUE_FAMILY_IGNORED, the other must be VK_QUEUE_FAMILY_IGNORED or a special queue family reserved for external memory transfers, as described in Queue Family Ownership Transfer.

  • If image was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and srcQueueFamilyIndex is VK_QUEUE_FAMILY_IGNORED, dstQueueFamilyIndex must also be VK_QUEUE_FAMILY_IGNORED.

  • If image was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and srcQueueFamilyIndex is not VK_QUEUE_FAMILY_IGNORED, it must be a valid queue family or a special queue family reserved for external memory transfers, as described in Queue Family Ownership Transfer.

  • If image was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE and dstQueueFamilyIndex is not VK_QUEUE_FAMILY_IGNORED, it must be a valid queue family or a special queue family reserved for external memory transfers, as described in Queue Family Ownership Transfer.

  • If image was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex and dstQueueFamilyIndex are not VK_QUEUE_FAMILY_IGNORED, at least one of them must be the same as the family of the queue that will execute this barrier

  • subresourceRange.baseMipLevel must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • If subresourceRange.levelCount is not VK_REMAINING_MIP_LEVELS, subresourceRange.baseMipLevel + subresourceRange.levelCount must be less than or equal to the mipLevels specified in VkImageCreateInfo when image was created

  • subresourceRange.baseArrayLayer must be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • If subresourceRange.layerCount is not VK_REMAINING_ARRAY_LAYERS, subresourceRange.baseArrayLayer + subresourceRange.layerCount must be less than or equal to the arrayLayers specified in VkImageCreateInfo when image was created

  • If image has a depth/stencil format with both depth and stencil components, then the aspectMask member of subresourceRange must include both VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT

  • If image has a single-plane color format or is not disjoint, then the aspectMask member of subresourceRange must be VK_IMAGE_ASPECT_COLOR_BIT

  • If image has a multi-planar format and the image is disjoint, then the aspectMask member of subresourceRange must include either at least one of VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, and VK_IMAGE_ASPECT_PLANE_2_BIT; or must include VK_IMAGE_ASPECT_COLOR_BIT

  • If image has a multi-planar format with only two planes, then the aspectMask member of subresourceRange must not include VK_IMAGE_ASPECT_PLANE_2_BIT

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL then image must have been created with VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL then image must have been created with VK_IMAGE_USAGE_SAMPLED_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL then image must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT set

  • If either oldLayout or newLayout is VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL then image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT set

  • If image is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER

  • pNext must be NULL or a pointer to a valid instance of VkSampleLocationsInfoEXT

  • srcAccessMask must be a valid combination of VkAccessFlagBits values

  • dstAccessMask must be a valid combination of VkAccessFlagBits values

  • oldLayout must be a valid VkImageLayout value

  • newLayout must be a valid VkImageLayout value

  • image must be a valid VkImage handle

  • subresourceRange must be a valid VkImageSubresourceRange structure

6.7.4. Queue Family Ownership Transfer

Resources created with a VkSharingMode of VK_SHARING_MODE_EXCLUSIVE must have their ownership explicitly transferred from one queue family to another in order to access their content in a well-defined manner on a queue in a different queue family. Resources shared with external APIs or instances using external memory must also explicitly manage ownership transfers between local and external queues (or equivalent constructs in external APIs) regardless of the VkSharingMode specified when creating them. The special queue family index VK_QUEUE_FAMILY_EXTERNAL represents any queue external to the resource’s current Vulkan instance, as long as the queue uses the same underlying physical device or device group and uses the same driver version as the resource’s VkDevice, as indicated by VkPhysicalDeviceIDProperties::deviceUUID and VkPhysicalDeviceIDProperties::driverUUID. The special queue family index VK_QUEUE_FAMILY_FOREIGN_EXT represents any queue external to the resource’s current Vulkan instance, regardless of the queue’s underlying physical device or driver version. This includes, for example, queues for fixed-function image processing devices, media codec devices, and display devices, as well as all queues that use the same underlying physical device (or device group) and driver version as the resource’s VkDevice. If memory dependencies are correctly expressed between uses of such a resource between two queues in different families, but no ownership transfer is defined, the contents of that resource are undefined for any read accesses performed by the second queue family.

Note

If an application does not need the contents of a resource to remain valid when transferring from one queue family to another, then the ownership transfer should be skipped.

Note

Applications should expect transfers to/from VK_QUEUE_FAMILY_FOREIGN_EXT to be more expensive than transfers to/from VK_QUEUE_FAMILY_EXTERNAL_KHR.

A queue family ownership transfer consists of two distinct parts:

  1. Release exclusive ownership from the source queue family

  2. Acquire exclusive ownership for the destination queue family

An application must ensure that these operations occur in the correct order by defining an execution dependency between them, e.g. using a semaphore.

A release operation is used to release exclusive ownership of a range of a buffer or image subresource range. A release operation is defined by executing a buffer memory barrier (for a buffer range) or an image memory barrier (for an image subresource range), on a queue from the source queue family. The srcQueueFamilyIndex parameter of the barrier must be set to the source queue family index, and the dstQueueFamilyIndex parameter to the destination queue family index. dstStageMask is ignored for such a barrier, such that no visibility operation is executed - the value of this mask does not affect the validity of the barrier. The release operation happens-after the availability operation.

An acquire operation is used to acquire exclusive ownership of a range of a buffer or image subresource range. An acquire operation is defined by executing a buffer memory barrier (for a buffer range) or an image memory barrier (for an image subresource range), on a queue from the destination queue family. The srcQueueFamilyIndex parameter of the barrier must be set to the source queue family index, and the dstQueueFamilyIndex parameter to the destination queue family index. srcStageMask is ignored for such a barrier, such that no availability operation is executed - the value of this mask does not affect the validity of the barrier. The acquire operation happens-before the visibility operation.

Note

Whilst it is not invalid to provide destination or source access masks for memory barriers used for release or acquire operations, respectively, they have no practical effect. Access after a release operation has undefined results, and so visibility for those accesses has no practical effect. Similarly, write access before an acquire operation will produce undefined results for future access, so availability of those writes has no practical use. In an earlier version of the specification, these were required to match on both sides - but this was subsequently relaxed. These masks should be set to 0.

If the transfer is via an image memory barrier, and an image layout transition is desired, then the values of oldLayout and newLayout in the release memory barrier must be equal to values of oldLayout and newLayout in the acquire memory barrier. Although the image layout transition is submitted twice, it will only be executed once. A layout transition specified in this way happens-after the release operation and happens-before the acquire operation.

If the values of srcQueueFamilyIndex and dstQueueFamilyIndex are equal, no ownership transfer is performed, and the barrier operates as if they were both set to VK_QUEUE_FAMILY_IGNORED.

Queue family ownership transfers may perform read and write accesses on all memory bound to the image subresource or buffer range, so applications must ensure that all memory writes have been made available before a queue family ownership transfer is executed. Available memory is automatically made visible to queue family release and acquire operations, and writes performed by those operations are automatically made available.

Once a queue family has acquired ownership of a buffer range or image subresource range of an VK_SHARING_MODE_EXCLUSIVE resource, its contents are undefined to other queue families unless ownership is transferred. The contents of any portion of another resource which aliases memory that is bound to the transferred buffer or image subresource range are undefined after a release or acquire operation.

6.8. Wait Idle Operations

To wait on the host for the completion of outstanding queue operations for a given queue, call:

VkResult vkQueueWaitIdle(
    VkQueue                                     queue);
  • queue is the queue on which to wait.

vkQueueWaitIdle is equivalent to submitting a fence to a queue and waiting with an infinite timeout for that fence to signal.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

To wait on the host for the completion of outstanding queue operations for all queues on a given logical device, call:

VkResult vkDeviceWaitIdle(
    VkDevice                                    device);
  • device is the logical device to idle.

vkDeviceWaitIdle is equivalent to calling vkQueueWaitIdle for all queues owned by device.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

Host Synchronization
  • Host access to all VkQueue objects created from device must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

6.9. Host Write Ordering Guarantees

When batches of command buffers are submitted to a queue via vkQueueSubmit, it defines a memory dependency with prior host operations, and execution of command buffers submitted to the queue.

The first synchronization scope is defined by the host execution model, but includes execution of vkQueueSubmit on the host and anything that happened-before it.

The second synchronization scope includes all commands submitted in the same queue submission, and all commands that occur later in submission order.

The first access scope includes all host writes to mappable device memory that are either coherent, or have been flushed with vkFlushMappedMemoryRanges.

The second access scope includes all memory access performed by the device.

6.10. Synchronization and Multiple Physical Devices

If a logical device includes more than one physical device, then fences, semaphores, and events all still have a single instance of the signaled state.

A fence becomes signaled when all physical devices complete the necessary queue operations.

Semaphore wait and signal operations all include a device index that is the sole physical device that performs the operation. These indices are provided in the VkDeviceGroupSubmitInfo and VkDeviceGroupBindSparseInfo structures. Semaphores are not exclusively owned by any physical device. For example, a semaphore can be signaled by one physical device and then waited on by a different physical device.

An event can only be waited on by the same physical device that signaled it (or the host).

7. Render Pass

A render pass represents a collection of attachments, subpasses, and dependencies between the subpasses, and describes how the attachments are used over the course of the subpasses. The use of a render pass in a command buffer is a render pass instance.

Render passes are represented by VkRenderPass handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkRenderPass)

An attachment description describes the properties of an attachment including its format, sample count, and how its contents are treated at the beginning and end of each render pass instance.

A subpass represents a phase of rendering that reads and writes a subset of the attachments in a render pass. Rendering commands are recorded into a particular subpass of a render pass instance.

A subpass description describes the subset of attachments that is involved in the execution of a subpass. Each subpass can read from some attachments as input attachments, write to some as color attachments or depth/stencil attachments, and perform multisample resolve operations to resolve attachments. A subpass description can also include a set of preserve attachments, which are attachments that are not read or written by the subpass but whose contents must be preserved throughout the subpass.

A subpass uses an attachment if the attachment is a color, depth/stencil, resolve, or input attachment for that subpass (as determined by the pColorAttachments, pDepthStencilAttachment, pResolveAttachments, and pInputAttachments members of VkSubpassDescription, respectively). A subpass does not use an attachment if that attachment is preserved by the subpass. The first use of an attachment is in the lowest numbered subpass that uses that attachment. Similarly, the last use of an attachment is in the highest numbered subpass that uses that attachment.

The subpasses in a render pass all render to the same dimensions, and fragments for pixel (x,y,layer) in one subpass can only read attachment contents written by previous subpasses at that same (x,y,layer) location.

Note

By describing a complete set of subpasses in advance, render passes provide the implementation an opportunity to optimize the storage and transfer of attachment data between subpasses.

In practice, this means that subpasses with a simple framebuffer-space dependency may be merged into a single tiled rendering pass, keeping the attachment data on-chip for the duration of a render pass instance. However, it is also quite common for a render pass to only contain a single subpass.

Subpass dependencies describe execution and memory dependencies between subpasses.

A subpass dependency chain is a sequence of subpass dependencies in a render pass, where the source subpass of each subpass dependency (after the first) equals the destination subpass of the previous dependency.

Execution of subpasses may overlap or execute out of order with regards to other subpasses, unless otherwise enforced by an execution dependency. Each subpass only respects submission order for commands recorded in the same subpass, and the vkCmdBeginRenderPass and vkCmdEndRenderPass commands that delimit the render pass - commands within other subpasses are not included. This affects most other implicit ordering guarantees.

A render pass describes the structure of subpasses and attachments independent of any specific image views for the attachments. The specific image views that will be used for the attachments, and their dimensions, are specified in VkFramebuffer objects. Framebuffers are created with respect to a specific render pass that the framebuffer is compatible with (see Render Pass Compatibility). Collectively, a render pass and a framebuffer define the complete render target state for one or more subpasses as well as the algorithmic dependencies between the subpasses.

The various pipeline stages of the drawing commands for a given subpass may execute concurrently and/or out of order, both within and across drawing commands, whilst still respecting pipeline order. However for a given (x,y,layer,sample) sample location, certain per-sample operations are performed in rasterization order.

7.1. Render Pass Creation

To create a render pass, call:

VkResult vkCreateRenderPass(
    VkDevice                                    device,
    const VkRenderPassCreateInfo*               pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkRenderPass*                               pRenderPass);
  • device is the logical device that creates the render pass.

  • pCreateInfo is a pointer to an instance of the VkRenderPassCreateInfo structure that describes the parameters of the render pass.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pRenderPass points to a VkRenderPass handle in which the resulting render pass object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkRenderPassCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pRenderPass must be a valid pointer to a VkRenderPass handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkRenderPassCreateInfo structure is defined as:

typedef struct VkRenderPassCreateInfo {
    VkStructureType                   sType;
    const void*                       pNext;
    VkRenderPassCreateFlags           flags;
    uint32_t                          attachmentCount;
    const VkAttachmentDescription*    pAttachments;
    uint32_t                          subpassCount;
    const VkSubpassDescription*       pSubpasses;
    uint32_t                          dependencyCount;
    const VkSubpassDependency*        pDependencies;
} VkRenderPassCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • attachmentCount is the number of attachments used by this render pass, or zero indicating no attachments. Attachments are referred to by zero-based indices in the range [0,attachmentCount).

  • pAttachments points to an array of attachmentCount number of VkAttachmentDescription structures describing properties of the attachments, or NULL if attachmentCount is zero.

  • subpassCount is the number of subpasses to create for this render pass. Subpasses are referred to by zero-based indices in the range [0,subpassCount). A render pass must have at least one subpass.

  • pSubpasses points to an array of subpassCount number of VkSubpassDescription structures describing properties of the subpasses.

  • dependencyCount is the number of dependencies between pairs of subpasses, or zero indicating no dependencies.

  • pDependencies points to an array of dependencyCount number of VkSubpassDependency structures describing dependencies between pairs of subpasses, or NULL if dependencyCount is zero.

Valid Usage
  • If any two subpasses operate on attachments with overlapping ranges of the same VkDeviceMemory object, and at least one subpass writes to that area of VkDeviceMemory, a subpass dependency must be included (either directly or via some intermediate subpasses) between them

  • If the attachment member of any element of pInputAttachments, pColorAttachments, pResolveAttachments or pDepthStencilAttachment, or the attachment indexed by any element of pPreserveAttachments in any element of pSubpasses is bound to a range of a VkDeviceMemory object that overlaps with any other attachment in any subpass (including the same subpass), the VkAttachmentDescription structures describing them must include VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT in flags

  • If the attachment member of any element of pInputAttachments, pColorAttachments, pResolveAttachments or pDepthStencilAttachment, or any element of pPreserveAttachments in any element of pSubpasses is not VK_ATTACHMENT_UNUSED, it must be less than attachmentCount

  • The value of each element of the pPreserveAttachments member in each element of pSubpasses must not be VK_ATTACHMENT_UNUSED

  • For any member of pAttachments with a loadOp equal to VK_ATTACHMENT_LOAD_OP_CLEAR, the first use of that attachment must not specify a layout equal to VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL.

  • For any member of pAttachments with a loadOp equal to VK_ATTACHMENT_LOAD_OP_CLEAR, the first use of that attachment must not specify a layout equal to VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL.

  • For any member of pAttachments with a stencilLoadOp equal to VK_ATTACHMENT_LOAD_OP_CLEAR, the first use of that attachment must not specify a layout equal to VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL.

  • If the pNext chain includes an instance of VkRenderPassInputAttachmentAspectCreateInfo, the subpass member of each element of its pAspectReferences member must be less than subpassCount

  • If the pNext chain includes an instance of VkRenderPassInputAttachmentAspectCreateInfo, the inputAttachmentIndex member of each element of its pAspectReferences member must be less than the value of inputAttachmentCount in the member of pSubpasses identified by its subpass member

  • If the pNext chain includes an instance of VkRenderPassInputAttachmentAspectCreateInfo, the aspectMask member of any element of pAspectReferences must only include aspects that are present in images of the format of the input attachment specified by the subpass and inputAttachment of the same element of pAspectReferences

  • If the pNext chain includes an instance of VkRenderPassMultiviewCreateInfo, and its subpassCount member is not zero, that member must be equal to the value of subpassCount

  • If the pNext chain includes an instance of VkRenderPassMultiviewCreateInfo, if its dependencyCount member is not zero, it must be equal to dependencyCount

  • If the pNext chain includes an instance of VkRenderPassMultiviewCreateInfo, for each non-zero element of pViewOffsets, the srcSubpass and dstSubpass members of pDependencies at the same index must not be equal

  • For any element of pDependencies, if the srcSubpass is not VK_SUBPASS_EXTERNAL, all stage flags included in the srcStageMask member of that dependency must be a pipeline stage supported by the pipeline identified by the pipelineBindPoint member of the source subpass.

  • For any element of pDependencies, if the dstSubpass is not VK_SUBPASS_EXTERNAL, all stage flags included in the dstStageMask member of that dependency must be a pipeline stage supported by the pipeline identified by the pipelineBindPoint member of the source subpass.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkRenderPassInputAttachmentAspectCreateInfo or VkRenderPassMultiviewCreateInfo

  • Each sType member in the pNext chain must be unique

  • flags must be 0

  • If attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkAttachmentDescription structures

  • pSubpasses must be a valid pointer to an array of subpassCount valid VkSubpassDescription structures

  • If dependencyCount is not 0, pDependencies must be a valid pointer to an array of dependencyCount valid VkSubpassDependency structures

  • subpassCount must be greater than 0

typedef VkFlags VkRenderPassCreateFlags;

VkRenderPassCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

If the VkRenderPassCreateInfo::pNext chain includes a VkRenderPassMultiviewCreateInfo structure, then that structure includes an array of view masks, view offsets, and correlation masks for the render pass.

The VkRenderPassMultiviewCreateInfo structure is defined as:

typedef struct VkRenderPassMultiviewCreateInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           subpassCount;
    const uint32_t*    pViewMasks;
    uint32_t           dependencyCount;
    const int32_t*     pViewOffsets;
    uint32_t           correlationMaskCount;
    const uint32_t*    pCorrelationMasks;
} VkRenderPassMultiviewCreateInfo;

or the equivalent

typedef VkRenderPassMultiviewCreateInfo VkRenderPassMultiviewCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • subpassCount is zero or is the number of subpasses in the render pass.

  • pViewMasks points to an array of subpassCount number of view masks, where each mask is a bitfield of view indices describing which views rendering is broadcast to in each subpass, when multiview is enabled. If subpassCount is zero, each view mask is treated as zero.

  • dependencyCount is zero or the number of dependencies in the render pass.

  • pViewOffsets points to an array of dependencyCount view offsets, one for each dependency. If dependencyCount is zero, each dependency’s view offset is treated as zero. Each view offset controls which views in the source subpass the views in the destination subpass depend on.

  • correlationMaskCount is zero or a number of correlation masks.

  • pCorrelationMasks is an array of view masks indicating sets of views that may be more efficient to render concurrently.

When a subpass uses a non-zero view mask, multiview functionality is considered to be enabled. Multiview is all-or-nothing for a render pass - that is, either all subpasses must have a non-zero view mask (though some subpasses may have only one view) or all must be zero. Multiview causes all drawing and clear commands in the subpass to behave as if they were broadcast to each view, where a view is represented by one layer of the framebuffer attachments. All draws and clears are broadcast to each view index whose bit is set in the view mask. The view index is provided in the ViewIndex shader input variable, and color, depth/stencil, and input attachments all read/write the layer of the framebuffer corresponding to the view index.

If the view mask is zero for all subpasses, multiview is considered to be disabled and all drawing commands execute normally, without this additional broadcasting.

Some implementations may not support multiview in conjunction with geometry shaders or tessellation shaders.

When multiview is enabled, the VK_DEPENDENCY_VIEW_LOCAL_BIT bit in a dependency can be used to express a view-local dependency, meaning that each view in the destination subpass depends on a single view in the source subpass. Unlike pipeline barriers, a subpass dependency can potentially have a different view mask in the source subpass and the destination subpass. If the dependency is view-local, then each view (dstView) in the destination subpass depends on the view dstView + pViewOffsets[dependency] in the source subpass. If there is not such a view in the source subpass, then this dependency does not affect that view in the destination subpass. If the dependency is not view-local, then all views in the destination subpass depend on all views in the source subpass, and the view offset is ignored. A non-zero view offset is not allowed in a self-dependency.

The elements of pCorrelationMasks are a set of masks of views indicating that views in the same mask may exhibit spatial coherency between the views, making it more efficient to render them concurrently. Correlation masks must not have a functional effect on the results of the multiview rendering.

When multiview is enabled, at the beginning of each subpass all non-render pass state is undefined. In particular, each time vkCmdBeginRenderPass or vkCmdNextSubpass is called the graphics pipeline must be bound, any relevant descriptor sets or vertex/index buffers must be bound, and any relevant dynamic state or push constants must be set before they are used.

A multiview subpass can declare that its shaders will write per-view attributes for all views in a single invocation, by setting the VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX bit in the subpass description. The only supported per-view attributes are position and viewport mask, and per-view position and viewport masks are written to output array variables decorated with PositionPerViewNV and ViewportMaskPerViewNV, respectively. If VK_NV_viewport_array2 is not supported and enabled, ViewportMaskPerViewNV must not be used. Values written to elements of PositionPerViewNV and ViewportMaskPerViewNV must not depend on the ViewIndex. The shader must also write to an output variable decorated with Position, and the value written to Position must equal the value written to PositionPerViewNV[ViewIndex]. Similarly, if ViewportMaskPerViewNV is written to then the shader must also write to an output variable decorated with ViewportMaskNV, and the value written to ViewportMaskNV must equal the value written to ViewportMaskPerViewNV[ViewIndex]. Implementations will either use values taken from Position and ViewportMaskNV and invoke the shader once for each view, or will use values taken from PositionPerViewNV and ViewportMaskPerViewNV and invoke the shader fewer times. The values written to Position and ViewportMaskNV must not depend on the values written to PositionPerViewNV and ViewportMaskPerViewNV, or vice versa (to allow compilers to eliminate the unused outputs). All attributes that do not have *PerViewNV counterparts must not depend on ViewIndex.

Per-view attributes are all-or-nothing for a subpass. That is, all pipelines compiled against a subpass that includes the VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX bit must write per-view attributes to the *PerViewNV[] shader outputs, in addition to the non-per-view (e.g. Position) outputs. Pipelines compiled against a subpass that does not include this bit must not include the *PerViewNV[] outputs in their interfaces.

Valid Usage
  • Each view index must not be set in more than one element of pCorrelationMasks

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO

  • If subpassCount is not 0, pViewMasks must be a valid pointer to an array of subpassCount uint32_t values

  • If dependencyCount is not 0, pViewOffsets must be a valid pointer to an array of dependencyCount int32_t values

  • If correlationMaskCount is not 0, pCorrelationMasks must be a valid pointer to an array of correlationMaskCount uint32_t values

The VkAttachmentDescription structure is defined as:

typedef struct VkAttachmentDescription {
    VkAttachmentDescriptionFlags    flags;
    VkFormat                        format;
    VkSampleCountFlagBits           samples;
    VkAttachmentLoadOp              loadOp;
    VkAttachmentStoreOp             storeOp;
    VkAttachmentLoadOp              stencilLoadOp;
    VkAttachmentStoreOp             stencilStoreOp;
    VkImageLayout                   initialLayout;
    VkImageLayout                   finalLayout;
} VkAttachmentDescription;
  • flags is a bitmask of VkAttachmentDescriptionFlagBits specifying additional properties of the attachment.

  • format is a VkFormat value specifying the format of the image view that will be used for the attachment.

  • samples is the number of samples of the image as defined in VkSampleCountFlagBits.

  • loadOp is a VkAttachmentLoadOp value specifying how the contents of color and depth components of the attachment are treated at the beginning of the subpass where it is first used.

  • storeOp is a VkAttachmentStoreOp value specifying how the contents of color and depth components of the attachment are treated at the end of the subpass where it is last used.

  • stencilLoadOp is a VkAttachmentLoadOp value specifying how the contents of stencil components of the attachment are treated at the beginning of the subpass where it is first used.

  • stencilStoreOp is a VkAttachmentStoreOp value specifying how the contents of stencil components of the attachment are treated at the end of the last subpass where it is used.

  • initialLayout is the layout the attachment image subresource will be in when a render pass instance begins.

  • finalLayout is the layout the attachment image subresource will be transitioned to when a render pass instance ends. During a render pass instance, an attachment can use a different layout in each subpass, if desired.

If the attachment uses a color format, then loadOp and storeOp are used, and stencilLoadOp and stencilStoreOp are ignored. If the format has depth and/or stencil components, loadOp and storeOp apply only to the depth data, while stencilLoadOp and stencilStoreOp define how the stencil data is handled. loadOp and stencilLoadOp define the load operations that execute as part of the first subpass that uses the attachment. storeOp and stencilStoreOp define the store operations that execute as part of the last subpass that uses the attachment.

The load operation for each sample in an attachment happens-before any recorded command which accesses the sample in the first subpass where the attachment is used. Load operations for attachments with a depth/stencil format execute in the VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT pipeline stage. Load operations for attachments with a color format execute in the VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT pipeline stage.

The store operation for each sample in an attachment happens-after any recorded command which accesses the sample in the last subpass where the attachment is used. Store operations for attachments with a depth/stencil format execute in the VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT pipeline stage. Store operations for attachments with a color format execute in the VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT pipeline stage.

If an attachment is not used by any subpass, then loadOp, storeOp, stencilStoreOp, and stencilLoadOp are ignored, and the attachment’s memory contents will not be modified by execution of a render pass instance.

The load and store operations apply on the first and last use of each view in the render pass, respectively. If a view index of an attachment is not included in the view mask in any subpass that uses it, then the load and store operations are ignored, and the attachment’s memory contents will not be modified by execution of a render pass instance.

During a render pass instance, input/color attachments with color formats that have a component size of 8, 16, or 32 bits must be represented in the attachment’s format throughout the instance. Attachments with other floating- or fixed-point color formats, or with depth components may be represented in a format with a precision higher than the attachment format, but must be represented with the same range. When such a component is loaded via the loadOp, it will be converted into an implementation-dependent format used by the render pass. Such components must be converted from the render pass format, to the format of the attachment, before they are resolved or stored at the end of a render pass instance via storeOp. Conversions occur as described in Numeric Representation and Computation and Fixed-Point Data Conversions.

If flags includes VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT, then the attachment is treated as if it shares physical memory with another attachment in the same render pass. This information limits the ability of the implementation to reorder certain operations (like layout transitions and the loadOp) such that it is not improperly reordered against other uses of the same physical memory via a different attachment. This is described in more detail below.

Valid Usage
  • finalLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED

Valid Usage (Implicit)

To specify which aspects of an input attachment can be read add a VkRenderPassInputAttachmentAspectCreateInfo structure to the pNext chain of the VkRenderPassCreateInfo structure:

The VkRenderPassInputAttachmentAspectCreateInfo structure is defined as:

typedef struct VkRenderPassInputAttachmentAspectCreateInfo {
    VkStructureType                            sType;
    const void*                                pNext;
    uint32_t                                   aspectReferenceCount;
    const VkInputAttachmentAspectReference*    pAspectReferences;
} VkRenderPassInputAttachmentAspectCreateInfo;

or the equivalent

typedef VkRenderPassInputAttachmentAspectCreateInfo VkRenderPassInputAttachmentAspectCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • aspectReferenceCount is the number of elements in the pAspectReferences array.

  • pAspectReferences points to an array of aspectReferenceCount number of VkInputAttachmentAspectReference structures describing which aspect(s) can be accessed for a given input attachment within a given subpass.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO

  • pAspectReferences must be a valid pointer to an array of aspectReferenceCount valid VkInputAttachmentAspectReference structures

  • aspectReferenceCount must be greater than 0

The VkInputAttachmentAspectReference structure specifies an aspect mask for a specific input attachment of a specific subpass in the render pass.

subpass and inputAttachmentIndex index into the render pass as:

pCreateInfo::pSubpasses[subpass].pInputAttachments[inputAttachmentIndex]

typedef struct VkInputAttachmentAspectReference {
    uint32_t              subpass;
    uint32_t              inputAttachmentIndex;
    VkImageAspectFlags    aspectMask;
} VkInputAttachmentAspectReference;

or the equivalent

typedef VkInputAttachmentAspectReference VkInputAttachmentAspectReferenceKHR;
  • subpass is an index into the pSubpasses array of the parent VkRenderPassCreateInfo structure.

  • inputAttachmentIndex is an index into the pInputAttachments of the specified subpass.

  • aspectMask is a mask of which aspect(s) can be accessed within the specified subpass.

Valid Usage
  • aspectMask must not include VK_IMAGE_ASPECT_METADATA_BIT

Valid Usage (Implicit)
editing-note

TODO (Jon) - it’s unclear whether the following two paragraphs are intended to apply to VkAttachmentDescription, one of the extension structures described immediately above, or something else. The following description of VkAttachmentDescriptionFlagBits should probably be moved up to near VkAttachmentDescription.

An application must only access the specified aspect(s).

An application can access any aspect of an input attachment that does not have a specified aspect mask.

Bits which can be set in VkAttachmentDescription::flags describing additional properties of the attachment are:

typedef enum VkAttachmentDescriptionFlagBits {
    VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT = 0x00000001,
} VkAttachmentDescriptionFlagBits;
  • VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT specifies that the attachment aliases the same device memory as other attachments.

typedef VkFlags VkAttachmentDescriptionFlags;

VkAttachmentDescriptionFlags is a bitmask type for setting a mask of zero or more VkAttachmentDescriptionFlagBits.

Possible values of VkAttachmentDescription::loadOp and stencilLoadOp, specifying how the contents of the attachment are treated, are:

typedef enum VkAttachmentLoadOp {
    VK_ATTACHMENT_LOAD_OP_LOAD = 0,
    VK_ATTACHMENT_LOAD_OP_CLEAR = 1,
    VK_ATTACHMENT_LOAD_OP_DONT_CARE = 2,
} VkAttachmentLoadOp;
  • VK_ATTACHMENT_LOAD_OP_LOAD specifies that the previous contents of the image within the render area will be preserved. For attachments with a depth/stencil format, this uses the access type VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT. For attachments with a color format, this uses the access type VK_ACCESS_COLOR_ATTACHMENT_READ_BIT.

  • VK_ATTACHMENT_LOAD_OP_CLEAR specifies that the contents within the render area will be cleared to a uniform value, which is specified when a render pass instance is begun. For attachments with a depth/stencil format, this uses the access type VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT. For attachments with a color format, this uses the access type VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT.

  • VK_ATTACHMENT_LOAD_OP_DONT_CARE specifies that the previous contents within the area need not be preserved; the contents of the attachment will be undefined inside the render area. For attachments with a depth/stencil format, this uses the access type VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT. For attachments with a color format, this uses the access type VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT.

Possible values of VkAttachmentDescription::storeOp and stencilStoreOp, specifying how the contents of the attachment are treated, are:

typedef enum VkAttachmentStoreOp {
    VK_ATTACHMENT_STORE_OP_STORE = 0,
    VK_ATTACHMENT_STORE_OP_DONT_CARE = 1,
} VkAttachmentStoreOp;
  • VK_ATTACHMENT_STORE_OP_STORE specifies the contents generated during the render pass and within the render area are written to memory. For attachments with a depth/stencil format, this uses the access type VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT. For attachments with a color format, this uses the access type VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT.

  • VK_ATTACHMENT_STORE_OP_DONT_CARE specifies the contents within the render area are not needed after rendering, and may be discarded; the contents of the attachment will be undefined inside the render area. For attachments with a depth/stencil format, this uses the access type VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT. For attachments with a color format, this uses the access type VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT.

editing-note

TODO (Jon) - the following text may need to be moved back to combine with vkCreateRenderPass above for automatic ref page generation.

If a render pass uses multiple attachments that alias the same device memory, those attachments must each include the VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT bit in their attachment description flags. Attachments aliasing the same memory occurs in multiple ways:

  • Multiple attachments being assigned the same image view as part of framebuffer creation.

  • Attachments using distinct image views that correspond to the same image subresource of an image.

  • Attachments using views of distinct image subresources which are bound to overlapping memory ranges.

Note

Render passes must include subpass dependencies (either directly or via a subpass dependency chain) between any two subpasses that operate on the same attachment or aliasing attachments and those subpass dependencies must include execution and memory dependencies separating uses of the aliases, if at least one of those subpasses writes to one of the aliases. These dependencies must not include the VK_DEPENDENCY_BY_REGION_BIT if the aliases are views of distinct image subresources which overlap in memory.

Multiple attachments that alias the same memory must not be used in a single subpass. A given attachment index must not be used multiple times in a single subpass, with one exception: two subpass attachments can use the same attachment index if at least one use is as an input attachment and neither use is as a resolve or preserve attachment. In other words, the same view can be used simultaneously as an input and color or depth/stencil attachment, but must not be used as multiple color or depth/stencil attachments nor as resolve or preserve attachments. The precise set of valid scenarios is described in more detail below.

If a set of attachments alias each other, then all except the first to be used in the render pass must use an initialLayout of VK_IMAGE_LAYOUT_UNDEFINED, since the earlier uses of the other aliases make their contents undefined. Once an alias has been used and a different alias has been used after it, the first alias must not be used in any later subpasses. However, an application can assign the same image view to multiple aliasing attachment indices, which allows that image view to be used multiple times even if other aliases are used in between.

Note

Once an attachment needs the VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT bit, there should be no additional cost of introducing additional aliases, and using these additional aliases may allow more efficient clearing of the attachments on multiple uses via VK_ATTACHMENT_LOAD_OP_CLEAR.

The VkSubpassDescription structure is defined as:

typedef struct VkSubpassDescription {
    VkSubpassDescriptionFlags       flags;
    VkPipelineBindPoint             pipelineBindPoint;
    uint32_t                        inputAttachmentCount;
    const VkAttachmentReference*    pInputAttachments;
    uint32_t                        colorAttachmentCount;
    const VkAttachmentReference*    pColorAttachments;
    const VkAttachmentReference*    pResolveAttachments;
    const VkAttachmentReference*    pDepthStencilAttachment;
    uint32_t                        preserveAttachmentCount;
    const uint32_t*                 pPreserveAttachments;
} VkSubpassDescription;
  • flags is a bitmask of VkSubpassDescriptionFlagBits specifying usage of the subpass.

  • pipelineBindPoint is a VkPipelineBindPoint value specifying whether this is a compute or graphics subpass. Currently, only graphics subpasses are supported.

  • inputAttachmentCount is the number of input attachments.

  • pInputAttachments is an array of VkAttachmentReference structures (defined below) that lists which of the render pass’s attachments can be read in the fragment shader stage during the subpass, and what layout each attachment will be in during the subpass. Each element of the array corresponds to an input attachment unit number in the shader, i.e. if the shader declares an input variable layout(input_attachment_index=X, set=Y, binding=Z) then it uses the attachment provided in pInputAttachments[X]. Input attachments must also be bound to the pipeline with a descriptor set, with the input attachment descriptor written in the location (set=Y, binding=Z). Fragment shaders can use subpass input variables to access the contents of an input attachment at the fragment’s (x, y, layer) framebuffer coordinates.

  • colorAttachmentCount is the number of color attachments.

  • pColorAttachments is an array of colorAttachmentCount VkAttachmentReference structures that lists which of the render pass’s attachments will be used as color attachments in the subpass, and what layout each attachment will be in during the subpass. Each element of the array corresponds to a fragment shader output location, i.e. if the shader declared an output variable layout(location=X) then it uses the attachment provided in pColorAttachments[X].

  • pResolveAttachments is NULL or an array of colorAttachmentCount VkAttachmentReference structures that lists which of the render pass’s attachments are resolved to at the end of the subpass, and what layout each attachment will be in during the multisample resolve operation. If pResolveAttachments is not NULL, each of its elements corresponds to a color attachment (the element in pColorAttachments at the same index), and a multisample resolve operation is defined for each attachment. At the end of each subpass, multisample resolve operations read the subpass’s color attachments, and resolve the samples for each pixel to the same pixel location in the corresponding resolve attachments, unless the resolve attachment index is VK_ATTACHMENT_UNUSED. If the first use of an attachment in a render pass is as a resolve attachment, then the loadOp is effectively ignored as the resolve is guaranteed to overwrite all pixels in the render area.

  • pDepthStencilAttachment is a pointer to a VkAttachmentReference specifying which attachment will be used for depth/stencil data and the layout it will be in during the subpass. Setting the attachment index to VK_ATTACHMENT_UNUSED or leaving this pointer as NULL indicates that no depth/stencil attachment will be used in the subpass.

  • preserveAttachmentCount is the number of preserved attachments.

  • pPreserveAttachments is an array of preserveAttachmentCount render pass attachment indices describing the attachments that are not used by a subpass, but whose contents must be preserved throughout the subpass.

The contents of an attachment within the render area become undefined at the start of a subpass S if all of the following conditions are true:

  • The attachment is used as a color, depth/stencil, or resolve attachment in any subpass in the render pass.

  • There is a subpass S1 that uses or preserves the attachment, and a subpass dependency from S1 to S.

  • The attachment is not used or preserved in subpass S.

Once the contents of an attachment become undefined in subpass S, they remain undefined for subpasses in subpass dependency chains starting with subpass S until they are written again. However, they remain valid for subpasses in other subpass dependency chains starting with subpass S1 if those subpasses use or preserve the attachment.

Valid Usage
  • pipelineBindPoint must be VK_PIPELINE_BIND_POINT_GRAPHICS

  • colorAttachmentCount must be less than or equal to VkPhysicalDeviceLimits::maxColorAttachments

  • If the first use of an attachment in this render pass is as an input attachment, and the attachment is not also used as a color or depth/stencil attachment in the same subpass, then loadOp must not be VK_ATTACHMENT_LOAD_OP_CLEAR

  • If pResolveAttachments is not NULL, for each resolve attachment that does not have the value VK_ATTACHMENT_UNUSED, the corresponding color attachment must not have the value VK_ATTACHMENT_UNUSED

  • If pResolveAttachments is not NULL, the sample count of each element of pColorAttachments must be anything other than VK_SAMPLE_COUNT_1_BIT

  • Each element of pResolveAttachments must have a sample count of VK_SAMPLE_COUNT_1_BIT

  • Each element of pResolveAttachments must have the same VkFormat as its corresponding color attachment

  • All attachments in pColorAttachments that are not VK_ATTACHMENT_UNUSED must have the same sample count

  • All attachments in pColorAttachments that are not VK_ATTACHMENT_UNUSED must have a sample count that is smaller than or equal to the sample count of pDepthStencilAttachment if it is not VK_ATTACHMENT_UNUSED

  • If any input attachments are VK_ATTACHMENT_UNUSED, then any pipelines bound during the subpass must not access those input attachments from the fragment shader

  • The attachment member of each element of pPreserveAttachments must not be VK_ATTACHMENT_UNUSED

  • Each element of pPreserveAttachments must not also be an element of any other member of the subpass description

  • If any attachment is used as both an input attachment and a color or depth/stencil attachment, then each use must use the same layout

  • If flags includes VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX, it must also include VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX.

Valid Usage (Implicit)
  • flags must be a valid combination of VkSubpassDescriptionFlagBits values

  • pipelineBindPoint must be a valid VkPipelineBindPoint value

  • If inputAttachmentCount is not 0, pInputAttachments must be a valid pointer to an array of inputAttachmentCount valid VkAttachmentReference structures

  • If colorAttachmentCount is not 0, pColorAttachments must be a valid pointer to an array of colorAttachmentCount valid VkAttachmentReference structures

  • If colorAttachmentCount is not 0, and pResolveAttachments is not NULL, pResolveAttachments must be a valid pointer to an array of colorAttachmentCount valid VkAttachmentReference structures

  • If pDepthStencilAttachment is not NULL, pDepthStencilAttachment must be a valid pointer to a valid VkAttachmentReference structure

  • If preserveAttachmentCount is not 0, pPreserveAttachments must be a valid pointer to an array of preserveAttachmentCount uint32_t values

Bits which can be set in VkSubpassDescription::flags, specifying usage of the subpass, are:

typedef enum VkSubpassDescriptionFlagBits {
    VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX = 0x00000001,
    VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX = 0x00000002,
} VkSubpassDescriptionFlagBits;
  • VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX specifies that shaders compiled for this subpass write the attributes for all views in a single invocation of each vertex processing stage. All pipelines compiled against a subpass that includes this bit must write per-view attributes to the code:*PerViewNV[] shader outputs, in addition to the non-per-view (e.g. Position) outputs.

  • VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX specifies that shaders compiled for this subpass use per-view positions which only differ in value in the x component. Per-view viewport mask can also be used.

typedef VkFlags VkSubpassDescriptionFlags;

VkSubpassDescriptionFlags is a bitmask type for setting a mask of zero or more VkSubpassDescriptionFlagBits.

The VkAttachmentReference structure is defined as:

typedef struct VkAttachmentReference {
    uint32_t         attachment;
    VkImageLayout    layout;
} VkAttachmentReference;
  • attachment is the index of the attachment of the render pass, and corresponds to the index of the corresponding element in the pAttachments array of the VkRenderPassCreateInfo structure. If any color or depth/stencil attachments are VK_ATTACHMENT_UNUSED, then no writes occur for those attachments.

  • layout is a VkImageLayout value specifying the layout the attachment uses during the subpass.

Valid Usage
  • layout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED

Valid Usage (Implicit)

The VkSubpassDependency structure is defined as:

typedef struct VkSubpassDependency {
    uint32_t                srcSubpass;
    uint32_t                dstSubpass;
    VkPipelineStageFlags    srcStageMask;
    VkPipelineStageFlags    dstStageMask;
    VkAccessFlags           srcAccessMask;
    VkAccessFlags           dstAccessMask;
    VkDependencyFlags       dependencyFlags;
} VkSubpassDependency;

If srcSubpass is equal to dstSubpass then the VkSubpassDependency describes a subpass self-dependency, and only constrains the pipeline barriers allowed within a subpass instance. Otherwise, when a render pass instance which includes a subpass dependency is submitted to a queue, it defines a memory dependency between the subpasses identified by srcSubpass and dstSubpass.

If srcSubpass is equal to VK_SUBPASS_EXTERNAL, the first synchronization scope includes commands that occur earlier in submission order than the vkCmdBeginRenderPass used to begin the render pass instance. Otherwise, the first set of commands includes all commands submitted as part of the subpass instance identified by srcSubpass and any load, store or multisample resolve operations on attachments used in srcSubpass. In either case, the first synchronization scope is limited to operations on the pipeline stages determined by the source stage mask specified by srcStageMask.

If dstSubpass is equal to VK_SUBPASS_EXTERNAL, the second synchronization scope includes commands that occur later in submission order than the vkCmdEndRenderPass used to end the render pass instance. Otherwise, the second set of commands includes all commands submitted as part of the subpass instance identified by dstSubpass and any load, store or multisample resolve operations on attachments used in dstSubpass. In either case, the second synchronization scope is limited to operations on the pipeline stages determined by the destination stage mask specified by dstStageMask.

The first access scope is limited to access in the pipeline stages determined by the source stage mask specified by srcStageMask. It is also limited to access types in the source access mask specified by srcAccessMask.

The second access scope is limited to access in the pipeline stages determined by the destination stage mask specified by dstStageMask. It is also limited to access types in the destination access mask specified by dstAccessMask.

The availability and visibility operations defined by a subpass dependency affect the execution of image layout transitions within the render pass.

Note

For non-attachment resources, the memory dependency expressed by subpass dependency is nearly identical to that of a VkMemoryBarrier (with matching srcAccessMask/dstAccessMask parameters) submitted as a part of a vkCmdPipelineBarrier (with matching srcStageMask/dstStageMask parameters). The only difference being that its scopes are limited to the identified subpasses rather than potentially affecting everything before and after.

For attachments however, subpass dependencies work more like an VkImageMemoryBarrier defined similarly to the VkMemoryBarrier above, the queue family indices set to VK_QUEUE_FAMILY_IGNORED, and layouts as follows:

  • The equivalent to oldLayout is the attachment’s layout according to the subpass description for srcSubpass.

  • The equivalent to newLayout is the attachment’s layout according to the subpass description for dstSubpass.

Valid Usage
  • If srcSubpass is not VK_SUBPASS_EXTERNAL, srcStageMask must not include VK_PIPELINE_STAGE_HOST_BIT

  • If dstSubpass is not VK_SUBPASS_EXTERNAL, dstStageMask must not include VK_PIPELINE_STAGE_HOST_BIT

  • If the geometry shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the geometry shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT

  • If the tessellation shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • If the tessellation shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT

  • srcSubpass must be less than or equal to dstSubpass, unless one of them is VK_SUBPASS_EXTERNAL, to avoid cyclic dependencies and ensure a valid execution order

  • srcSubpass and dstSubpass must not both be equal to VK_SUBPASS_EXTERNAL

  • If srcSubpass is equal to dstSubpass, srcStageMask and dstStageMask must only contain one of VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT, VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT, VK_PIPELINE_STAGE_VERTEX_INPUT_BIT, VK_PIPELINE_STAGE_VERTEX_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT, VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT, VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT, VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT, VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT, or VK_PIPELINE_STAGE_ALL_GRAPHICS_BIT

  • If srcSubpass is equal to dstSubpass and not all of the stages in srcStageMask and dstStageMask are framebuffer-space stages, the logically latest pipeline stage in srcStageMask must be logically earlier than or equal to the logically earliest pipeline stage in dstStageMask

  • Any access flag included in srcAccessMask must be supported by one of the pipeline stages in srcStageMask, as specified in the table of supported access types.

  • Any access flag included in dstAccessMask must be supported by one of the pipeline stages in dstStageMask, as specified in the table of supported access types.

  • If dependencyFlags includes VK_DEPENDENCY_VIEW_LOCAL_BIT, then both srcSubpass and dstSubpass must not equal VK_SUBPASS_EXTERNAL

  • If dependencyFlags includes VK_DEPENDENCY_VIEW_LOCAL_BIT, then the render pass must have multiview enabled

  • If srcSubpass equals dstSubpass and that subpass has more than one bit set in the view mask, then dependencyFlags must include VK_DEPENDENCY_VIEW_LOCAL_BIT

Valid Usage (Implicit)

When multiview is enabled, the execution of the multiple views of one subpass may not occur simultaneously or even back-to-back, and rather may be interleaved with the execution of other subpasses. The load and store operations apply to attachments on a per-view basis. For example, an attachment using VK_ATTACHMENT_LOAD_OP_CLEAR will have each view cleared on first use, but the first use of one view may be temporally distant from the first use of another view.

Note

A good mental model for multiview is to think of a multiview subpass as if it were a collection of individual (per-view) subpasses that are logically grouped together and described as a single multiview subpass in the API. Similarly, a multiview attachment can be thought of like several individual attachments that happen to be layers in a single image. A view-local dependency between two multiview subpasses acts like a set of one-to-one dependencies between corresponding pairs of per-view subpasses. A view-global dependency between two multiview subpasses acts like a set of N × M dependencies between all pairs of per-view subpasses in the source and destination. Thus, it is a more compact representation which also makes clear the commonality and reuse that is present between views in a subpass. This interpretation motivates the answers to questions like “when does the load op apply” - it is on the first use of each view of an attachment, as if each view were a separate attachment.

editing-note

The following two alleged implicit dependencies are practically no-ops, as the operations they describe are already guaranteed by semaphores and submission order (so they’re almost entirely no-ops on their own). The only reason they exist is because it simplifies reasoning about where automatic layout transitions happen. Further rewrites of this chapter could potentially remove the need for these.

If there is no subpass dependency from VK_SUBPASS_EXTERNAL to the first subpass that uses an attachment, then an implicit subpass dependency exists from VK_SUBPASS_EXTERNAL to the first subpass it is used in. The subpass dependency operates as if defined with the following parameters:

VkSubpassDependency implicitDependency = {
    .srcSubpass = VK_SUBPASS_EXTERNAL;
    .dstSubpass = firstSubpass; // First subpass attachment is used in
    .srcStageMask = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT;
    .dstStageMask = VK_PIPELINE_STAGE_ALL_COMMANDS_BIT;
    .srcAccessMask = 0;
    .dstAccessMask = VK_ACCESS_INPUT_ATTACHMENT_READ_BIT |
                     VK_ACCESS_COLOR_ATTACHMENT_READ_BIT |
                     VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT |
                     VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT |
                     VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT;
    .dependencyFlags = 0;
};

Similarly, if there is no subpass dependency from the last subpass that uses an attachment to VK_SUBPASS_EXTERNAL, then an implicit subpass dependency exists from the last subpass it is used in to VK_SUBPASS_EXTERNAL. The subpass dependency operates as if defined with the following parameters:

VkSubpassDependency implicitDependency = {
    .srcSubpass = lastSubpass; // Last subpass attachment is used in
    .dstSubpass = VK_SUBPASS_EXTERNAL;
    .srcStageMask = VK_PIPELINE_STAGE_ALL_COMMANDS_BIT;
    .dstStageMask = VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT;
    .srcAccessMask = VK_ACCESS_INPUT_ATTACHMENT_READ_BIT |
                     VK_ACCESS_COLOR_ATTACHMENT_READ_BIT |
                     VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT |
                     VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT |
                     VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT;
    .dstAccessMask = 0;
    .dependencyFlags = 0;
};

As subpasses may overlap or execute out of order with regards to other subpasses unless a subpass dependency chain describes otherwise, the layout transitions required between subpasses cannot be known to an application. Instead, an application provides the layout that each attachment must be in at the start and end of a render pass, and the layout it must be in during each subpass it is used in. The implementation then must execute layout transitions between subpasses in order to guarantee that the images are in the layouts required by each subpass, and in the final layout at the end of the render pass.

Automatic layout transitions apply to the entire image subresource attached to the framebuffer. If the attachment view is a 2D or 2D array view of a 3D image, even if the attachment view only refers to a subset of the slices of the selected mip level of the 3D image, automatic layout transitions apply to the entire subresource referenced which is the entire mip level in this case.

Automatic layout transitions away from the layout used in a subpass happen-after the availability operations for all dependencies with that subpass as the srcSubpass.

Automatic layout transitions into the layout used in a subpass happen-before the visibility operations for all dependencies with that subpass as the dstSubpass.

Automatic layout transitions away from initialLayout happens-after the availability operations for all dependencies with a srcSubpass equal to VK_SUBPASS_EXTERNAL, where dstSubpass uses the attachment that will be transitioned. For attachments created with VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT, automatic layout transitions away from initialLayout happen-after the availability operations for all dependencies with a srcSubpass equal to VK_SUBPASS_EXTERNAL, where dstSubpass uses any aliased attachment.

Automatic layout transitions into finalLayout happens-before the visibility operations for all dependencies with a dstSubpass equal to VK_SUBPASS_EXTERNAL, where srcSubpass uses the attachment that will be transitioned. For attachments created with VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT, automatic layout transitions into finalLayout happen-before the visibility operations for all dependencies with a dstSubpass equal to VK_SUBPASS_EXTERNAL, where srcSubpass uses any aliased attachment.

The image layout of the depth aspect of a depth/stencil attachment referring to an image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT is dependent on the last sample locations used to render to the attachment, thus automatic layout transitions use the sample locations state specified in VkRenderPassSampleLocationsBeginInfoEXT.

Automatic layout transitions of an attachment referring to a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT use the sample locations the image subresource range referenced by the attachment was last rendered with. If the current render pass does not use the attachment as a depth/stencil attachment in any subpass that happens-before, the automatic layout transition uses the sample locations state specified in the sampleLocationsInfo member of the element of the VkRenderPassSampleLocationsBeginInfoEXT::pAttachmentInitialSampleLocations array for which the attachmentIndex member equals the attachment index of the attachment, if one is specified. Otherwise, the automatic layout transition uses the sample locations state specified in the sampleLocationsInfo member of the element of the VkRenderPassSampleLocationsBeginInfoEXT::pPostSubpassSampleLocations array for which the subpassIndex member equals the index of the subpass that last used the attachment as a depth/stencil attachment, if one is specified.

If no sample locations state has been specified for an automatic layout transition performed on an attachment referring to a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT the contents of the depth aspect of the depth/stencil attachment become undefined as if the layout of the attachment was transitioned from the VK_IMAGE_LAYOUT_UNDEFINED layout.

If two subpasses use the same attachment in different layouts, and both layouts are read-only, no subpass dependency needs to be specified between those subpasses. If an implementation treats those layouts separately, it must insert an implicit subpass dependency between those subpasses to separate the uses in each layout. The subpass dependency operates as if defined with the following parameters:

// Used for input attachments
VkPipelineStageFlags inputAttachmentStages = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT;
VkAccessFlags inputAttachmentAccess = VK_ACCESS_INPUT_ATTACHMENT_READ_BIT;

// Used for depth/stencil attachments
VkPipelineStageFlags depthStencilAttachmentStages = VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT | VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT;
VkAccessFlags depthStencilAttachmentAccess = VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT;

VkSubpassDependency implicitDependency = {
    .srcSubpass = firstSubpass;
    .dstSubpass = secondSubpass;
    .srcStageMask = inputAttachmentStages | depthStencilAttachmentStages;
    .dstStageMask = inputAttachmentStages | depthStencilAttachmentStages;
    .srcAccessMask = inputAttachmentAccess | depthStencilAttachmentAccess;
    .dstAccessMask = inputAttachmentAccess | depthStencilAttachmentAccess;
    .dependencyFlags = 0;
};

If a subpass uses the same attachment as both an input attachment and either a color attachment or a depth/stencil attachment, writes via the color or depth/stencil attachment are not automatically made visible to reads via the input attachment, causing a feedback loop, except in any of the following conditions:

  • If the color components or depth/stencil components read by the input attachment are mutually exclusive with the components written by the color or depth/stencil attachments, then there is no feedback loop. This requires the graphics pipelines used by the subpass to disable writes to color components that are read as inputs via the colorWriteMask, and to disable writes to depth/stencil components that are read as inputs via depthWriteEnable or stencilTestEnable.

  • If the attachment is used as an input attachment and depth/stencil attachment only, and the depth/stencil attachment is not written to.

  • If a memory dependency is inserted between when the attachment is written and when it is subsequently read by later fragments. Pipeline barriers expressing a subpass self-dependency are the only way to achieve this, and one must be inserted every time a fragment will read values at a particular sample (x, y, layer, sample) coordinate, if those values have been written since the most recent pipeline barrier; or the since start of the subpass if there have been no pipeline barriers since the start of the subpass.

An attachment used as both an input attachment and a color attachment must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR or VK_IMAGE_LAYOUT_GENERAL layout. An attachment used as an input attachment and depth/stencil attachment must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_GENERAL layout. An attachment must not be used as both a depth/stencil attachment and a color attachment.

To destroy a render pass, call:

void vkDestroyRenderPass(
    VkDevice                                    device,
    VkRenderPass                                renderPass,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the render pass.

  • renderPass is the handle of the render pass to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to renderPass must have completed execution

  • If VkAllocationCallbacks were provided when renderPass was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when renderPass was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If renderPass is not VK_NULL_HANDLE, renderPass must be a valid VkRenderPass handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If renderPass is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to renderPass must be externally synchronized

7.2. Render Pass Compatibility

Framebuffers and graphics pipelines are created based on a specific render pass object. They must only be used with that render pass object, or one compatible with it.

Two attachment references are compatible if they have matching format and sample count, or are both VK_ATTACHMENT_UNUSED or the pointer that would contain the reference is NULL.

Two arrays of attachment references are compatible if all corresponding pairs of attachments are compatible. If the arrays are of different lengths, attachment references not present in the smaller array are treated as VK_ATTACHMENT_UNUSED.

Two render passes are compatible if their corresponding color, input, resolve, and depth/stencil attachment references are compatible and if they are otherwise identical except for:

  • Initial and final image layout in attachment descriptions

  • Load and store operations in attachment descriptions

  • Image layout in attachment references

A framebuffer is compatible with a render pass if it was created using the same render pass or a compatible render pass.

7.3. Framebuffers

Render passes operate in conjunction with framebuffers. Framebuffers represent a collection of specific memory attachments that a render pass instance uses.

Framebuffers are represented by VkFramebuffer handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkFramebuffer)

To create a framebuffer, call:

VkResult vkCreateFramebuffer(
    VkDevice                                    device,
    const VkFramebufferCreateInfo*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkFramebuffer*                              pFramebuffer);
  • device is the logical device that creates the framebuffer.

  • pCreateInfo points to a VkFramebufferCreateInfo structure which describes additional information about framebuffer creation.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pFramebuffer points to a VkFramebuffer handle in which the resulting framebuffer object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkFramebufferCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pFramebuffer must be a valid pointer to a VkFramebuffer handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkFramebufferCreateInfo structure is defined as:

typedef struct VkFramebufferCreateInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkFramebufferCreateFlags    flags;
    VkRenderPass                renderPass;
    uint32_t                    attachmentCount;
    const VkImageView*          pAttachments;
    uint32_t                    width;
    uint32_t                    height;
    uint32_t                    layers;
} VkFramebufferCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • renderPass is a render pass that defines what render passes the framebuffer will be compatible with. See Render Pass Compatibility for details.

  • attachmentCount is the number of attachments.

  • pAttachments is an array of VkImageView handles, each of which will be used as the corresponding attachment in a render pass instance.

  • width, height and layers define the dimensions of the framebuffer. If the render pass uses multiview, then layers must be one and each attachment requires a number of layers that is greater than the maximum bit index set in the view mask in the subpasses in which it is used.

Applications must ensure that all accesses to memory that backs image subresources used as attachments in a given renderpass instance either happen-before the load operations for those attachments, or happen-after the store operations for those attachments.

For depth/stencil attachments, each aspect can be used separately as attachments and non-attachments as long as the non-attachment accesses are also via an image subresource in either the VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL layout or the VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL layout, and the attachment resource uses whichever of those two layouts the image accesses do not. Use of non-attachment aspects in this case is only well defined if the attachment is used in the subpass where the non-attachment access is being made, or the layout of the image subresource is constant throughout the entire render pass instance, including the initialLayout and finalLayout.

Note

These restrictions mean that the render pass has full knowledge of all uses of all of the attachments, so that the implementation is able to make correct decisions about when and how to perform layout transitions, when to overlap execution of subpasses, etc.

It is legal for a subpass to use no color or depth/stencil attachments, and rather use shader side effects such as image stores and atomics to produce an output. In this case, the subpass continues to use the width, height, and layers of the framebuffer to define the dimensions of the rendering area, and the rasterizationSamples from each pipeline’s VkPipelineMultisampleStateCreateInfo to define the number of samples used in rasterization; however, if VkPhysicalDeviceFeatures::variableMultisampleRate is VK_FALSE, then all pipelines to be bound with a given zero-attachment subpass must have the same value for VkPipelineMultisampleStateCreateInfo::rasterizationSamples.

Valid Usage
  • attachmentCount must be equal to the attachment count specified in renderPass

  • Each element of pAttachments that is used as a color attachment or resolve attachment by renderPass must have been created with a usage value including VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT

  • Each element of pAttachments that is used as a depth/stencil attachment by renderPass must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT

  • Each element of pAttachments that is used as an input attachment by renderPass must have been created with a usage value including VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT

  • Each element of pAttachments must have been created with an VkFormat value that matches the VkFormat specified by the corresponding VkAttachmentDescription in renderPass

  • Each element of pAttachments must have been created with a samples value that matches the samples value specified by the corresponding VkAttachmentDescription in renderPass

  • Each element of pAttachments must have dimensions at least as large as the corresponding framebuffer dimension

  • Each element of pAttachments must only specify a single mip level

  • Each element of pAttachments must have been created with the identity swizzle

  • width must be greater than 0.

  • width must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferWidth

  • height must be greater than 0.

  • height must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferHeight

  • layers must be greater than 0.

  • layers must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferLayers

  • Each element of pAttachments that is a 2D or 2D array image view taken from a 3D image must not be a depth/stencil format

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • renderPass must be a valid VkRenderPass handle

  • If attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkImageView handles

  • Both of renderPass, and the elements of pAttachments that are valid handles must have been created, allocated, or retrieved from the same VkDevice

typedef VkFlags VkFramebufferCreateFlags;

VkFramebufferCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To destroy a framebuffer, call:

void vkDestroyFramebuffer(
    VkDevice                                    device,
    VkFramebuffer                               framebuffer,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the framebuffer.

  • framebuffer is the handle of the framebuffer to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to framebuffer must have completed execution

  • If VkAllocationCallbacks were provided when framebuffer was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when framebuffer was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If framebuffer is not VK_NULL_HANDLE, framebuffer must be a valid VkFramebuffer handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If framebuffer is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to framebuffer must be externally synchronized

7.4. Render Pass Commands

An application records the commands for a render pass instance one subpass at a time, by beginning a render pass instance, iterating over the subpasses to record commands for that subpass, and then ending the render pass instance.

To begin a render pass instance, call:

void vkCmdBeginRenderPass(
    VkCommandBuffer                             commandBuffer,
    const VkRenderPassBeginInfo*                pRenderPassBegin,
    VkSubpassContents                           contents);
  • commandBuffer is the command buffer in which to record the command.

  • pRenderPassBegin is a pointer to a VkRenderPassBeginInfo structure (defined below) which specifies the render pass to begin an instance of, and the framebuffer the instance uses.

  • contents is a VkSubpassContents value specifying how the commands in the first subpass will be provided.

After beginning a render pass instance, the command buffer is ready to record the commands for the first subpass of that render pass.

Valid Usage
  • If any of the initialLayout or finalLayout member of the VkAttachmentDescription structures or the layout member of the VkAttachmentReference structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL then the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT set

  • If any of the initialLayout or finalLayout member of the VkAttachmentDescription structures or the layout member of the VkAttachmentReference structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL then the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set

  • If any of the initialLayout or finalLayout member of the VkAttachmentDescription structures or the layout member of the VkAttachmentReference structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL then the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with VK_IMAGE_USAGE_SAMPLED_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT set

  • If any of the initialLayout or finalLayout member of the VkAttachmentDescription structures or the layout member of the VkAttachmentReference structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL then the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT set

  • If any of the initialLayout or finalLayout member of the VkAttachmentDescription structures or the layout member of the VkAttachmentReference structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL then the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT set

  • If any of the initialLayout members of the VkAttachmentDescription structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is not VK_IMAGE_LAYOUT_UNDEFINED, then each such initialLayout must be equal to the current layout of the corresponding attachment image subresource of the framebuffer specified in the framebuffer member of pRenderPassBegin

  • The srcStageMask and dstStageMask members of any element of the pDependencies member of VkRenderPassCreateInfo used to create renderPass must be supported by the capabilities of the queue family identified by the queueFamilyIndex member of the VkCommandPoolCreateInfo used to create the command pool which commandBuffer was allocated from.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pRenderPassBegin must be a valid pointer to a valid VkRenderPassBeginInfo structure

  • contents must be a valid VkSubpassContents value

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called outside of a render pass instance

  • commandBuffer must be a primary VkCommandBuffer

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary

Outside

Graphics

Graphics

The VkRenderPassBeginInfo structure is defined as:

typedef struct VkRenderPassBeginInfo {
    VkStructureType        sType;
    const void*            pNext;
    VkRenderPass           renderPass;
    VkFramebuffer          framebuffer;
    VkRect2D               renderArea;
    uint32_t               clearValueCount;
    const VkClearValue*    pClearValues;
} VkRenderPassBeginInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • renderPass is the render pass to begin an instance of.

  • framebuffer is the framebuffer containing the attachments that are used with the render pass.

  • renderArea is the render area that is affected by the render pass instance, and is described in more detail below.

  • clearValueCount is the number of elements in pClearValues.

  • pClearValues is an array of VkClearValue structures that contains clear values for each attachment, if the attachment uses a loadOp value of VK_ATTACHMENT_LOAD_OP_CLEAR or if the attachment has a depth/stencil format and uses a stencilLoadOp value of VK_ATTACHMENT_LOAD_OP_CLEAR. The array is indexed by attachment number. Only elements corresponding to cleared attachments are used. Other elements of pClearValues are ignored.

renderArea is the render area that is affected by the render pass instance. The effects of attachment load, store and multisample resolve operations are restricted to the pixels whose x and y coordinates fall within the render area on all attachments. The render area extends to all layers of framebuffer. The application must ensure (using scissor if necessary) that all rendering is contained within the render area, otherwise the pixels outside of the render area become undefined and shader side effects may occur for fragments outside the render area. The render area must be contained within the framebuffer dimensions.

When multiview is enabled, the resolve operation at the end of a subpass applies to all views in the view mask.

Note

There may be a performance cost for using a render area smaller than the framebuffer, unless it matches the render area granularity for the render pass.

Valid Usage
  • clearValueCount must be greater than the largest attachment index in renderPass that specifies a loadOp (or stencilLoadOp, if the attachment has a depth/stencil format) of VK_ATTACHMENT_LOAD_OP_CLEAR

  • If clearValueCount is not 0, pClearValues must be a valid pointer to an array of clearValueCount valid VkClearValue unions

  • renderPass must be compatible with the renderPass member of the VkFramebufferCreateInfo structure specified when creating framebuffer.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkDeviceGroupRenderPassBeginInfo or VkRenderPassSampleLocationsBeginInfoEXT

  • Each sType member in the pNext chain must be unique

  • renderPass must be a valid VkRenderPass handle

  • framebuffer must be a valid VkFramebuffer handle

  • Both of framebuffer, and renderPass must have been created, allocated, or retrieved from the same VkDevice

The image layout of the depth aspect of a depth/stencil attachment referring to an image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT is dependent on the last sample locations used to render to the image subresource, thus preserving the contents of such depth/stencil attachments across subpass boundaries requires the application to specify these sample locations whenever a layout transition of the attachment may occur. This information can be provided by chaining an instance of the VkRenderPassSampleLocationsBeginInfoEXT structure to the pNext chain of VkRenderPassBeginInfo.

The VkRenderPassSampleLocationsBeginInfoEXT structure is defined as:

typedef struct VkRenderPassSampleLocationsBeginInfoEXT {
    VkStructureType                          sType;
    const void*                              pNext;
    uint32_t                                 attachmentInitialSampleLocationsCount;
    const VkAttachmentSampleLocationsEXT*    pAttachmentInitialSampleLocations;
    uint32_t                                 postSubpassSampleLocationsCount;
    const VkSubpassSampleLocationsEXT*       pPostSubpassSampleLocations;
} VkRenderPassSampleLocationsBeginInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • attachmentInitialSampleLocationsCount is the number of elements in the pAttachmentInitialSampleLocations array.

  • pAttachmentInitialSampleLocations is an array of attachmentInitialSampleLocationsCount VkAttachmentSampleLocationsEXT structures specifying the attachment indices and their corresponding sample location state. Each element of pAttachmentInitialSampleLocations can specify the sample location state to use in the automatic layout transition performed to transition a depth/stencil attachment from the initial layout of the attachment to the image layout specified for the attachment in the first subpass using it.

  • postSubpassSampleLocationsCount is the number of elements in the pPostSubpassSampleLocations array.

  • pPostSubpassSampleLocations is an array of postSubpassSampleLocationsCount VkSubpassSampleLocationsEXT structures specifying the subpass indices and their corresponding sample location state. Each element of pPostSubpassSampleLocations can specify the sample location state to use in the automatic layout transition performed to transition the depth/stencil attachment used by the specified subpass to the image layout specified in a dependent subpass or to the final layout of the attachment in case the specified subpass is the last subpass using that attachment. In addition, if VkPhysicalDeviceSampleLocationsPropertiesEXT::variableSampleLocations is VK_FALSE, each element of pPostSubpassSampleLocations must specify the sample location state that matches the sample locations used by all pipelines that will be bound to a command buffer during the specified subpass. If variableSampleLocations is VK_TRUE, the sample locations used for rasterization do not depend on pPostSubpassSampleLocations.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_RENDER_PASS_SAMPLE_LOCATIONS_BEGIN_INFO_EXT

  • If attachmentInitialSampleLocationsCount is not 0, pAttachmentInitialSampleLocations must be a valid pointer to an array of attachmentInitialSampleLocationsCount valid VkAttachmentSampleLocationsEXT structures

  • If postSubpassSampleLocationsCount is not 0, pPostSubpassSampleLocations must be a valid pointer to an array of postSubpassSampleLocationsCount valid VkSubpassSampleLocationsEXT structures

The VkAttachmentSampleLocationsEXT structure is defined as:

typedef struct VkAttachmentSampleLocationsEXT {
    uint32_t                    attachmentIndex;
    VkSampleLocationsInfoEXT    sampleLocationsInfo;
} VkAttachmentSampleLocationsEXT;
  • attachmentIndex is the index of the attachment for which the sample locations state is provided.

  • sampleLocationsInfo is the sample locations state to use for the layout transition of the given attachment from the initial layout of the attachment to the image layout specified for the attachment in the first subpass using it.

If the image referenced by the framebuffer attachment at index attachmentIndex was not created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT then the values specified in sampleLocationsInfo are ignored.

Valid Usage
Valid Usage (Implicit)
  • sampleLocationsInfo must be a valid VkSampleLocationsInfoEXT structure

The VkSubpassSampleLocationsEXT structure is defined as:

typedef struct VkSubpassSampleLocationsEXT {
    uint32_t                    subpassIndex;
    VkSampleLocationsInfoEXT    sampleLocationsInfo;
} VkSubpassSampleLocationsEXT;
  • subpassIndex is the index of the subpass for which the sample locations state is provided.

  • sampleLocationsInfo is the sample locations state to use for the layout transition of the depth/stencil attachment away from the image layout the attachment is used with in the subpass specified in subpassIndex.

If the image referenced by the depth/stencil attachment used in the subpass identified by subpassIndex was not created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT or if the subpass does not use a depth/stencil attachment, and VkPhysicalDeviceSampleLocationsPropertiesEXT::variableSampleLocations is VK_TRUE then the values specified in sampleLocationsInfo are ignored.

Valid Usage
Valid Usage (Implicit)
  • sampleLocationsInfo must be a valid VkSampleLocationsInfoEXT structure

Possible values of vkCmdBeginRenderPass::contents, specifying how the commands in the first subpass will be provided, are:

typedef enum VkSubpassContents {
    VK_SUBPASS_CONTENTS_INLINE = 0,
    VK_SUBPASS_CONTENTS_SECONDARY_COMMAND_BUFFERS = 1,
} VkSubpassContents;
  • VK_SUBPASS_CONTENTS_INLINE specifies that the contents of the subpass will be recorded inline in the primary command buffer, and secondary command buffers must not be executed within the subpass.

  • VK_SUBPASS_CONTENTS_SECONDARY_COMMAND_BUFFERS specifies that the contents are recorded in secondary command buffers that will be called from the primary command buffer, and vkCmdExecuteCommands is the only valid command on the command buffer until vkCmdNextSubpass or vkCmdEndRenderPass.

If the pNext chain of VkRenderPassBeginInfo includes a VkDeviceGroupRenderPassBeginInfo structure, then that structure includes a device mask and set of render areas for the render pass instance.

The VkDeviceGroupRenderPassBeginInfo structure is defined as:

typedef struct VkDeviceGroupRenderPassBeginInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceMask;
    uint32_t           deviceRenderAreaCount;
    const VkRect2D*    pDeviceRenderAreas;
} VkDeviceGroupRenderPassBeginInfo;

or the equivalent

typedef VkDeviceGroupRenderPassBeginInfo VkDeviceGroupRenderPassBeginInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • deviceMask is the device mask for the render pass instance.

  • deviceRenderAreaCount is the number of elements in the pDeviceRenderAreas array.

  • pDeviceRenderAreas is an array of structures of type VkRect2D defining the render area for each physical device.

The deviceMask serves several purposes. It is an upper bound on the set of physical devices that can be used during the render pass instance, and the initial device mask when the render pass instance begins. Render pass attachment load, store, and resolve operations only apply to physical devices included in the device mask. Subpass dependencies only apply to the physical devices in the device mask.

If deviceRenderAreaCount is not zero, then the elements of pDeviceRenderAreas override the value of VkRenderPassBeginInfo::renderArea, and provide a render area specific to each physical device. These render areas serve the same purpose as VkRenderPassBeginInfo::renderArea, including controlling the region of attachments that are cleared by VK_ATTACHMENT_LOAD_OP_CLEAR and that are resolved into resolve attachments.

If this structure is not present, the render pass instance’s device mask is the value of VkDeviceGroupCommandBufferBeginInfo::deviceMask. If this structure is not present or if deviceRenderAreaCount is zero, VkRenderPassBeginInfo::renderArea is used for all physical devices.

Valid Usage
  • deviceMask must be a valid device mask value

  • deviceMask must not be zero

  • deviceMask must be a subset of the command buffer’s initial device mask

  • deviceRenderAreaCount must either be zero or equal to the number of physical devices in the logical device.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO

  • If deviceRenderAreaCount is not 0, pDeviceRenderAreas must be a valid pointer to an array of deviceRenderAreaCount VkRect2D structures

To query the render area granularity, call:

void vkGetRenderAreaGranularity(
    VkDevice                                    device,
    VkRenderPass                                renderPass,
    VkExtent2D*                                 pGranularity);
  • device is the logical device that owns the render pass.

  • renderPass is a handle to a render pass.

  • pGranularity points to a VkExtent2D structure in which the granularity is returned.

The conditions leading to an optimal renderArea are:

  • the offset.x member in renderArea is a multiple of the width member of the returned VkExtent2D (the horizontal granularity).

  • the offset.y member in renderArea is a multiple of the height of the returned VkExtent2D (the vertical granularity).

  • either the offset.width member in renderArea is a multiple of the horizontal granularity or offset.x+offset.width is equal to the width of the framebuffer in the VkRenderPassBeginInfo.

  • either the offset.height member in renderArea is a multiple of the vertical granularity or offset.y+offset.height is equal to the height of the framebuffer in the VkRenderPassBeginInfo.

Subpass dependencies are not affected by the render area, and apply to the entire image subresources attached to the framebuffer as specified in the description of automatic layout transitions. Similarly, pipeline barriers are valid even if their effect extends outside the render area.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • renderPass must be a valid VkRenderPass handle

  • pGranularity must be a valid pointer to a VkExtent2D structure

  • renderPass must have been created, allocated, or retrieved from device

To transition to the next subpass in the render pass instance after recording the commands for a subpass, call:

void vkCmdNextSubpass(
    VkCommandBuffer                             commandBuffer,
    VkSubpassContents                           contents);
  • commandBuffer is the command buffer in which to record the command.

  • contents specifies how the commands in the next subpass will be provided, in the same fashion as the corresponding parameter of vkCmdBeginRenderPass.

The subpass index for a render pass begins at zero when vkCmdBeginRenderPass is recorded, and increments each time vkCmdNextSubpass is recorded.

Moving to the next subpass automatically performs any multisample resolve operations in the subpass being ended. End-of-subpass multisample resolves are treated as color attachment writes for the purposes of synchronization. That is, they are considered to execute in the VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT pipeline stage and their writes are synchronized with VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT. Synchronization between rendering within a subpass and any resolve operations at the end of the subpass occurs automatically, without need for explicit dependencies or pipeline barriers. However, if the resolve attachment is also used in a different subpass, an explicit dependency is needed.

After transitioning to the next subpass, the application can record the commands for that subpass.

Valid Usage
  • The current subpass index must be less than the number of subpasses in the render pass minus one

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • contents must be a valid VkSubpassContents value

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • commandBuffer must be a primary VkCommandBuffer

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary

Inside

Graphics

Graphics

To record a command to end a render pass instance after recording the commands for the last subpass, call:

void vkCmdEndRenderPass(
    VkCommandBuffer                             commandBuffer);
  • commandBuffer is the command buffer in which to end the current render pass instance.

Ending a render pass instance performs any multisample resolve operations on the final subpass.

Valid Usage
  • The current subpass index must be equal to the number of subpasses in the render pass minus one

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • commandBuffer must be a primary VkCommandBuffer

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary

Inside

Graphics

Graphics

8. Shaders

A shader specifies programmable operations that execute for each vertex, control point, tessellated vertex, primitive, fragment, or workgroup in the corresponding stage(s) of the graphics and compute pipelines.

Graphics pipelines include vertex shader execution as a result of primitive assembly, followed, if enabled, by tessellation control and evaluation shaders operating on patches, geometry shaders, if enabled, operating on primitives, and fragment shaders, if present, operating on fragments generated by Rasterization. In this specification, vertex, tessellation control, tessellation evaluation and geometry shaders are collectively referred to as vertex processing stages and occur in the logical pipeline before rasterization. The fragment shader occurs logically after rasterization.

Only the compute shader stage is included in a compute pipeline. Compute shaders operate on compute invocations in a workgroup.

Shaders can read from input variables, and read from and write to output variables. Input and output variables can be used to transfer data between shader stages, or to allow the shader to interact with values that exist in the execution environment. Similarly, the execution environment provides constants that describe capabilities.

Shader variables are associated with execution environment-provided inputs and outputs using built-in decorations in the shader. The available decorations for each stage are documented in the following subsections.

8.1. Shader Modules

Shader modules contain shader code and one or more entry points. Shaders are selected from a shader module by specifying an entry point as part of pipeline creation. The stages of a pipeline can use shaders that come from different modules. The shader code defining a shader module must be in the SPIR-V format, as described by the Vulkan Environment for SPIR-V appendix.

Shader modules are represented by VkShaderModule handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkShaderModule)

To create a shader module, call:

VkResult vkCreateShaderModule(
    VkDevice                                    device,
    const VkShaderModuleCreateInfo*             pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkShaderModule*                             pShaderModule);
  • device is the logical device that creates the shader module.

  • pCreateInfo is a pointer to an instance of the VkShaderModuleCreateInfo structure.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pShaderModule points to a VkShaderModule handle in which the resulting shader module object is returned.

Once a shader module has been created, any entry points it contains can be used in pipeline shader stages as described in Compute Pipelines and Graphics Pipelines.

If the shader stage fails to compile VK_ERROR_INVALID_SHADER_NV will be generated and the compile log will be reported back to the application by VK_EXT_debug_report if enabled.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkShaderModuleCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pShaderModule must be a valid pointer to a VkShaderModule handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INVALID_SHADER_NV

The VkShaderModuleCreateInfo structure is defined as:

typedef struct VkShaderModuleCreateInfo {
    VkStructureType              sType;
    const void*                  pNext;
    VkShaderModuleCreateFlags    flags;
    size_t                       codeSize;
    const uint32_t*              pCode;
} VkShaderModuleCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • codeSize is the size, in bytes, of the code pointed to by pCode.

  • pCode points to code that is used to create the shader module. The type and format of the code is determined from the content of the memory addressed by pCode.

Valid Usage
  • codeSize must be greater than 0

  • If pCode points to SPIR-V code, codeSize must be a multiple of 4

  • pCode must point to either valid SPIR-V code, formatted and packed as described by the Khronos SPIR-V Specification or valid GLSL code which must be written to the GL_KHR_vulkan_glsl extension specification

  • If pCode points to SPIR-V code, that code must adhere to the validation rules described by the Validation Rules within a Module section of the SPIR-V Environment appendix

  • If pCode points to GLSL code, it must be valid GLSL code written to the GL_KHR_vulkan_glsl GLSL extension specification

  • pCode must declare the Shader capability for SPIR-V code

  • pCode must not declare any capability that is not supported by the API, as described by the Capabilities section of the SPIR-V Environment appendix

  • If pCode declares any of the capabilities listed as optional in the SPIR-V Environment appendix, the corresponding feature(s) must be enabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SHADER_MODULE_CREATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkShaderModuleValidationCacheCreateInfoEXT

  • flags must be 0

  • pCode must be a valid pointer to an array of \(codeSize \over 4\) uint32_t values

typedef VkFlags VkShaderModuleCreateFlags;

VkShaderModuleCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To use a VkValidationCacheEXT to cache shader validation results, add a VkShaderModuleValidationCacheCreateInfoEXT to the pNext chain of the VkShaderModuleCreateInfo structure, specifying the cache object to use.

The VkShaderModuleValidationCacheCreateInfoEXT struct is defined as:

typedef struct VkShaderModuleValidationCacheCreateInfoEXT {
    VkStructureType         sType;
    const void*             pNext;
    VkValidationCacheEXT    validationCache;
} VkShaderModuleValidationCacheCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • validationCache is the validation cache object from which the results of prior validation attempts will be written, and to which new validation results for this VkShaderModule will be written (if not already present).

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SHADER_MODULE_VALIDATION_CACHE_CREATE_INFO_EXT

  • validationCache must be a valid VkValidationCacheEXT handle

To destroy a shader module, call:

void vkDestroyShaderModule(
    VkDevice                                    device,
    VkShaderModule                              shaderModule,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the shader module.

  • shaderModule is the handle of the shader module to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

A shader module can be destroyed while pipelines created using its shaders are still in use.

Valid Usage
  • If VkAllocationCallbacks were provided when shaderModule was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when shaderModule was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If shaderModule is not VK_NULL_HANDLE, shaderModule must be a valid VkShaderModule handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If shaderModule is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to shaderModule must be externally synchronized

8.2. Shader Execution

At each stage of the pipeline, multiple invocations of a shader may execute simultaneously. Further, invocations of a single shader produced as the result of different commands may execute simultaneously. The relative execution order of invocations of the same shader type is undefined. Shader invocations may complete in a different order than that in which the primitives they originated from were drawn or dispatched by the application. However, fragment shader outputs are written to attachments in rasterization order.

The relative order of invocations of different shader types is largely undefined. However, when invoking a shader whose inputs are generated from a previous pipeline stage, the shader invocations from the previous stage are guaranteed to have executed far enough to generate input values for all required inputs.

8.3. Shader Memory Access Ordering

The order in which image or buffer memory is read or written by shaders is largely undefined. For some shader types (vertex, tessellation evaluation, and in some cases, fragment), even the number of shader invocations that may perform loads and stores is undefined.

In particular, the following rules apply:

  • Vertex and tessellation evaluation shaders will be invoked at least once for each unique vertex, as defined in those sections.

  • Fragment shaders will be invoked zero or more times, as defined in that section.

  • The relative order of invocations of the same shader type are undefined. A store issued by a shader when working on primitive B might complete prior to a store for primitive A, even if primitive A is specified prior to primitive B. This applies even to fragment shaders; while fragment shader outputs are always written to the framebuffer in rasterization order, stores executed by fragment shader invocations are not.

  • The relative order of invocations of different shader types is largely undefined.

Note

The above limitations on shader invocation order make some forms of synchronization between shader invocations within a single set of primitives unimplementable. For example, having one invocation poll memory written by another invocation assumes that the other invocation has been launched and will complete its writes in finite time.

Stores issued to different memory locations within a single shader invocation may not be visible to other invocations, or may not become visible in the order they were performed.

The OpMemoryBarrier instruction can be used to provide stronger ordering of reads and writes performed by a single invocation. OpMemoryBarrier guarantees that any memory transactions issued by the shader invocation prior to the instruction complete prior to the memory transactions issued after the instruction. Memory barriers are needed for algorithms that require multiple invocations to access the same memory and require the operations to be performed in a partially-defined relative order. For example, if one shader invocation does a series of writes, followed by an OpMemoryBarrier instruction, followed by another write, then the results of the series of writes before the barrier become visible to other shader invocations at a time earlier or equal to when the results of the final write become visible to those invocations. In practice it means that another invocation that sees the results of the final write would also see the previous writes. Without the memory barrier, the final write may be visible before the previous writes.

Writes that are the result of shader stores through a variable decorated with Coherent automatically have available writes to the same buffer, buffer view, or image view made visible to them, and are themselves automatically made available to access by the same buffer, buffer view, or image view. Reads that are the result of shader loads through a variable decorated with Coherent automatically have available writes to the same buffer, buffer view, or image view made visible to them. The order that coherent writes to different locations become available is undefined, unless enforced by a memory barrier instruction or other memory dependency.

Note

Explicit memory dependencies must still be used to guarantee availability and visibility for access via other buffers, buffer views, or image views.

The built-in atomic memory transaction instructions can be used to read and write a given memory address atomically. While built-in atomic functions issued by multiple shader invocations are executed in undefined order relative to each other, these functions perform both a read and a write of a memory address and guarantee that no other memory transaction will write to the underlying memory between the read and write. Atomic operations ensure automatic availability and visibility for writes and reads in the same way as those to Coherent variables.

Note

Memory accesses performed on different resource descriptors with the same memory backing may not be well-defined even with the Coherent decoration or via atomics, due to things such as image layouts or ownership of the resource - as described in the Synchronization and Cache Control chapter.

Note

Atomics allow shaders to use shared global addresses for mutual exclusion or as counters, among other uses.

8.4. Shader Inputs and Outputs

Data is passed into and out of shaders using variables with input or output storage class, respectively. User-defined inputs and outputs are connected between stages by matching their Location decorations. Additionally, data can be provided by or communicated to special functions provided by the execution environment using BuiltIn decorations.

In many cases, the same BuiltIn decoration can be used in multiple shader stages with similar meaning. The specific behavior of variables decorated as BuiltIn is documented in the following sections.

8.5. Vertex Shaders

Each vertex shader invocation operates on one vertex and its associated vertex attribute data, and outputs one vertex and associated data. Graphics pipelines must include a vertex shader, and the vertex shader stage is always the first shader stage in the graphics pipeline.

8.5.1. Vertex Shader Execution

A vertex shader must be executed at least once for each vertex specified by a draw command. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view. During execution, the shader is presented with the index of the vertex and instance for which it has been invoked. Input variables declared in the vertex shader are filled by the implementation with the values of vertex attributes associated with the invocation being executed.

If the same vertex is specified multiple times in a draw command (e.g. by including the same index value multiple times in an index buffer) the implementation may reuse the results of vertex shading if it can statically determine that the vertex shader invocations will produce identical results.

Note

It is implementation-dependent when and if results of vertex shading are reused, and thus how many times the vertex shader will be executed. This is true also if the vertex shader contains stores or atomic operations (see vertexPipelineStoresAndAtomics).

8.6. Tessellation Control Shaders

The tessellation control shader is used to read an input patch provided by the application and to produce an output patch. Each tessellation control shader invocation operates on an input patch (after all control points in the patch are processed by a vertex shader) and its associated data, and outputs a single control point of the output patch and its associated data, and can also output additional per-patch data. The input patch is sized according to the patchControlPoints member of VkPipelineTessellationStateCreateInfo, as part of input assembly. The size of the output patch is controlled by the OpExecutionMode OutputVertices specified in the tessellation control or tessellation evaluation shaders, which must be specified in at least one of the shaders. The size of the input and output patches must each be greater than zero and less than or equal to VkPhysicalDeviceLimits::maxTessellationPatchSize.

8.6.1. Tessellation Control Shader Execution

A tessellation control shader is invoked at least once for each output vertex in a patch. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view.

Inputs to the tessellation control shader are generated by the vertex shader. Each invocation of the tessellation control shader can read the attributes of any incoming vertices and their associated data. The invocations corresponding to a given patch execute logically in parallel, with undefined relative execution order. However, the OpControlBarrier instruction can be used to provide limited control of the execution order by synchronizing invocations within a patch, effectively dividing tessellation control shader execution into a set of phases. Tessellation control shaders will read undefined values if one invocation reads a per-vertex or per-patch attribute written by another invocation at any point during the same phase, or if two invocations attempt to write different values to the same per-patch output in a single phase.

8.7. Tessellation Evaluation Shaders

The Tessellation Evaluation Shader operates on an input patch of control points and their associated data, and a single input barycentric coordinate indicating the invocation’s relative position within the subdivided patch, and outputs a single vertex and its associated data.

8.7.1. Tessellation Evaluation Shader Execution

A tessellation evaluation shader is invoked at least once for each unique vertex generated by the tessellator. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view.

8.8. Geometry Shaders

The geometry shader operates on a group of vertices and their associated data assembled from a single input primitive, and emits zero or more output primitives and the group of vertices and their associated data required for each output primitive.

8.8.1. Geometry Shader Execution

A geometry shader is invoked at least once for each primitive produced by the tessellation stages, or at least once for each primitive generated by primitive assembly when tessellation is not in use. The number of geometry shader invocations per input primitive is determined from the invocation count of the geometry shader specified by the OpExecutionMode Invocations in the geometry shader. If the invocation count is not specified, then a default of one invocation is executed. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view.

8.9. Fragment Shaders

Fragment shaders are invoked as the result of rasterization in a graphics pipeline. Each fragment shader invocation operates on a single fragment and its associated data. With few exceptions, fragment shaders do not have access to any data associated with other fragments and are considered to execute in isolation of fragment shader invocations associated with other fragments.

8.9.1. Fragment Shader Execution

For each fragment generated by rasterization, a fragment shader may be invoked. A fragment shader must not be invoked if the Early Per-Fragment Tests cause it to have no coverage. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view.

Furthermore, if it is determined that a fragment generated as the result of rasterizing a first primitive will have its outputs entirely overwritten by a fragment generated as the result of rasterizing a second primitive in the same subpass, and the fragment shader used for the fragment has no other side effects, then the fragment shader may not be executed for the fragment from the first primitive.

Relative ordering of execution of different fragment shader invocations is not defined.

When a primitive (partially or fully) covers a pixel, the number of times the fragment shader is invoked is implementation-dependent, but must obey the following constraints:

  • Each covered sample is included in a single fragment shader invocation.

  • When sample shading is not enabled, there is at least one fragment shader invocation.

  • When sample shading is enabled, the minimum number of fragment shader invocations is as defined in Sample Shading.

When there is more than one fragment shader invocation per pixel, the association of samples to invocations is implementation-dependent.

In addition to the conditions outlined above for the invocation of a fragment shader, a fragment shader invocation may be produced as a helper invocation. A helper invocation is a fragment shader invocation that is created solely for the purposes of evaluating derivatives for use in non-helper fragment shader invocations. Stores and atomics performed by helper invocations must not have any effect on memory, and values returned by atomic instructions in helper invocations are undefined.

8.9.2. Early Fragment Tests

An explicit control is provided to allow fragment shaders to enable early fragment tests. If the fragment shader specifies the EarlyFragmentTests OpExecutionMode, the per-fragment tests described in Early Fragment Test Mode are performed prior to fragment shader execution. Otherwise, they are performed after fragment shader execution.

If the fragment shader additionally specifies the PostDepthCoverage OpExecutionMode, the value of a variable decorated with the SampleMask built-in reflects the coverage after the early fragment tests. Otherwise, it reflects the coverage before the early fragment tests.

8.10. Compute Shaders

Compute shaders are invoked via vkCmdDispatch and vkCmdDispatchIndirect commands. In general, they have access to similar resources as shader stages executing as part of a graphics pipeline.

Compute workloads are formed from groups of work items called workgroups and processed by the compute shader in the current compute pipeline. A workgroup is a collection of shader invocations that execute the same shader, potentially in parallel. Compute shaders execute in global workgroups which are divided into a number of local workgroups with a size that can be set by assigning a value to the LocalSize execution mode or via an object decorated by the WorkgroupSize decoration. An invocation within a local workgroup can share data with other members of the local workgroup through shared variables and issue memory and control flow barriers to synchronize with other members of the local workgroup.

8.11. Interpolation Decorations

Interpolation decorations control the behavior of attribute interpolation in the fragment shader stage. Interpolation decorations can be applied to Input storage class variables in the fragment shader stage’s interface, and control the interpolation behavior of those variables.

Inputs that could be interpolated can be decorated by at most one of the following decorations:

  • Flat: no interpolation

  • NoPerspective: linear interpolation (for lines and polygons).

Fragment input variables decorated with neither Flat nor NoPerspective use perspective-correct interpolation (for lines and polygons).

The presence of and type of interpolation is controlled by the above interpolation decorations as well as the auxiliary decorations Centroid and Sample.

A variable decorated with Flat will not be interpolated. Instead, it will have the same value for every fragment within a triangle. This value will come from a single provoking vertex. A variable decorated with Flat can also be decorated with Centroid or Sample, which will mean the same thing as decorating it only as Flat.

For fragment shader input variables decorated with neither Centroid nor Sample, the assigned variable may be interpolated anywhere within the pixel and a single value may be assigned to each sample within the pixel.

Centroid and Sample can be used to control the location and frequency of the sampling of the decorated fragment shader input. If a fragment shader input is decorated with Centroid, a single value may be assigned to that variable for all samples in the pixel, but that value must be interpolated to a location that lies in both the pixel and in the primitive being rendered, including any of the pixel’s samples covered by the primitive. Because the location at which the variable is interpolated may be different in neighboring pixels, and derivatives may be computed by computing differences between neighboring pixels, derivatives of centroid-sampled inputs may be less accurate than those for non-centroid interpolated variables. The PostDepthCoverage execution mode does not affect the determination of the centroid location. If a fragment shader input is decorated with Sample, a separate value must be assigned to that variable for each covered sample in the pixel, and that value must be sampled at the location of the individual sample. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the pixel center must be used for Centroid, Sample, and undecorated attribute interpolation.

Fragment shader inputs that are signed or unsigned integers, integer vectors, or any double-precision floating-point type must be decorated with Flat.

When the VK_AMD_shader_explicit_vertex_parameter device extension is enabled inputs can be also decorated with the CustomInterpAMD interpolation decoration, including fragment shader inputs that are signed or unsigned integers, integer vectors, or any double-precision floating-point type. Inputs decorated with CustomInterpAMD can only be accessed by the extended instruction InterpolateAtVertexAMD and allows accessing the value of the input for individual vertices of the primitive.

8.12. Static Use

A SPIR-V module declares a global object in memory using the OpVariable instruction, which results in a pointer x to that object. A specific entry point in a SPIR-V module is said to statically use that object if that entry point’s call tree contains a function that contains a memory instruction or image instruction with x as an id operand. See the “Memory Instructions” and “Image Instructions” subsections of section 3 “Binary Form” of the SPIR-V specification for the complete list of SPIR-V memory instructions.

Static use is not used to control the behavior of variables with Input and Output storage. The effects of those variables are applied based only on whether they are present in a shader entry point’s interface.

8.13. Invocation and Derivative Groups

An invocation group (see the subsection “Control Flow” of section 2 of the SPIR-V specification) for a compute shader is the set of invocations in a single local workgroup. For graphics shaders, an invocation group is an implementation-dependent subset of the set of shader invocations of a given shader stage which are produced by a single drawing command. For indirect drawing commands with drawCount greater than one, invocations from separate draws are in distinct invocation groups.

Note

Because the partitioning of invocations into invocation groups is implementation-dependent and not observable, applications generally need to assume the worst case of all invocations in a draw belonging to a single invocation group.

A derivative group (see the subsection “Control Flow” of section 2 of the SPIR-V 1.00 Revision 4 specification) for a fragment shader is the set of invocations generated by a single primitive (point, line, or triangle), including any helper invocations generated by that primitive. Derivatives are undefined for a sampled image instruction if the instruction is in flow control that is not uniform across the derivative group.

8.14. Subgroups

A subgroup (see the subsection ``Control Flow'' of section 2 of the SPIR-V 1.3 Revision 1 specification) is a set of invocations that can synchronize and share data with each other efficiently. An invocation group is partitioned into one or more subgroups.

Subgroup operations are divided into various categories as described in VkSubgroupFeatureFlagBits.

8.14.1. Basic Subgroup Operations

The basic subgroup operations allow two classes of functionality within shaders - elect and barrier. Invocations within a subgroup can choose a single invocation to perform some task for the subgroup as a whole using elect. Invocations within a subgroup can perform a subgroup barrier to ensure the ordering of execution or memory accesses within a subgroup. Barriers can be performed on buffer memory accesses, WorkgroupLocal memory accesses, and image memory accesses to ensure that any results written are visible by other invocations within the subgroup. An OpControlBarrier can also be used to perform a full execution control barrier. A full execution control barrier will ensure that each active invocation within the subgroup reaches a point of execution before any are allowed to continue.

8.14.2. Vote Subgroup Operations

The vote subgroup operations allow invocations within a subgroup to compare values across a subgroup. The types of votes enabled are:

  • Do all active subgroup invocations agree that an expression is true?

  • Do any active subgroup invocations evaluate an expression to true?

  • Do all active subgroup invocations have the same value of an expression?

Note

These operations are useful in combination with control flow in that they allow for developers to check whether conditions match across the subgroup and choose potentially faster code-paths in these cases.

8.14.3. Arithmetic Subgroup Operations

The arithmetic subgroup operations allow invocations to perform scan and reduction operations across a subgroup. For reduction operations, each invocation in a subgroup will obtain the same result of these arithmetic operations applied across the subgroup. For scan operations, each invocation in the subgroup will perform an inclusive or exclusive scan, cumulatively applying the operation across the invocations in a subgroup in linear order. The operations supported are add, mul, min, max, and, or, xor.

8.14.4. Ballot Subgroup Operations

The ballot subgroup operations allow invocations to perform more complex votes across the subgroup. The ballot functionality allows all invocations within a subgroup to provide a boolean value and get as a result what each invocation provided as their boolean value. The broadcast functionality allows values to be broadcast from an invocation to all other invocations within the subgroup, given that the invocation to be broadcast from is known at pipeline creation time.

8.14.5. Shuffle Subgroup Operations

The shuffle subgroup operations allow invocations to read values from other invocations within a subgroup.

8.14.6. Shuffle Relative Subgroup Operations

The shuffle relative subgroup operations allow invocations to read values from other invocations within the subgroup relative to the current invocation in the group. The relative operations supported allow data to be shifted up and down through the invocations within a subgroup.

8.14.7. Clustered Subgroup Operations

The clustered subgroup operations allow invocations to perform an operation among partitions of a subgroup, such that the operation is only performed within the subgroup invocations within a partition. The partitions for clustered subgroup operations are consecutive power-of-two size groups of invocations and the cluster size must be known at pipeline creation time. The operations supported are add, mul, min, max, and, or, xor.

8.14.8. Quad Subgroup Operations

The quad subgroup operations allow clusters of 4 invocations (a quad), to share data efficiently with each other.

8.14.9. Partitioned Subgroup Operations

The partitioned subgroup operations allow invocations to perform an operation among partitions of a subgroup, such that the operation is only performed within the subgroup invocations within a partition. The partitions for partitioned subgroup operations can group the invocations into arbitrary subsets and can be computed at runtime. The operations supported are add, mul, min, max, and, or, xor.

8.15. Validation Cache

Validation cache objects allow the result of internal validation to be reused, both within a single application run and between multiple runs. Reuse within a single run is achieved by passing the same validation cache object when creating supported Vulkan objects. Reuse across runs of an application is achieved by retrieving validation cache contents in one run of an application, saving the contents, and using them to preinitialize a validation cache on a subsequent run. The contents of the validation cache objects are managed by the validation layers. Applications can manage the host memory consumed by a validation cache object and control the amount of data retrieved from a validation cache object.

Validation cache objects are represented by VkValidationCacheEXT handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkValidationCacheEXT)

To create validation cache objects, call:

VkResult vkCreateValidationCacheEXT(
    VkDevice                                    device,
    const VkValidationCacheCreateInfoEXT*       pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkValidationCacheEXT*                       pValidationCache);
  • device is the logical device that creates the validation cache object.

  • pCreateInfo is a pointer to a VkValidationCacheCreateInfoEXT structure that contains the initial parameters for the validation cache object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pValidationCache is a pointer to a VkValidationCacheEXT handle in which the resulting validation cache object is returned.

Note

Applications can track and manage the total host memory size of a validation cache object using the pAllocator. Applications can limit the amount of data retrieved from a validation cache object in vkGetValidationCacheDataEXT. Implementations should not internally limit the total number of entries added to a validation cache object or the total host memory consumed.

Once created, a validation cache can be passed to the vkCreateShaderModule command as part of the VkShaderModuleCreateInfo pNext chain. If a VkShaderModuleValidationCacheCreateInfoEXT object is part of the VkShaderModuleCreateInfo::pNext chain, and its validationCache field is not VK_NULL_HANDLE, the implementation will query it for possible reuse opportunities and update it with new content. The use of the validation cache object in these commands is internally synchronized, and the same validation cache object can be used in multiple threads simultaneously.

Note

Implementations should make every effort to limit any critical sections to the actual accesses to the cache, which is expected to be significantly shorter than the duration of the vkCreateShaderModule command.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkValidationCacheCreateInfoEXT structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pValidationCache must be a valid pointer to a VkValidationCacheEXT handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkValidationCacheCreateInfoEXT structure is defined as:

typedef struct VkValidationCacheCreateInfoEXT {
    VkStructureType                    sType;
    const void*                        pNext;
    VkValidationCacheCreateFlagsEXT    flags;
    size_t                             initialDataSize;
    const void*                        pInitialData;
} VkValidationCacheCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • initialDataSize is the number of bytes in pInitialData. If initialDataSize is zero, the validation cache will initially be empty.

  • pInitialData is a pointer to previously retrieved validation cache data. If the validation cache data is incompatible (as defined below) with the device, the validation cache will be initially empty. If initialDataSize is zero, pInitialData is ignored.

Valid Usage
  • If initialDataSize is not 0, it must be equal to the size of pInitialData, as returned by vkGetValidationCacheDataEXT when pInitialData was originally retrieved

  • If initialDataSize is not 0, pInitialData must have been retrieved from a previous call to vkGetValidationCacheDataEXT

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_VALIDATION_CACHE_CREATE_INFO_EXT

  • pNext must be NULL

  • flags must be 0

  • If initialDataSize is not 0, pInitialData must be a valid pointer to an array of initialDataSize bytes

typedef VkFlags VkValidationCacheCreateFlagsEXT;

VkValidationCacheCreateFlagsEXT is a bitmask type for setting a mask, but is currently reserved for future use.

Validation cache objects can be merged using the command:

VkResult vkMergeValidationCachesEXT(
    VkDevice                                    device,
    VkValidationCacheEXT                        dstCache,
    uint32_t                                    srcCacheCount,
    const VkValidationCacheEXT*                 pSrcCaches);
  • device is the logical device that owns the validation cache objects.

  • dstCache is the handle of the validation cache to merge results into.

  • srcCacheCount is the length of the pSrcCaches array.

  • pSrcCaches is an array of validation cache handles, which will be merged into dstCache. The previous contents of dstCache are included after the merge.

Note

The details of the merge operation are implementation dependent, but implementations should merge the contents of the specified validation caches and prune duplicate entries.

Valid Usage
  • dstCache must not appear in the list of source caches

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • dstCache must be a valid VkValidationCacheEXT handle

  • pSrcCaches must be a valid pointer to an array of srcCacheCount valid VkValidationCacheEXT handles

  • srcCacheCount must be greater than 0

  • dstCache must have been created, allocated, or retrieved from device

  • Each element of pSrcCaches must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to dstCache must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Data can be retrieved from a validation cache object using the command:

VkResult vkGetValidationCacheDataEXT(
    VkDevice                                    device,
    VkValidationCacheEXT                        validationCache,
    size_t*                                     pDataSize,
    void*                                       pData);
  • device is the logical device that owns the validation cache.

  • validationCache is the validation cache to retrieve data from.

  • pDataSize is a pointer to a value related to the amount of data in the validation cache, as described below.

  • pData is either NULL or a pointer to a buffer.

If pData is NULL, then the maximum size of the data that can be retrieved from the validation cache, in bytes, is returned in pDataSize. Otherwise, pDataSize must point to a variable set by the user to the size of the buffer, in bytes, pointed to by pData, and on return the variable is overwritten with the amount of data actually written to pData.

If pDataSize is less than the maximum size that can be retrieved by the validation cache, at most pDataSize bytes will be written to pData, and vkGetValidationCacheDataEXT will return VK_INCOMPLETE. Any data written to pData is valid and can be provided as the pInitialData member of the VkValidationCacheCreateInfoEXT structure passed to vkCreateValidationCacheEXT.

Two calls to vkGetValidationCacheDataEXT with the same parameters must retrieve the same data unless a command that modifies the contents of the cache is called between them.

Applications can store the data retrieved from the validation cache, and use these data, possibly in a future run of the application, to populate new validation cache objects. The results of validation, however, may depend on the vendor ID, device ID, driver version, and other details of the device. To enable applications to detect when previously retrieved data is incompatible with the device, the initial bytes written to pData must be a header consisting of the following members:

Table 11. Layout for validation cache header version VK_VALIDATION_CACHE_HEADER_VERSION_ONE_EXT
Offset Size Meaning

0

4

length in bytes of the entire validation cache header written as a stream of bytes, with the least significant byte first

4

4

a VkValidationCacheHeaderVersionEXT value written as a stream of bytes, with the least significant byte first

8

VK_UUID_SIZE

a layer commit ID expressed as a UUID, which uniquely identifies the version of the validation layers used to generate these validation results

The first four bytes encode the length of the entire validation cache header, in bytes. This value includes all fields in the header including the validation cache version field and the size of the length field.

The next four bytes encode the validation cache version, as described for VkValidationCacheHeaderVersionEXT. A consumer of the validation cache should use the cache version to interpret the remainder of the cache header.

If pDataSize is less than what is necessary to store this header, nothing will be written to pData and zero will be written to pDataSize.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • validationCache must be a valid VkValidationCacheEXT handle

  • pDataSize must be a valid pointer to a size_t value

  • If the value referenced by pDataSize is not 0, and pData is not NULL, pData must be a valid pointer to an array of pDataSize bytes

  • validationCache must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Possible values of the second group of four bytes in the header returned by vkGetValidationCacheDataEXT, encoding the validation cache version, are:

typedef enum VkValidationCacheHeaderVersionEXT {
    VK_VALIDATION_CACHE_HEADER_VERSION_ONE_EXT = 1,
} VkValidationCacheHeaderVersionEXT;
  • VK_VALIDATION_CACHE_HEADER_VERSION_ONE_EXT specifies version one of the validation cache.

To destroy a validation cache, call:

void vkDestroyValidationCacheEXT(
    VkDevice                                    device,
    VkValidationCacheEXT                        validationCache,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the validation cache object.

  • validationCache is the handle of the validation cache to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when validationCache was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when validationCache was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If validationCache is not VK_NULL_HANDLE, validationCache must be a valid VkValidationCacheEXT handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If validationCache is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to validationCache must be externally synchronized

9. Pipelines

The following figure shows a block diagram of the Vulkan pipelines. Some Vulkan commands specify geometric objects to be drawn or computational work to be performed, while others specify state controlling how objects are handled by the various pipeline stages, or control data transfer between memory organized as images and buffers. Commands are effectively sent through a processing pipeline, either a graphics pipeline or a compute pipeline.

The first stage of the graphics pipeline (Input Assembler) assembles vertices to form geometric primitives such as points, lines, and triangles, based on a requested primitive topology. In the next stage (Vertex Shader) vertices can be transformed, computing positions and attributes for each vertex. If tessellation and/or geometry shaders are supported, they can then generate multiple primitives from a single input primitive, possibly changing the primitive topology or generating additional attribute data in the process.

The final resulting primitives are clipped to a clip volume in preparation for the next stage, Rasterization. The rasterizer produces a series of framebuffer addresses and values using a two-dimensional description of a point, line segment, or triangle. Each fragment so produced is fed to the next stage (Fragment Shader) that performs operations on individual fragments before they finally alter the framebuffer. These operations include conditional updates into the framebuffer based on incoming and previously stored depth values (to effect depth buffering), blending of incoming fragment colors with stored colors, as well as masking, stenciling, and other logical operations on fragment values.

Framebuffer operations read and write the color and depth/stencil attachments of the framebuffer for a given subpass of a render pass instance. The attachments can be used as input attachments in the fragment shader in a later subpass of the same render pass.

The compute pipeline is a separate pipeline from the graphics pipeline, which operates on one-, two-, or three-dimensional workgroups which can read from and write to buffer and image memory.

This ordering is meant only as a tool for describing Vulkan, not as a strict rule of how Vulkan is implemented, and we present it only as a means to organize the various operations of the pipelines. Actual ordering guarantees between pipeline stages are explained in detail in the synchronization chapter.

pipeline
Figure 2. Block diagram of the Vulkan pipeline

Each pipeline is controlled by a monolithic object created from a description of all of the shader stages and any relevant fixed-function stages. Linking the whole pipeline together allows the optimization of shaders based on their input/outputs and eliminates expensive draw time state validation.

A pipeline object is bound to the current state using vkCmdBindPipeline. Any pipeline object state that is specified as dynamic is not applied to the current state when the pipeline object is bound, but is instead set by dynamic state setting commands.

No state, including dynamic state, is inherited from one command buffer to another.

Compute and graphics pipelines are each represented by VkPipeline handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPipeline)

9.1. Compute Pipelines

Compute pipelines consist of a single static compute shader stage and the pipeline layout.

The compute pipeline represents a compute shader and is created by calling vkCreateComputePipelines with module and pName selecting an entry point from a shader module, where that entry point defines a valid compute shader, in the VkPipelineShaderStageCreateInfo structure contained within the VkComputePipelineCreateInfo structure.

To create compute pipelines, call:

VkResult vkCreateComputePipelines(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    uint32_t                                    createInfoCount,
    const VkComputePipelineCreateInfo*          pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkPipeline*                                 pPipelines);
  • device is the logical device that creates the compute pipelines.

  • pipelineCache is either VK_NULL_HANDLE, indicating that pipeline caching is disabled; or the handle of a valid pipeline cache object, in which case use of that cache is enabled for the duration of the command.

  • createInfoCount is the length of the pCreateInfos and pPipelines arrays.

  • pCreateInfos is an array of VkComputePipelineCreateInfo structures.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pPipelines is a pointer to an array in which the resulting compute pipeline objects are returned.

    editing-note

    TODO (Jon) - Should we say something like “the i’th element of the pPipelines array is created based on the corresponding element of the pCreateInfos array”? Also for vkCreateGraphicsPipelines below.

Valid Usage
  • If the flags member of any element of pCreateInfos contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and the basePipelineIndex member of that same element is not -1, basePipelineIndex must be less than the index into pCreateInfos that corresponds to that element

  • If the flags member of any element of pCreateInfos contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, the base pipeline must have been created with the VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT flag set

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If pipelineCache is not VK_NULL_HANDLE, pipelineCache must be a valid VkPipelineCache handle

  • pCreateInfos must be a valid pointer to an array of createInfoCount valid VkComputePipelineCreateInfo structures

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pPipelines must be a valid pointer to an array of createInfoCount VkPipeline handles

  • createInfoCount must be greater than 0

  • If pipelineCache is a valid handle, it must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INVALID_SHADER_NV

The VkComputePipelineCreateInfo structure is defined as:

typedef struct VkComputePipelineCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkPipelineCreateFlags              flags;
    VkPipelineShaderStageCreateInfo    stage;
    VkPipelineLayout                   layout;
    VkPipeline                         basePipelineHandle;
    int32_t                            basePipelineIndex;
} VkComputePipelineCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkPipelineCreateFlagBits specifying how the pipeline will be generated.

  • stage is a VkPipelineShaderStageCreateInfo describing the compute shader.

  • layout is the description of binding locations used by both the pipeline and descriptor sets used with the pipeline.

  • basePipelineHandle is a pipeline to derive from

  • basePipelineIndex is an index into the pCreateInfos parameter to use as a pipeline to derive from

The parameters basePipelineHandle and basePipelineIndex are described in more detail in Pipeline Derivatives.

stage points to a structure of type VkPipelineShaderStageCreateInfo.

Valid Usage
  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid handle to a compute VkPipeline

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineHandle is VK_NULL_HANDLE, basePipelineIndex must be a valid index into the calling command’s pCreateInfos parameter

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is not -1, basePipelineHandle must be VK_NULL_HANDLE

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineHandle is not VK_NULL_HANDLE, basePipelineIndex must be -1

  • The stage member of stage must be VK_SHADER_STAGE_COMPUTE_BIT

  • The shader code for the entry point identified by stage and the rest of the state identified by this structure must adhere to the pipeline linking rules described in the Shader Interfaces chapter

  • layout must be consistent with the layout of the compute shader specified in stage

  • The number of resources in layout accessible to the compute shader stage must be less than or equal to VkPhysicalDeviceLimits::maxPerStageResources

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_CREATE_INFO

  • pNext must be NULL

  • flags must be a valid combination of VkPipelineCreateFlagBits values

  • stage must be a valid VkPipelineShaderStageCreateInfo structure

  • layout must be a valid VkPipelineLayout handle

  • Both of basePipelineHandle, and layout that are valid handles must have been created, allocated, or retrieved from the same VkDevice

The VkPipelineShaderStageCreateInfo structure is defined as:

typedef struct VkPipelineShaderStageCreateInfo {
    VkStructureType                     sType;
    const void*                         pNext;
    VkPipelineShaderStageCreateFlags    flags;
    VkShaderStageFlagBits               stage;
    VkShaderModule                      module;
    const char*                         pName;
    const VkSpecializationInfo*         pSpecializationInfo;
} VkPipelineShaderStageCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • stage is a VkShaderStageFlagBits value specifying a single pipeline stage.

  • module is a VkShaderModule object that contains the shader for this stage.

  • pName is a pointer to a null-terminated UTF-8 string specifying the entry point name of the shader for this stage.

  • pSpecializationInfo is a pointer to VkSpecializationInfo, as described in Specialization Constants, and can be NULL.

Valid Usage
  • If the geometry shaders feature is not enabled, stage must not be VK_SHADER_STAGE_GEOMETRY_BIT

  • If the tessellation shaders feature is not enabled, stage must not be VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT or VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT

  • stage must not be VK_SHADER_STAGE_ALL_GRAPHICS, or VK_SHADER_STAGE_ALL

  • pName must be the name of an OpEntryPoint in module with an execution model that matches stage

  • If the identified entry point includes any variable in its interface that is declared with the ClipDistance BuiltIn decoration, that variable must not have an array size greater than VkPhysicalDeviceLimits::maxClipDistances

  • If the identified entry point includes any variable in its interface that is declared with the CullDistance BuiltIn decoration, that variable must not have an array size greater than VkPhysicalDeviceLimits::maxCullDistances

  • If the identified entry point includes any variables in its interface that are declared with the ClipDistance or CullDistance BuiltIn decoration, those variables must not have array sizes which sum to more than VkPhysicalDeviceLimits::maxCombinedClipAndCullDistances

  • If the identified entry point includes any variable in its interface that is declared with the SampleMask BuiltIn decoration, that variable must not have an array size greater than VkPhysicalDeviceLimits::maxSampleMaskWords

  • If stage is VK_SHADER_STAGE_VERTEX_BIT, the identified entry point must not include any input variable in its interface that is decorated with CullDistance

  • If stage is VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT or VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT, and the identified entry point has an OpExecutionMode instruction that specifies a patch size with OutputVertices, the patch size must be greater than 0 and less than or equal to VkPhysicalDeviceLimits::maxTessellationPatchSize

  • If stage is VK_SHADER_STAGE_GEOMETRY_BIT, the identified entry point must have an OpExecutionMode instruction that specifies a maximum output vertex count that is greater than 0 and less than or equal to VkPhysicalDeviceLimits::maxGeometryOutputVertices

  • If stage is VK_SHADER_STAGE_GEOMETRY_BIT, the identified entry point must have an OpExecutionMode instruction that specifies an invocation count that is greater than 0 and less than or equal to VkPhysicalDeviceLimits::maxGeometryShaderInvocations

  • If stage is VK_SHADER_STAGE_GEOMETRY_BIT, and the identified entry point writes to Layer for any primitive, it must write the same value to Layer for all vertices of a given primitive

  • If stage is VK_SHADER_STAGE_GEOMETRY_BIT, and the identified entry point writes to ViewportIndex for any primitive, it must write the same value to ViewportIndex for all vertices of a given primitive

  • If stage is VK_SHADER_STAGE_FRAGMENT_BIT, the identified entry point must not include any output variables in its interface decorated with CullDistance

  • If stage is VK_SHADER_STAGE_FRAGMENT_BIT, and the identified entry point writes to FragDepth in any execution path, it must write to FragDepth in all execution paths

  • If stage is VK_SHADER_STAGE_FRAGMENT_BIT, and the identified entry point writes to FragStencilRefEXT in any execution path, it must write to FragStencilRefEXT in all execution paths

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • stage must be a valid VkShaderStageFlagBits value

  • module must be a valid VkShaderModule handle

  • pName must be a null-terminated UTF-8 string

  • If pSpecializationInfo is not NULL, pSpecializationInfo must be a valid pointer to a valid VkSpecializationInfo structure

typedef VkFlags VkPipelineShaderStageCreateFlags;

VkPipelineShaderStageCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

Commands and structures which need to specify one or more shader stages do so using a bitmask whose bits correspond to stages. Bits which can be set to specify shader stages are:

typedef enum VkShaderStageFlagBits {
    VK_SHADER_STAGE_VERTEX_BIT = 0x00000001,
    VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT = 0x00000002,
    VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT = 0x00000004,
    VK_SHADER_STAGE_GEOMETRY_BIT = 0x00000008,
    VK_SHADER_STAGE_FRAGMENT_BIT = 0x00000010,
    VK_SHADER_STAGE_COMPUTE_BIT = 0x00000020,
    VK_SHADER_STAGE_ALL_GRAPHICS = 0x0000001F,
    VK_SHADER_STAGE_ALL = 0x7FFFFFFF,
} VkShaderStageFlagBits;
  • VK_SHADER_STAGE_VERTEX_BIT specifies the vertex stage.

  • VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT specifies the tessellation control stage.

  • VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT specifies the tessellation evaluation stage.

  • VK_SHADER_STAGE_GEOMETRY_BIT specifies the geometry stage.

  • VK_SHADER_STAGE_FRAGMENT_BIT specifies the fragment stage.

  • VK_SHADER_STAGE_COMPUTE_BIT specifies the compute stage.

  • VK_SHADER_STAGE_ALL_GRAPHICS is a combination of bits used as shorthand to specify all graphics stages defined above (excluding the compute stage).

  • VK_SHADER_STAGE_ALL is a combination of bits used as shorthand to specify all shader stages supported by the device, including all additional stages which are introduced by extensions.

typedef VkFlags VkShaderStageFlags;

VkShaderStageFlags is a bitmask type for setting a mask of zero or more VkShaderStageFlagBits.

9.2. Graphics Pipelines

Graphics pipelines consist of multiple shader stages, multiple fixed-function pipeline stages, and a pipeline layout.

To create graphics pipelines, call:

VkResult vkCreateGraphicsPipelines(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    uint32_t                                    createInfoCount,
    const VkGraphicsPipelineCreateInfo*         pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkPipeline*                                 pPipelines);
  • device is the logical device that creates the graphics pipelines.

  • pipelineCache is either VK_NULL_HANDLE, indicating that pipeline caching is disabled; or the handle of a valid pipeline cache object, in which case use of that cache is enabled for the duration of the command.

  • createInfoCount is the length of the pCreateInfos and pPipelines arrays.

  • pCreateInfos is an array of VkGraphicsPipelineCreateInfo structures.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pPipelines is a pointer to an array in which the resulting graphics pipeline objects are returned.

The VkGraphicsPipelineCreateInfo structure includes an array of shader create info structures containing all the desired active shader stages, as well as creation info to define all relevant fixed-function stages, and a pipeline layout.

Valid Usage
  • If the flags member of any element of pCreateInfos contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and the basePipelineIndex member of that same element is not -1, basePipelineIndex must be less than the index into pCreateInfos that corresponds to that element

  • If the flags member of any element of pCreateInfos contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, the base pipeline must have been created with the VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT flag set

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If pipelineCache is not VK_NULL_HANDLE, pipelineCache must be a valid VkPipelineCache handle

  • pCreateInfos must be a valid pointer to an array of createInfoCount valid VkGraphicsPipelineCreateInfo structures

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pPipelines must be a valid pointer to an array of createInfoCount VkPipeline handles

  • createInfoCount must be greater than 0

  • If pipelineCache is a valid handle, it must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INVALID_SHADER_NV

The VkGraphicsPipelineCreateInfo structure is defined as:

typedef struct VkGraphicsPipelineCreateInfo {
    VkStructureType                                  sType;
    const void*                                      pNext;
    VkPipelineCreateFlags                            flags;
    uint32_t                                         stageCount;
    const VkPipelineShaderStageCreateInfo*           pStages;
    const VkPipelineVertexInputStateCreateInfo*      pVertexInputState;
    const VkPipelineInputAssemblyStateCreateInfo*    pInputAssemblyState;
    const VkPipelineTessellationStateCreateInfo*     pTessellationState;
    const VkPipelineViewportStateCreateInfo*         pViewportState;
    const VkPipelineRasterizationStateCreateInfo*    pRasterizationState;
    const VkPipelineMultisampleStateCreateInfo*      pMultisampleState;
    const VkPipelineDepthStencilStateCreateInfo*     pDepthStencilState;
    const VkPipelineColorBlendStateCreateInfo*       pColorBlendState;
    const VkPipelineDynamicStateCreateInfo*          pDynamicState;
    VkPipelineLayout                                 layout;
    VkRenderPass                                     renderPass;
    uint32_t                                         subpass;
    VkPipeline                                       basePipelineHandle;
    int32_t                                          basePipelineIndex;
} VkGraphicsPipelineCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkPipelineCreateFlagBits specifying how the pipeline will be generated.

  • stageCount is the number of entries in the pStages array.

  • pStages is an array of size stageCount structures of type VkPipelineShaderStageCreateInfo describing the set of the shader stages to be included in the graphics pipeline.

  • pVertexInputState is a pointer to an instance of the VkPipelineVertexInputStateCreateInfo structure.

  • pInputAssemblyState is a pointer to an instance of the VkPipelineInputAssemblyStateCreateInfo structure which determines input assembly behavior, as described in Drawing Commands.

  • pTessellationState is a pointer to an instance of the VkPipelineTessellationStateCreateInfo structure, and is ignored if the pipeline does not include a tessellation control shader stage and tessellation evaluation shader stage.

  • pViewportState is a pointer to an instance of the VkPipelineViewportStateCreateInfo structure, and is ignored if the pipeline has rasterization disabled.

  • pRasterizationState is a pointer to an instance of the VkPipelineRasterizationStateCreateInfo structure.

  • pMultisampleState is a pointer to an instance of the VkPipelineMultisampleStateCreateInfo, and is ignored if the pipeline has rasterization disabled.

  • pDepthStencilState is a pointer to an instance of the VkPipelineDepthStencilStateCreateInfo structure, and is ignored if the pipeline has rasterization disabled or if the subpass of the render pass the pipeline is created against does not use a depth/stencil attachment.

  • pColorBlendState is a pointer to an instance of the VkPipelineColorBlendStateCreateInfo structure, and is ignored if the pipeline has rasterization disabled or if the subpass of the render pass the pipeline is created against does not use any color attachments.

  • pDynamicState is a pointer to VkPipelineDynamicStateCreateInfo and is used to indicate which properties of the pipeline state object are dynamic and can be changed independently of the pipeline state. This can be NULL, which means no state in the pipeline is considered dynamic.

  • layout is the description of binding locations used by both the pipeline and descriptor sets used with the pipeline.

  • renderPass is a handle to a render pass object describing the environment in which the pipeline will be used; the pipeline must only be used with an instance of any render pass compatible with the one provided. See Render Pass Compatibility for more information.

  • subpass is the index of the subpass in the render pass where this pipeline will be used.

  • basePipelineHandle is a pipeline to derive from.

  • basePipelineIndex is an index into the pCreateInfos parameter to use as a pipeline to derive from.

The parameters basePipelineHandle and basePipelineIndex are described in more detail in Pipeline Derivatives.

pStages points to an array of VkPipelineShaderStageCreateInfo structures, which were previously described in Compute Pipelines.

pDynamicState points to a structure of type VkPipelineDynamicStateCreateInfo.

If any shader stage fails to compile, the compile log will be reported back to the application, and VK_ERROR_INVALID_SHADER_NV will be generated.

Valid Usage
  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid handle to a graphics VkPipeline

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineHandle is VK_NULL_HANDLE, basePipelineIndex must be a valid index into the calling command’s pCreateInfos parameter

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is not -1, basePipelineHandle must be VK_NULL_HANDLE

  • If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineHandle is not VK_NULL_HANDLE, basePipelineIndex must be -1

  • The stage member of each element of pStages must be unique

  • The stage member of one element of pStages must be VK_SHADER_STAGE_VERTEX_BIT

  • The stage member of each element of pStages must not be VK_SHADER_STAGE_COMPUTE_BIT

  • If pStages includes a tessellation control shader stage, it must include a tessellation evaluation shader stage

  • If pStages includes a tessellation evaluation shader stage, it must include a tessellation control shader stage

  • If pStages includes a tessellation control shader stage and a tessellation evaluation shader stage, pTessellationState must be a valid pointer to a valid VkPipelineTessellationStateCreateInfo structure

  • If pStages includes tessellation shader stages, the shader code of at least one stage must contain an OpExecutionMode instruction that specifies the type of subdivision in the pipeline

  • If pStages includes tessellation shader stages, and the shader code of both stages contain an OpExecutionMode instruction that specifies the type of subdivision in the pipeline, they must both specify the same subdivision mode

  • If pStages includes tessellation shader stages, the shader code of at least one stage must contain an OpExecutionMode instruction that specifies the output patch size in the pipeline

  • If pStages includes tessellation shader stages, and the shader code of both contain an OpExecutionMode instruction that specifies the out patch size in the pipeline, they must both specify the same patch size

  • If pStages includes tessellation shader stages, the topology member of pInputAssembly must be VK_PRIMITIVE_TOPOLOGY_PATCH_LIST

  • If the topology member of pInputAssembly is VK_PRIMITIVE_TOPOLOGY_PATCH_LIST, pStages must include tessellation shader stages

  • If pStages includes a geometry shader stage, and does not include any tessellation shader stages, its shader code must contain an OpExecutionMode instruction that specifies an input primitive type that is compatible with the primitive topology specified in pInputAssembly

  • If pStages includes a geometry shader stage, and also includes tessellation shader stages, its shader code must contain an OpExecutionMode instruction that specifies an input primitive type that is compatible with the primitive topology that is output by the tessellation stages

  • If pStages includes a fragment shader stage and a geometry shader stage, and the fragment shader code reads from an input variable that is decorated with PrimitiveID, then the geometry shader code must write to a matching output variable, decorated with PrimitiveID, in all execution paths

  • If pStages includes a fragment shader stage, its shader code must not read from any input attachment that is defined as VK_ATTACHMENT_UNUSED in subpass

  • The shader code for the entry points identified by pStages, and the rest of the state identified by this structure must adhere to the pipeline linking rules described in the Shader Interfaces chapter

  • If rasterization is not disabled and subpass uses a depth/stencil attachment in renderPass that has a layout of VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL in the VkAttachmentReference defined by subpass, the depthWriteEnable member of pDepthStencilState must be VK_FALSE

  • If rasterization is not disabled and subpass uses a depth/stencil attachment in renderPass that has a layout of VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL in the VkAttachmentReference defined by subpass, the failOp, passOp and depthFailOp members of each of the front and back members of pDepthStencilState must be VK_STENCIL_OP_KEEP

  • If rasterization is not disabled and the subpass uses color attachments, then for each color attachment in the subpass the blendEnable member of the corresponding element of the pAttachment member of pColorBlendState must be VK_FALSE if the format of the attachment does not support color blend operations, as specified by the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT flag in VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties

  • If rasterization is not disabled and the subpass uses color attachments, the attachmentCount member of pColorBlendState must be equal to the colorAttachmentCount used to create subpass

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_VIEWPORT, the pViewports member of pViewportState must be a valid pointer to an array of pViewportState::viewportCount VkViewport structures

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_SCISSOR, the pScissors member of pViewportState must be a valid pointer to an array of pViewportState::scissorCount VkRect2D structures

  • If the wide lines feature is not enabled, and no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_LINE_WIDTH, the lineWidth member of pRasterizationState must be 1.0

  • If the rasterizerDiscardEnable member of pRasterizationState is VK_FALSE, pViewportState must be a valid pointer to a valid VkPipelineViewportStateCreateInfo structure

  • If the rasterizerDiscardEnable member of pRasterizationState is VK_FALSE, pMultisampleState must be a valid pointer to a valid VkPipelineMultisampleStateCreateInfo structure

  • If the rasterizerDiscardEnable member of pRasterizationState is VK_FALSE, and subpass uses a depth/stencil attachment, pDepthStencilState must be a valid pointer to a valid VkPipelineDepthStencilStateCreateInfo structure

  • If the rasterizerDiscardEnable member of pRasterizationState is VK_FALSE, and subpass uses color attachments, pColorBlendState must be a valid pointer to a valid VkPipelineColorBlendStateCreateInfo structure

  • If the depth bias clamping feature is not enabled, no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_DEPTH_BIAS, and the depthBiasEnable member of pRasterizationState is VK_TRUE, the depthBiasClamp member of pRasterizationState must be 0.0

  • If the VK_EXT_depth_range_unrestricted extension is not enabled and no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_DEPTH_BOUNDS, and the depthBoundsTestEnable member of pDepthStencilState is VK_TRUE, the minDepthBounds and maxDepthBounds members of pDepthStencilState must be between 0.0 and 1.0, inclusive

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT, and the sampleLocationsEnable member of a VkPipelineSampleLocationsStateCreateInfoEXT structure chained to the pNext chain of pMultisampleState is VK_TRUE, sampleLocationsInfo.sampleLocationGridSize.width must evenly divide VkMultisamplePropertiesEXT::sampleLocationGridSize.width as returned by vkGetPhysicalDeviceMultisamplePropertiesEXT with a samples parameter equaling rasterizationSamples

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT, and the sampleLocationsEnable member of a VkPipelineSampleLocationsStateCreateInfoEXT structure chained to the pNext chain of pMultisampleState is VK_TRUE, sampleLocationsInfo.sampleLocationGridSize.height must evenly divide VkMultisamplePropertiesEXT::sampleLocationGridSize.height as returned by vkGetPhysicalDeviceMultisamplePropertiesEXT with a samples parameter equaling rasterizationSamples

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT, and the sampleLocationsEnable member of a VkPipelineSampleLocationsStateCreateInfoEXT structure chained to the pNext chain of pMultisampleState is VK_TRUE, sampleLocationsInfo.sampleLocationsPerPixel must equal rasterizationSamples

  • If the sampleLocationsEnable member of a VkPipelineSampleLocationsStateCreateInfoEXT structure chained to the pNext chain of pMultisampleState is VK_TRUE, the fragment shader code must not statically use the extended instruction InterpolateAtSample

  • layout must be consistent with all shaders specified in pStages

  • If subpass uses color and/or depth/stencil attachments, then the rasterizationSamples member of pMultisampleState must equal the maximum of the sample counts of those subpass attachments

  • If subpass has a depth/stencil attachment and depth test, stencil test, or depth bounds test are enabled, then the rasterizationSamples member of pMultisampleState must be the same as the sample count of the depth/stencil attachment

  • If subpass has any color attachments, then the rasterizationSamples member of pMultisampleState must be greater than or equal to the sample count for those subpass attachments

  • If subpass does not use any color and/or depth/stencil attachments, then the rasterizationSamples member of pMultisampleState must follow the rules for a zero-attachment subpass

  • subpass must be a valid subpass within renderPass

  • If the renderPass has multiview enabled and subpass has more than one bit set in the view mask and multiviewTessellationShader is not enabled, then pStages must not include tessellation shaders.

  • If the renderPass has multiview enabled and subpass has more than one bit set in the view mask and multiviewGeometryShader is not enabled, then pStages must not include a geometry shader.

  • If the renderPass has multiview enabled and subpass has more than one bit set in the view mask, shaders in the pipeline must not write to the Layer built-in output

  • If the renderPass has multiview enabled, then all shaders must not include variables decorated with the Layer built-in decoration in their interfaces.

  • flags must not contain the VK_PIPELINE_CREATE_DISPATCH_BASE flag.

  • If pStages includes a fragment shader stage and an input attachment was referenced by the VkRenderPassInputAttachmentAspectCreateInfo at renderPass create time, its shader code must not read from any aspect that was not specified in the aspectMask of the corresponding VkInputAttachmentAspectReference structure.

  • The number of resources in layout accessible to each shader stage that is used by the pipeline must be less than or equal to VkPhysicalDeviceLimits::maxPerStageResources

  • If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV, and the viewportWScalingEnable member of a VkPipelineViewportWScalingStateCreateInfoNV structure, chained to the pNext chain of pViewportState, is VK_TRUE, the pViewportWScalings member of the VkPipelineViewportWScalingStateCreateInfoNV must be a pointer to an array of VkPipelineViewportWScalingStateCreateInfoNV::viewportCount valid VkViewportWScalingNV structures

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkPipelineDiscardRectangleStateCreateInfoEXT

  • flags must be a valid combination of VkPipelineCreateFlagBits values

  • pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures

  • pVertexInputState must be a valid pointer to a valid VkPipelineVertexInputStateCreateInfo structure

  • pInputAssemblyState must be a valid pointer to a valid VkPipelineInputAssemblyStateCreateInfo structure

  • pRasterizationState must be a valid pointer to a valid VkPipelineRasterizationStateCreateInfo structure

  • If pDynamicState is not NULL, pDynamicState must be a valid pointer to a valid VkPipelineDynamicStateCreateInfo structure

  • layout must be a valid VkPipelineLayout handle

  • renderPass must be a valid VkRenderPass handle

  • stageCount must be greater than 0

  • Each of basePipelineHandle, layout, and renderPass that are valid handles must have been created, allocated, or retrieved from the same VkDevice

Possible values of the flags member of VkGraphicsPipelineCreateInfo and VkComputePipelineCreateInfo, specifying how a pipeline is created, are:

typedef enum VkPipelineCreateFlagBits {
    VK_PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT = 0x00000001,
    VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT = 0x00000002,
    VK_PIPELINE_CREATE_DERIVATIVE_BIT = 0x00000004,
    VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT = 0x00000008,
    VK_PIPELINE_CREATE_DISPATCH_BASE = 0x00000010,
    VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT_KHR = VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT,
    VK_PIPELINE_CREATE_DISPATCH_BASE_KHR = VK_PIPELINE_CREATE_DISPATCH_BASE,
} VkPipelineCreateFlagBits;
  • VK_PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT specifies that the created pipeline will not be optimized. Using this flag may reduce the time taken to create the pipeline.

  • VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT specifies that the pipeline to be created is allowed to be the parent of a pipeline that will be created in a subsequent call to vkCreateGraphicsPipelines or vkCreateComputePipelines.

  • VK_PIPELINE_CREATE_DERIVATIVE_BIT specifies that the pipeline to be created will be a child of a previously created parent pipeline.

  • VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT specifies that any shader input variables decorated as ViewIndex will be assigned values as if they were decorated as DeviceIndex.

  • VK_PIPELINE_CREATE_DISPATCH_BASE specifies that a compute pipeline can be used with vkCmdDispatchBase with a non-zero base workgroup.

It is valid to set both VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT and VK_PIPELINE_CREATE_DERIVATIVE_BIT. This allows a pipeline to be both a parent and possibly a child in a pipeline hierarchy. See Pipeline Derivatives for more information.

typedef VkFlags VkPipelineCreateFlags;

VkPipelineCreateFlags is a bitmask type for setting a mask of zero or more VkPipelineCreateFlagBits.

The VkPipelineDynamicStateCreateInfo structure is defined as:

typedef struct VkPipelineDynamicStateCreateInfo {
    VkStructureType                      sType;
    const void*                          pNext;
    VkPipelineDynamicStateCreateFlags    flags;
    uint32_t                             dynamicStateCount;
    const VkDynamicState*                pDynamicStates;
} VkPipelineDynamicStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • dynamicStateCount is the number of elements in the pDynamicStates array.

  • pDynamicStates is an array of VkDynamicState values specifying which pieces of pipeline state will use the values from dynamic state commands rather than from pipeline state creation info.

Valid Usage
  • Each element of pDynamicStates must be unique

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_DYNAMIC_STATE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • pDynamicStates must be a valid pointer to an array of dynamicStateCount valid VkDynamicState values

  • dynamicStateCount must be greater than 0

typedef VkFlags VkPipelineDynamicStateCreateFlags;

VkPipelineDynamicStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The source of different pieces of dynamic state is specified by the VkPipelineDynamicStateCreateInfo::pDynamicStates property of the currently active pipeline, each of whose elements must be one of the values:

typedef enum VkDynamicState {
    VK_DYNAMIC_STATE_VIEWPORT = 0,
    VK_DYNAMIC_STATE_SCISSOR = 1,
    VK_DYNAMIC_STATE_LINE_WIDTH = 2,
    VK_DYNAMIC_STATE_DEPTH_BIAS = 3,
    VK_DYNAMIC_STATE_BLEND_CONSTANTS = 4,
    VK_DYNAMIC_STATE_DEPTH_BOUNDS = 5,
    VK_DYNAMIC_STATE_STENCIL_COMPARE_MASK = 6,
    VK_DYNAMIC_STATE_STENCIL_WRITE_MASK = 7,
    VK_DYNAMIC_STATE_STENCIL_REFERENCE = 8,
    VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV = 1000087000,
    VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT = 1000099000,
    VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT = 1000143000,
} VkDynamicState;
  • VK_DYNAMIC_STATE_VIEWPORT specifies that the pViewports state in VkPipelineViewportStateCreateInfo will be ignored and must be set dynamically with vkCmdSetViewport before any draw commands. The number of viewports used by a pipeline is still specified by the viewportCount member of VkPipelineViewportStateCreateInfo.

  • VK_DYNAMIC_STATE_SCISSOR specifies that the pScissors state in VkPipelineViewportStateCreateInfo will be ignored and must be set dynamically with vkCmdSetScissor before any draw commands. The number of scissor rectangles used by a pipeline is still specified by the scissorCount member of VkPipelineViewportStateCreateInfo.

  • VK_DYNAMIC_STATE_LINE_WIDTH specifies that the lineWidth state in VkPipelineRasterizationStateCreateInfo will be ignored and must be set dynamically with vkCmdSetLineWidth before any draw commands that generate line primitives for the rasterizer.

  • VK_DYNAMIC_STATE_DEPTH_BIAS specifies that the depthBiasConstantFactor, depthBiasClamp and depthBiasSlopeFactor states in VkPipelineRasterizationStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthBias before any draws are performed with depthBiasEnable in VkPipelineRasterizationStateCreateInfo set to VK_TRUE.

  • VK_DYNAMIC_STATE_BLEND_CONSTANTS specifies that the blendConstants state in VkPipelineColorBlendStateCreateInfo will be ignored and must be set dynamically with vkCmdSetBlendConstants before any draws are performed with a pipeline state with VkPipelineColorBlendAttachmentState member blendEnable set to VK_TRUE and any of the blend functions using a constant blend color.

  • VK_DYNAMIC_STATE_DEPTH_BOUNDS specifies that the minDepthBounds and maxDepthBounds states of VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthBounds before any draws are performed with a pipeline state with VkPipelineDepthStencilStateCreateInfo member depthBoundsTestEnable set to VK_TRUE.

  • VK_DYNAMIC_STATE_STENCIL_COMPARE_MASK specifies that the compareMask state in VkPipelineDepthStencilStateCreateInfo for both front and back will be ignored and must be set dynamically with vkCmdSetStencilCompareMask before any draws are performed with a pipeline state with VkPipelineDepthStencilStateCreateInfo member stencilTestEnable set to VK_TRUE

  • VK_DYNAMIC_STATE_STENCIL_WRITE_MASK specifies that the writeMask state in VkPipelineDepthStencilStateCreateInfo for both front and back will be ignored and must be set dynamically with vkCmdSetStencilWriteMask before any draws are performed with a pipeline state with VkPipelineDepthStencilStateCreateInfo member stencilTestEnable set to VK_TRUE

  • VK_DYNAMIC_STATE_STENCIL_REFERENCE specifies that the reference state in VkPipelineDepthStencilStateCreateInfo for both front and back will be ignored and must be set dynamically with vkCmdSetStencilReference before any draws are performed with a pipeline state with VkPipelineDepthStencilStateCreateInfo member stencilTestEnable set to VK_TRUE

  • VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV specifies that the pViewportScalings state in VkPipelineViewportWScalingStateCreateInfoNV will be ignored and must be set dynamically with vkCmdSetViewportWScalingNV before any draws are performed with a pipeline state with VkPipelineViewportWScalingStateCreateInfo member viewportScalingEnable set to VK_TRUE

  • VK_DYNAMIC_STATE_DISCARD_RECTANGLES_EXT specifies that the pDiscardRectangles state in VkPipelineDiscardRectangleStateCreateInfoEXT will be ignored and must be set dynamically with vkCmdSetDiscardRectangleEXT before any draw or clear commands. The VkDiscardRectangleModeEXT and the number of active discard rectangles is still specified by the discardRectangleMode and discardRectangleCount members of VkPipelineDiscardRectangleStateCreateInfoEXT.

  • VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT specifies that the sampleLocationsInfo state in VkPipelineSampleLocationsStateCreateInfoEXT will be ignored and must be set dynamically with vkCmdSetSampleLocationsEXT before any draw or clear commands. Enabling custom sample locations is still indicated by the sampleLocationsEnable member of VkPipelineSampleLocationsStateCreateInfoEXT.

9.2.1. Valid Combinations of Stages for Graphics Pipelines

If tessellation shader stages are omitted, the tessellation shading and fixed-function stages of the pipeline are skipped.

If a geometry shader is omitted, the geometry shading stage is skipped.

If a fragment shader is omitted, the results of fragment processing are undefined. Specifically, any fragment color outputs are considered to have undefined values, and the fragment depth is considered to be unmodified. This can be useful for depth-only rendering.

Presence of a shader stage in a pipeline is indicated by including a valid VkPipelineShaderStageCreateInfo with module and pName selecting an entry point from a shader module, where that entry point is valid for the stage specified by stage.

Presence of some of the fixed-function stages in the pipeline is implicitly derived from enabled shaders and provided state. For example, the fixed-function tessellator is always present when the pipeline has valid Tessellation Control and Tessellation Evaluation shaders.

For example:

9.3. Pipeline destruction

To destroy a graphics or compute pipeline, call:

void vkDestroyPipeline(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the pipeline.

  • pipeline is the handle of the pipeline to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to pipeline must have completed execution

  • If VkAllocationCallbacks were provided when pipeline was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when pipeline was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If pipeline is not VK_NULL_HANDLE, pipeline must be a valid VkPipeline handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If pipeline is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to pipeline must be externally synchronized

9.4. Multiple Pipeline Creation

Multiple pipelines can be created simultaneously by passing an array of VkGraphicsPipelineCreateInfo or VkComputePipelineCreateInfo structures into the vkCreateGraphicsPipelines and vkCreateComputePipelines commands, respectively. Applications can group together similar pipelines to be created in a single call, and implementations are encouraged to look for reuse opportunities within a group-create.

When an application attempts to create many pipelines in a single command, it is possible that some subset may fail creation. In that case, the corresponding entries in the pPipelines output array will be filled with VK_NULL_HANDLE values. If any pipeline fails creation (for example, due to out of memory errors), the vkCreate*Pipelines commands will return an error code. The implementation will attempt to create all pipelines, and only return VK_NULL_HANDLE values for those that actually failed.

9.5. Pipeline Derivatives

A pipeline derivative is a child pipeline created from a parent pipeline, where the child and parent are expected to have much commonality. The goal of derivative pipelines is that they be cheaper to create using the parent as a starting point, and that it be more efficient (on either host or device) to switch/bind between children of the same parent.

A derivative pipeline is created by setting the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag in the Vk*PipelineCreateInfo structure. If this is set, then exactly one of basePipelineHandle or basePipelineIndex members of the structure must have a valid handle/index, and specifies the parent pipeline. If basePipelineHandle is used, the parent pipeline must have already been created. If basePipelineIndex is used, then the parent is being created in the same command. VK_NULL_HANDLE acts as the invalid handle for basePipelineHandle, and -1 is the invalid index for basePipelineIndex. If basePipelineIndex is used, the base pipeline must appear earlier in the array. The base pipeline must have been created with the VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT flag set.

9.6. Pipeline Cache

Pipeline cache objects allow the result of pipeline construction to be reused between pipelines and between runs of an application. Reuse between pipelines is achieved by passing the same pipeline cache object when creating multiple related pipelines. Reuse across runs of an application is achieved by retrieving pipeline cache contents in one run of an application, saving the contents, and using them to preinitialize a pipeline cache on a subsequent run. The contents of the pipeline cache objects are managed by the implementation. Applications can manage the host memory consumed by a pipeline cache object and control the amount of data retrieved from a pipeline cache object.

Pipeline cache objects are represented by VkPipelineCache handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPipelineCache)

To create pipeline cache objects, call:

VkResult vkCreatePipelineCache(
    VkDevice                                    device,
    const VkPipelineCacheCreateInfo*            pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkPipelineCache*                            pPipelineCache);
  • device is the logical device that creates the pipeline cache object.

  • pCreateInfo is a pointer to a VkPipelineCacheCreateInfo structure that contains the initial parameters for the pipeline cache object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pPipelineCache is a pointer to a VkPipelineCache handle in which the resulting pipeline cache object is returned.

Note

Applications can track and manage the total host memory size of a pipeline cache object using the pAllocator. Applications can limit the amount of data retrieved from a pipeline cache object in vkGetPipelineCacheData. Implementations should not internally limit the total number of entries added to a pipeline cache object or the total host memory consumed.

Once created, a pipeline cache can be passed to the vkCreateGraphicsPipelines and vkCreateComputePipelines commands. If the pipeline cache passed into these commands is not VK_NULL_HANDLE, the implementation will query it for possible reuse opportunities and update it with new content. The use of the pipeline cache object in these commands is internally synchronized, and the same pipeline cache object can be used in multiple threads simultaneously.

Note

Implementations should make every effort to limit any critical sections to the actual accesses to the cache, which is expected to be significantly shorter than the duration of the vkCreateGraphicsPipelines and vkCreateComputePipelines commands.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkPipelineCacheCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pPipelineCache must be a valid pointer to a VkPipelineCache handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineCacheCreateInfo structure is defined as:

typedef struct VkPipelineCacheCreateInfo {
    VkStructureType               sType;
    const void*                   pNext;
    VkPipelineCacheCreateFlags    flags;
    size_t                        initialDataSize;
    const void*                   pInitialData;
} VkPipelineCacheCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • initialDataSize is the number of bytes in pInitialData. If initialDataSize is zero, the pipeline cache will initially be empty.

  • pInitialData is a pointer to previously retrieved pipeline cache data. If the pipeline cache data is incompatible (as defined below) with the device, the pipeline cache will be initially empty. If initialDataSize is zero, pInitialData is ignored.

Valid Usage
  • If initialDataSize is not 0, it must be equal to the size of pInitialData, as returned by vkGetPipelineCacheData when pInitialData was originally retrieved

  • If initialDataSize is not 0, pInitialData must have been retrieved from a previous call to vkGetPipelineCacheData

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_CACHE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • If initialDataSize is not 0, pInitialData must be a valid pointer to an array of initialDataSize bytes

typedef VkFlags VkPipelineCacheCreateFlags;

VkPipelineCacheCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

Pipeline cache objects can be merged using the command:

VkResult vkMergePipelineCaches(
    VkDevice                                    device,
    VkPipelineCache                             dstCache,
    uint32_t                                    srcCacheCount,
    const VkPipelineCache*                      pSrcCaches);
  • device is the logical device that owns the pipeline cache objects.

  • dstCache is the handle of the pipeline cache to merge results into.

  • srcCacheCount is the length of the pSrcCaches array.

  • pSrcCaches is an array of pipeline cache handles, which will be merged into dstCache. The previous contents of dstCache are included after the merge.

Note

The details of the merge operation are implementation dependent, but implementations should merge the contents of the specified pipelines and prune duplicate entries.

Valid Usage
  • dstCache must not appear in the list of source caches

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • dstCache must be a valid VkPipelineCache handle

  • pSrcCaches must be a valid pointer to an array of srcCacheCount valid VkPipelineCache handles

  • srcCacheCount must be greater than 0

  • dstCache must have been created, allocated, or retrieved from device

  • Each element of pSrcCaches must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to dstCache must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Data can be retrieved from a pipeline cache object using the command:

VkResult vkGetPipelineCacheData(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    size_t*                                     pDataSize,
    void*                                       pData);
  • device is the logical device that owns the pipeline cache.

  • pipelineCache is the pipeline cache to retrieve data from.

  • pDataSize is a pointer to a value related to the amount of data in the pipeline cache, as described below.

  • pData is either NULL or a pointer to a buffer.

If pData is NULL, then the maximum size of the data that can be retrieved from the pipeline cache, in bytes, is returned in pDataSize. Otherwise, pDataSize must point to a variable set by the user to the size of the buffer, in bytes, pointed to by pData, and on return the variable is overwritten with the amount of data actually written to pData.

If pDataSize is less than the maximum size that can be retrieved by the pipeline cache, at most pDataSize bytes will be written to pData, and vkGetPipelineCacheData will return VK_INCOMPLETE. Any data written to pData is valid and can be provided as the pInitialData member of the VkPipelineCacheCreateInfo structure passed to vkCreatePipelineCache.

Two calls to vkGetPipelineCacheData with the same parameters must retrieve the same data unless a command that modifies the contents of the cache is called between them.

Applications can store the data retrieved from the pipeline cache, and use these data, possibly in a future run of the application, to populate new pipeline cache objects. The results of pipeline compiles, however, may depend on the vendor ID, device ID, driver version, and other details of the device. To enable applications to detect when previously retrieved data is incompatible with the device, the initial bytes written to pData must be a header consisting of the following members:

Table 12. Layout for pipeline cache header version VK_PIPELINE_CACHE_HEADER_VERSION_ONE
Offset Size Meaning

0

4

length in bytes of the entire pipeline cache header written as a stream of bytes, with the least significant byte first

4

4

a VkPipelineCacheHeaderVersion value written as a stream of bytes, with the least significant byte first

8

4

a vendor ID equal to VkPhysicalDeviceProperties::vendorID written as a stream of bytes, with the least significant byte first

12

4

a device ID equal to VkPhysicalDeviceProperties::deviceID written as a stream of bytes, with the least significant byte first

16

VK_UUID_SIZE

a pipeline cache ID equal to VkPhysicalDeviceProperties::pipelineCacheUUID

The first four bytes encode the length of the entire pipeline cache header, in bytes. This value includes all fields in the header including the pipeline cache version field and the size of the length field.

The next four bytes encode the pipeline cache version, as described for VkPipelineCacheHeaderVersion. A consumer of the pipeline cache should use the cache version to interpret the remainder of the cache header.

If pDataSize is less than what is necessary to store this header, nothing will be written to pData and zero will be written to pDataSize.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pipelineCache must be a valid VkPipelineCache handle

  • pDataSize must be a valid pointer to a size_t value

  • If the value referenced by pDataSize is not 0, and pData is not NULL, pData must be a valid pointer to an array of pDataSize bytes

  • pipelineCache must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Possible values of the second group of four bytes in the header returned by vkGetPipelineCacheData, encoding the pipeline cache version, are:

typedef enum VkPipelineCacheHeaderVersion {
    VK_PIPELINE_CACHE_HEADER_VERSION_ONE = 1,
} VkPipelineCacheHeaderVersion;
  • VK_PIPELINE_CACHE_HEADER_VERSION_ONE specifies version one of the pipeline cache.

To destroy a pipeline cache, call:

void vkDestroyPipelineCache(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the pipeline cache object.

  • pipelineCache is the handle of the pipeline cache to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when pipelineCache was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when pipelineCache was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If pipelineCache is not VK_NULL_HANDLE, pipelineCache must be a valid VkPipelineCache handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If pipelineCache is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to pipelineCache must be externally synchronized

9.7. Specialization Constants

Specialization constants are a mechanism whereby constants in a SPIR-V module can have their constant value specified at the time the VkPipeline is created. This allows a SPIR-V module to have constants that can be modified while executing an application that uses the Vulkan API.

Note

Specialization constants are useful to allow a compute shader to have its local workgroup size changed at runtime by the user, for example.

Each instance of the VkPipelineShaderStageCreateInfo structure contains a parameter pSpecializationInfo, which can be NULL to indicate no specialization constants, or point to a VkSpecializationInfo structure.

The VkSpecializationInfo structure is defined as:

typedef struct VkSpecializationInfo {
    uint32_t                           mapEntryCount;
    const VkSpecializationMapEntry*    pMapEntries;
    size_t                             dataSize;
    const void*                        pData;
} VkSpecializationInfo;
  • mapEntryCount is the number of entries in the pMapEntries array.

  • pMapEntries is a pointer to an array of VkSpecializationMapEntry which maps constant IDs to offsets in pData.

  • dataSize is the byte size of the pData buffer.

  • pData contains the actual constant values to specialize with.

pMapEntries points to a structure of type VkSpecializationMapEntry.

Valid Usage
  • The offset member of each element of pMapEntries must be less than dataSize

  • The size member of each element of pMapEntries must be less than or equal to dataSize minus offset

  • If mapEntryCount is not 0, pMapEntries must be a valid pointer to an array of mapEntryCount valid VkSpecializationMapEntry structures

Valid Usage (Implicit)
  • If dataSize is not 0, pData must be a valid pointer to an array of dataSize bytes

The VkSpecializationMapEntry structure is defined as:

typedef struct VkSpecializationMapEntry {
    uint32_t    constantID;
    uint32_t    offset;
    size_t      size;
} VkSpecializationMapEntry;
  • constantID is the ID of the specialization constant in SPIR-V.

  • offset is the byte offset of the specialization constant value within the supplied data buffer.

  • size is the byte size of the specialization constant value within the supplied data buffer.

If a constantID value is not a specialization constant ID used in the shader, that map entry does not affect the behavior of the pipeline.

Valid Usage
  • For a constantID specialization constant declared in a shader, size must match the byte size of the constantID. If the specialization constant is of type boolean, size must be the byte size of VkBool32

In human readable SPIR-V:

OpDecorate %x SpecId 13 ; decorate .x component of WorkgroupSize with ID 13
OpDecorate %y SpecId 42 ; decorate .y component of WorkgroupSize with ID 42
OpDecorate %z SpecId 3  ; decorate .z component of WorkgroupSize with ID 3
OpDecorate %wgsize BuiltIn WorkgroupSize ; decorate WorkgroupSize onto constant
%i32 = OpTypeInt 32 0 ; declare an unsigned 32-bit type
%uvec3 = OpTypeVector %i32 3 ; declare a 3 element vector type of unsigned 32-bit
%x = OpSpecConstant %i32 1 ; declare the .x component of WorkgroupSize
%y = OpSpecConstant %i32 1 ; declare the .y component of WorkgroupSize
%z = OpSpecConstant %i32 1 ; declare the .z component of WorkgroupSize
%wgsize = OpSpecConstantComposite %uvec3 %x %y %z ; declare WorkgroupSize

From the above we have three specialization constants, one for each of the x, y & z elements of the WorkgroupSize vector.

Now to specialize the above via the specialization constants mechanism:

const VkSpecializationMapEntry entries[] =
{
    {
        13,                             // constantID
        0 * sizeof(uint32_t),           // offset
        sizeof(uint32_t)                // size
    },
    {
        42,                             // constantID
        1 * sizeof(uint32_t),           // offset
        sizeof(uint32_t)                // size
    },
    {
        3,                              // constantID
        2 * sizeof(uint32_t),           // offset
        sizeof(uint32_t)                // size
    }
};

const uint32_t data[] = { 16, 8, 4 }; // our workgroup size is 16x8x4

const VkSpecializationInfo info =
{
    3,                                  // mapEntryCount
    entries,                            // pMapEntries
    3 * sizeof(uint32_t),               // dataSize
    data,                               // pData
};

Then when calling vkCreateComputePipelines, and passing the VkSpecializationInfo we defined as the pSpecializationInfo parameter of VkPipelineShaderStageCreateInfo, we will create a compute pipeline with the runtime specified local workgroup size.

Another example would be that an application has a SPIR-V module that has some platform-dependent constants they wish to use.

In human readable SPIR-V:

OpDecorate %1 SpecId 0  ; decorate our signed 32-bit integer constant
OpDecorate %2 SpecId 12 ; decorate our 32-bit floating-point constant
%i32 = OpTypeInt 32 1   ; declare a signed 32-bit type
%float = OpTypeFloat 32 ; declare a 32-bit floating-point type
%1 = OpSpecConstant %i32 -1 ; some signed 32-bit integer constant
%2 = OpSpecConstant %float 0.5 ; some 32-bit floating-point constant

From the above we have two specialization constants, one is a signed 32-bit integer and the second is a 32-bit floating-point.

Now to specialize the above via the specialization constants mechanism:

struct SpecializationData {
    int32_t data0;
    float data1;
};

const VkSpecializationMapEntry entries[] =
{
    {
        0,                                    // constantID
        offsetof(SpecializationData, data0),  // offset
        sizeof(SpecializationData::data0)     // size
    },
    {
        12,                                   // constantID
        offsetof(SpecializationData, data1),  // offset
        sizeof(SpecializationData::data1)     // size
    }
};

SpecializationData data;
data.data0 = -42;    // set the data for the 32-bit integer
data.data1 = 42.0f;  // set the data for the 32-bit floating-point

const VkSpecializationInfo info =
{
    2,                                  // mapEntryCount
    entries,                            // pMapEntries
    sizeof(data),                       // dataSize
    &data,                              // pData
};

It is legal for a SPIR-V module with specializations to be compiled into a pipeline where no specialization info was provided. SPIR-V specialization constants contain default values such that if a specialization is not provided, the default value will be used. In the examples above, it would be valid for an application to only specialize some of the specialization constants within the SPIR-V module, and let the other constants use their default values encoded within the OpSpecConstant declarations.

9.8. Pipeline Binding

Once a pipeline has been created, it can be bound to the command buffer using the command:

void vkCmdBindPipeline(
    VkCommandBuffer                             commandBuffer,
    VkPipelineBindPoint                         pipelineBindPoint,
    VkPipeline                                  pipeline);
  • commandBuffer is the command buffer that the pipeline will be bound to.

  • pipelineBindPoint is a VkPipelineBindPoint value specifying whether to bind to the compute or graphics bind point. Binding one does not disturb the other.

  • pipeline is the pipeline to be bound.

Once bound, a pipeline binding affects subsequent graphics or compute commands in the command buffer until a different pipeline is bound to the bind point. The pipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE controls the behavior of vkCmdDispatch and vkCmdDispatchIndirect. The pipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS controls the behavior of all drawing commands. No other commands are affected by the pipeline state.

Valid Usage
  • If pipelineBindPoint is VK_PIPELINE_BIND_POINT_COMPUTE, the VkCommandPool that commandBuffer was allocated from must support compute operations

  • If pipelineBindPoint is VK_PIPELINE_BIND_POINT_GRAPHICS, the VkCommandPool that commandBuffer was allocated from must support graphics operations

  • If pipelineBindPoint is VK_PIPELINE_BIND_POINT_COMPUTE, pipeline must be a compute pipeline

  • If pipelineBindPoint is VK_PIPELINE_BIND_POINT_GRAPHICS, pipeline must be a graphics pipeline

  • If the variable multisample rate feature is not supported, pipeline is a graphics pipeline, the current subpass has no attachments, and this is not the first call to this function with a graphics pipeline after transitioning to the current subpass, then the sample count specified by this pipeline must match that set in the previous pipeline

  • If VkPhysicalDeviceSampleLocationsPropertiesEXT::variableSampleLocations is VK_FALSE, and pipeline is a graphics pipeline created with a VkPipelineSampleLocationsStateCreateInfoEXT structure having its sampleLocationsEnable member set to VK_TRUE but without VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT enabled then the current render pass instance must have been begun by specifying a VkRenderPassSampleLocationsBeginInfoEXT structure whose pPostSubpassSampleLocations member contains an element with a subpassIndex matching the current subpass index and the sampleLocationsInfo member of that element must match the sampleLocationsInfo specified in VkPipelineSampleLocationsStateCreateInfoEXT when the pipeline was created

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pipelineBindPoint must be a valid VkPipelineBindPoint value

  • pipeline must be a valid VkPipeline handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • Both of commandBuffer, and pipeline must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

Possible values of vkCmdBindPipeline::pipelineBindPoint, specifying the bind point of a pipeline object, are:

typedef enum VkPipelineBindPoint {
    VK_PIPELINE_BIND_POINT_GRAPHICS = 0,
    VK_PIPELINE_BIND_POINT_COMPUTE = 1,
} VkPipelineBindPoint;
  • VK_PIPELINE_BIND_POINT_COMPUTE specifies binding as a compute pipeline.

  • VK_PIPELINE_BIND_POINT_GRAPHICS specifies binding as a graphics pipeline.

9.9. Dynamic State

When a pipeline object is bound, any pipeline object state that is not specified as dynamic is applied to the command buffer state. Pipeline object state that is specified as dynamic is not applied to the command buffer state at this time. Instead, dynamic state can be modified at any time and persists for the lifetime of the command buffer, or until modified by another dynamic state setting command or another pipeline bind.

When a pipeline object is bound, the following applies to each state parameter:

  • If the state is not specified as dynamic in the new pipeline object, then that command buffer state is overwritten by the state in the new pipeline object.

  • If the state is specified as dynamic in both the new and the previous pipeline object, then that command buffer state is not disturbed.

  • If the state is specified as dynamic in the new pipeline object but is not specified as dynamic in the previous pipeline object, then that command buffer state becomes undefined. If the state is an array, then the entire array becomes undefined.

  • If the state is an array specified as dynamic in both the new and the previous pipeline object, and the array size is not the same in both pipeline objects, then that command buffer state becomes undefined.

Dynamic state setting commands must not be issued for state that is not specified as dynamic in the bound pipeline object.

Dynamic state that does not affect the result of operations can be left undefined.

Note

For example, if blending is disabled by the pipeline object state then the dynamic color blend constants do not need to be specified in the command buffer, even if this state is specified as dynamic in the pipeline object.

9.10. Pipeline Shader Information

Information about a particular shader that has been compiled as part of a pipeline object can be extracted by calling:

VkResult vkGetShaderInfoAMD(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    VkShaderStageFlagBits                       shaderStage,
    VkShaderInfoTypeAMD                         infoType,
    size_t*                                     pInfoSize,
    void*                                       pInfo);
  • device is the device that created pipeline.

  • pipeline is the target of the query.

  • shaderStage identifies the particular shader within the pipeline about which information is being queried.

  • infoType describes what kind of information is being queried.

  • pInfoSize is a pointer to a value related to the amount of data the query returns, as described below.

  • pInfo is either NULL or a pointer to a buffer.

If pInfo is NULL, then the maximum size of the information that can be retrieved about the shader, in bytes, is returned in pInfoSize. Otherwise, pInfoSize must point to a variable set by the user to the size of the buffer, in bytes, pointed to by pInfo, and on return the variable is overwritten with the amount of data actually written to pInfo.

If pInfoSize is less than the maximum size that can be retrieved by the pipeline cache, then at most pInfoSize bytes will be written to pInfo, and vkGetShaderInfoAMD will return VK_INCOMPLETE.

Not all information is available for every shader and implementations may not support all kinds of information for any shader. When a certain type of information is unavailable, the function returns VK_ERROR_FEATURE_NOT_PRESENT.

If information is successfully and fully queried, the function will return VK_SUCCESS.

For VK_SHADER_INFO_TYPE_STATISTICS_AMD, an instance of VkShaderStatisticsInfoAMD will be written to the buffer pointed to by pInfo. This structure will be populated with statistics regarding the physical device resources used by that shader along with other miscellaneous information and is described in further detail below.

For VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD, pInfo points to a UTF-8 null-terminated string containing human-readable disassembly. The exact formatting and contents of the disassembly string are vendor-specific.

The formatting and contents of all other types of information, including VK_SHADER_INFO_TYPE_BINARY_AMD, are left to the vendor and are not further specified by this extension.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pipeline must be a valid VkPipeline handle

  • shaderStage must be a valid VkShaderStageFlagBits value

  • infoType must be a valid VkShaderInfoTypeAMD value

  • pInfoSize must be a valid pointer to a size_t value

  • If the value referenced by pInfoSize is not 0, and pInfo is not NULL, pInfo must be a valid pointer to an array of pInfoSize bytes

  • pipeline must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_FEATURE_NOT_PRESENT

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkShaderStatisticsInfoAMD structure is defined as:

typedef struct VkShaderStatisticsInfoAMD {
    VkShaderStageFlags          shaderStageMask;
    VkShaderResourceUsageAMD    resourceUsage;
    uint32_t                    numPhysicalVgprs;
    uint32_t                    numPhysicalSgprs;
    uint32_t                    numAvailableVgprs;
    uint32_t                    numAvailableSgprs;
    uint32_t                    computeWorkGroupSize[3];
} VkShaderStatisticsInfoAMD;
  • shaderStageMask are the combination of logical shader stages contained within this shader.

  • resourceUsage is an instance of VkShaderResourceUsageAMD describing internal physical device resources used by this shader.

  • numPhysicalVgprs is the maximum number of vector instruction general-purpose registers (VGPRs) available to the physical device.

  • numPhysicalSgprs is the maximum number of scalar instruction general-purpose registers (SGPRs) available to the physical device.

  • numAvailableVgprs is the maximum limit of VGPRs made available to the shader compiler.

  • numAvailableSgprs is the maximum limit of SGPRs made available to the shader compiler.

  • computeWorkGroupSize is the local workgroup size of this shader in { X, Y, Z } dimensions.

Some implementations may merge multiple logical shader stages together in a single shader. In such cases, shaderStageMask will contain a bitmask of all of the stages that are active within that shader. Consequently, if specifying those stages as input to vkGetShaderInfoAMD, the same output information may be returned for all such shader stage queries.

The number of available VGPRs and SGPRs (numAvailableVgprs and numAvailableSgprs respectively) are the shader-addressable subset of physical registers that is given as a limit to the compiler for register assignment. These values may further be limited by implementations due to performance optimizations where register pressure is a bottleneck.

The VkShaderResourceUsageAMD structure is defined as:

typedef struct VkShaderResourceUsageAMD {
    uint32_t    numUsedVgprs;
    uint32_t    numUsedSgprs;
    uint32_t    ldsSizePerLocalWorkGroup;
    size_t      ldsUsageSizeInBytes;
    size_t      scratchMemUsageInBytes;
} VkShaderResourceUsageAMD;
  • numUsedVgprs is the number of vector instruction general-purpose registers used by this shader.

  • numUsedSgprs is the number of scalar instruction general-purpose registers used by this shader.

  • ldsSizePerLocalWorkGroup is the maximum local data store size per work group in bytes.

  • ldsUsageSizeInBytes is the LDS usage size in bytes per work group by this shader.

  • scratchMemUsageInBytes is the scratch memory usage in bytes by this shader.

10. Memory Allocation

Vulkan memory is broken up into two categories, host memory and device memory.

10.1. Host Memory

Host memory is memory needed by the Vulkan implementation for non-device-visible storage.

Note

This memory may be used to store the implementation’s representation and state of Vulkan objects.

Vulkan provides applications the opportunity to perform host memory allocations on behalf of the Vulkan implementation. If this feature is not used, the implementation will perform its own memory allocations. Since most memory allocations are off the critical path, this is not meant as a performance feature. Rather, this can be useful for certain embedded systems, for debugging purposes (e.g. putting a guard page after all host allocations), or for memory allocation logging.

Allocators are provided by the application as a pointer to a VkAllocationCallbacks structure:

typedef struct VkAllocationCallbacks {
    void*                                   pUserData;
    PFN_vkAllocationFunction                pfnAllocation;
    PFN_vkReallocationFunction              pfnReallocation;
    PFN_vkFreeFunction                      pfnFree;
    PFN_vkInternalAllocationNotification    pfnInternalAllocation;
    PFN_vkInternalFreeNotification          pfnInternalFree;
} VkAllocationCallbacks;
  • pUserData is a value to be interpreted by the implementation of the callbacks. When any of the callbacks in VkAllocationCallbacks are called, the Vulkan implementation will pass this value as the first parameter to the callback. This value can vary each time an allocator is passed into a command, even when the same object takes an allocator in multiple commands.

  • pfnAllocation is a pointer to an application-defined memory allocation function of type PFN_vkAllocationFunction.

  • pfnReallocation is a pointer to an application-defined memory reallocation function of type PFN_vkReallocationFunction.

  • pfnFree is a pointer to an application-defined memory free function of type PFN_vkFreeFunction.

  • pfnInternalAllocation is a pointer to an application-defined function that is called by the implementation when the implementation makes internal allocations, and it is of type PFN_vkInternalAllocationNotification.

  • pfnInternalFree is a pointer to an application-defined function that is called by the implementation when the implementation frees internal allocations, and it is of type PFN_vkInternalFreeNotification.

Valid Usage
  • pfnAllocation must be a valid pointer to a valid user-defined PFN_vkAllocationFunction

  • pfnReallocation must be a valid pointer to a valid user-defined PFN_vkReallocationFunction

  • pfnFree must be a valid pointer to a valid user-defined PFN_vkFreeFunction

  • If either of pfnInternalAllocation or pfnInternalFree is not NULL, both must be valid callbacks

The type of pfnAllocation is:

typedef void* (VKAPI_PTR *PFN_vkAllocationFunction)(
    void*                                       pUserData,
    size_t                                      size,
    size_t                                      alignment,
    VkSystemAllocationScope                     allocationScope);
  • pUserData is the value specified for VkAllocationCallbacks::pUserData in the allocator specified by the application.

  • size is the size in bytes of the requested allocation.

  • alignment is the requested alignment of the allocation in bytes and must be a power of two.

  • allocationScope is a VkSystemAllocationScope value specifying the allocation scope of the lifetime of the allocation, as described here.

If pfnAllocation is unable to allocate the requested memory, it must return NULL. If the allocation was successful, it must return a valid pointer to memory allocation containing at least size bytes, and with the pointer value being a multiple of alignment.

Note

Correct Vulkan operation cannot be assumed if the application does not follow these rules.

For example, pfnAllocation (or pfnReallocation) could cause termination of running Vulkan instance(s) on a failed allocation for debugging purposes, either directly or indirectly. In these circumstances, it cannot be assumed that any part of any affected VkInstance objects are going to operate correctly (even vkDestroyInstance), and the application must ensure it cleans up properly via other means (e.g. process termination).

If pfnAllocation returns NULL, and if the implementation is unable to continue correct processing of the current command without the requested allocation, it must treat this as a run-time error, and generate VK_ERROR_OUT_OF_HOST_MEMORY at the appropriate time for the command in which the condition was detected, as described in Return Codes.

If the implementation is able to continue correct processing of the current command without the requested allocation, then it may do so, and must not generate VK_ERROR_OUT_OF_HOST_MEMORY as a result of this failed allocation.

The type of pfnReallocation is:

typedef void* (VKAPI_PTR *PFN_vkReallocationFunction)(
    void*                                       pUserData,
    void*                                       pOriginal,
    size_t                                      size,
    size_t                                      alignment,
    VkSystemAllocationScope                     allocationScope);
  • pUserData is the value specified for VkAllocationCallbacks::pUserData in the allocator specified by the application.

  • pOriginal must be either NULL or a pointer previously returned by pfnReallocation or pfnAllocation of the same allocator.

  • size is the size in bytes of the requested allocation.

  • alignment is the requested alignment of the allocation in bytes and must be a power of two.

  • allocationScope is a VkSystemAllocationScope value specifying the allocation scope of the lifetime of the allocation, as described here.

pfnReallocation must return an allocation with enough space for size bytes, and the contents of the original allocation from bytes zero to min(original size, new size) - 1 must be preserved in the returned allocation. If size is larger than the old size, the contents of the additional space are undefined. If satisfying these requirements involves creating a new allocation, then the old allocation should be freed.

If pOriginal is NULL, then pfnReallocation must behave equivalently to a call to PFN_vkAllocationFunction with the same parameter values (without pOriginal).

If size is zero, then pfnReallocation must behave equivalently to a call to PFN_vkFreeFunction with the same pUserData parameter value, and pMemory equal to pOriginal.

If pOriginal is non-NULL, the implementation must ensure that alignment is equal to the alignment used to originally allocate pOriginal.

If this function fails and pOriginal is non-NULL the application must not free the old allocation.

pfnReallocation must follow the same rules for return values as PFN_vkAllocationFunction.

The type of pfnFree is:

typedef void (VKAPI_PTR *PFN_vkFreeFunction)(
    void*                                       pUserData,
    void*                                       pMemory);
  • pUserData is the value specified for VkAllocationCallbacks::pUserData in the allocator specified by the application.

  • pMemory is the allocation to be freed.

pMemory may be NULL, which the callback must handle safely. If pMemory is non-NULL, it must be a pointer previously allocated by pfnAllocation or pfnReallocation. The application should free this memory.

The type of pfnInternalAllocation is:

typedef void (VKAPI_PTR *PFN_vkInternalAllocationNotification)(
    void*                                       pUserData,
    size_t                                      size,
    VkInternalAllocationType                    allocationType,
    VkSystemAllocationScope                     allocationScope);
  • pUserData is the value specified for VkAllocationCallbacks::pUserData in the allocator specified by the application.

  • size is the requested size of an allocation.

  • allocationType is a VkInternalAllocationType value specifying the requested type of an allocation.

  • allocationScope is a VkSystemAllocationScope value specifying the allocation scope of the lifetime of the allocation, as described here.

This is a purely informational callback.

The type of pfnInternalFree is:

typedef void (VKAPI_PTR *PFN_vkInternalFreeNotification)(
    void*                                       pUserData,
    size_t                                      size,
    VkInternalAllocationType                    allocationType,
    VkSystemAllocationScope                     allocationScope);
  • pUserData is the value specified for VkAllocationCallbacks::pUserData in the allocator specified by the application.

  • size is the requested size of an allocation.

  • allocationType is a VkInternalAllocationType value specifying the requested type of an allocation.

  • allocationScope is a VkSystemAllocationScope value specifying the allocation scope of the lifetime of the allocation, as described here.

Each allocation has an allocation scope which defines its lifetime and which object it is associated with. Possible values passed to the allocationScope parameter of the callback functions specified by VkAllocationCallbacks, indicating the allocation scope, are:

typedef enum VkSystemAllocationScope {
    VK_SYSTEM_ALLOCATION_SCOPE_COMMAND = 0,
    VK_SYSTEM_ALLOCATION_SCOPE_OBJECT = 1,
    VK_SYSTEM_ALLOCATION_SCOPE_CACHE = 2,
    VK_SYSTEM_ALLOCATION_SCOPE_DEVICE = 3,
    VK_SYSTEM_ALLOCATION_SCOPE_INSTANCE = 4,
} VkSystemAllocationScope;
  • VK_SYSTEM_ALLOCATION_SCOPE_COMMAND specifies that the allocation is scoped to the duration of the Vulkan command.

  • VK_SYSTEM_ALLOCATION_SCOPE_OBJECT specifies that the allocation is scoped to the lifetime of the Vulkan object that is being created or used.

  • VK_SYSTEM_ALLOCATION_SCOPE_CACHE specifies that the allocation is scoped to the lifetime of a VkPipelineCache or VkValidationCacheEXT object.

  • VK_SYSTEM_ALLOCATION_SCOPE_DEVICE specifies that the allocation is scoped to the lifetime of the Vulkan device.

  • VK_SYSTEM_ALLOCATION_SCOPE_INSTANCE specifies that the allocation is scoped to the lifetime of the Vulkan instance.

Most Vulkan commands operate on a single object, or there is a sole object that is being created or manipulated. When an allocation uses an allocation scope of VK_SYSTEM_ALLOCATION_SCOPE_OBJECT or VK_SYSTEM_ALLOCATION_SCOPE_CACHE, the allocation is scoped to the object being created or manipulated.

When an implementation requires host memory, it will make callbacks to the application using the most specific allocator and allocation scope available:

  • If an allocation is scoped to the duration of a command, the allocator will use the VK_SYSTEM_ALLOCATION_SCOPE_COMMAND allocation scope. The most specific allocator available is used: if the object being created or manipulated has an allocator, that object’s allocator will be used, else if the parent VkDevice has an allocator it will be used, else if the parent VkInstance has an allocator it will be used. Else,

  • If an allocation is associated with an object of type VkValidationCacheEXT or VkPipelineCache, the allocator will use the VK_SYSTEM_ALLOCATION_SCOPE_CACHE allocation scope. The most specific allocator available is used (cache, else device, else instance). Else,

  • If an allocation is scoped to the lifetime of an object, that object is being created or manipulated by the command, and that object’s type is not VkDevice or VkInstance, the allocator will use an allocation scope of VK_SYSTEM_ALLOCATION_SCOPE_OBJECT. The most specific allocator available is used (object, else device, else instance). Else,

  • If an allocation is scoped to the lifetime of a device, the allocator will use an allocation scope of VK_SYSTEM_ALLOCATION_SCOPE_DEVICE. The most specific allocator available is used (device, else instance). Else,

  • If the allocation is scoped to the lifetime of an instance and the instance has an allocator, its allocator will be used with an allocation scope of VK_SYSTEM_ALLOCATION_SCOPE_INSTANCE.

  • Otherwise an implementation will allocate memory through an alternative mechanism that is unspecified.

Objects that are allocated from pools do not specify their own allocator. When an implementation requires host memory for such an object, that memory is sourced from the object’s parent pool’s allocator.

The application is not expected to handle allocating memory that is intended for execution by the host due to the complexities of differing security implementations across multiple platforms. The implementation will allocate such memory internally and invoke an application provided informational callback when these internal allocations are allocated and freed. Upon allocation of executable memory, pfnInternalAllocation will be called. Upon freeing executable memory, pfnInternalFree will be called. An implementation will only call an informational callback for executable memory allocations and frees.

The allocationType parameter to the pfnInternalAllocation and pfnInternalFree functions may be one of the following values:

typedef enum VkInternalAllocationType {
    VK_INTERNAL_ALLOCATION_TYPE_EXECUTABLE = 0,
} VkInternalAllocationType;
  • VK_INTERNAL_ALLOCATION_TYPE_EXECUTABLE specifies that the allocation is intended for execution by the host.

An implementation must only make calls into an application-provided allocator during the execution of an API command. An implementation must only make calls into an application-provided allocator from the same thread that called the provoking API command. The implementation should not synchronize calls to any of the callbacks. If synchronization is needed, the callbacks must provide it themselves. The informational callbacks are subject to the same restrictions as the allocation callbacks.

If an implementation intends to make calls through an VkAllocationCallbacks structure between the time a vkCreate* command returns and the time a corresponding vkDestroy* command begins, that implementation must save a copy of the allocator before the vkCreate* command returns. The callback functions and any data structures they rely upon must remain valid for the lifetime of the object they are associated with.

If an allocator is provided to a vkCreate* command, a compatible allocator must be provided to the corresponding vkDestroy* command. Two VkAllocationCallbacks structures are compatible if memory allocated with pfnAllocation or pfnReallocation in each can be freed with pfnReallocation or pfnFree in the other. An allocator must not be provided to a vkDestroy* command if an allocator was not provided to the corresponding vkCreate* command.

If a non-NULL allocator is used, the pfnAllocation, pfnReallocation and pfnFree members must be non-NULL and point to valid implementations of the callbacks. An application can choose to not provide informational callbacks by setting both pfnInternalAllocation and pfnInternalFree to NULL. pfnInternalAllocation and pfnInternalFree must either both be NULL or both be non-NULL.

If pfnAllocation or pfnReallocation fail, the implementation may fail object creation and/or generate an VK_ERROR_OUT_OF_HOST_MEMORY error, as appropriate.

Allocation callbacks must not call any Vulkan commands.

The following sets of rules define when an implementation is permitted to call the allocator callbacks.

pfnAllocation or pfnReallocation may be called in the following situations:

  • Allocations scoped to a VkDevice or VkInstance may be allocated from any API command.

  • Allocations scoped to a command may be allocated from any API command.

  • Allocations scoped to a VkPipelineCache may only be allocated from:

    • vkCreatePipelineCache

    • vkMergePipelineCaches for dstCache

    • vkCreateGraphicsPipelines for pipelineCache

    • vkCreateComputePipelines for pipelineCache

  • Allocations scoped to a VkValidationCacheEXT may only be allocated from:

    • vkCreateValidationCacheEXT

    • vkMergeValidationCachesEXT for dstCache

    • vkCreateShaderModule for validationCache in VkShaderModuleValidationCacheCreateInfoEXT

  • Allocations scoped to a VkDescriptorPool may only be allocated from:

    • any command that takes the pool as a direct argument

    • vkAllocateDescriptorSets for the descriptorPool member of its pAllocateInfo parameter

    • vkCreateDescriptorPool

  • Allocations scoped to a VkCommandPool may only be allocated from:

    • any command that takes the pool as a direct argument

    • vkCreateCommandPool

    • vkAllocateCommandBuffers for the commandPool member of its pAllocateInfo parameter

    • any vkCmd* command whose commandBuffer was allocated from that VkCommandPool

  • Allocations scoped to any other object may only be allocated in that object’s vkCreate* command.

pfnFree may be called in the following situations:

  • Allocations scoped to a VkDevice or VkInstance may be freed from any API command.

  • Allocations scoped to a command must be freed by any API command which allocates such memory.

  • Allocations scoped to a VkPipelineCache may be freed from vkDestroyPipelineCache.

  • Allocations scoped to a VkValidationCacheEXT may be freed from vkDestroyValidationCacheEXT.

  • Allocations scoped to a VkDescriptorPool may be freed from

    • any command that takes the pool as a direct argument

  • Allocations scoped to a VkCommandPool may be freed from:

    • any command that takes the pool as a direct argument

    • vkResetCommandBuffer whose commandBuffer was allocated from that VkCommandPool

  • Allocations scoped to any other object may be freed in that object’s vkDestroy* command.

  • Any command that allocates host memory may also free host memory of the same scope.

10.2. Device Memory

Device memory is memory that is visible to the device — for example the contents of the image or buffer objects, which can be natively used by the device.

Memory properties of a physical device describe the memory heaps and memory types available.

To query memory properties, call:

void vkGetPhysicalDeviceMemoryProperties(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties*           pMemoryProperties);
  • physicalDevice is the handle to the device to query.

  • pMemoryProperties points to an instance of VkPhysicalDeviceMemoryProperties structure in which the properties are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pMemoryProperties must be a valid pointer to a VkPhysicalDeviceMemoryProperties structure

The VkPhysicalDeviceMemoryProperties structure is defined as:

typedef struct VkPhysicalDeviceMemoryProperties {
    uint32_t        memoryTypeCount;
    VkMemoryType    memoryTypes[VK_MAX_MEMORY_TYPES];
    uint32_t        memoryHeapCount;
    VkMemoryHeap    memoryHeaps[VK_MAX_MEMORY_HEAPS];
} VkPhysicalDeviceMemoryProperties;
  • memoryTypeCount is the number of valid elements in the memoryTypes array.

  • memoryTypes is an array of VkMemoryType structures describing the memory types that can be used to access memory allocated from the heaps specified by memoryHeaps.

  • memoryHeapCount is the number of valid elements in the memoryHeaps array.

  • memoryHeaps is an array of VkMemoryHeap structures describing the memory heaps from which memory can be allocated.

The VkPhysicalDeviceMemoryProperties structure describes a number of memory heaps as well as a number of memory types that can be used to access memory allocated in those heaps. Each heap describes a memory resource of a particular size, and each memory type describes a set of memory properties (e.g. host cached vs uncached) that can be used with a given memory heap. Allocations using a particular memory type will consume resources from the heap indicated by that memory type’s heap index. More than one memory type may share each heap, and the heaps and memory types provide a mechanism to advertise an accurate size of the physical memory resources while allowing the memory to be used with a variety of different properties.

The number of memory heaps is given by memoryHeapCount and is less than or equal to VK_MAX_MEMORY_HEAPS. Each heap is described by an element of the memoryHeaps array as a VkMemoryHeap structure. The number of memory types available across all memory heaps is given by memoryTypeCount and is less than or equal to VK_MAX_MEMORY_TYPES. Each memory type is described by an element of the memoryTypes array as a VkMemoryType structure.

At least one heap must include VK_MEMORY_HEAP_DEVICE_LOCAL_BIT in VkMemoryHeap::flags. If there are multiple heaps that all have similar performance characteristics, they may all include VK_MEMORY_HEAP_DEVICE_LOCAL_BIT. In a unified memory architecture (UMA) system there is often only a single memory heap which is considered to be equally “local” to the host and to the device, and such an implementation must advertise the heap as device-local.

Each memory type returned by vkGetPhysicalDeviceMemoryProperties must have its propertyFlags set to one of the following values:

  • 0

  • VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_COHERENT_BIT

  • VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_CACHED_BIT

  • VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_CACHED_BIT |
    VK_MEMORY_PROPERTY_HOST_COHERENT_BIT

  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT

  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
    VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_COHERENT_BIT

  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
    VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_CACHED_BIT

  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
    VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
    VK_MEMORY_PROPERTY_HOST_CACHED_BIT |
    VK_MEMORY_PROPERTY_HOST_COHERENT_BIT

  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
    VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT

  • VK_MEMORY_PROPERTY_PROTECTED_BIT

  • VK_MEMORY_PROPERTY_PROTECTED_BIT | VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT

There must be at least one memory type with both the VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT and VK_MEMORY_PROPERTY_HOST_COHERENT_BIT bits set in its propertyFlags. There must be at least one memory type with the VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT bit set in its propertyFlags.

For each pair of elements X and Y returned in memoryTypes, X must be placed at a lower index position than Y if:

  • either the set of bit flags returned in the propertyFlags member of X is a strict subset of the set of bit flags returned in the propertyFlags member of Y.

  • or the propertyFlags members of X and Y are equal, and X belongs to a memory heap with greater performance (as determined in an implementation-specific manner).

Note

There is no ordering requirement between X and Y elements for the case their propertyFlags members are not in a subset relation. That potentially allows more than one possible way to order the same set of memory types. Notice that the list of all allowed memory property flag combinations is written in the required order. But if instead VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT was before VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, the list would still be in the required order.

This ordering requirement enables applications to use a simple search loop to select the desired memory type along the lines of:

// Find a memory in `memoryTypeBitsRequirement` that includes all of `requiredProperties`
int32_t findProperties(const VkPhysicalDeviceMemoryProperties* pMemoryProperties,
                       uint32_t memoryTypeBitsRequirement,
                       VkMemoryPropertyFlags requiredProperties) {
    const uint32_t memoryCount = pMemoryProperties->memoryTypeCount;
    for (uint32_t memoryIndex = 0; memoryIndex < memoryCount; ++memoryIndex) {
        const uint32_t memoryTypeBits = (1 << memoryIndex);
        const bool isRequiredMemoryType = memoryTypeBitsRequirement & memoryTypeBits;

        const VkMemoryPropertyFlags properties =
            pMemoryProperties->memoryTypes[memoryIndex].propertyFlags;
        const bool hasRequiredProperties =
            (properties & requiredProperties) == requiredProperties;

        if (isRequiredMemoryType && hasRequiredProperties)
            return static_cast<int32_t>(memoryIndex);
    }

    // failed to find memory type
    return -1;
}

// Try to find an optimal memory type, or if it does not exist try fallback memory type
// `device` is the VkDevice
// `image` is the VkImage that requires memory to be bound
// `memoryProperties` properties as returned by vkGetPhysicalDeviceMemoryProperties
// `requiredProperties` are the property flags that must be present
// `optimalProperties` are the property flags that are preferred by the application
VkMemoryRequirements memoryRequirements;
vkGetImageMemoryRequirements(device, image, &memoryRequirements);
int32_t memoryType =
    findProperties(&memoryProperties, memoryRequirements.memoryTypeBits, optimalProperties);
if (memoryType == -1) // not found; try fallback properties
    memoryType =
        findProperties(&memoryProperties, memoryRequirements.memoryTypeBits, requiredProperties);

To query memory properties, call:

void vkGetPhysicalDeviceMemoryProperties2(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties2*          pMemoryProperties);

or the equivalent command

void vkGetPhysicalDeviceMemoryProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties2*          pMemoryProperties);
  • physicalDevice is the handle to the device to query.

  • pMemoryProperties points to an instance of VkPhysicalDeviceMemoryProperties2 structure in which the properties are returned.

vkGetPhysicalDeviceMemoryProperties2 behaves similarly to vkGetPhysicalDeviceMemoryProperties, with the ability to return extended information in a pNext chain of output structures.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pMemoryProperties must be a valid pointer to a VkPhysicalDeviceMemoryProperties2 structure

The VkPhysicalDeviceMemoryProperties2 structure is defined as:

typedef struct VkPhysicalDeviceMemoryProperties2 {
    VkStructureType                     sType;
    void*                               pNext;
    VkPhysicalDeviceMemoryProperties    memoryProperties;
} VkPhysicalDeviceMemoryProperties2;

or the equivalent

typedef VkPhysicalDeviceMemoryProperties2 VkPhysicalDeviceMemoryProperties2KHR;
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2

  • pNext must be NULL

The VkMemoryHeap structure is defined as:

typedef struct VkMemoryHeap {
    VkDeviceSize         size;
    VkMemoryHeapFlags    flags;
} VkMemoryHeap;
  • size is the total memory size in bytes in the heap.

  • flags is a bitmask of VkMemoryHeapFlagBits specifying attribute flags for the heap.

Bits which may be set in VkMemoryHeap::flags, indicating attribute flags for the heap, are:

typedef enum VkMemoryHeapFlagBits {
    VK_MEMORY_HEAP_DEVICE_LOCAL_BIT = 0x00000001,
    VK_MEMORY_HEAP_MULTI_INSTANCE_BIT = 0x00000002,
    VK_MEMORY_HEAP_MULTI_INSTANCE_BIT_KHR = VK_MEMORY_HEAP_MULTI_INSTANCE_BIT,
} VkMemoryHeapFlagBits;
  • VK_MEMORY_HEAP_DEVICE_LOCAL_BIT specifies that the heap corresponds to device local memory. Device local memory may have different performance characteristics than host local memory, and may support different memory property flags.

  • VK_MEMORY_HEAP_MULTI_INSTANCE_BIT specifies that in a logical device representing more than one physical device, there is a per-physical device instance of the heap memory. By default, an allocation from such a heap will be replicated to each physical device’s instance of the heap.

typedef VkFlags VkMemoryHeapFlags;

VkMemoryHeapFlags is a bitmask type for setting a mask of zero or more VkMemoryHeapFlagBits.

The VkMemoryType structure is defined as:

typedef struct VkMemoryType {
    VkMemoryPropertyFlags    propertyFlags;
    uint32_t                 heapIndex;
} VkMemoryType;

Bits which may be set in VkMemoryType::propertyFlags, indicating properties of a memory heap, are:

typedef enum VkMemoryPropertyFlagBits {
    VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT = 0x00000001,
    VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT = 0x00000002,
    VK_MEMORY_PROPERTY_HOST_COHERENT_BIT = 0x00000004,
    VK_MEMORY_PROPERTY_HOST_CACHED_BIT = 0x00000008,
    VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT = 0x00000010,
    VK_MEMORY_PROPERTY_PROTECTED_BIT = 0x00000020,
} VkMemoryPropertyFlagBits;
  • VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT bit specifies that memory allocated with this type is the most efficient for device access. This property will be set if and only if the memory type belongs to a heap with the VK_MEMORY_HEAP_DEVICE_LOCAL_BIT set.

  • VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT bit specifies that memory allocated with this type can be mapped for host access using vkMapMemory.

  • VK_MEMORY_PROPERTY_HOST_COHERENT_BIT bit specifies that the host cache management commands vkFlushMappedMemoryRanges and vkInvalidateMappedMemoryRanges are not needed to flush host writes to the device or make device writes visible to the host, respectively.

  • VK_MEMORY_PROPERTY_HOST_CACHED_BIT bit specifies that memory allocated with this type is cached on the host. Host memory accesses to uncached memory are slower than to cached memory, however uncached memory is always host coherent.

  • VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT bit specifies that the memory type only allows device access to the memory. Memory types must not have both VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT and VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT set. Additionally, the object’s backing memory may be provided by the implementation lazily as specified in Lazily Allocated Memory.

  • VK_MEMORY_PROPERTY_PROTECTED_BIT bit specifies that the memory type only allows device access to the memory, and allows protected queue operations to access the memory. Memory types must not have VK_MEMORY_PROPERTY_PROTECTED_BIT set and any of VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT set, or VK_MEMORY_PROPERTY_HOST_COHERENT_BIT set, or VK_MEMORY_PROPERTY_HOST_CACHED_BIT set.

typedef VkFlags VkMemoryPropertyFlags;

VkMemoryPropertyFlags is a bitmask type for setting a mask of zero or more VkMemoryPropertyFlagBits.

A Vulkan device operates on data in device memory via memory objects that are represented in the API by a VkDeviceMemory handle:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDeviceMemory)

To allocate memory objects, call:

VkResult vkAllocateMemory(
    VkDevice                                    device,
    const VkMemoryAllocateInfo*                 pAllocateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDeviceMemory*                             pMemory);
  • device is the logical device that owns the memory.

  • pAllocateInfo is a pointer to an instance of the VkMemoryAllocateInfo structure describing parameters of the allocation. A successful returned allocation must use the requested parameters — no substitution is permitted by the implementation.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pMemory is a pointer to a VkDeviceMemory handle in which information about the allocated memory is returned.

Allocations returned by vkAllocateMemory are guaranteed to meet any alignment requirement of the implementation. For example, if an implementation requires 128 byte alignment for images and 64 byte alignment for buffers, the device memory returned through this mechanism would be 128-byte aligned. This ensures that applications can correctly suballocate objects of different types (with potentially different alignment requirements) in the same memory object.

When memory is allocated, its contents are undefined with the following constraint:

  • The contents of unprotected memory must not be a function of data protected memory objects, even if those memory objects were previously freed.

Note

The contents of memory allocated by one application should not be a function of data from protected memory objects of another application, even if those memory objects were previously freed.

The maximum number of valid memory allocations that can exist simultaneously within a VkDevice may be restricted by implementation- or platform-dependent limits. If a call to vkAllocateMemory would cause the total number of allocations to exceed these limits, such a call will fail and must return VK_ERROR_TOO_MANY_OBJECTS. The maxMemoryAllocationCount feature describes the number of allocations that can exist simultaneously before encountering these internal limits.

Some platforms may have a limit on the maximum size of a single allocation. For example, certain systems may fail to create allocations with a size greater than or equal to 4GB. Such a limit is implementation-dependent, and if such a failure occurs then the error VK_ERROR_OUT_OF_DEVICE_MEMORY must be returned. This limit is advertised in VkPhysicalDeviceMaintenance3Properties::maxMemoryAllocationSize.

Valid Usage
Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pAllocateInfo must be a valid pointer to a valid VkMemoryAllocateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pMemory must be a valid pointer to a VkDeviceMemory handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryAllocateInfo structure is defined as:

typedef struct VkMemoryAllocateInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkDeviceSize       allocationSize;
    uint32_t           memoryTypeIndex;
} VkMemoryAllocateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • allocationSize is the size of the allocation in bytes

  • memoryTypeIndex is an index identifying a memory type from the memoryTypes array of the VkPhysicalDeviceMemoryProperties structure

An instance of the VkMemoryAllocateInfo structure defines a memory import operation if the pNext chain contains an instance of one of the following structures: * VkImportMemoryWin32HandleInfoKHR with non-zero handleType value * VkImportMemoryFdInfoKHR with a non-zero handleType value * VkImportMemoryHostPointerInfoEXT with a non-zero handleType value * VkImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value

Importing memory must not modify the content of the memory. Implementations must ensure that importing memory does not enable the importing Vulkan instance to access any memory or resources in other Vulkan instances other than that corresponding to the memory object imported. Implementations must also ensure accessing imported memory which has not been initialized does not allow the importing Vulkan instance to obtain data from the exporting Vulkan instance or vice-versa.

Note

How exported and imported memory is isolated is left to the implementation, but applications should be aware that such isolation may prevent implementations from placing multiple exportable memory objects in the same physical or virtual page. Hence, applications should avoid creating many small external memory objects whenever possible.

When performing a memory import operation, it is the responsibility of the application to ensure the external handles meet all valid usage requirements. However, implementations must perform sufficient validation of external handles to ensure that the operation results in a valid memory object which will not cause program termination, device loss, queue stalls, or corruption of other resources when used as allowed according to its allocation parameters. If the external handle provided does not meet these requirements, the implementation must fail the memory import operation with the error code VK_ERROR_INVALID_EXTERNAL_HANDLE.

Valid Usage
  • If the pNext chain contains an instance of VkExportMemoryAllocateInfo, and any of the handle types specified in VkExportMemoryAllocateInfo::handleTypes require a dedicated allocation, as reported by vkGetPhysicalDeviceImageFormatProperties2 in VkExternalImageFormatProperties::externalMemoryProperties::externalMemoryFeatures or VkExternalBufferProperties::externalMemoryProperties::externalMemoryFeatures, the pNext chain must contain an instance of VkMemoryDedicatedAllocateInfo or VkDedicatedAllocationMemoryAllocateInfoNV with either its image or buffer field set to a value other than VK_NULL_HANDLE.

  • If the pNext chain contains an instance of VkExportMemoryAllocateInfo, it must not contain an instance of VkExportMemoryAllocateInfoNV or VkExportMemoryWin32HandleInfoNV.

  • If the pNext chain contains an instance of VkImportMemoryWin32HandleInfoKHR, it must not contain an instance of VkImportMemoryWin32HandleInfoNV.

  • If the parameters define an import operation, the external handle specified was created by the Vulkan API, and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT_KHR, then the values of allocationSize and memoryTypeIndex must match those specified when the memory object being imported was created.

  • If the parameters define an import operation and the external handle specified was created by the Vulkan API, the device mask specified by VkMemoryAllocateFlagsInfo must match that specified when the memory object being imported was allocated.

  • If the parameters define an import operation and the external handle specified was created by the Vulkan API, the list of physical devices that comprise the logical device passed to vkAllocateMemory must match the list of physical devices that comprise the logical device on which the memory was originally allocated.

  • If the parameters define an import operation and the external handle is an NT handle or a global share handle created outside of the Vulkan API, the value of memoryTypeIndex must be one of those returned by vkGetMemoryWin32HandlePropertiesKHR.

  • If the parameters define an import operation, the external handle was created by the Vulkan API, and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR or VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_KHR, then the values of allocationSize and memoryTypeIndex must match those specified when the memory object being imported was created.

  • If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT, or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT, allocationSize must match the size reported in the memory requirements of the image or buffer member of the instance of VkDedicatedAllocationMemoryAllocateInfoNV included in the pNext chain.

  • If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT, allocationSize must match the size specified when creating the Direct3D 12 heap from which the external handle was extracted.

  • If the parameters define an import operation and the external handle is a POSIX file descriptor created outside of the Vulkan API, the value of memoryTypeIndex must be one of those returned by vkGetMemoryFdPropertiesKHR.

  • If the protected memory feature is not enabled, the VkMemoryAllocateInfo::memoryTypeIndex must not indicate a memory type that reports VK_MEMORY_PROPERTY_PROTECTED_BIT.

  • If the parameters define an import operation and the external handle is a host pointer, the value of memoryTypeIndex must be one of those returned by vkGetMemoryHostPointerPropertiesEXT

  • If the parameters define an import operation and the external handle is a host pointer, allocationSize must be an integer multiple of VkPhysicalDeviceExternalMemoryHostPropertiesEXT::minImportedHostPointerAlignment

  • If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BIT_ANDROID:

  • If the parameters do not define an import operation, and the pNext chain contains an instance of VkExportMemoryAllocateInfo with VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID included in its handleTypes member, and the pNext contains an instance of VkMemoryDedicatedAllocateInfo with image not equal to VK_NULL_HANDLE, then allocationSize must be 0, otherwise allocationSize must be greater than 0.

  • If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes an instance of VkMemoryDedicatedAllocateInfo with image that is not VK_NULL_HANDLE:

    • The Android hardware buffer’s usage must include at least one of AHARDWAREBUFFER_USAGE_GPU_COLOR_OUTPUT or AHARDWAREBUFFER_USAGE_GPU_SAMPLED_IMAGE

    • The image’s format must be VK_FORMAT_UNDEFINED or the format returned by vkGetAndroidHardwareBufferPropertiesANDROID in VkAndroidHardwareBufferFormatPropertiesANDROID::format for the Android hardware buffer.

    • The image’s and Android hardware buffer’s width, height, and array layer dimensions must be the same

    • If the Android hardware buffer’s usage includes AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE, the image must have ⌊log2(max(width, height))⌋ + 1 mip levels, otherwise it must have exactly 1 mip level.

    • Each bit set in the image’s usage must be listed in AHardwareBuffer Usage Equivalence, and if there is a corresponding AHARDWAREBUFFER_USAGE bit listed that bit must be included in the Android hardware buffer’s usage

Valid Usage (Implicit)

If the pNext chain includes a VkMemoryDedicatedAllocateInfo structure, then that structure includes a handle of the sole buffer or image resource that the memory can be bound to.

The VkMemoryDedicatedAllocateInfo structure is defined as:

typedef struct VkMemoryDedicatedAllocateInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
    VkBuffer           buffer;
} VkMemoryDedicatedAllocateInfo;

or the equivalent

typedef VkMemoryDedicatedAllocateInfo VkMemoryDedicatedAllocateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • image is VK_NULL_HANDLE or a handle of an image which this memory will be bound to.

  • buffer is VK_NULL_HANDLE or a handle of a buffer which this memory will be bound to.

Valid Usage
  • At least one of image and buffer must be VK_NULL_HANDLE

  • If image is not VK_NULL_HANDLE, VkMemoryAllocateInfo::allocationSize must equal the VkMemoryRequirements::size of the image

  • If image is not VK_NULL_HANDLE, image must have been created without VK_IMAGE_CREATE_SPARSE_BINDING_BIT set in VkImageCreateInfo::flags

  • If buffer is not VK_NULL_HANDLE, VkMemoryAllocateInfo::allocationSize must equal the VkMemoryRequirements::size of the buffer

  • If buffer is not VK_NULL_HANDLE, buffer must have been created without VK_BUFFER_CREATE_SPARSE_BINDING_BIT set in VkBufferCreateInfo::flags

  • If image is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation with handle type VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT, or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT, and the external handle was created by the Vulkan API, then the memory being imported must also be a dedicated image allocation and image must be identical to the image associated with the imported memory.

  • If buffer is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation with handle type VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT, or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT, and the external handle was created by the Vulkan API, then the memory being imported must also be a dedicated buffer allocation and buffer must be identical to the buffer associated with the imported memory.

  • If image is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation with handle type VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT, the memory being imported must also be a dedicated image allocation and image must be identical to the image associated with the imported memory.

  • If buffer is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation with handle type VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT, the memory being imported must also be a dedicated buffer allocation and buffer must be identical to the buffer associated with the imported memory.

  • If image is not VK_NULL_HANDLE, image must not have been created with VK_IMAGE_CREATE_DISJOINT_BIT set in VkImageCreateInfo::flags

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO

  • If image is not VK_NULL_HANDLE, image must be a valid VkImage handle

  • If buffer is not VK_NULL_HANDLE, buffer must be a valid VkBuffer handle

  • Both of buffer, and image that are valid handles must have been created, allocated, or retrieved from the same VkDevice

If the pNext chain includes a VkDedicatedAllocationMemoryAllocateInfoNV structure, then that structure includes a handle of the sole buffer or image resource that the memory can be bound to.

The VkDedicatedAllocationMemoryAllocateInfoNV structure is defined as:

typedef struct VkDedicatedAllocationMemoryAllocateInfoNV {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
    VkBuffer           buffer;
} VkDedicatedAllocationMemoryAllocateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • image is VK_NULL_HANDLE or a handle of an image which this memory will be bound to.

  • buffer is VK_NULL_HANDLE or a handle of a buffer which this memory will be bound to.

Valid Usage
  • At least one of image and buffer must be VK_NULL_HANDLE

  • If image is not VK_NULL_HANDLE, the image must have been created with VkDedicatedAllocationImageCreateInfoNV::dedicatedAllocation equal to VK_TRUE

  • If buffer is not VK_NULL_HANDLE, the buffer must have been created with VkDedicatedAllocationBufferCreateInfoNV::dedicatedAllocation equal to VK_TRUE

  • If image is not VK_NULL_HANDLE, VkMemoryAllocateInfo::allocationSize must equal the VkMemoryRequirements::size of the image

  • If buffer is not VK_NULL_HANDLE, VkMemoryAllocateInfo::allocationSize must equal the VkMemoryRequirements::size of the buffer

  • If image is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation, the memory being imported must also be a dedicated image allocation and image must be identical to the image associated with the imported memory.

  • If buffer is not VK_NULL_HANDLE and VkMemoryAllocateInfo defines a memory import operation, the memory being imported must also be a dedicated buffer allocation and buffer must be identical to the buffer associated with the imported memory.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_MEMORY_ALLOCATE_INFO_NV

  • If image is not VK_NULL_HANDLE, image must be a valid VkImage handle

  • If buffer is not VK_NULL_HANDLE, buffer must be a valid VkBuffer handle

  • Both of buffer, and image that are valid handles must have been created, allocated, or retrieved from the same VkDevice

When allocating memory that may be exported to another process or Vulkan instance, add a VkExportMemoryAllocateInfo structure to the pNext chain of the VkMemoryAllocateInfo structure, specifying the handle types that may be exported.

The VkExportMemoryAllocateInfo structure is defined as:

typedef struct VkExportMemoryAllocateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExportMemoryAllocateInfo;

or the equivalent

typedef VkExportMemoryAllocateInfo VkExportMemoryAllocateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBits specifying one or more memory handle types the application can export from the resulting allocation. The application can request multiple handle types for the same allocation.

Valid Usage
Valid Usage (Implicit)

To specify additional attributes of NT handles exported from a memory object, add the VkExportMemoryWin32HandleInfoKHR structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkExportMemoryWin32HandleInfoKHR structure is defined as:

typedef struct VkExportMemoryWin32HandleInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    const SECURITY_ATTRIBUTES*    pAttributes;
    DWORD                         dwAccess;
    LPCWSTR                       name;
} VkExportMemoryWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pAttributes is a pointer to a Windows SECURITY_ATTRIBUTES structure specifying security attributes of the handle.

  • dwAccess is a DWORD specifying access rights of the handle.

  • name is a NULL-terminated UTF-16 string to associate with the underlying resource referenced by NT handles exported from the created memory.

If this structure is not present, or if pAttributes is set to NULL, default security descriptor values will be used, and child processes created by the application will not inherit the handle, as described in the MSDN documentation for “Synchronization Object Security and Access Rights”1. Further, if the structure is not present, the access rights will be

DXGI_SHARED_RESOURCE_READ | DXGI_SHARED_RESOURCE_WRITE

for handles of the following types:

VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT

And

GENERIC_ALL

for handles of the following types:

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT

Valid Usage
  • If VkExportMemoryAllocateInfo::handleTypes does not include VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT, or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT, VkExportMemoryWin32HandleInfoKHR must not be in the pNext chain of VkMemoryAllocateInfo.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_KHR

  • If pAttributes is not NULL, pAttributes must be a valid pointer to a valid SECURITY_ATTRIBUTES value

To import memory from a Windows handle, add a VkImportMemoryWin32HandleInfoKHR structure to the pNext chain of the VkMemoryAllocateInfo structure.

The VkImportMemoryWin32HandleInfoKHR structure is defined as:

typedef struct VkImportMemoryWin32HandleInfoKHR {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalMemoryHandleTypeFlagBits    handleType;
    HANDLE                                handle;
    LPCWSTR                               name;
} VkImportMemoryWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType specifies the type of handle or name.

  • handle is the external handle to import, or NULL.

  • name is a NULL-terminated UTF-16 string naming the underlying memory resource to import, or NULL.

Importing memory objects from Windows handles does not transfer ownership of the handle to the Vulkan implementation. For handle types defined as NT handles, the application must release ownership using the CloseHandle system call when the handle is no longer needed.

Applications can import the same underlying memory into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance. In all cases, each import operation must create a distinct VkDeviceMemory object.

Valid Usage
  • If handleType is not 0, it must be supported for import, as reported by VkExternalImageFormatProperties or VkExternalBufferProperties.

  • The memory from which handle was exported, or the memory named by name must have been created on the same underlying physical device as device.

  • If handleType is not 0, it must be defined as an NT handle or a global share handle.

  • If handleType is not VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT, VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT, or VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT, name must be NULL.

  • If handleType is not 0 and handle is NULL, name must name a valid memory resource of the type specified by handleType.

  • If handleType is not 0 and name is NULL, handle must be a valid handle of the type specified by handleType.

  • if handle is not NULL, name must be NULL.

  • If handle is not NULL, it must obey any requirements listed for handleType in external memory handle types compatibility.

  • If name is not NULL, it must obey any requirements listed for handleType in external memory handle types compatibility.

Valid Usage (Implicit)

To export a Windows handle representing the underlying resources of a Vulkan device memory object, call:

VkResult vkGetMemoryWin32HandleKHR(
    VkDevice                                    device,
    const VkMemoryGetWin32HandleInfoKHR*        pGetWin32HandleInfo,
    HANDLE*                                     pHandle);
  • device is the logical device that created the device memory being exported.

  • pGetWin32HandleInfo is a pointer to an instance of the VkMemoryGetWin32HandleInfoKHR structure containing parameters of the export operation.

  • pHandle will return the Windows handle representing the underlying resources of the device memory object.

For handle types defined as NT handles, the handles returned by vkGetMemoryWin32HandleKHR are owned by the application. To avoid leaking resources, the application must release ownership of them using the CloseHandle system call when they are no longer needed.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetWin32HandleInfo must be a valid pointer to a valid VkMemoryGetWin32HandleInfoKHR structure

  • pHandle must be a valid pointer to a HANDLE value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetWin32HandleInfoKHR structure is defined as:

typedef struct VkMemoryGetWin32HandleInfoKHR {
    VkStructureType                       sType;
    const void*                           pNext;
    VkDeviceMemory                        memory;
    VkExternalMemoryHandleTypeFlagBits    handleType;
} VkMemoryGetWin32HandleInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memory is the memory object from which the handle will be exported.

  • handleType is the type of handle requested.

The properties of the handle returned depend on the value of handleType. See VkExternalMemoryHandleTypeFlagBits for a description of the properties of the defined external memory handle types.

Valid Usage
  • handleType must have been included in VkExportMemoryAllocateInfo::handleTypes when memory was created.

  • If handleType is defined as an NT handle, vkGetMemoryWin32HandleKHR must be called no more than once for each valid unique combination of memory and handleType.

  • handleType must be defined as an NT handle or a global share handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_GET_WIN32_HANDLE_INFO_KHR

  • pNext must be NULL

  • memory must be a valid VkDeviceMemory handle

  • handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

Windows memory handles compatible with Vulkan may also be created by non-Vulkan APIs using methods beyond the scope of this specification. To determine the correct parameters to use when importing such handles, call:

VkResult vkGetMemoryWin32HandlePropertiesKHR(
    VkDevice                                    device,
    VkExternalMemoryHandleTypeFlagBits          handleType,
    HANDLE                                      handle,
    VkMemoryWin32HandlePropertiesKHR*           pMemoryWin32HandleProperties);
  • device is the logical device that will be importing handle.

  • handleType is the type of the handle handle.

  • handle is the handle which will be imported.

  • pMemoryWin32HandleProperties will return properties of handle.

Valid Usage
  • handle must be an external memory handle created outside of the Vulkan API.

  • handleType must not be one of the handle types defined as opaque.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

  • pMemoryWin32HandleProperties must be a valid pointer to a VkMemoryWin32HandlePropertiesKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryWin32HandlePropertiesKHR structure returned is defined as:

typedef struct VkMemoryWin32HandlePropertiesKHR {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           memoryTypeBits;
} VkMemoryWin32HandlePropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memoryTypeBits is a bitmask containing one bit set for every memory type which the specified windows handle can be imported as.

To import memory from a POSIX file descriptor handle, add a VkImportMemoryFdInfoKHR structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkImportMemoryFdInfoKHR structure is defined as:

typedef struct VkImportMemoryFdInfoKHR {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalMemoryHandleTypeFlagBits    handleType;
    int                                   fd;
} VkImportMemoryFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType specifies the handle type of fd.

  • fd is the external handle to import.

Importing memory from a file descriptor transfers ownership of the file descriptor from the application to the Vulkan implementation. The application must not perform any operations on the file descriptor after a successful import.

Applications can import the same underlying memory into multiple instances of Vulkan, into the same instance from which it was exported, and multiple times into a given Vulkan instance. In all cases, each import operation must create a distinct VkDeviceMemory object.

Valid Usage
Valid Usage (Implicit)

To export a POSIX file descriptor representing the underlying resources of a Vulkan device memory object, call:

VkResult vkGetMemoryFdKHR(
    VkDevice                                    device,
    const VkMemoryGetFdInfoKHR*                 pGetFdInfo,
    int*                                        pFd);
  • device is the logical device that created the device memory being exported.

  • pGetFdInfo is a pointer to an instance of the VkMemoryGetFdInfoKHR structure containing parameters of the export operation.

  • pFd will return a file descriptor representing the underlying resources of the device memory object.

Each call to vkGetMemoryFdKHR must create a new file descriptor and transfer ownership of it to the application. To avoid leaking resources, the application must release ownership of the file descriptor using the close system call when it is no longer needed, or by importing a Vulkan memory object from it. Where supported by the operating system, the implementation must set the file descriptor to be closed automatically when an execve system call is made.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pGetFdInfo must be a valid pointer to a valid VkMemoryGetFdInfoKHR structure

  • pFd must be a valid pointer to a int value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetFdInfoKHR structure is defined as:

typedef struct VkMemoryGetFdInfoKHR {
    VkStructureType                       sType;
    const void*                           pNext;
    VkDeviceMemory                        memory;
    VkExternalMemoryHandleTypeFlagBits    handleType;
} VkMemoryGetFdInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memory is the memory object from which the handle will be exported.

  • handleType is the type of handle requested.

The properties of the file descriptor exported depend on the value of handleType. See VkExternalMemoryHandleTypeFlagBits for a description of the properties of the defined external memory handle types.

Note

The size of the exported file may be larger than the size requested by VkMemoryAllocateInfo::allocationSize. If handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT, then the application can query the file’s actual size with lseek(2).

Valid Usage
  • handleType must have been included in VkExportMemoryAllocateInfo::handleTypes when memory was created.

  • handleType must be defined as a POSIX file descriptor handle.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_GET_FD_INFO_KHR

  • pNext must be NULL

  • memory must be a valid VkDeviceMemory handle

  • handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

POSIX file descriptor memory handles compatible with Vulkan may also be created by non-Vulkan APIs using methods beyond the scope of this specification. To determine the correct parameters to use when importing such handles, call:

VkResult vkGetMemoryFdPropertiesKHR(
    VkDevice                                    device,
    VkExternalMemoryHandleTypeFlagBits          handleType,
    int                                         fd,
    VkMemoryFdPropertiesKHR*                    pMemoryFdProperties);
  • device is the logical device that will be importing fd.

  • handleType is the type of the handle fd.

  • fd is the handle which will be imported.

  • pMemoryFdProperties will return properties of the handle fd.

Valid Usage
  • fd must be an external memory handle created outside of the Vulkan API.

  • handleType must not be VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT_KHR.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

  • pMemoryFdProperties must be a valid pointer to a VkMemoryFdPropertiesKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryFdPropertiesKHR structure returned is defined as:

typedef struct VkMemoryFdPropertiesKHR {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           memoryTypeBits;
} VkMemoryFdPropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memoryTypeBits is a bitmask containing one bit set for every memory type which the specified file descriptor can be imported as.

To import memory from a host pointer, add a VkImportMemoryHostPointerInfoEXT structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkImportMemoryHostPointerInfoEXT structure is defined as:

typedef struct VkImportMemoryHostPointerInfoEXT {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalMemoryHandleTypeFlagBits    handleType;
    void*                                 pHostPointer;
} VkImportMemoryHostPointerInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType specifies the handle type.

  • pHostPointer is the host pointer to import from.

Importing memory from a host pointer shares ownership of the memory between the host and the Vulkan implementation. The application can continue to access the memory through the host pointer but it is the application’s responsibility to synchronize device and non-device access to the underlying memory as defined in Host Access to Device Memory Objects.

Applications can import the same underlying memory into multiple instances of Vulkan and multiple times into a given Vulkan instance. However, implementations may fail to import the same underlying memory multiple times into a given physical device due to platform constraints.

Importing memory from a particular host pointer may not be possible due to additional platform-specific restrictions beyond the scope of this specification in which case the implementation must fail the memory import operation with the error code VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR.

The application must ensure that the imported memory range remains valid and accessible for the lifetime of the imported memory object.

Valid Usage
  • If handleType is not 0, it must be supported for import, as reported in VkExternalMemoryPropertiesKHR

  • If handleType is not 0, it must be VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT or VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT

  • pHostPointer must be a pointer aligned to an integer multiple of VkPhysicalDeviceExternalMemoryHostPropertiesEXT::minImportedHostPointerAlignment

  • If handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT, pHostPointer must be a pointer to allocationSize number of bytes of host memory, where allocationSize is the member of the VkMemoryAllocateInfo structure this structure is chained to

  • If handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT, pHostPointer must be a pointer to allocationSize number of bytes of host mapped foreign memory, where allocationSize is the member of the VkMemoryAllocateInfo structure this structure is chained to

Valid Usage (Implicit)

To determine the correct parameters to use when importing host pointers, call:

VkResult vkGetMemoryHostPointerPropertiesEXT(
    VkDevice                                    device,
    VkExternalMemoryHandleTypeFlagBits          handleType,
    const void*                                 pHostPointer,
    VkMemoryHostPointerPropertiesEXT*           pMemoryHostPointerProperties);
  • device is the logical device that will be importing pHostPointer.

  • handleType is the type of the handle pHostPointer.

  • pHostPointer is the host pointer to import from.

Valid Usage
  • handleType must be VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT or VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT

  • pHostPointer must be a pointer aligned to an integer multiple of VkPhysicalDeviceExternalMemoryHostPropertiesEXT::minImportedHostPointerAlignment

  • If handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT, pHostPointer must be a pointer to host memory

  • If handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT, pHostPointer must be a pointer to host mapped foreign memory

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

  • pMemoryHostPointerProperties must be a valid pointer to a VkMemoryHostPointerPropertiesEXT structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryHostPointerPropertiesEXT structure is defined as:

typedef struct VkMemoryHostPointerPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           memoryTypeBits;
} VkMemoryHostPointerPropertiesEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memoryTypeBits is a bitmask containing one bit set for every memory type which the specified host pointer can be imported as.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_HOST_POINTER_PROPERTIES_EXT

  • pNext must be NULL

To import memory created outside of the current Vulkan instance from an Android hardware buffer, add a VkImportAndroidHardwareBufferInfoANDROID structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkImportAndroidHardwareBufferInfoANDROID structure is defined as:

typedef struct VkImportAndroidHardwareBufferInfoANDROID {
    VkStructureType            sType;
    const void*                pNext;
    struct AHardwareBuffer*    buffer;
} VkImportAndroidHardwareBufferInfoANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • buffer is the Android hardware buffer to import.

If the vkAllocateMemory command succeeds, the implementation must acquire a reference to the imported hardware buffer, which it must release when the device memory object is freed. If the command fails, the implementation must not retain a reference.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMPORT_ANDROID_HARDWARE_BUFFER_INFO_ANDROID

  • buffer must be a valid pointer to a AHardwareBuffer value

To export an Android hardware buffer representing the underlying resources of a Vulkan device memory object, call:

VkResult vkGetMemoryAndroidHardwareBufferANDROID(
    VkDevice                                    device,
    const VkMemoryGetAndroidHardwareBufferInfoANDROID* pInfo,
    struct AHardwareBuffer**                    pBuffer);
  • device is the logical device that created the device memory being exported.

  • pInfo is a pointer to an instance of the VkMemoryGetAndroidHardwareBufferInfoANDROID structure containing parameters of the export operation.

  • pBuffer will return an Android hardware buffer representing the underlying resources of the device memory object.

Each call to vkGetMemoryAndroidHardwareBufferANDROID must return an Android hardware buffer with a new reference acquired in addition to the reference held by the VkDeviceMemory. To avoid leaking resources, the application must release the reference by calling AHardwareBuffer_release when it is no longer needed. When called with the same handle in VkMemoryGetAndroidHardwareBufferInfoANDROID::memory, vkGetMemoryAndroidHardwareBufferANDROID must return the same Android hardware buffer object. If the device memory was created by importing an Android hardware buffer, vkGetMemoryAndroidHardwareBufferANDROID must return that same Android hardware buffer object.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pInfo must be a valid pointer to a valid VkMemoryGetAndroidHardwareBufferInfoANDROID structure

  • pBuffer must be a valid pointer to a valid pointer to a AHardwareBuffer value

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetAndroidHardwareBufferInfoANDROID structure is defined as:

typedef struct VkMemoryGetAndroidHardwareBufferInfoANDROID {
    VkStructureType    sType;
    const void*        pNext;
    VkDeviceMemory     memory;
} VkMemoryGetAndroidHardwareBufferInfoANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memory is the memory object from which the Android hardware buffer will be exported.

Valid Usage
  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must have been included in VkExportMemoryAllocateInfoKHR::handleTypes when memory was created.

  • If the pNext chain of the VkMemoryAllocateInfo used to allocate memory included a VkMemoryDedicatedAllocateInfo with non-NULL image member, then that image must already be bound to memory.

To determine the memory parameters to use when importing an Android hardware buffer, call:

VkResult vkGetAndroidHardwareBufferPropertiesANDROID(
    VkDevice                                    device,
    const struct AHardwareBuffer*               buffer,
    VkAndroidHardwareBufferPropertiesANDROID*   pProperties);
  • device is the logical device that will be importing buffer.

  • buffer is the Android hardware buffer which will be imported.

  • pProperties will return properties of buffer.

Valid Usage
  • buffer must be a valid Android hardware buffer object with at least one of the AHARDWAREBUFFER_USAGE_GPU_* usage flags.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • buffer must be a valid pointer to a valid AHardwareBuffer value

  • pProperties must be a valid pointer to a VkAndroidHardwareBufferPropertiesANDROID structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR

The VkAndroidHardwareBufferPropertiesANDROID structure returned is defined as:

typedef struct VkAndroidHardwareBufferPropertiesANDROID {
    VkStructureType    sType;
    void*              pNext;
    VkDeviceSize       allocationSize;
    uint32_t           memoryTypeBits;
} VkAndroidHardwareBufferPropertiesANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • allocationSize is the size of the external memory

  • memoryTypeBits is a bitmask containing one bit set for every memory type which the specified Android hardware buffer can be imported as.

To obtain format properties of an Android hardware buffer, include an instance of VkAndroidHardwareBufferFormatPropertiesANDROID in the pNext chain of the VkAndroidHardwareBufferPropertiesANDROID instance passed to vkGetAndroidHardwareBufferPropertiesANDROID. This structure is defined as:

typedef struct VkAndroidHardwareBufferFormatPropertiesANDROID {
    VkStructureType                  sType;
    void*                            pNext;
    VkFormat                         format;
    uint64_t                         externalFormat;
    VkFormatFeatureFlags             formatFeatures;
    VkComponentMapping               samplerYcbcrConversionComponents;
    VkSamplerYcbcrModelConversion    suggestedYcbcrModel;
    VkSamplerYcbcrRange              suggestedYcbcrRange;
    VkChromaLocation                 suggestedXChromaOffset;
    VkChromaLocation                 suggestedYChromaOffset;
} VkAndroidHardwareBufferFormatPropertiesANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • format is the Vulkan format corresponding to the Android hardware buffer’s format, or VK_FORMAT_UNDEFINED if there isn’t an equivalent Vulkan format.

  • externalFormat is an implementation-defined external format identifier for use with VkExternalFormatANDROID. It must not be zero.

  • formatFeatures describes the capabilities of this external format when used with an image bound to memory imported from buffer.

  • samplerYcbcrConversionComponents is the component swizzle that should be used in VkSamplerYcbcrConversionCreateInfo.

  • suggestedYcbcrModel is a suggested color model to use in the VkSamplerYcbcrConversionCreateInfo.

  • suggestedYcbcrRange is a suggested numerical value range to use in VkSamplerYcbcrConversionCreateInfo.

  • suggestedXChromaOffset is a suggested X chroma offset to use in VkSamplerYcbcrConversionCreateInfo.

  • suggestedYChromaOffset is a suggested Y chroma offset to use in VkSamplerYcbcrConversionCreateInfo.

If the Android hardware buffer has one of the formats listed in the Format Equivalence table, then format must have the equivalent Vulkan format listed in the table. Otherwise, format may be VK_FORMAT_UNDEFINED, indicating the Android hardware buffer can only be used with an external format.

The formatFeatures member must include VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT and at least one of VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT or VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT, and should include VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT.

Note

The formatFeatures member only indicates the features available when using an <memory-external-android-hardware-buffer-external-formats,external-format image>> created from the Android hardware buffer. Images from Android hardware buffers with a format other than VK_FORMAT_UNDEFINED are subject to the format capabilities obtained from vkGetPhysicalDeviceFormatProperties2, and vkGetPhysicalDeviceImageFormatProperties2 with appropriate parameters. These sets of features are independent of each other, e.g. the external format will support sampler Y’CBCR conversion even if the non-external format doesn’t, and writing to non-external format images is possible but writing to external format images is not.

Android hardware buffers with the same external format must have the same support for VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT, VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT, VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT, VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT, VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT, and VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT. in formatFeatures. Other format features may differ between Android hardware buffers that have the same external format. This allows applications to use the same VkSamplerYcbcrConversion object (and samplers and pipelines created from them) for any Android hardware buffers that have the same external format.

If format is not VK_FORMAT_UNDEFINED, then the value of samplerYcbcrConversionComponents must be valid when used as the components member of VkSamplerYcbcrConversionCreateInfo with that format. If format is VK_FORMAT_UNDEFINED, all members of samplerYcbcrConversionComponents must be VK_COMPONENT_SWIZZLE_IDENTITY.

Implementations may not always be able to determine the color model, numerical range, or chroma offsets of the image contents, so the values in VkAndroidHardwareBufferFormatPropertiesANDROID are only suggestions. Applications should treat these values as sensible defaults to use in the absence of more reliable information obtained through some other means. If the underlying physical device is also usable via OpenGL ES with the GL_OES_EGL_image_external extension, the implementation should suggest values that will produce similar sampled values as would be obtained by sampling the same external image via samplerExternalOES in OpenGL ES using equivalent sampler parameters.

Note

Since GL_OES_EGL_image_external does not require the same sampling and conversion calculations as Vulkan does, achieving identical results between APIs may not be possible on some implementations.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_FORMAT_PROPERTIES_ANDROID

When allocating memory that may be exported to another process or Vulkan instance, add a VkExportMemoryAllocateInfoNV structure to the pNext chain of the VkMemoryAllocateInfo structure, specifying the handle types that may be exported.

The VkExportMemoryAllocateInfoNV structure is defined as:

typedef struct VkExportMemoryAllocateInfoNV {
    VkStructureType                      sType;
    const void*                          pNext;
    VkExternalMemoryHandleTypeFlagsNV    handleTypes;
} VkExportMemoryAllocateInfoNV;
Valid Usage (Implicit)

When VkExportMemoryAllocateInfoNV::handleTypes includes VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV, add a VkExportMemoryWin32HandleInfoNV to the pNext chain of the VkExportMemoryAllocateInfoNV structure to specify security attributes and access rights for the memory object’s external handle.

The VkExportMemoryWin32HandleInfoNV structure is defined as:

typedef struct VkExportMemoryWin32HandleInfoNV {
    VkStructureType               sType;
    const void*                   pNext;
    const SECURITY_ATTRIBUTES*    pAttributes;
    DWORD                         dwAccess;
} VkExportMemoryWin32HandleInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pAttributes is a pointer to a Windows SECURITY_ATTRIBUTES structure specifying security attributes of the handle.

  • dwAccess is a DWORD specifying access rights of the handle.

If this structure is not present, or if pAttributes is set to NULL, default security descriptor values will be used, and child processes created by the application will not inherit the handle, as described in the MSDN documentation for “Synchronization Object Security and Access Rights”[1]. Further, if the structure is not present, the access rights will be

DXGI_SHARED_RESOURCE_READ | DXGI_SHARED_RESOURCE_WRITE
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_NV

  • If pAttributes is not NULL, pAttributes must be a valid pointer to a valid SECURITY_ATTRIBUTES value

To import memory created on the same physical device but outside of the current Vulkan instance, add a VkImportMemoryWin32HandleInfoNV structure to the pNext chain of the VkMemoryAllocateInfo structure, specifying a handle to and the type of the memory.

The VkImportMemoryWin32HandleInfoNV structure is defined as:

typedef struct VkImportMemoryWin32HandleInfoNV {
    VkStructureType                      sType;
    const void*                          pNext;
    VkExternalMemoryHandleTypeFlagsNV    handleType;
    HANDLE                               handle;
} VkImportMemoryWin32HandleInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType is 0 or a VkExternalMemoryHandleTypeFlagBitsNV value specifying the type of memory handle in handle.

  • handle is a Windows HANDLE referring to the memory.

If handleType is 0, this structure is ignored by consumers of the VkMemoryAllocateInfo structure it is chained from.

Valid Usage
  • handleType must not have more than one bit set.

  • handle must be a valid handle to memory, obtained as specified by handleType.

Valid Usage (Implicit)

Bits which can be set in handleType are:

Possible values of VkImportMemoryWin32HandleInfoNV::handleType, specifying the type of an external memory handle, are:

typedef enum VkExternalMemoryHandleTypeFlagBitsNV {
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV = 0x00000001,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_NV = 0x00000002,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_IMAGE_BIT_NV = 0x00000004,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_IMAGE_KMT_BIT_NV = 0x00000008,
} VkExternalMemoryHandleTypeFlagBitsNV;
  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_NV specifies a handle to memory returned by vkGetMemoryWin32HandleNV.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV specifies a handle to memory returned by vkGetMemoryWin32HandleNV, or one duplicated from such a handle using DuplicateHandle().

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_IMAGE_BIT_NV specifies a valid NT handle to memory returned by IDXGIResource1::CreateSharedHandle(), or a handle duplicated from such a handle using DuplicateHandle().

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_IMAGE_KMT_BIT_NV specifies a handle to memory returned by IDXGIResource::GetSharedHandle().

editing-note

(Jon) If additional (non-Win32) bits are added to the possible memory types, this type should move to the VK_NV_external_memory_capabilities section, and each bit would then be protected by ifdefs for the extension it’s defined by.

typedef VkFlags VkExternalMemoryHandleTypeFlagsNV;

VkExternalMemoryHandleTypeFlagsNV is a bitmask type for setting a mask of zero or more VkExternalMemoryHandleTypeFlagBitsNV.

To retrieve the handle corresponding to a device memory object created with VkExportMemoryAllocateInfoNV::handleTypes set to include VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV or VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_NV, call:

VkResult vkGetMemoryWin32HandleNV(
    VkDevice                                    device,
    VkDeviceMemory                              memory,
    VkExternalMemoryHandleTypeFlagsNV           handleType,
    HANDLE*                                     pHandle);
  • device is the logical device that owns the memory.

  • memory is the VkDeviceMemory object.

  • handleType is a bitmask of VkExternalMemoryHandleTypeFlagBitsNV containing a single bit specifying the type of handle requested.

  • handle points to a Windows HANDLE in which the handle is returned.

Valid Usage
Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • memory must be a valid VkDeviceMemory handle

  • handleType must be a valid combination of VkExternalMemoryHandleTypeFlagBitsNV values

  • handleType must not be 0

  • pHandle must be a valid pointer to a HANDLE value

  • memory must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_TOO_MANY_OBJECTS

  • VK_ERROR_OUT_OF_HOST_MEMORY

If the pNext chain of VkMemoryAllocateInfo includes a VkMemoryAllocateFlagsInfo structure, then that structure includes flags and a device mask controlling how many instances of the memory will be allocated.

The VkMemoryAllocateFlagsInfo structure is defined as:

typedef struct VkMemoryAllocateFlagsInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkMemoryAllocateFlags    flags;
    uint32_t                 deviceMask;
} VkMemoryAllocateFlagsInfo;

or the equivalent

typedef VkMemoryAllocateFlagsInfo VkMemoryAllocateFlagsInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkMemoryAllocateFlagBits controlling the allocation.

  • deviceMask is a mask of physical devices in the logical device, indicating that memory must be allocated on each device in the mask, if VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT is set in flags.

If VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT is not set, the number of instances allocated depends on whether VK_MEMORY_HEAP_MULTI_INSTANCE_BIT is set in the memory heap. If VK_MEMORY_HEAP_MULTI_INSTANCE_BIT is set, then memory is allocated for every physical device in the logical device (as if deviceMask has bits set for all device indices). If VK_MEMORY_HEAP_MULTI_INSTANCE_BIT is not set, then a single instance of memory is allocated (as if deviceMask is set to one).

On some implementations, allocations from a multi-instance heap may consume memory on all physical devices even if the deviceMask excludes some devices. If VkPhysicalDeviceGroupProperties::subsetAllocation is VK_TRUE, then memory is only consumed for the devices in the device mask.

Note

In practice, most allocations on a multi-instance heap will be allocated across all physical devices. Unicast allocation support is an optional optimization for a minority of allocations.

Valid Usage
  • If VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT is set, deviceMask must be a valid device mask.

  • If VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT is set, deviceMask must not be zero

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO

  • flags must be a valid combination of VkMemoryAllocateFlagBits values

Bits which can be set in VkMemoryAllocateFlagsInfo::flags, controlling device memory allocation, are:

typedef enum VkMemoryAllocateFlagBits {
    VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT = 0x00000001,
    VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT_KHR = VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT,
} VkMemoryAllocateFlagBits;

or the equivalent

typedef VkMemoryAllocateFlagBits VkMemoryAllocateFlagBitsKHR;
  • VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT specifies that memory will be allocated for the devices in VkMemoryAllocateFlagsInfo::deviceMask.

typedef VkFlags VkMemoryAllocateFlags;

or the equivalent

typedef VkMemoryAllocateFlags VkMemoryAllocateFlagsKHR;

VkMemoryAllocateFlags is a bitmask type for setting a mask of zero or more VkMemoryAllocateFlagBits.

To free a memory object, call:

void vkFreeMemory(
    VkDevice                                    device,
    VkDeviceMemory                              memory,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that owns the memory.

  • memory is the VkDeviceMemory object to be freed.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Before freeing a memory object, an application must ensure the memory object is no longer in use by the device—​for example by command buffers in the pending state. The memory can remain bound to images or buffers at the time the memory object is freed, but any further use of them (on host or device) for anything other than destroying those objects will result in undefined behavior. If there are still any bound images or buffers, the memory may not be immediately released by the implementation, but must be released by the time all bound images and buffers have been destroyed. Once memory is released, it is returned to the heap from which it was allocated.

How memory objects are bound to Images and Buffers is described in detail in the Resource Memory Association section.

If a memory object is mapped at the time it is freed, it is implicitly unmapped.

Note

As described below, host writes are not implicitly flushed when the memory object is unmapped, but the implementation must guarantee that writes that have not been flushed do not affect any other memory.

Valid Usage
  • All submitted commands that refer to memory (via images or buffers) must have completed execution

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If memory is not VK_NULL_HANDLE, memory must be a valid VkDeviceMemory handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If memory is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to memory must be externally synchronized

10.2.1. Host Access to Device Memory Objects

Memory objects created with vkAllocateMemory are not directly host accessible.

Memory objects created with the memory property VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT are considered mappable. Memory objects must be mappable in order to be successfully mapped on the host.

To retrieve a host virtual address pointer to a region of a mappable memory object, call:

VkResult vkMapMemory(
    VkDevice                                    device,
    VkDeviceMemory                              memory,
    VkDeviceSize                                offset,
    VkDeviceSize                                size,
    VkMemoryMapFlags                            flags,
    void**                                      ppData);
  • device is the logical device that owns the memory.

  • memory is the VkDeviceMemory object to be mapped.

  • offset is a zero-based byte offset from the beginning of the memory object.

  • size is the size of the memory range to map, or VK_WHOLE_SIZE to map from offset to the end of the allocation.

  • flags is reserved for future use.

  • ppData points to a pointer in which is returned a host-accessible pointer to the beginning of the mapped range. This pointer minus offset must be aligned to at least VkPhysicalDeviceLimits::minMemoryMapAlignment.

It is an application error to call vkMapMemory on a memory object that is already mapped.

Note

vkMapMemory will fail if the implementation is unable to allocate an appropriately sized contiguous virtual address range, e.g. due to virtual address space fragmentation or platform limits. In such cases, vkMapMemory must return VK_ERROR_MEMORY_MAP_FAILED. The application can improve the likelihood of success by reducing the size of the mapped range and/or removing unneeded mappings using VkUnmapMemory.

vkMapMemory does not check whether the device memory is currently in use before returning the host-accessible pointer. The application must guarantee that any previously submitted command that writes to this range has completed before the host reads from or writes to that range, and that any previously submitted command that reads from that range has completed before the host writes to that region (see here for details on fulfilling such a guarantee). If the device memory was allocated without the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT set, these guarantees must be made for an extended range: the application must round down the start of the range to the nearest multiple of VkPhysicalDeviceLimits::nonCoherentAtomSize, and round the end of the range up to the nearest multiple of VkPhysicalDeviceLimits::nonCoherentAtomSize.

While a range of device memory is mapped for host access, the application is responsible for synchronizing both device and host access to that memory range.

Note

It is important for the application developer to become meticulously familiar with all of the mechanisms described in the chapter on Synchronization and Cache Control as they are crucial to maintaining memory access ordering.

Valid Usage
  • memory must not be currently mapped

  • offset must be less than the size of memory

  • If size is not equal to VK_WHOLE_SIZE, size must be greater than 0

  • If size is not equal to VK_WHOLE_SIZE, size must be less than or equal to the size of the memory minus offset

  • memory must have been created with a memory type that reports VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT

  • memory must not have been allocated with multiple instances.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • memory must be a valid VkDeviceMemory handle

  • flags must be 0

  • ppData must be a valid pointer to a pointer value

  • memory must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to memory must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_MEMORY_MAP_FAILED

typedef VkFlags VkMemoryMapFlags;

VkMemoryMapFlags is a bitmask type for setting a mask, but is currently reserved for future use.

Two commands are provided to enable applications to work with non-coherent memory allocations: vkFlushMappedMemoryRanges and vkInvalidateMappedMemoryRanges.

Note

If the memory object was created with the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT set, vkFlushMappedMemoryRanges and vkInvalidateMappedMemoryRanges are unnecessary and may have a performance cost. However, availability and visibility operations still need to be managed on the device. See the description of host access types for more information.

To flush ranges of non-coherent memory from the host caches, call:

VkResult vkFlushMappedMemoryRanges(
    VkDevice                                    device,
    uint32_t                                    memoryRangeCount,
    const VkMappedMemoryRange*                  pMemoryRanges);
  • device is the logical device that owns the memory ranges.

  • memoryRangeCount is the length of the pMemoryRanges array.

  • pMemoryRanges is a pointer to an array of VkMappedMemoryRange structures describing the memory ranges to flush.

vkFlushMappedMemoryRanges guarantees that host writes to the memory ranges described by pMemoryRanges can be made available to device access, via availability operations from the VK_ACCESS_HOST_WRITE_BIT access type.

Within each range described by pMemoryRanges, each set of nonCoherentAtomSize bytes in that range is flushed if any byte in that set has been written by the host since it was first mapped, or the last time it was flushed. If pMemoryRanges includes sets of nonCoherentAtomSize bytes where no bytes have been written by the host, those bytes must not be flushed.

Unmapping non-coherent memory does not implicitly flush the mapped memory, and host writes that have not been flushed may not ever be visible to the device. However, implementations must ensure that writes that have not been flushed do not become visible to any other memory.

Note

The above guarantee avoids a potential memory corruption in scenarios where host writes to a mapped memory object have not been flushed before the memory is unmapped (or freed), and the virtual address range is subsequently reused for a different mapping (or memory allocation).

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pMemoryRanges must be a valid pointer to an array of memoryRangeCount valid VkMappedMemoryRange structures

  • memoryRangeCount must be greater than 0

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

To invalidate ranges of non-coherent memory from the host caches, call:

VkResult vkInvalidateMappedMemoryRanges(
    VkDevice                                    device,
    uint32_t                                    memoryRangeCount,
    const VkMappedMemoryRange*                  pMemoryRanges);
  • device is the logical device that owns the memory ranges.

  • memoryRangeCount is the length of the pMemoryRanges array.

  • pMemoryRanges is a pointer to an array of VkMappedMemoryRange structures describing the memory ranges to invalidate.

vkInvalidateMappedMemoryRanges guarantees that device writes to the memory ranges described by pMemoryRanges, which have been made visible to the VK_ACCESS_HOST_WRITE_BIT and VK_ACCESS_HOST_READ_BIT access types, are made visible to the host. If a range of non-coherent memory is written by the host and then invalidated without first being flushed, its contents are undefined.

Within each range described by pMemoryRanges, each set of nonCoherentAtomSize bytes in that range is invalidated if any byte in that set has been written by the device since it was first mapped, or the last time it was invalidated.

Note

Mapping non-coherent memory does not implicitly invalidate the mapped memory, and device writes that have not been invalidated must be made visible before the host reads or overwrites them.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pMemoryRanges must be a valid pointer to an array of memoryRangeCount valid VkMappedMemoryRange structures

  • memoryRangeCount must be greater than 0

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkMappedMemoryRange structure is defined as:

typedef struct VkMappedMemoryRange {
    VkStructureType    sType;
    const void*        pNext;
    VkDeviceMemory     memory;
    VkDeviceSize       offset;
    VkDeviceSize       size;
} VkMappedMemoryRange;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memory is the memory object to which this range belongs.

  • offset is the zero-based byte offset from the beginning of the memory object.

  • size is either the size of range, or VK_WHOLE_SIZE to affect the range from offset to the end of the current mapping of the allocation.

Valid Usage
  • memory must be currently mapped

  • If size is not equal to VK_WHOLE_SIZE, offset and size must specify a range contained within the currently mapped range of memory

  • If size is equal to VK_WHOLE_SIZE, offset must be within the currently mapped range of memory

  • If size is equal to VK_WHOLE_SIZE, the end of the current mapping of memory must be a multiple of VkPhysicalDeviceLimits::nonCoherentAtomSize bytes from the beginning of the memory object.

  • offset must be a multiple of VkPhysicalDeviceLimits::nonCoherentAtomSize

  • If size is not equal to VK_WHOLE_SIZE, size must either be a multiple of VkPhysicalDeviceLimits::nonCoherentAtomSize, or offset plus size must equal the size of memory.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MAPPED_MEMORY_RANGE

  • pNext must be NULL

  • memory must be a valid VkDeviceMemory handle

To unmap a memory object once host access to it is no longer needed by the application, call:

void vkUnmapMemory(
    VkDevice                                    device,
    VkDeviceMemory                              memory);
  • device is the logical device that owns the memory.

  • memory is the memory object to be unmapped.

Valid Usage
  • memory must be currently mapped

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • memory must be a valid VkDeviceMemory handle

  • memory must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to memory must be externally synchronized

10.2.2. Lazily Allocated Memory

If the memory object is allocated from a heap with the VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT bit set, that object’s backing memory may be provided by the implementation lazily. The actual committed size of the memory may initially be as small as zero (or as large as the requested size), and monotonically increases as additional memory is needed.

A memory type with this flag set is only allowed to be bound to a VkImage whose usage flags include VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT.

Note

Using lazily allocated memory objects for framebuffer attachments that are not needed once a render pass instance has completed may allow some implementations to never allocate memory for such attachments.

To determine the amount of lazily-allocated memory that is currently committed for a memory object, call:

void vkGetDeviceMemoryCommitment(
    VkDevice                                    device,
    VkDeviceMemory                              memory,
    VkDeviceSize*                               pCommittedMemoryInBytes);
  • device is the logical device that owns the memory.

  • memory is the memory object being queried.

  • pCommittedMemoryInBytes is a pointer to a VkDeviceSize value in which the number of bytes currently committed is returned, on success.

The implementation may update the commitment at any time, and the value returned by this query may be out of date.

The implementation guarantees to allocate any committed memory from the heapIndex indicated by the memory type that the memory object was created with.

Valid Usage
  • memory must have been created with a memory type that reports VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • memory must be a valid VkDeviceMemory handle

  • pCommittedMemoryInBytes must be a valid pointer to a VkDeviceSize value

  • memory must have been created, allocated, or retrieved from device

10.2.3. Protected Memory

Protected memory divides device memory into protected device memory and unprotected device memory.

Protected memory adds the following concepts:

  • Memory:

    • Unprotected device memory, which can be visible to the device and can be visible to the host

    • Protected device memory, which can be visible to the device but must not be visible to the host

  • Resources:

    • Unprotected images and unprotected buffers, to which unprotected memory can be bound

    • Protected images and protected buffers, to which protected memory can be bound

  • Command buffers:

    • Unprotected command buffers, which can be submitted to a device queue to execute unprotected queue operations

    • Protected command buffers, which can be submitted to a protected-capable device queue to execute protected queue operations

  • Device queues:

    • Unprotected device queues, to which unprotected command buffers can be submitted

    • Protected-capable device queues, to which unprotected command buffers or protected command buffers can be submitted

  • Queue submissions

    • Unprotected queue submissions, through which unprotected command buffers can be submitted

    • Protected queue submissions, through which protected command buffers can be submitted

  • Queue operations

    • Unprotected queue operations

      • Any read from or write to protected memory during unprotected queue operations results in undefined behavior but is subject to the inviolable rules below.

    • Protected queue operations

      • Any write to unprotected memory during protected queue operations results in undefined behavior but is subject to the inviolable rules below.

      • Except for framebuffer-space pipeline stages, compute shader stage, and transfer stage, any read from or write to protected memory during protected queue operations results in undefined behavior but is subject to the inviolable rules below.

      • Any queries during protected queue operations results in undefined behavior but is subject to the inviolable rules below.

Protected memory inviolable rules

Implementations must ensure that correct usage or incorrect usage by an application does not affect the integrity of the memory protection system.

The implementation must guarantee that:

  • Protected device memory must not be visible to the host.

  • Values written to unprotected device memory must not be a function of data from protected memory.

Incorrect usage by an application of the memory protection system results in undefined behavior which may include process termination or device loss.

10.2.4. External Memory Handle Types

Android Hardware Buffer

Android’s NDK defines AHardwareBuffer objects which represent device memory that is shareable across processes and that can be accessed by a variety of media APIs and the hardware used to implement them. These Android hardware buffer objects may be imported into VkDeviceMemory objects for access via Vulkan, or exported from Vulkan.

Android hardware buffer objects are reference-counted using Android NDK functions outside the scope of this specification. A VkDeviceMemory imported from an Android hardware buffer or that can be exported to an Android hardware buffer must acquire a reference to its AHardwareBuffer object, and must release this reference when the device memory is freed. During the host execution of a Vulkan command which has an Android hardware buffer as a parameter (including indirect parameters via pNext chains), the application must not decrement the Android hardware buffer’s reference count to zero.

Android hardware buffers can be mapped and unmapped for CPU access using NDK functions. These lock and unlock APIs are considered to acquire and release ownership of the Android hardware buffer, and applications must follow the rules described in External Resource Sharing to transfer ownership between the Vulkan instance and these native APIs.

Android hardware buffers can be shared with external APIs and Vulkan instances on the same device and also with foreign devices. When transferring ownership of the Android hardware buffer, the external and foreign special queue families described in Queue Family Ownership Transfer are not identical. All APIs which produce or consume Android hardware buffers are considered to use foreign devices, except OpenGL ES contexts and Vulkan logical devices that have matching device and driver UUIDs. Implementations may treat a transfer to or from the foreign queue family as if it were a transfer to or from the external queue family when the Android hardware buffer’s usage only permits it to be used on the same physical device.

Android Hardware Buffer Optimal Usages

Vulkan buffer and image usage flags do not correspond exactly to Android hardware buffer usage flags. When allocating Android hardware buffers with non-Vulkan APIs, if any AHARDWAREBUFFER_USAGE_GPU_* usage bits are included, by default the allocator must allocate the memory in such a way that it can support Vulkan usages and creation flags in the usage equivalence table which don’t have Android hardware buffer equivalents.

The VkAndroidHardwareBufferUsageANDROID structure can be attached to the pNext chain of a VkImageFormatProperties2 instance passed to vkGetPhysicalDeviceImageFormatProperties2 to obtain optimal Android hardware buffer usage flags for specific Vulkan resource creation parameters. Some usage flags returned by these commands are required based on the input parameters, but additional vendor-specific usage flags (AHARDWAREBUFFER_USAGE_VENDOR_*) may also be returned. Any Android hardware buffer allocated with these vendor-specific usage flags and imported to Vulkan must only be bound to resources created with parameters that are a subset of the parameters used to obtain the Android hardware buffer usage, since the memory may have been allocated in a way incompatible with other parameters. If a Android hardware buffer is successfully allocated with additional non-vendor-specific usage flags in addition to the recommended usage, it must support being used in the same ways as a Android hardware buffer allocated with only the recommended usage, and also in ways indicated by the additional usage.

Android Hardware Buffer External Formats

Android hardware buffers can represent images using implementation-specific formats, layouts, color models, etc. which don’t have Vulkan equivalents. Such external formats are commonly used by external image sources such as video decoders or cameras. Vulkan can import Android hardware buffers that have external formats, but since the image contents are in an undiscoverable and possibly proprietary representation, images with external formats must only be used as sampled images, must only be sampled with a sampler that has Y’CBCR conversion enabled, and must have optimal tiling.

Images that will be backed by a Android hardware buffer can use an external format by setting VkImageCreateInfo::format to VK_FORMAT_UNDEFINED and including an instance of VkExternalFormatANDROID in the pNext chain. Images can be created with an external format even if the Android hardware buffer has a format which has an equivalent Vulkan format to enable consistent handling of images from sources that might use either category of format. However, all images created with an external format are subject to the valid usage requirements associated with external formats, even if the Android hardware buffer’s format has a Vulkan equivalent. The external format of an Android hardware buffer can be obtained by passing an instance of VkAndroidHardwareBufferFormatPropertiesANDROID to vkGetAndroidHardwareBufferPropertiesANDROID.

Android Hardware Buffer Image Resources

Android hardware buffers have intrinsic width, height, format, and usage properties, so Vulkan images bound to memory imported from an Android hardware buffer must use dedicated allocations: VkMemoryDedicatedRequirements::requiresDedicatedAllocation must be VK_TRUE for images created with VkExternalMemoryImageCreateInfo::handleTypes that includes VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID. When creating an image that will be bound to an imported Android hardware buffer, the image creation parameters must be equivalent to the AHardwareBuffer properties as described by the valid usage of VkMemoryAllocateInfo. Similarly, device memory allocated for a dedicated image must not be exported to an Android hardware buffer until it has been bound to that image, and the implementation must return an Android hardware buffer with properties derived from the image:

  • The width and height members of AHardwareBuffer_desc must be the same as the width and height members of VkImageCreateInfo::extent, respectively;

  • The layers member of AHardwareBuffer_desc must be the same as the arrayLayers member of VkImageCreateInfo;

  • The format member of AHardwareBuffer_desc must be equivalent to VkImageCreateInfo::format as defined by AHardwareBuffer Format Equivalence;

  • The usage member of AHardwareBuffer_desc must include bits corresponding to bits included in VkImageCreateInfo::usage and VkImageCreateInfo::flags where such a correspondence exists according to AHardwareBuffer Usage Equivalence. It may also include additional usage bits, including vendor-specific usages. Presence of vendor usage bits may make the Android hardware buffer only usable in ways indicated by the image creation parameters, even when used outside Vulkan, in a similar way that allocating the Android hardware buffer with usage returned in VkAndroidHardwareBufferUsageANDROID.

Implementations may support fewer combinations of image creation parameters for images with Android hardware buffer external handle type than for non-external images. Support for a given set of parameters can be determined by passing VkExternalImageFormatProperties to vkGetPhysicalDeviceImageFormatProperties2 with handleType set to VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID. Any Android hardware buffer successfully allocated outside Vulkan with usage that includes AHARDWAREBUFFER_USAGE_GPU_* must be supported when using equivalent Vulkan image parameters. If a given choice of image parameters are supported for import, they can also be used to create an image and memory that will be exported to an Android hardware buffer.

Table 13. AHardwareBuffer Format Equivalence
AHardwareBuffer Format Vulkan Format

AHARDWAREBUFFER_FORMAT_R8G8B8A8_UNORM

VK_FORMAT_R8G8B8A8_UNORM

AHARDWAREBUFFER_FORMAT_R8G8B8X8_UNORM

VK_FORMAT_R8G8B8A8_UNORM

AHARDWAREBUFFER_FORMAT_R8G8B8_UNORM

VK_FORMAT_R8G8B8_UNORM

AHARDWAREBUFFER_FORMAT_R5G6B5_UNORM

VK_FORMAT_R5G6B5_UNORM_PACK16

AHARDWAREBUFFER_FORMAT_R16G16B16A16_FLOAT

VK_FORMAT_R16G16B16A16_SFLOAT

AHARDWAREBUFFER_FORMAT_R10G10B10A2_UNORM

VK_FORMAT_A2B10G10R10_UNORM_PACK32

AHARDWAREBUFFER_FORMAT_D16_UNORM

VK_FORMAT_D16_UNORM

AHARDWAREBUFFER_FORMAT_D24_UNORM

VK_FORMAT_X8_D24_UNORM_PACK32

AHARDWAREBUFFER_FORMAT_D24_UNORM_S8_UINT

VK_FORMAT_D24_UNORM_S8_UINT

AHARDWAREBUFFER_FORMAT_D32_FLOAT

VK_FORMAT_D32_SFLOAT

AHARDWAREBUFFER_FORMAT_D32_FLOAT_S8_UINT

VK_FORMAT_D32_SFLOAT_S8_UINT

AHARDWAREBUFFER_FORMAT_S8_UINT

VK_FORMAT_S8_UINT

Table 14. AHardwareBuffer Usage Equivalence
AHardwareBuffer Usage Vulkan Usage or Creation Flag

None

VK_IMAGE_USAGE_TRANSFER_SRC_BIT

None

VK_IMAGE_USAGE_TRANSFER_DST_BIT

AHARDWAREBUFFER_USAGE_GPU_SAMPLED_IMAGE

VK_IMAGE_USAGE_SAMPLED_BIT

AHARDWAREBUFFER_USAGE_GPU_SAMPLED_IMAGE

VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT

AHARDWAREBUFFER_USAGE_GPU_COLOR_OUTPUT

VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT

AHARDWAREBUFFER_USAGE_GPU_CUBE_MAP

VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT

AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE

None 2

AHARDWAREBUFFER_USAGE_PROTECTED_CONTENT

VK_IMAGE_CREATE_PROTECTED_BIT

None

VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT

None

VK_IMAGE_CREATE_EXTENDED_USAGE_BIT

2

The AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE flag does not correspond to a Vulkan image usage or creation flag. Instead, it’s presence indicates that the Android hardware buffer contains a complete set of mip levels (VkImageCreateInfo::mipLevels is ⌈log2(max(width, height))⌉ + 1), and it’s absence indicates that the Android hardware buffer contains only a single mip level.

Note

When using VK_IMAGE_USAGE_MUTABLE_FORMAT_BIT with Android hardware buffer images, applications should use VkImageFormatListCreateInfoKHR to inform the implementation which view formats will be used with the image. For some common sets of format, this allows some implementations to provide significantly better performance when accessing the image via Vulkan.

Android Hardware Buffer Buffer Resources

Android hardware buffers with a format of AHARDWAREBUFFER_FORMAT_BLOB and usage that includes AHARDWAREBUFFER_USAGE_GPU_DATA_BUFFER may be used as the backing store for VkBuffer objects. Such Android hardware buffers have a size in bytes specified by their width; height and layers are 1.

Unlike images, buffer resources backed by Android hardware buffers do not require dedicated allocations.

Exported AHardwareBuffer objects that do not have dedicated images must have a format of AHARDWAREBUFFER_FORMAT_BLOB, usage must include AHARDWAREBUFFER_USAGE_GPU_DATA_BUFFER, width must equal the device memory allocation size, and height and layers must be 1.

10.2.5. Peer Memory Features

Peer memory is memory that is allocated for a given physical device and then bound to a resource and accessed by a different physical device, in a logical device that represents multiple physical devices. Some ways of reading and writing peer memory may not be supported by a device.

To determine how peer memory can be accessed, call:

void vkGetDeviceGroupPeerMemoryFeatures(
    VkDevice                                    device,
    uint32_t                                    heapIndex,
    uint32_t                                    localDeviceIndex,
    uint32_t                                    remoteDeviceIndex,
    VkPeerMemoryFeatureFlags*                   pPeerMemoryFeatures);

or the equivalent command

void vkGetDeviceGroupPeerMemoryFeaturesKHR(
    VkDevice                                    device,
    uint32_t                                    heapIndex,
    uint32_t                                    localDeviceIndex,
    uint32_t                                    remoteDeviceIndex,
    VkPeerMemoryFeatureFlags*                   pPeerMemoryFeatures);
  • device is the logical device that owns the memory.

  • heapIndex is the index of the memory heap from which the memory is allocated.

  • localDeviceIndex is the device index of the physical device that performs the memory access.

  • remoteDeviceIndex is the device index of the physical device that the memory is allocated for.

  • pPeerMemoryFeatures is a pointer to a bitmask of VkPeerMemoryFeatureFlagBits indicating which types of memory accesses are supported for the combination of heap, local, and remote devices.

Valid Usage
  • heapIndex must be less than memoryHeapCount

  • localDeviceIndex must be a valid device index

  • remoteDeviceIndex must be a valid device index

  • localDeviceIndex must not equal remoteDeviceIndex

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pPeerMemoryFeatures must be a valid pointer to a VkPeerMemoryFeatureFlags value

  • pPeerMemoryFeatures must not be 0

Bits which may be set in the value returned for vkGetDeviceGroupPeerMemoryFeatures::pPeerMemoryFeatures, indicating the supported peer memory features, are:

typedef enum VkPeerMemoryFeatureFlagBits {
    VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT = 0x00000001,
    VK_PEER_MEMORY_FEATURE_COPY_DST_BIT = 0x00000002,
    VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT = 0x00000004,
    VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT = 0x00000008,
    VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT_KHR = VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT,
    VK_PEER_MEMORY_FEATURE_COPY_DST_BIT_KHR = VK_PEER_MEMORY_FEATURE_COPY_DST_BIT,
    VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT_KHR = VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT,
    VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT_KHR = VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT,
} VkPeerMemoryFeatureFlagBits;

or the equivalent

typedef VkPeerMemoryFeatureFlagBits VkPeerMemoryFeatureFlagBitsKHR;
  • VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT specifies that the memory can be accessed as the source of a vkCmdCopyBuffer, vkCmdCopyImage, vkCmdCopyBufferToImage, or vkCmdCopyImageToBuffer command.

  • VK_PEER_MEMORY_FEATURE_COPY_DST_BIT specifies that the memory can be accessed as the destination of a vkCmdCopyBuffer, vkCmdCopyImage, vkCmdCopyBufferToImage, or vkCmdCopyImageToBuffer command.

  • VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT specifies that the memory can be read as any memory access type.

  • VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT specifies that the memory can be written as any memory access type. Shader atomics are considered to be writes.

Note

The peer memory features of a memory heap also apply to any accesses that may be performed during image layout transitions.

VK_PEER_MEMORY_FEATURE_COPY_DST_BIT must be supported for all host local heaps and for at least one device local heap.

If a device does not support a peer memory feature, it is still valid to use a resource that includes both local and peer memory bindings with the corresponding access type as long as only the local bindings are actually accessed. For example, an application doing split-frame rendering would use framebuffer attachments that include both local and peer memory bindings, but would scissor the rendering to only update local memory.

typedef VkFlags VkPeerMemoryFeatureFlags;

or the equivalent

typedef VkPeerMemoryFeatureFlags VkPeerMemoryFeatureFlagsKHR;

VkPeerMemoryFeatureFlags is a bitmask type for setting a mask of zero or more VkPeerMemoryFeatureFlagBits.

11. Resource Creation

Vulkan supports two primary resource types: buffers and images. Resources are views of memory with associated formatting and dimensionality. Buffers are essentially unformatted arrays of bytes whereas images contain format information, can be multidimensional and may have associated metadata.

11.1. Buffers

Buffers represent linear arrays of data which are used for various purposes by binding them to a graphics or compute pipeline via descriptor sets or via certain commands, or by directly specifying them as parameters to certain commands.

Buffers are represented by VkBuffer handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkBuffer)

To create buffers, call:

VkResult vkCreateBuffer(
    VkDevice                                    device,
    const VkBufferCreateInfo*                   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkBuffer*                                   pBuffer);
  • device is the logical device that creates the buffer object.

  • pCreateInfo is a pointer to an instance of the VkBufferCreateInfo structure containing parameters affecting creation of the buffer.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pBuffer points to a VkBuffer handle in which the resulting buffer object is returned.

Valid Usage
  • If the flags member of pCreateInfo includes VK_BUFFER_CREATE_SPARSE_BINDING_BIT, creating this VkBuffer must not cause the total required sparse memory for all currently valid sparse resources on the device to exceed VkPhysicalDeviceLimits::sparseAddressSpaceSize

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkBufferCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pBuffer must be a valid pointer to a VkBuffer handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkBufferCreateInfo structure is defined as:

typedef struct VkBufferCreateInfo {
    VkStructureType        sType;
    const void*            pNext;
    VkBufferCreateFlags    flags;
    VkDeviceSize           size;
    VkBufferUsageFlags     usage;
    VkSharingMode          sharingMode;
    uint32_t               queueFamilyIndexCount;
    const uint32_t*        pQueueFamilyIndices;
} VkBufferCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkBufferCreateFlagBits specifying additional parameters of the buffer.

  • size is the size in bytes of the buffer to be created.

  • usage is a bitmask of VkBufferUsageFlagBits specifying allowed usages of the buffer.

  • sharingMode is a VkSharingMode value specifying the sharing mode of the buffer when it will be accessed by multiple queue families.

  • queueFamilyIndexCount is the number of entries in the pQueueFamilyIndices array.

  • pQueueFamilyIndices is a list of queue families that will access this buffer (ignored if sharingMode is not VK_SHARING_MODE_CONCURRENT).

editing-note

(Jon) Should the constraint on usage != 0 be converted to a Valid Usage statement? See gitlab #854.

Valid Usage
  • size must be greater than 0

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, pQueueFamilyIndices must be a valid pointer to an array of queueFamilyIndexCount uint32_t values

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, queueFamilyIndexCount must be greater than 1

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, each element of pQueueFamilyIndices must be unique and must be less than pQueueFamilyPropertyCount returned by either vkGetPhysicalDeviceQueueFamilyProperties or vkGetPhysicalDeviceQueueFamilyProperties2 for the physicalDevice that was used to create device

  • If the sparse bindings feature is not enabled, flags must not contain VK_BUFFER_CREATE_SPARSE_BINDING_BIT

  • If the sparse buffer residency feature is not enabled, flags must not contain VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse aliased residency feature is not enabled, flags must not contain VK_BUFFER_CREATE_SPARSE_ALIASED_BIT

  • If flags contains VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT or VK_BUFFER_CREATE_SPARSE_ALIASED_BIT, it must also contain VK_BUFFER_CREATE_SPARSE_BINDING_BIT

  • If the pNext chain contains an instance of VkExternalMemoryBufferCreateInfo, its handleTypes member must only contain bits that are also in VkExternalBufferProperties::externalMemoryProperties.pname:compatibleHandleTypes, as returned by vkGetPhysicalDeviceExternalBufferProperties with pExternalBufferInfo->handleType equal to any one of the handle types specified in VkExternalMemoryBufferCreateInfo::handleTypes

  • If the protected memory feature is not enabled, flags must not contain VK_BUFFER_CREATE_PROTECTED_BIT

  • If any of the bits VK_BUFFER_CREATE_SPARSE_BINDING_BIT, VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT, or VK_BUFFER_CREATE_SPARSE_ALIASED_BIT are set, VK_BUFFER_CREATE_PROTECTED_BIT must not also be set

  • If the pNext chain contains an instance of VkDedicatedAllocationBufferCreateInfoNV, and the dedicatedAllocation member of the chained structure is VK_TRUE, then flags must not include VK_BUFFER_CREATE_SPARSE_BINDING_BIT, VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT, or VK_BUFFER_CREATE_SPARSE_ALIASED_BIT

Valid Usage (Implicit)

Bits which can be set in VkBufferCreateInfo::usage, specifying usage behavior of a buffer, are:

typedef enum VkBufferUsageFlagBits {
    VK_BUFFER_USAGE_TRANSFER_SRC_BIT = 0x00000001,
    VK_BUFFER_USAGE_TRANSFER_DST_BIT = 0x00000002,
    VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT = 0x00000004,
    VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT = 0x00000008,
    VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT = 0x00000010,
    VK_BUFFER_USAGE_STORAGE_BUFFER_BIT = 0x00000020,
    VK_BUFFER_USAGE_INDEX_BUFFER_BIT = 0x00000040,
    VK_BUFFER_USAGE_VERTEX_BUFFER_BIT = 0x00000080,
    VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT = 0x00000100,
} VkBufferUsageFlagBits;
  • VK_BUFFER_USAGE_TRANSFER_SRC_BIT specifies that the buffer can be used as the source of a transfer command (see the definition of VK_PIPELINE_STAGE_TRANSFER_BIT).

  • VK_BUFFER_USAGE_TRANSFER_DST_BIT specifies that the buffer can be used as the destination of a transfer command.

  • VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT specifies that the buffer can be used to create a VkBufferView suitable for occupying a VkDescriptorSet slot of type VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER.

  • VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT specifies that the buffer can be used to create a VkBufferView suitable for occupying a VkDescriptorSet slot of type VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER.

  • VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT specifies that the buffer can be used in a VkDescriptorBufferInfo suitable for occupying a VkDescriptorSet slot either of type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC.

  • VK_BUFFER_USAGE_STORAGE_BUFFER_BIT specifies that the buffer can be used in a VkDescriptorBufferInfo suitable for occupying a VkDescriptorSet slot either of type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC.

  • VK_BUFFER_USAGE_INDEX_BUFFER_BIT specifies that the buffer is suitable for passing as the buffer parameter to vkCmdBindIndexBuffer.

  • VK_BUFFER_USAGE_VERTEX_BUFFER_BIT specifies that the buffer is suitable for passing as an element of the pBuffers array to vkCmdBindVertexBuffers.

  • VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT specifies that the buffer is suitable for passing as the buffer parameter to vkCmdDrawIndirect, vkCmdDrawIndexedIndirect, or vkCmdDispatchIndirect. It is also suitable for passing as the buffer member of VkIndirectCommandsTokenNVX, or sequencesCountBuffer or sequencesIndexBuffer member of VkCmdProcessCommandsInfoNVX

typedef VkFlags VkBufferUsageFlags;

VkBufferUsageFlags is a bitmask type for setting a mask of zero or more VkBufferUsageFlagBits.

Bits which can be set in VkBufferCreateInfo::flags, specifying additional parameters of a buffer, are:

typedef enum VkBufferCreateFlagBits {
    VK_BUFFER_CREATE_SPARSE_BINDING_BIT = 0x00000001,
    VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT = 0x00000002,
    VK_BUFFER_CREATE_SPARSE_ALIASED_BIT = 0x00000004,
    VK_BUFFER_CREATE_PROTECTED_BIT = 0x00000008,
} VkBufferCreateFlagBits;
  • VK_BUFFER_CREATE_SPARSE_BINDING_BIT specifies that the buffer will be backed using sparse memory binding.

  • VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT specifies that the buffer can be partially backed using sparse memory binding. Buffers created with this flag must also be created with the VK_BUFFER_CREATE_SPARSE_BINDING_BIT flag.

  • VK_BUFFER_CREATE_SPARSE_ALIASED_BIT specifies that the buffer will be backed using sparse memory binding with memory ranges that might also simultaneously be backing another buffer (or another portion of the same buffer). Buffers created with this flag must also be created with the VK_BUFFER_CREATE_SPARSE_BINDING_BIT flag.

  • VK_BUFFER_CREATE_PROTECTED_BIT specifies that the buffer is a protected buffer.

See Sparse Resource Features and Physical Device Features for details of the sparse memory features supported on a device.

typedef VkFlags VkBufferCreateFlags;

VkBufferCreateFlags is a bitmask type for setting a mask of zero or more VkBufferCreateFlagBits.

If the pNext chain includes a VkDedicatedAllocationBufferCreateInfoNV structure, then that structure includes an enable controlling whether the buffer will have a dedicated memory allocation bound to it.

The VkDedicatedAllocationBufferCreateInfoNV structure is defined as:

typedef struct VkDedicatedAllocationBufferCreateInfoNV {
    VkStructureType    sType;
    const void*        pNext;
    VkBool32           dedicatedAllocation;
} VkDedicatedAllocationBufferCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • dedicatedAllocation specifies whether the buffer will have a dedicated allocation bound to it.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_BUFFER_CREATE_INFO_NV

To define a set of external memory handle types that may be used as backing store for a buffer, add a VkExternalMemoryBufferCreateInfo structure to the pNext chain of the VkBufferCreateInfo structure. The VkExternalMemoryBufferCreateInfo structure is defined as:

typedef struct VkExternalMemoryBufferCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExternalMemoryBufferCreateInfo;

or the equivalent

typedef VkExternalMemoryBufferCreateInfo VkExternalMemoryBufferCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBits specifying one or more external memory handle types.

Valid Usage (Implicit)

To destroy a buffer, call:

void vkDestroyBuffer(
    VkDevice                                    device,
    VkBuffer                                    buffer,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the buffer.

  • buffer is the buffer to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to buffer, either directly or via a VkBufferView, must have completed execution

  • If VkAllocationCallbacks were provided when buffer was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when buffer was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If buffer is not VK_NULL_HANDLE, buffer must be a valid VkBuffer handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If buffer is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to buffer must be externally synchronized

11.2. Buffer Views

A buffer view represents a contiguous range of a buffer and a specific format to be used to interpret the data. Buffer views are used to enable shaders to access buffer contents interpreted as formatted data. In order to create a valid buffer view, the buffer must have been created with at least one of the following usage flags:

  • VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT

  • VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT

Buffer views are represented by VkBufferView handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkBufferView)

To create a buffer view, call:

VkResult vkCreateBufferView(
    VkDevice                                    device,
    const VkBufferViewCreateInfo*               pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkBufferView*                               pView);
  • device is the logical device that creates the buffer view.

  • pCreateInfo is a pointer to an instance of the VkBufferViewCreateInfo structure containing parameters to be used to create the buffer.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pView points to a VkBufferView handle in which the resulting buffer view object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkBufferViewCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pView must be a valid pointer to a VkBufferView handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkBufferViewCreateInfo structure is defined as:

typedef struct VkBufferViewCreateInfo {
    VkStructureType            sType;
    const void*                pNext;
    VkBufferViewCreateFlags    flags;
    VkBuffer                   buffer;
    VkFormat                   format;
    VkDeviceSize               offset;
    VkDeviceSize               range;
} VkBufferViewCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • buffer is a VkBuffer on which the view will be created.

  • format is a VkFormat describing the format of the data elements in the buffer.

  • offset is an offset in bytes from the base address of the buffer. Accesses to the buffer view from shaders use addressing that is relative to this starting offset.

  • range is a size in bytes of the buffer view. If range is equal to VK_WHOLE_SIZE, the range from offset to the end of the buffer is used. If VK_WHOLE_SIZE is used and the remaining size of the buffer is not a multiple of the element size of format, then the nearest smaller multiple is used.

Valid Usage
  • offset must be less than the size of buffer

  • offset must be a multiple of VkPhysicalDeviceLimits::minTexelBufferOffsetAlignment

  • If range is not equal to VK_WHOLE_SIZE, range must be greater than 0

  • If range is not equal to VK_WHOLE_SIZE, range must be a multiple of the element size of format

  • If range is not equal to VK_WHOLE_SIZE, range divided by the element size of format must be less than or equal to VkPhysicalDeviceLimits::maxTexelBufferElements

  • If range is not equal to VK_WHOLE_SIZE, the sum of offset and range must be less than or equal to the size of buffer

  • buffer must have been created with a usage value containing at least one of VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT

  • If buffer was created with usage containing VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT, format must be supported for uniform texel buffers, as specified by the VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT flag in VkFormatProperties::bufferFeatures returned by vkGetPhysicalDeviceFormatProperties

  • If buffer was created with usage containing VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT, format must be supported for storage texel buffers, as specified by the VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT flag in VkFormatProperties::bufferFeatures returned by vkGetPhysicalDeviceFormatProperties

  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BUFFER_VIEW_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • buffer must be a valid VkBuffer handle

  • format must be a valid VkFormat value

typedef VkFlags VkBufferViewCreateFlags;

VkBufferViewCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To destroy a buffer view, call:

void vkDestroyBufferView(
    VkDevice                                    device,
    VkBufferView                                bufferView,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the buffer view.

  • bufferView is the buffer view to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to bufferView must have completed execution

  • If VkAllocationCallbacks were provided when bufferView was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when bufferView was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If bufferView is not VK_NULL_HANDLE, bufferView must be a valid VkBufferView handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If bufferView is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to bufferView must be externally synchronized

11.3. Images

Images represent multidimensional - up to 3 - arrays of data which can be used for various purposes (e.g. attachments, textures), by binding them to a graphics or compute pipeline via descriptor sets, or by directly specifying them as parameters to certain commands.

Images are represented by VkImage handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkImage)

To create images, call:

VkResult vkCreateImage(
    VkDevice                                    device,
    const VkImageCreateInfo*                    pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkImage*                                    pImage);
  • device is the logical device that creates the image.

  • pCreateInfo is a pointer to an instance of the VkImageCreateInfo structure containing parameters to be used to create the image.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pImage points to a VkImage handle in which the resulting image object is returned.

Valid Usage
  • If the flags member of pCreateInfo includes VK_IMAGE_CREATE_SPARSE_BINDING_BIT, creating this VkImage must not cause the total required sparse memory for all currently valid sparse resources on the device to exceed VkPhysicalDeviceLimits::sparseAddressSpaceSize

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkImageCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pImage must be a valid pointer to a VkImage handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkImageCreateInfo structure is defined as:

typedef struct VkImageCreateInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkImageCreateFlags       flags;
    VkImageType              imageType;
    VkFormat                 format;
    VkExtent3D               extent;
    uint32_t                 mipLevels;
    uint32_t                 arrayLayers;
    VkSampleCountFlagBits    samples;
    VkImageTiling            tiling;
    VkImageUsageFlags        usage;
    VkSharingMode            sharingMode;
    uint32_t                 queueFamilyIndexCount;
    const uint32_t*          pQueueFamilyIndices;
    VkImageLayout            initialLayout;
} VkImageCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkImageCreateFlagBits describing additional parameters of the image.

  • imageType is a VkImageType value specifying the basic dimensionality of the image. Layers in array textures do not count as a dimension for the purposes of the image type.

  • format is a VkFormat describing the format and type of the data elements that will be contained in the image.

  • extent is a VkExtent3D describing the number of data elements in each dimension of the base level.

  • mipLevels describes the number of levels of detail available for minified sampling of the image.

  • arrayLayers is the number of layers in the image.

  • samples is the number of sub-data element samples in the image as defined in VkSampleCountFlagBits. See Multisampling.

  • tiling is a VkImageTiling value specifying the tiling arrangement of the data elements in memory.

  • usage is a bitmask of VkImageUsageFlagBits describing the intended usage of the image.

  • sharingMode is a VkSharingMode value specifying the sharing mode of the image when it will be accessed by multiple queue families.

  • queueFamilyIndexCount is the number of entries in the pQueueFamilyIndices array.

  • pQueueFamilyIndices is a list of queue families that will access this image (ignored if sharingMode is not VK_SHARING_MODE_CONCURRENT).

  • initialLayout is a VkImageLayout value specifying the initial VkImageLayout of all image subresources of the image. See Image Layouts.

Images created with tiling equal to VK_IMAGE_TILING_LINEAR have further restrictions on their limits and capabilities compared to images created with tiling equal to VK_IMAGE_TILING_OPTIMAL. Creation of images with tiling VK_IMAGE_TILING_LINEAR may not be supported unless other parameters meet all of the constraints:

  • imageType is VK_IMAGE_TYPE_2D

  • format is not a depth/stencil format

  • mipLevels is 1

  • arrayLayers is 1

  • samples is VK_SAMPLE_COUNT_1_BIT

  • usage only includes VK_IMAGE_USAGE_TRANSFER_SRC_BIT and/or VK_IMAGE_USAGE_TRANSFER_DST_BIT

Implementations may support additional limits and capabilities beyond those listed above.

To query an implementation’s specific capabilities for a given combination of format, imageType, tiling, usage, VkExternalMemoryImageCreateInfo::handleTypes and flags, call vkGetPhysicalDeviceImageFormatProperties2. The return value specifies whether that combination of image settings is supported. On success, the VkImageFormatProperties output parameter specifies the set of valid samples bits and the limits for extent, mipLevels, arrayLayers, and maxResourceSize. Even if vkGetPhysicalDeviceImageFormatProperties2. returns success and the parameters to vkCreateImage are all within the returned limits, vkCreateImage must fail and return VK_ERROR_OUT_OF_DEVICE_MEMORY if the resulting size of the image would be larger than maxResourceSize.

To determine the set of valid usage bits for a given format, call vkGetPhysicalDeviceFormatProperties.

Note

For images created without VK_IMAGE_CREATE_EXTENDED_USAGE_BIT a usage bit is valid if it is supported for the format the image is created with.

For images created with VK_IMAGE_CREATE_EXTENDED_USAGE_BIT a usage bit is valid if it is supported for at least one of the formats a VkImageView created from the image can have (see Image Views for more detail).

Valid Usage
  • If the pNext chain doesn’t contain an instance of VkExternalFormatANDROID, or if format is not VK_FORMAT_UNDEFINED, the combination of format, imageType, tiling, usage, and flags must be supported, as indicated by a VK_SUCCESS return value from vkGetPhysicalDeviceImageFormatProperties invoked with the same values passed to the corresponding parameters.

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, pQueueFamilyIndices must be a valid pointer to an array of queueFamilyIndexCount uint32_t values

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, queueFamilyIndexCount must be greater than 1

  • If sharingMode is VK_SHARING_MODE_CONCURRENT, each element of pQueueFamilyIndices must be unique and must be less than pQueueFamilyPropertyCount returned by either vkGetPhysicalDeviceQueueFamilyProperties or vkGetPhysicalDeviceQueueFamilyProperties2 for the physicalDevice that was used to create device

  • format must not be VK_FORMAT_UNDEFINED

  • extent::width must be greater than 0.

  • extent::height must be greater than 0.

  • extent::depth must be greater than 0.

  • mipLevels must be greater than 0

  • arrayLayers must be greater than 0

  • If flags contains VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, imageType must be VK_IMAGE_TYPE_2D

  • If flags contains VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT, imageType must be VK_IMAGE_TYPE_3D

  • If imageType is VK_IMAGE_TYPE_1D, extent.width must be less than or equal to VkPhysicalDeviceLimits::maxImageDimension1D, or VkImageFormatProperties::maxExtent.width (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure) - whichever is higher

  • If imageType is VK_IMAGE_TYPE_2D and flags does not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, extent.width and extent.height must be less than or equal to VkPhysicalDeviceLimits::maxImageDimension2D, or VkImageFormatProperties::maxExtent.width/height (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure) - whichever is higher

  • If imageType is VK_IMAGE_TYPE_2D and flags contains VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, extent.width and extent.height must be less than or equal to VkPhysicalDeviceLimits::maxImageDimensionCube, or VkImageFormatProperties::maxExtent.width/height (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure) - whichever is higher

  • If imageType is VK_IMAGE_TYPE_2D and flags contains VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, extent.width and extent.height must be equal and arrayLayers must be greater than or equal to 6

  • If imageType is VK_IMAGE_TYPE_3D, extent.width, extent.height and extent.depth must be less than or equal to VkPhysicalDeviceLimits::maxImageDimension3D, or VkImageFormatProperties::maxExtent.width/height/depth (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure) - whichever is higher

  • If imageType is VK_IMAGE_TYPE_1D, both extent.height and extent.depth must be 1

  • If imageType is VK_IMAGE_TYPE_2D, extent.depth must be 1

  • mipLevels must be less than or equal to ⌊log2(max(extent.width, extent.height, extent.depth))⌋ + 1.

  • mipLevels must be less than or equal to VkImageFormatProperties::maxMipLevels (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure)

  • arrayLayers must be less than or equal to VkImageFormatProperties::maxArrayLayers (as returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure)

  • If imageType is VK_IMAGE_TYPE_3D, arrayLayers must be 1.

  • If samples is not VK_SAMPLE_COUNT_1_BIT, imageType must be VK_IMAGE_TYPE_2D, flags must not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, tiling must be VK_IMAGE_TILING_OPTIMAL, and mipLevels must be equal to 1

  • If usage includes VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, then bits other than VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, and VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT must not be set

  • If usage includes VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, extent.width must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferWidth

  • If usage includes VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, extent.height must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferHeight

  • If usage includes VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, usage must also contain at least one of VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT.

  • samples must be a bit value that is set in VkImageFormatProperties::sampleCounts returned by vkGetPhysicalDeviceImageFormatProperties with format, imageType, tiling, usage, and flags equal to those in this structure

  • If the multisampled storage images feature is not enabled, and usage contains VK_IMAGE_USAGE_STORAGE_BIT, samples must be VK_SAMPLE_COUNT_1_BIT

  • If the sparse bindings feature is not enabled, flags must not contain VK_IMAGE_CREATE_SPARSE_BINDING_BIT

  • If the sparse aliased residency feature is not enabled, flags must not contain VK_IMAGE_CREATE_SPARSE_ALIASED_BIT

  • If imageType is VK_IMAGE_TYPE_1D, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for 2D images feature is not enabled, and imageType is VK_IMAGE_TYPE_2D, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for 3D images feature is not enabled, and imageType is VK_IMAGE_TYPE_3D, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for images with 2 samples feature is not enabled, imageType is VK_IMAGE_TYPE_2D, and samples is VK_SAMPLE_COUNT_2_BIT, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for images with 4 samples feature is not enabled, imageType is VK_IMAGE_TYPE_2D, and samples is VK_SAMPLE_COUNT_4_BIT, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for images with 8 samples feature is not enabled, imageType is VK_IMAGE_TYPE_2D, and samples is VK_SAMPLE_COUNT_8_BIT, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If the sparse residency for images with 16 samples feature is not enabled, imageType is VK_IMAGE_TYPE_2D, and samples is VK_SAMPLE_COUNT_16_BIT, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT

  • If flags contains VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT or VK_IMAGE_CREATE_SPARSE_ALIASED_BIT, it must also contain VK_IMAGE_CREATE_SPARSE_BINDING_BIT

  • If any of the bits VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, or VK_IMAGE_CREATE_SPARSE_ALIASED_BIT are set, VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT must not also be set

  • If the protected memory feature is not enabled, flags must not contain VK_IMAGE_CREATE_PROTECTED_BIT.

  • If any of the bits VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, or VK_IMAGE_CREATE_SPARSE_ALIASED_BIT are set, VK_IMAGE_CREATE_PROTECTED_BIT must not also be set.

  • If the pNext chain contains an instance of VkExternalMemoryImageCreateInfoNV, it must not contain an instance of VkExternalMemoryImageCreateInfo.

  • If the pNext chain contains an instance of VkExternalMemoryImageCreateInfo, its handleTypes member must only contain bits that are also in VkExternalImageFormatProperties::externalMemoryProperties::compatibleHandleTypes, as returned by vkGetPhysicalDeviceImageFormatProperties2 with format, imageType, tiling, usage, and flags equal to those in this structure, and with an instance of VkPhysicalDeviceExternalImageFormatInfo in the pNext chain, with a handleType equal to any one of the handle types specified in VkExternalMemoryImageCreateInfo::handleTypes

  • If the pNext chain contains an instance of VkExternalMemoryImageCreateInfoNV, its handleTypes member must only contain bits that are also in VkExternalImageFormatPropertiesNV::externalMemoryProperties::compatibleHandleTypes, as returned by vkGetPhysicalDeviceExternalImageFormatPropertiesNV with format, imageType, tiling, usage, and flags equal to those in this structure, and with externalHandleType equal to any one of the handle types specified in VkExternalMemoryImageCreateInfoNV::handleTypes

  • If the logical device was created with VkDeviceGroupDeviceCreateInfo::physicalDeviceCount equal to 1, flags must not contain VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT

  • If flags contains VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT, then mipLevels must be one, arrayLayers must be one, imageType must be VK_IMAGE_TYPE_2D, and tiling must be VK_IMAGE_TILING_OPTIMAL

  • If flags contains VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT, then format must be a block-compressed image format, an ETC compressed image format, or an ASTC compressed image format.

  • If flags contains VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT, then flags must also contain VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT.

  • initialLayout must be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED.

  • If the pNext chain includes a VkExternalMemoryImageCreateInfo or VkExternalMemoryImageCreateInfoNV structure whose handleTypes member is not 0, initialLayout must be VK_IMAGE_LAYOUT_UNDEFINED

  • If the image format is one of those listed in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views:

    • mipLevels must be 1

    • samples must be VK_SAMPLE_COUNT_1_BIT

    • imageType must be VK_IMAGE_TYPE_2D

    • arrayLayers must be 1

  • If tiling is VK_IMAGE_TILING_OPTIMAL, format is a multi-planar format, and VkFormatProperties::optimalTilingFeatures (as returned by vkGetPhysicalDeviceFormatProperties with the same value of format) does not include VK_FORMAT_FEATURE_DISJOINT_BIT, flags must not contain VK_IMAGE_CREATE_DISJOINT_BIT

  • If tiling is VK_IMAGE_TILING_LINEAR, format is a multi-planar format, and VkFormatProperties::linearTilingFeatures (as returned by vkGetPhysicalDeviceFormatProperties with the same value of format) does not include VK_FORMAT_FEATURE_DISJOINT_BIT, flags must not contain VK_IMAGE_CREATE_DISJOINT_BIT

  • If format is not a multi-planar format, and flags does not include VK_IMAGE_CREATE_ALIAS_BIT, flags must not contain VK_IMAGE_CREATE_DISJOINT_BIT

  • If flags contains VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT format must be a depth or depth/stencil format

  • If the pNext chain includes a VkExternalMemoryImageCreateInfo structure whose handleTypes member includes VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID:

    • imageType must be VK_IMAGE_TYPE_2D

    • mipLevels must either be 1 or equal to ⌊log2(max(extent.width, extent.height, extent.depth))⌋ + 1.

    • If format is not VK_FORMAT_UNDEFINED, then format, imageType, tiling, usage, flags, mipLevels, and samples must be supported with VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID external memory handle types according to vkGetPhysicalDeviceImageFormatProperties2

    • If format is VK_FORMAT_UNDEFINED, then the pNext chain must include a VkExternalFormatANDROID structure whose externalFormat member is not 0

  • If the pNext chain includes a VkExternalFormatANDROID structure whose externalFormat member is not 0:

    • format must be VK_FORMAT_UNDEFINED

    • flags must not include VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT

    • usage must not include any usages except VK_IMAGE_USAGE_SAMPLED_BIT

    • tiling must be VK_IMAGE_TILING_OPTIMAL

Valid Usage (Implicit)

If the pNext chain includes a VkDedicatedAllocationImageCreateInfoNV structure, then that structure includes an enable controlling whether the image will have a dedicated memory allocation bound to it.

The VkDedicatedAllocationImageCreateInfoNV structure is defined as:

typedef struct VkDedicatedAllocationImageCreateInfoNV {
    VkStructureType    sType;
    const void*        pNext;
    VkBool32           dedicatedAllocation;
} VkDedicatedAllocationImageCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • dedicatedAllocation specifies whether the image will have a dedicated allocation bound to it.

Note

Using a dedicated allocation for color and depth/stencil attachments or other large images may improve performance on some devices.

Valid Usage
  • If dedicatedAllocation is VK_TRUE, VkImageCreateInfo::flags must not include VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, or VK_IMAGE_CREATE_SPARSE_ALIASED_BIT

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_IMAGE_CREATE_INFO_NV

To define a set of external memory handle types that may be used as backing store for an image, add a VkExternalMemoryImageCreateInfo structure to the pNext chain of the VkImageCreateInfo structure. The VkExternalMemoryImageCreateInfo structure is defined as:

typedef struct VkExternalMemoryImageCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExternalMemoryImageCreateInfo;

or the equivalent

typedef VkExternalMemoryImageCreateInfo VkExternalMemoryImageCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBits specifying one or more external memory handle types.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO

  • handleTypes must be a valid combination of VkExternalMemoryHandleTypeFlagBits values

  • handleTypes must not be 0

If the pNext chain includes a VkExternalMemoryImageCreateInfoNV structure, then that structure defines a set of external memory handle types that may be used as backing store for the image.

The VkExternalMemoryImageCreateInfoNV structure is defined as:

typedef struct VkExternalMemoryImageCreateInfoNV {
    VkStructureType                      sType;
    const void*                          pNext;
    VkExternalMemoryHandleTypeFlagsNV    handleTypes;
} VkExternalMemoryImageCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBitsNV specifying one or more external memory handle types.

Valid Usage (Implicit)

To create an image with an external format, include an instance of VkExternalFormatANDROID in the pNext chain of VkImageCreateInfo. VkExternalFormatANDROID is defined as:

typedef struct VkExternalFormatANDROID {
    VkStructureType    sType;
    void*              pNext;
    uint64_t           externalFormat;
} VkExternalFormatANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • externalFormat is an implementation-defined identifier for the external format

If externalFormat is zero, the effect is as if the VkExternalFormatANDROID structure was not present. Otherwise, the image will have the specified external format, and VkImageCreateInfo::format must be VK_FORMAT_UNDEFINED.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_FORMAT_ANDROID

If the pNext chain of VkImageCreateInfo includes a VkImageSwapchainCreateInfoKHR structure, then that structure includes a swapchain handle indicating that the image will be bound to memory from that swapchain.

The VkImageSwapchainCreateInfoKHR structure is defined as:

typedef struct VkImageSwapchainCreateInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkSwapchainKHR     swapchain;
} VkImageSwapchainCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchain is VK_NULL_HANDLE or a handle of a swapchain that the image will be bound to.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_SWAPCHAIN_CREATE_INFO_KHR

  • If swapchain is not VK_NULL_HANDLE, swapchain must be a valid VkSwapchainKHR handle

If the pNext list of VkImageCreateInfo includes a VkImageFormatListCreateInfoKHR structure, then that structure contains a list of all formats that can be used when creating views of this image.

The VkImageFormatListCreateInfoKHR structure is defined as:

typedef struct VkImageFormatListCreateInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           viewFormatCount;
    const VkFormat*    pViewFormats;
} VkImageFormatListCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • viewFormatCount is the number of entries in the pViewFormats array.

  • pViewFormats is an array which lists of all formats which can be used when creating views of this image.

If viewFormatCount is zero, pViewFormats is ignored and the image is created as if the VkImageFormatListCreateInfoKHR structure were not included in the pNext list of VkImageCreateInfo.

Valid Usage
  • If viewFormatCount is not 0, all of the formats in the pViewFormats array must be compatible with the format specified in the format field of VkImageCreateInfo, as described in the compatibility table.

  • If VkImageCreateInfo::flags does not contain VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT, viewFormatCount must be 0 or 1.

  • If viewFormatCount is not 0, VkImageCreateInfo::format must be in pViewFormats.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_FORMAT_LIST_CREATE_INFO_KHR

  • If viewFormatCount is not 0, pViewFormats must be a valid pointer to an array of viewFormatCount valid VkFormat values

Bits which can be set in VkImageCreateInfo::usage, specifying intended usage of an image, are:

typedef enum VkImageUsageFlagBits {
    VK_IMAGE_USAGE_TRANSFER_SRC_BIT = 0x00000001,
    VK_IMAGE_USAGE_TRANSFER_DST_BIT = 0x00000002,
    VK_IMAGE_USAGE_SAMPLED_BIT = 0x00000004,
    VK_IMAGE_USAGE_STORAGE_BIT = 0x00000008,
    VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT = 0x00000010,
    VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT = 0x00000020,
    VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT = 0x00000040,
    VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT = 0x00000080,
} VkImageUsageFlagBits;
  • VK_IMAGE_USAGE_TRANSFER_SRC_BIT specifies that the image can be used as the source of a transfer command.

  • VK_IMAGE_USAGE_TRANSFER_DST_BIT specifies that the image can be used as the destination of a transfer command.

  • VK_IMAGE_USAGE_SAMPLED_BIT specifies that the image can be used to create a VkImageView suitable for occupying a VkDescriptorSet slot either of type VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and be sampled by a shader.

  • VK_IMAGE_USAGE_STORAGE_BIT specifies that the image can be used to create a VkImageView suitable for occupying a VkDescriptorSet slot of type VK_DESCRIPTOR_TYPE_STORAGE_IMAGE.

  • VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT specifies that the image can be used to create a VkImageView suitable for use as a color or resolve attachment in a VkFramebuffer.

  • VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT specifies that the image can be used to create a VkImageView suitable for use as a depth/stencil attachment in a VkFramebuffer.

  • VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT specifies that the memory bound to this image will have been allocated with the VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT (see Memory Allocation for more detail). This bit can be set for any image that can be used to create a VkImageView suitable for use as a color, resolve, depth/stencil, or input attachment.

  • VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT specifies that the image can be used to create a VkImageView suitable for occupying VkDescriptorSet slot of type VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT; be read from a shader as an input attachment; and be used as an input attachment in a framebuffer.

typedef VkFlags VkImageUsageFlags;

VkImageUsageFlags is a bitmask type for setting a mask of zero or more VkImageUsageFlagBits.

Bits which can be set in VkImageCreateInfo::flags, specifying additional parameters of an image, are:

typedef enum VkImageCreateFlagBits {
    VK_IMAGE_CREATE_SPARSE_BINDING_BIT = 0x00000001,
    VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT = 0x00000002,
    VK_IMAGE_CREATE_SPARSE_ALIASED_BIT = 0x00000004,
    VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT = 0x00000008,
    VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT = 0x00000010,
    VK_IMAGE_CREATE_ALIAS_BIT = 0x00000400,
    VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT = 0x00000040,
    VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT = 0x00000020,
    VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT = 0x00000080,
    VK_IMAGE_CREATE_EXTENDED_USAGE_BIT = 0x00000100,
    VK_IMAGE_CREATE_PROTECTED_BIT = 0x00000800,
    VK_IMAGE_CREATE_DISJOINT_BIT = 0x00000200,
    VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT = 0x00001000,
    VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR = VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT,
    VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT_KHR = VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT,
    VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT_KHR = VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT,
    VK_IMAGE_CREATE_EXTENDED_USAGE_BIT_KHR = VK_IMAGE_CREATE_EXTENDED_USAGE_BIT,
    VK_IMAGE_CREATE_DISJOINT_BIT_KHR = VK_IMAGE_CREATE_DISJOINT_BIT,
    VK_IMAGE_CREATE_ALIAS_BIT_KHR = VK_IMAGE_CREATE_ALIAS_BIT,
} VkImageCreateFlagBits;
  • VK_IMAGE_CREATE_SPARSE_BINDING_BIT specifies that the image will be backed using sparse memory binding.

  • VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT specifies that the image can be partially backed using sparse memory binding. Images created with this flag must also be created with the VK_IMAGE_CREATE_SPARSE_BINDING_BIT flag.

  • VK_IMAGE_CREATE_SPARSE_ALIASED_BIT specifies that the image will be backed using sparse memory binding with memory ranges that might also simultaneously be backing another image (or another portion of the same image). Images created with this flag must also be created with the VK_IMAGE_CREATE_SPARSE_BINDING_BIT flag

  • VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT specifies that the image can be used to create a VkImageView with a different format from the image. For multi-planar formats, VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT specifies that a VkImageView can be created of a plane of the image.

  • VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT specifies that the image can be used to create a VkImageView of type VK_IMAGE_VIEW_TYPE_CUBE or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY.

  • VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT specifies that the image can be used to create a VkImageView of type VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY.

  • VK_IMAGE_CREATE_PROTECTED_BIT specifies that the image is a protected image.

  • VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT specifies that the image can be used with a non-zero value of the splitInstanceBindRegionCount member of a VkBindImageMemoryDeviceGroupInfo structure passed into vkBindImageMemory2. This flag also has the effect of making the image use the standard sparse image block dimensions.

  • VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT specifies that the image having a compressed format can be used to create a VkImageView with an uncompressed format where each texel in the image view corresponds to a compressed texel block of the image.

  • VK_IMAGE_CREATE_EXTENDED_USAGE_BIT specifies that the image can be created with usage flags that are not supported for the format the image is created with but are supported for at least one format a VkImageView created from the image can have.

  • VK_IMAGE_CREATE_DISJOINT_BIT specifies that an image with a multi-planar format must have each plane separately bound to memory, rather than having a single memory binding for the whole image; the presence of this bit distinguishes a disjoint image from an image without this bit set.

  • VK_IMAGE_CREATE_ALIAS_BIT specifies that two images created with the same creation parameters and aliased to the same memory can interpret the contents of the memory consistently with each other, subject to the rules described in the Memory Aliasing section. This flag further specifies that each plane of a disjoint image can share an in-memory non-linear representation with single-plane images, and that a single-plane image can share an in-memory non-linear representation with a plane of a multi-planar disjoint image, according to the rules in Compatible formats of planes of multi-planar formats. If the pNext chain includes a VkExternalMemoryImageCreateInfo or VkExternalMemoryImageCreateInfoNV structure whose handleTypes member is not 0, it is as if VK_IMAGE_CREATE_ALIAS_BIT is set.

  • VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT specifies that an image with a depth or depth/stencil format can be used with custom sample locations when used as a depth/stencil attachment.

typedef VkFlags VkImageCreateFlags;

VkImageCreateFlags is a bitmask type for setting a mask of zero or more VkImageCreateFlagBits.

Possible values of VkImageCreateInfo::imageType, specifying the basic dimensionality of an image, are:

typedef enum VkImageType {
    VK_IMAGE_TYPE_1D = 0,
    VK_IMAGE_TYPE_2D = 1,
    VK_IMAGE_TYPE_3D = 2,
} VkImageType;
  • VK_IMAGE_TYPE_1D specifies a one-dimensional image.

  • VK_IMAGE_TYPE_2D specifies a two-dimensional image.

  • VK_IMAGE_TYPE_3D specifies a three-dimensional image.

Possible values of VkImageCreateInfo::tiling, specifying the tiling arrangement of data elements in an image, are:

typedef enum VkImageTiling {
    VK_IMAGE_TILING_OPTIMAL = 0,
    VK_IMAGE_TILING_LINEAR = 1,
} VkImageTiling;
  • VK_IMAGE_TILING_OPTIMAL specifies optimal tiling (texels are laid out in an implementation-dependent arrangement, for more optimal memory access).

  • VK_IMAGE_TILING_LINEAR specifies linear tiling (texels are laid out in memory in row-major order, possibly with some padding on each row).

To query the host access layout of an image subresource, for an image created with linear tiling, call:

void vkGetImageSubresourceLayout(
    VkDevice                                    device,
    VkImage                                     image,
    const VkImageSubresource*                   pSubresource,
    VkSubresourceLayout*                        pLayout);
  • device is the logical device that owns the image.

  • image is the image whose layout is being queried.

  • pSubresource is a pointer to a VkImageSubresource structure selecting a specific image for the image subresource.

  • pLayout points to a VkSubresourceLayout structure in which the layout is returned.

If the VkFormat of image is a multi-planar format, vkGetImageSubresourceLayout describes one plane of the image.

vkGetImageSubresourceLayout is invariant for the lifetime of a single image. However, the subresource layout of images in Android hardware buffer external memory isn’t known until the image has been bound to memory, so calling vkGetImageSubresourceLayout for such an image before it has been bound will result in undefined behavior.

Valid Usage
  • image must have been created with tiling equal to VK_IMAGE_TILING_LINEAR

  • The aspectMask member of pSubresource must only have a single bit set

  • The mipLevel member of pSubresource must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • The arrayLayer member of pSubresource must be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • If the format of image is a multi-planar format with two planes, the aspectMask member of pSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT or VK_IMAGE_ASPECT_PLANE_1_BIT

  • If the format of image is a multi-planar format with three planes, the aspectMask member of pSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT or VK_IMAGE_ASPECT_PLANE_2_BIT

  • If image was created with the VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID external memory handle type, then image must be bound to memory.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • image must be a valid VkImage handle

  • pSubresource must be a valid pointer to a valid VkImageSubresource structure

  • pLayout must be a valid pointer to a VkSubresourceLayout structure

  • image must have been created, allocated, or retrieved from device

The VkImageSubresource structure is defined as:

typedef struct VkImageSubresource {
    VkImageAspectFlags    aspectMask;
    uint32_t              mipLevel;
    uint32_t              arrayLayer;
} VkImageSubresource;
  • aspectMask is a VkImageAspectFlags selecting the image aspect.

  • mipLevel selects the mipmap level.

  • arrayLayer selects the array layer.

Valid Usage (Implicit)

Information about the layout of the image subresource is returned in a VkSubresourceLayout structure:

typedef struct VkSubresourceLayout {
    VkDeviceSize    offset;
    VkDeviceSize    size;
    VkDeviceSize    rowPitch;
    VkDeviceSize    arrayPitch;
    VkDeviceSize    depthPitch;
} VkSubresourceLayout;
  • offset is the byte offset from the start of the image where the image subresource begins.

  • size is the size in bytes of the image subresource. size includes any extra memory that is required based on rowPitch.

  • rowPitch describes the number of bytes between each row of texels in an image.

  • arrayPitch describes the number of bytes between each array layer of an image.

  • depthPitch describes the number of bytes between each slice of 3D image.

For images created with linear tiling, rowPitch, arrayPitch and depthPitch describe the layout of the image subresource in linear memory. For uncompressed formats, rowPitch is the number of bytes between texels with the same x coordinate in adjacent rows (y coordinates differ by one). arrayPitch is the number of bytes between texels with the same x and y coordinate in adjacent array layers of the image (array layer values differ by one). depthPitch is the number of bytes between texels with the same x and y coordinate in adjacent slices of a 3D image (z coordinates differ by one). Expressed as an addressing formula, the starting byte of a texel in the image subresource has address:

// (x,y,z,layer) are in texel coordinates
address(x,y,z,layer) = layer*arrayPitch + z*depthPitch + y*rowPitch + x*elementSize + offset

For compressed formats, the rowPitch is the number of bytes between compressed texel blocks in adjacent rows. arrayPitch is the number of bytes between compressed texel blocks in adjacent array layers. depthPitch is the number of bytes between compressed texel blocks in adjacent slices of a 3D image.

// (x,y,z,layer) are in compressed texel block coordinates
address(x,y,z,layer) = layer*arrayPitch + z*depthPitch + y*rowPitch + x*compressedTexelBlockByteSize + offset;

arrayPitch is undefined for images that were not created as arrays. depthPitch is defined only for 3D images.

For single-plane color formats, the aspectMask member of VkImageSubresource must be VK_IMAGE_ASPECT_COLOR_BIT. For depth/stencil formats, aspectMask must be either VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT. On implementations that store depth and stencil aspects separately, querying each of these image subresource layouts will return a different offset and size representing the region of memory used for that aspect. On implementations that store depth and stencil aspects interleaved, the same offset and size are returned and represent the interleaved memory allocation.

For multi-planar formats, the aspectMask member of VkImageSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or (for 3-plane formats only) VK_IMAGE_ASPECT_PLANE_2_BIT. Querying each of these image subresource layouts will return a different offset and size representing the region of memory used for that plane.

To destroy an image, call:

void vkDestroyImage(
    VkDevice                                    device,
    VkImage                                     image,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the image.

  • image is the image to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to image, either directly or via a VkImageView, must have completed execution

  • If VkAllocationCallbacks were provided when image was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when image was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If image is not VK_NULL_HANDLE, image must be a valid VkImage handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If image is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to image must be externally synchronized

11.4. Image Layouts

Images are stored in implementation-dependent opaque layouts in memory. Each layout has limitations on what kinds of operations are supported for image subresources using the layout. At any given time, the data representing an image subresource in memory exists in a particular layout which is determined by the most recent layout transition that was performed on that image subresource. Applications have control over which layout each image subresource uses, and can transition an image subresource from one layout to another. Transitions can happen with an image memory barrier, included as part of a vkCmdPipelineBarrier or a vkCmdWaitEvents command buffer command (see Image Memory Barriers), or as part of a subpass dependency within a render pass (see VkSubpassDependency). The image layout is per-image subresource, and separate image subresources of the same image can be in different layouts at the same time with one exception - depth and stencil aspects of a given image subresource must always be in the same layout.

Note

Each layout may offer optimal performance for a specific usage of image memory. For example, an image with a layout of VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL may provide optimal performance for use as a color attachment, but be unsupported for use in transfer commands. Applications can transition an image subresource from one layout to another in order to achieve optimal performance when the image subresource is used for multiple kinds of operations. After initialization, applications need not use any layout other than the general layout, though this may produce suboptimal performance on some implementations.

Upon creation, all image subresources of an image are initially in the same layout, where that layout is selected by the VkImageCreateInfo::initialLayout member. The initialLayout must be either VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED. If it is VK_IMAGE_LAYOUT_PREINITIALIZED, then the image data can be preinitialized by the host while using this layout, and the transition away from this layout will preserve that data. If it is VK_IMAGE_LAYOUT_UNDEFINED, then the contents of the data are considered to be undefined, and the transition away from this layout is not guaranteed to preserve that data. For either of these initial layouts, any image subresources must be transitioned to another layout before they are accessed by the device.

Host access to image memory is only well-defined for images created with VK_IMAGE_TILING_LINEAR tiling and for image subresources of those images which are currently in either the VK_IMAGE_LAYOUT_PREINITIALIZED or VK_IMAGE_LAYOUT_GENERAL layout. Calling vkGetImageSubresourceLayout for a linear image returns a subresource layout mapping that is valid for either of those image layouts.

The set of image layouts consists of:

typedef enum VkImageLayout {
    VK_IMAGE_LAYOUT_UNDEFINED = 0,
    VK_IMAGE_LAYOUT_GENERAL = 1,
    VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL = 2,
    VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL = 3,
    VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL = 4,
    VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL = 5,
    VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL = 6,
    VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL = 7,
    VK_IMAGE_LAYOUT_PREINITIALIZED = 8,
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL = 1000117000,
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL = 1000117001,
    VK_IMAGE_LAYOUT_PRESENT_SRC_KHR = 1000001002,
    VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR = 1000111000,
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL,
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL,
} VkImageLayout;

The type(s) of device access supported by each layout are:

  • VK_IMAGE_LAYOUT_UNDEFINED does not support device access. This layout must only be used as the initialLayout member of VkImageCreateInfo or VkAttachmentDescription, or as the oldLayout in an image transition. When transitioning out of this layout, the contents of the memory are not guaranteed to be preserved.

  • VK_IMAGE_LAYOUT_PREINITIALIZED does not support device access. This layout must only be used as the initialLayout member of VkImageCreateInfo or VkAttachmentDescription, or as the oldLayout in an image transition. When transitioning out of this layout, the contents of the memory are preserved. This layout is intended to be used as the initial layout for an image whose contents are written by the host, and hence the data can be written to memory immediately, without first executing a layout transition. Currently, VK_IMAGE_LAYOUT_PREINITIALIZED is only useful with VK_IMAGE_TILING_LINEAR images because there is not a standard layout defined for VK_IMAGE_TILING_OPTIMAL images.

  • VK_IMAGE_LAYOUT_GENERAL supports all types of device access.

  • VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL must only be used as a color or resolve attachment in a VkFramebuffer. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT usage bit enabled.

  • VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL must only be used as a depth/stencil attachment in a VkFramebuffer. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT usage bit enabled.

  • VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL must only be used as a read-only depth/stencil attachment in a VkFramebuffer and/or as a read-only image in a shader (which can be read as a sampled image, combined image/sampler and/or input attachment). This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT usage bit enabled. Only image subresources of images created with VK_IMAGE_USAGE_SAMPLED_BIT can be used as a sampled image or combined image/sampler in a shader. Similarly, only image subresources of images created with VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT can be used as input attachments.

  • VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL: must only be used as a depth/stencil attachment in a VkFramebuffer, where the depth aspect is read-only, and/or as a read-only image in a shader (which can be read as a sampled image, combined image/sampler and/or input attachment) where only the depth aspect is accessed. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT usage bit enabled. Only image subresources of images created with VK_IMAGE_USAGE_SAMPLED_BIT can be used as a sampled image or combined image/sampler in a shader. Similarly, only image subresources of images created with VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT can be used as input attachments.

  • VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL: must only be used as a depth/stencil attachment in a VkFramebuffer, where the stencil aspect is read-only, and/or as a read-only image in a shader (which can be read as a sampled image, combined image/sampler and/or input attachment) where only the stencil aspect is accessed. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT usage bit enabled. Only image subresources of images created with VK_IMAGE_USAGE_SAMPLED_BIT can be used as a sampled image or combined image/sampler in a shader. Similarly, only image subresources of images created with VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT can be used as input attachments.

  • VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL must only be used as a read-only image in a shader (which can be read as a sampled image, combined image/sampler and/or input attachment). This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_SAMPLED_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT usage bit enabled.

  • VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL must only be used as a source image of a transfer command (see the definition of VK_PIPELINE_STAGE_TRANSFER_BIT). This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_TRANSFER_SRC_BIT usage bit enabled.

  • VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL must only be used as a destination image of a transfer command. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_TRANSFER_DST_BIT usage bit enabled.

  • VK_IMAGE_LAYOUT_PRESENT_SRC_KHR must only be used for presenting a presentable image for display. A swapchain’s image must be transitioned to this layout before calling vkQueuePresentKHR, and must be transitioned away from this layout after calling vkAcquireNextImageKHR.

  • VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR is valid only for shared presentable images, and must be used for any usage the image supports.

The layout of each image subresource is not a state of the image subresource itself, but is rather a property of how the data in memory is organized, and thus for each mechanism of accessing an image in the API the application must specify a parameter or structure member that indicates which image layout the image subresource(s) are considered to be in when the image will be accessed. For transfer commands, this is a parameter to the command (see Clear Commands and Copy Commands). For use as a framebuffer attachment, this is a member in the substructures of the VkRenderPassCreateInfo (see Render Pass). For use in a descriptor set, this is a member in the VkDescriptorImageInfo structure (see Descriptor Set Updates). At the time that any command buffer command accessing an image executes on any queue, the layouts of the image subresources that are accessed must all match the layout specified via the API controlling those accesses.

When performing a layout transition on an image subresource, the old layout value must either equal the current layout of the image subresource (at the time the transition executes), or else be VK_IMAGE_LAYOUT_UNDEFINED (implying that the contents of the image subresource need not be preserved). The new layout used in a transition must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED.

The image layout of each image subresource of a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT is dependent on the last sample locations used to render to the image subresource as a depth/stencil attachment, thus applications must provide the same sample locations that were last used to render to the given image subresource whenever a layout transition of the image subresource happens, otherwise the contents of the depth aspect of the image subresource become undefined.

In addition, depth reads from a depth/stencil attachment referring to an image subresource range of a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT using different sample locations than what have been last used to perform depth writes to the image subresources of the same image subresource range produce undefined results.

Similarly, depth writes to a depth/stencil attachment referring to an image subresource range of a depth/stencil image created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT using different sample locations than what have been last used to perform depth writes to the image subresources of the same image subresource range make the contents of the depth aspect of those image subresources undefined.

11.5. Image Views

Image objects are not directly accessed by pipeline shaders for reading or writing image data. Instead, image views representing contiguous ranges of the image subresources and containing additional metadata are used for that purpose. Views must be created on images of compatible types, and must represent a valid subset of image subresources.

Image views are represented by VkImageView handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkImageView)

The types of image views that can be created are:

typedef enum VkImageViewType {
    VK_IMAGE_VIEW_TYPE_1D = 0,
    VK_IMAGE_VIEW_TYPE_2D = 1,
    VK_IMAGE_VIEW_TYPE_3D = 2,
    VK_IMAGE_VIEW_TYPE_CUBE = 3,
    VK_IMAGE_VIEW_TYPE_1D_ARRAY = 4,
    VK_IMAGE_VIEW_TYPE_2D_ARRAY = 5,
    VK_IMAGE_VIEW_TYPE_CUBE_ARRAY = 6,
} VkImageViewType;

The exact image view type is partially implicit, based on the image’s type and sample count, as well as the view creation parameters as described in the image view compatibility table for vkCreateImageView. This table also shows which SPIR-V OpTypeImage Dim and Arrayed parameters correspond to each image view type.

To create an image view, call:

VkResult vkCreateImageView(
    VkDevice                                    device,
    const VkImageViewCreateInfo*                pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkImageView*                                pView);
  • device is the logical device that creates the image view.

  • pCreateInfo is a pointer to an instance of the VkImageViewCreateInfo structure containing parameters to be used to create the image view.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pView points to a VkImageView handle in which the resulting image view object is returned.

Some of the image creation parameters are inherited by the view. In particular, image view creation inherits the implicit parameter usage specifying the allowed usages of the image view that, by default, takes the value of the corresponding usage parameter specified in VkImageCreateInfo at image creation time. This implicit parameter can be overriden by chaining a VkImageViewUsageCreateInfo structure through the pNext member to VkImageViewCreateInfo as described later in this section.

The remaining parameters are contained in the pCreateInfo.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkImageViewCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pView must be a valid pointer to a VkImageView handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkImageViewCreateInfo structure is defined as:

typedef struct VkImageViewCreateInfo {
    VkStructureType            sType;
    const void*                pNext;
    VkImageViewCreateFlags     flags;
    VkImage                    image;
    VkImageViewType            viewType;
    VkFormat                   format;
    VkComponentMapping         components;
    VkImageSubresourceRange    subresourceRange;
} VkImageViewCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • image is a VkImage on which the view will be created.

  • viewType is an VkImageViewType value specifying the type of the image view.

  • format is a VkFormat describing the format and type used to interpret data elements in the image.

  • components is a VkComponentMapping specifies a remapping of color components (or of depth or stencil components after they have been converted into color components).

  • subresourceRange is a VkImageSubresourceRange selecting the set of mipmap levels and array layers to be accessible to the view.

If image was created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT flag, and if the format of the image is not multi-planar, format can be different from the image’s format, but if image was created without the VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT flag and they are not equal they must be compatible. Image format compatibility is defined in the Format Compatibility Classes section. Views of compatible formats will have the same mapping between texel coordinates and memory locations irrespective of the format, with only the interpretation of the bit pattern changing.

Note

Values intended to be used with one view format may not be exactly preserved when written or read through a different format. For example, an integer value that happens to have the bit pattern of a floating point denorm or NaN may be flushed or canonicalized when written or read through a view with a floating point format. Similarly, a value written through a signed normalized format that has a bit pattern exactly equal to -2b may be changed to -2b + 1 as described in Conversion from Normalized Fixed-Point to Floating-Point.

If image was created with the VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT flag, format must be compatible with the image’s format as described above, or must be an uncompressed format in which case it must be size-compatible with the image’s format, as defined for copying data between images In this case the resulting image view’s texel dimensions equal the dimensions of the selected mip level divided by the compressed texel block size and rounded up.

If the image view is to be used with a sampler which supports sampler Y’CBCR conversion, an identically defined object of type VkSamplerYcbcrConversion to that used to create the sampler must be passed to vkCreateImageView in a VkSamplerYcbcrConversionInfo added to the pNext chain of VkImageViewCreateInfo.

If the image has a multi-planar format and subresourceRange.aspectMask is VK_IMAGE_ASPECT_COLOR_BIT, format must be identical to the image format, and the sampler to be used with the image view must enable sampler Y’CBCR conversion.

If image was created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT and the image has a multi-planar format, and if subresourceRange.aspectMask is VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT, format must be compatible with the corresponding plane of the image, and the sampler to be used with the image view must not enable sampler Y’CBCR conversion. The width and height of the single-plane image view must be derived from the multi-planar image’s dimensions in the manner listed for plane compatibility for the plane.

Any view of an image plane will have the same mapping between texel coordinates and memory locations as used by the channels of the color aspect, subject to the formulae relating texel coordinates to lower-resolution planes as described in Chroma Reconstruction. That is, if an R or B plane has a reduced resolution relative to the G plane of the multi-planar image, the image view operates using the (uplane, vplane) unnormalized coordinates of the reduced-resolution plane, and these coordinates access the same memory locations as the (ucolor, vcolor) unnormalized coordinates of the color aspect for which chroma reconstruction operations operate on the same (uplane, vplane) or (iplane, jplane) coordinates.

Table 15. Image and image view parameter compatibility requirements
Dim, Arrayed, MS Image parameters View parameters

imageType = ci.imageType
width = ci.extent.width
height = ci.extent.height
depth = ci.extent.depth
arrayLayers = ci.arrayLayers
samples = ci.samples
flags = ci.flags
where ci is the VkImageCreateInfo used to create image.

baseArrayLayer, layerCount, and levelCount are members of the subresourceRange member.

1D, 0, 0

imageType = VK_IMAGE_TYPE_1D
width ≥ 1
height = 1
depth = 1
arrayLayers ≥ 1
samples = 1

viewType = VK_IMAGE_VIEW_TYPE_1D
baseArrayLayer ≥ 0
layerCount = 1

1D, 1, 0

imageType = VK_IMAGE_TYPE_1D
width ≥ 1
height = 1
depth = 1
arrayLayers ≥ 1
samples = 1

viewType = VK_IMAGE_VIEW_TYPE_1D_ARRAY
baseArrayLayer ≥ 0
layerCount ≥ 1

2D, 0, 0

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height ≥ 1
depth = 1
arrayLayers ≥ 1
samples = 1

viewType = VK_IMAGE_VIEW_TYPE_2D
baseArrayLayer ≥ 0
layerCount = 1

2D, 1, 0

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height ≥ 1
depth = 1
arrayLayers ≥ 1
samples = 1

viewType = VK_IMAGE_VIEW_TYPE_2D_ARRAY
baseArrayLayer ≥ 0
layerCount ≥ 1

2D, 0, 1

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height ≥ 1
depth = 1
arrayLayers ≥ 1
samples > 1

viewType = VK_IMAGE_VIEW_TYPE_2D
baseArrayLayer ≥ 0
layerCount = 1

2D, 1, 1

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height ≥ 1
depth = 1
arrayLayers ≥ 1
samples > 1

viewType = VK_IMAGE_VIEW_TYPE_2D_ARRAY
baseArrayLayer ≥ 0
layerCount ≥ 1

CUBE, 0, 0

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height = width
depth = 1
arrayLayers ≥ 6
samples = 1
flags includes VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT

viewType = VK_IMAGE_VIEW_TYPE_CUBE
baseArrayLayer ≥ 0
layerCount = 6

CUBE, 1, 0

imageType = VK_IMAGE_TYPE_2D
width ≥ 1
height = width
depth = 1
N ≥ 1
arrayLayers ≥ 6 × N
samples = 1
flags includes VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT

viewType = VK_IMAGE_VIEW_TYPE_CUBE_ARRAY
baseArrayLayer ≥ 0
layerCount = 6 × N, N ≥ 1

3D, 0, 0

imageType = VK_IMAGE_TYPE_3D
width ≥ 1
height ≥ 1
depth ≥ 1
arrayLayers = 1
samples = 1

viewType = VK_IMAGE_VIEW_TYPE_3D
baseArrayLayer = 0
layerCount = 1

3D, 0, 0

imageType = VK_IMAGE_TYPE_3D
width ≥ 1
height ≥ 1
depth ≥ 1
arrayLayers = 1
samples = 1
flags includes VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT
flags does not include VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, and VK_IMAGE_CREATE_SPARSE_ALIASED_BIT

viewType = VK_IMAGE_VIEW_TYPE_2D
levelCount = 1
baseArrayLayer ≥ 0
layerCount = 1

3D, 0, 0

imageType = VK_IMAGE_TYPE_3D
width ≥ 1
height ≥ 1
depth ≥ 1
arrayLayers = 1
samples = 1
flags includes VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT
flags does not include VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, and VK_IMAGE_CREATE_SPARSE_ALIASED_BIT

viewType = VK_IMAGE_VIEW_TYPE_2D_ARRAY
levelCount = 1
baseArrayLayer ≥ 0
layerCount ≥ 1

Valid Usage
  • If image was not created with VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT then viewType must not be VK_IMAGE_VIEW_TYPE_CUBE or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If the image cubemap arrays feature is not enabled, viewType must not be VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If image was created with VK_IMAGE_TYPE_3D but without VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT set then viewType must not be VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY

  • If image was created with VK_IMAGE_TILING_LINEAR, format must be format that has at least one supported feature bit present in the value of VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • image must have been created with a usage value containing at least one of VK_IMAGE_USAGE_SAMPLED_BIT, VK_IMAGE_USAGE_STORAGE_BIT, VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT

  • If image was created with VK_IMAGE_TILING_LINEAR and usage contains VK_IMAGE_USAGE_SAMPLED_BIT, format must be supported for sampled images, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT flag in VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_LINEAR and usage contains VK_IMAGE_USAGE_STORAGE_BIT, format must be supported for storage images, as specified by the VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT flag in VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_LINEAR and usage contains VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, format must be supported for color attachments, as specified by the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT flag in VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_LINEAR and usage contains VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, format must be supported for depth/stencil attachments, as specified by the VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT flag in VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_OPTIMAL and format is not VK_FORMAT_UNDEFINED, format must be format that has at least one supported feature bit present in the value of VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_OPTIMAL, and format is not VK_FORMAT_UNDEFINED, and usage contains VK_IMAGE_USAGE_SAMPLED_BIT, format must be supported for sampled images, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_OPTIMAL, and format is not VK_FORMAT_UNDEFINED, and usage contains VK_IMAGE_USAGE_STORAGE_BIT, format must be supported for storage images, as specified by the VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_OPTIMAL, and format is not VK_FORMAT_UNDEFINED, and usage contains VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, format must be supported for color attachments, as specified by the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • If image was created with VK_IMAGE_TILING_OPTIMAL, and format is not VK_FORMAT_UNDEFINED, and usage contains VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, format must be supported for depth/stencil attachments, as specified by the VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties with the same value of format

  • subresourceRange.baseMipLevel must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • If subresourceRange.levelCount is not VK_REMAINING_MIP_LEVELS, subresourceRange.baseMipLevel + subresourceRange.levelCount must be less than or equal to the mipLevels specified in VkImageCreateInfo when image was created

  • If image is not a 3D image created with VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT set, or viewType is not VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY, subresourceRange::baseArrayLayer must be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • If subresourceRange::layerCount is not VK_REMAINING_ARRAY_LAYERS, image is not a 3D image created with VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT set, or viewType is not VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY, subresourceRange::layerCount must be non-zero and subresourceRange::baseArrayLayer + subresourceRange::layerCount must be less than or equal to the arrayLayers specified in VkImageCreateInfo when image was created

  • If image is a 3D image created with VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT set, and viewType is VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY, subresourceRange::baseArrayLayer must be less than the extent.depth specified in VkImageCreateInfo when image was created

  • If subresourceRange::layerCount is not VK_REMAINING_ARRAY_LAYERS, image is a 3D image created with VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT set, and viewType is VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY, subresourceRange::layerCount must be non-zero and subresourceRange::baseArrayLayer + subresourceRange::layerCount must be less than or equal to the extent.depth specified in VkImageCreateInfo when image was created

  • If image was created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT flag, format must be compatible with the format used to create image, as defined in Format Compatibility Classes

  • If image was created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT flag, but without the VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT flag, and if the format of the image is not a multi-planar format, format must be compatible with the format used to create image, as defined in Format Compatibility Classes

  • If image was created with the VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT flag, format must be compatible with, or must be an uncompressed format that is size-compatible with, the format used to create image.

  • If image was created with the VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT flag, the levelCount and layerCount members of subresourceRange must both be 1.

  • If a VkImageFormatListCreateInfoKHR structure was included in the pNext chain of the VkImageCreateInfo struct used when creating image and the viewFormatCount field of VkImageFormatListCreateInfoKHR is not zero then format must be one of the formats in VkImageFormatListCreateInfoKHR::pViewFormats.

  • If image was created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT flag, if the format of the image is a multi-planar format, and if subresourceRange.aspectMask is one of VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT, then format must be compatible with the VkFormat for the plane of the image format indicated by subresourceRange.aspectMask, as defined in Compatible formats of planes of multi-planar formats

  • If image was not created with the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT flag, or if the format of the image is a multi-planar format and if subresourceRange.aspectMask is VK_IMAGE_ASPECT_COLOR_BIT, format must be identical to the format used to create image

  • If the pNext chain contains an instance of VkSamplerYcbcrConversionInfo with a conversion value other than VK_NULL_HANDLE, all members of components must have the value VK_COMPONENT_SWIZZLE_IDENTITY.

  • If image is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • subresourceRange and viewType must be compatible with the image, as described in the compatibility table

  • If image has an external format:

    • format must be VK_FORMAT_UNDEFINED

    • The pNext chain must contain an instance of VkSamplerYcbcrConversionInfo with a conversion object created with the same external format as image

    • All members of components must be VK_COMPONENT_SWIZZLE_IDENTITY

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkImageViewUsageCreateInfo or VkSamplerYcbcrConversionInfo

  • Each sType member in the pNext chain must be unique

  • flags must be 0

  • image must be a valid VkImage handle

  • viewType must be a valid VkImageViewType value

  • format must be a valid VkFormat value

  • components must be a valid VkComponentMapping structure

  • subresourceRange must be a valid VkImageSubresourceRange structure

typedef VkFlags VkImageViewCreateFlags;

VkImageViewCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The set of usages for the created image view can be restricted compared to the parent image’s usage flags by chaining a VkImageViewUsageCreateInfo structure through the pNext member to VkImageViewCreateInfo.

The VkImageViewUsageCreateInfo structure is defined as:

typedef struct VkImageViewUsageCreateInfo {
    VkStructureType      sType;
    const void*          pNext;
    VkImageUsageFlags    usage;
} VkImageViewUsageCreateInfo;

or the equivalent

typedef VkImageViewUsageCreateInfo VkImageViewUsageCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • usage is a bitmask describing the allowed usages of the image view. See VkImageUsageFlagBits for a description of the supported bits.

When this structure is chained to VkImageViewCreateInfo the usage field overrides the implicit usage parameter inherited from image creation time and its value is used instead for the purposes of determining the valid usage conditions of VkImageViewCreateInfo.

Valid Usage
  • usage must not include any set bits that were not set in the usage member of the VkImageCreateInfo structure used to create the image this image view is created from.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO

  • usage must be a valid combination of VkImageUsageFlagBits values

  • usage must not be 0

The VkImageSubresourceRange structure is defined as:

typedef struct VkImageSubresourceRange {
    VkImageAspectFlags    aspectMask;
    uint32_t              baseMipLevel;
    uint32_t              levelCount;
    uint32_t              baseArrayLayer;
    uint32_t              layerCount;
} VkImageSubresourceRange;
  • aspectMask is a bitmask of VkImageAspectFlagBits specifying which aspect(s) of the image are included in the view.

  • baseMipLevel is the first mipmap level accessible to the view.

  • levelCount is the number of mipmap levels (starting from baseMipLevel) accessible to the view.

  • baseArrayLayer is the first array layer accessible to the view.

  • layerCount is the number of array layers (starting from baseArrayLayer) accessible to the view.

The number of mipmap levels and array layers must be a subset of the image subresources in the image. If an application wants to use all mip levels or layers in an image after the baseMipLevel or baseArrayLayer, it can set levelCount and layerCount to the special values VK_REMAINING_MIP_LEVELS and VK_REMAINING_ARRAY_LAYERS without knowing the exact number of mip levels or layers.

For cube and cube array image views, the layers of the image view starting at baseArrayLayer correspond to faces in the order +X, -X, +Y, -Y, +Z, -Z. For cube arrays, each set of six sequential layers is a single cube, so the number of cube maps in a cube map array view is layerCount / 6, and image array layer (baseArrayLayer + i) is face index (i mod 6) of cube i / 6. If the number of layers in the view, whether set explicitly in layerCount or implied by VK_REMAINING_ARRAY_LAYERS, is not a multiple of 6, behavior when indexing the last cube is undefined.

aspectMask must be only VK_IMAGE_ASPECT_COLOR_BIT, VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT if format is a color, depth-only or stencil-only format, respectively, except if format is a multi-planar format. If using a depth/stencil format with both depth and stencil components, aspectMask must include at least one of VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT, and can include both.

When the VkImageSubresourceRange structure is used to select a subset of the slices of a 3D image’s mip level in order to create a 2D or 2D array image view of a 3D image created with VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT, baseArrayLayer and layerCount specify the first slice index and the number of slices to include in the created image view. Such an image view can be used as a framebuffer attachment that refers only to the specified range of slices of the selected mip level. However, any layout transitions performed on such an attachment view during a render pass instance still apply to the entire subresource referenced which includes all the slices of the selected mip level.

When using an imageView of a depth/stencil image to populate a descriptor set (e.g. for sampling in the shader, or for use as an input attachment), the aspectMask must only include one bit and selects whether the imageView is used for depth reads (i.e. using a floating-point sampler or input attachment in the shader) or stencil reads (i.e. using an unsigned integer sampler or input attachment in the shader). When an imageView of a depth/stencil image is used as a depth/stencil framebuffer attachment, the aspectMask is ignored and both depth and stencil image subresources are used.

The components member is of type VkComponentMapping, and describes a remapping from components of the image to components of the vector returned by shader image instructions. This remapping must be identity for storage image descriptors, input attachment descriptors, framebuffer attachments, and any VkImageView used with a combined image sampler that enables sampler Y’CBCR conversion.

When creating a VkImageView, if sampler Y’CBCR conversion is enabled in the sampler, the aspectMask of a subresourceRange used by the VkImageView must be VK_IMAGE_ASPECT_COLOR_BIT.

When creating a VkImageView, if sampler Y’CBCR conversion is not enabled in the sampler and the image format is multi-planar, the image must have been created with VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT, and the aspectMask of the VkImageView’s subresourceRange must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT or VK_IMAGE_ASPECT_PLANE_2_BIT.

Valid Usage
  • If levelCount is not VK_REMAINING_MIP_LEVELS, it must be greater than 0

  • If layerCount is not VK_REMAINING_ARRAY_LAYERS, it must be greater than 0

  • If aspectMask includes VK_IMAGE_ASPECT_COLOR_BIT, then it must not include any of VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT

Valid Usage (Implicit)

Bits which can be set in an aspect mask to specify aspects of an image for purposes such as identifying a subresource, are:

typedef enum VkImageAspectFlagBits {
    VK_IMAGE_ASPECT_COLOR_BIT = 0x00000001,
    VK_IMAGE_ASPECT_DEPTH_BIT = 0x00000002,
    VK_IMAGE_ASPECT_STENCIL_BIT = 0x00000004,
    VK_IMAGE_ASPECT_METADATA_BIT = 0x00000008,
    VK_IMAGE_ASPECT_PLANE_0_BIT = 0x00000010,
    VK_IMAGE_ASPECT_PLANE_1_BIT = 0x00000020,
    VK_IMAGE_ASPECT_PLANE_2_BIT = 0x00000040,
    VK_IMAGE_ASPECT_PLANE_0_BIT_KHR = VK_IMAGE_ASPECT_PLANE_0_BIT,
    VK_IMAGE_ASPECT_PLANE_1_BIT_KHR = VK_IMAGE_ASPECT_PLANE_1_BIT,
    VK_IMAGE_ASPECT_PLANE_2_BIT_KHR = VK_IMAGE_ASPECT_PLANE_2_BIT,
} VkImageAspectFlagBits;
  • VK_IMAGE_ASPECT_COLOR_BIT specifies the color aspect.

  • VK_IMAGE_ASPECT_DEPTH_BIT specifies the depth aspect.

  • VK_IMAGE_ASPECT_STENCIL_BIT specifies the stencil aspect.

  • VK_IMAGE_ASPECT_METADATA_BIT specifies the metadata aspect, used for sparse sparse resource operations.

typedef VkFlags VkImageAspectFlags;

VkImageAspectFlags is a bitmask type for setting a mask of zero or more VkImageAspectFlagBits.

The VkComponentMapping structure is defined as:

typedef struct VkComponentMapping {
    VkComponentSwizzle    r;
    VkComponentSwizzle    g;
    VkComponentSwizzle    b;
    VkComponentSwizzle    a;
} VkComponentMapping;
  • r is a VkComponentSwizzle specifying the component value placed in the R component of the output vector.

  • g is a VkComponentSwizzle specifying the component value placed in the G component of the output vector.

  • b is a VkComponentSwizzle specifying the component value placed in the B component of the output vector.

  • a is a VkComponentSwizzle specifying the component value placed in the A component of the output vector.

Valid Usage (Implicit)

Possible values of the members of VkComponentMapping, specifying the component values placed in each component of the output vector, are:

typedef enum VkComponentSwizzle {
    VK_COMPONENT_SWIZZLE_IDENTITY = 0,
    VK_COMPONENT_SWIZZLE_ZERO = 1,
    VK_COMPONENT_SWIZZLE_ONE = 2,
    VK_COMPONENT_SWIZZLE_R = 3,
    VK_COMPONENT_SWIZZLE_G = 4,
    VK_COMPONENT_SWIZZLE_B = 5,
    VK_COMPONENT_SWIZZLE_A = 6,
} VkComponentSwizzle;
  • VK_COMPONENT_SWIZZLE_IDENTITY specifies that the component is set to the identity swizzle.

  • VK_COMPONENT_SWIZZLE_ZERO specifies that the component is set to zero.

  • VK_COMPONENT_SWIZZLE_ONE specifies that the component is set to either 1 or 1.0, depending on whether the type of the image view format is integer or floating-point respectively, as determined by the Format Definition section for each VkFormat.

  • VK_COMPONENT_SWIZZLE_R specifies that the component is set to the value of the R component of the image.

  • VK_COMPONENT_SWIZZLE_G specifies that the component is set to the value of the G component of the image.

  • VK_COMPONENT_SWIZZLE_B specifies that the component is set to the value of the B component of the image.

  • VK_COMPONENT_SWIZZLE_A specifies that the component is set to the value of the A component of the image.

Setting the identity swizzle on a component is equivalent to setting the identity mapping on that component. That is:

Table 16. Component Mappings Equivalent To VK_COMPONENT_SWIZZLE_IDENTITY
Component Identity Mapping

components.r

VK_COMPONENT_SWIZZLE_R

components.g

VK_COMPONENT_SWIZZLE_G

components.b

VK_COMPONENT_SWIZZLE_B

components.a

VK_COMPONENT_SWIZZLE_A

To destroy an image view, call:

void vkDestroyImageView(
    VkDevice                                    device,
    VkImageView                                 imageView,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the image view.

  • imageView is the image view to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to imageView must have completed execution

  • If VkAllocationCallbacks were provided when imageView was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when imageView was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If imageView is not VK_NULL_HANDLE, imageView must be a valid VkImageView handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If imageView is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to imageView must be externally synchronized

11.6. Resource Memory Association

Resources are initially created as virtual allocations with no backing memory. Device memory is allocated separately (see Device Memory) and then associated with the resource. This association is done differently for sparse and non-sparse resources.

Resources created with any of the sparse creation flags are considered sparse resources. Resources created without these flags are non-sparse. The details on resource memory association for sparse resources is described in Sparse Resources.

Non-sparse resources must be bound completely and contiguously to a single VkDeviceMemory object before the resource is passed as a parameter to any of the following operations:

  • creating image or buffer views

  • updating descriptor sets

  • recording commands in a command buffer

Once bound, the memory binding is immutable for the lifetime of the resource.

In a logical device representing more than one physical device, buffer and image resources exist on all physical devices but can be bound to memory differently on each. Each such replicated resource is an instance of the resource. For sparse resources, each instance can be bound to memory arbitrarily differently. For non-sparse resources, each instance can either be bound to the local or a peer instance of the memory, or for images can be bound to rectangular regions from the local and/or peer instances. When a resource is used in a descriptor set, each physical device interprets the descriptor according to its own instance’s binding to memory.

Note

There are no new copy commands to transfer data between physical devices. Instead, an application can create a resource with a peer mapping and use it as the source or destination of a transfer command executed by a single physical device to copy the data from one physical device to another.

To determine the memory requirements for a buffer resource, call:

void vkGetBufferMemoryRequirements(
    VkDevice                                    device,
    VkBuffer                                    buffer,
    VkMemoryRequirements*                       pMemoryRequirements);
  • device is the logical device that owns the buffer.

  • buffer is the buffer to query.

  • pMemoryRequirements points to an instance of the VkMemoryRequirements structure in which the memory requirements of the buffer object are returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • buffer must be a valid VkBuffer handle

  • pMemoryRequirements must be a valid pointer to a VkMemoryRequirements structure

  • buffer must have been created, allocated, or retrieved from device

To determine the memory requirements for an image resource which is not created with the VK_IMAGE_CREATE_DISJOINT_BIT flag set, call:

void vkGetImageMemoryRequirements(
    VkDevice                                    device,
    VkImage                                     image,
    VkMemoryRequirements*                       pMemoryRequirements);
  • device is the logical device that owns the image.

  • image is the image to query.

  • pMemoryRequirements points to an instance of the VkMemoryRequirements structure in which the memory requirements of the image object are returned.

Valid Usage
  • image must not have been created with the VK_IMAGE_CREATE_DISJOINT_BIT flag set

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • image must be a valid VkImage handle

  • pMemoryRequirements must be a valid pointer to a VkMemoryRequirements structure

  • image must have been created, allocated, or retrieved from device

The VkMemoryRequirements structure is defined as:

typedef struct VkMemoryRequirements {
    VkDeviceSize    size;
    VkDeviceSize    alignment;
    uint32_t        memoryTypeBits;
} VkMemoryRequirements;
  • size is the size, in bytes, of the memory allocation required for the resource.

  • alignment is the alignment, in bytes, of the offset within the allocation required for the resource.

  • memoryTypeBits is a bitmask and contains one bit set for every supported memory type for the resource. Bit i is set if and only if the memory type i in the VkPhysicalDeviceMemoryProperties structure for the physical device is supported for the resource.

The precise size of images that will be bound to external Android hardware buffer memory is unknown until the memory has been imported or allocated, so calling vkGetImageMemoryRequirements with such an image before it has been bound to memory will result in undefined behavior. When importing Android hardware buffer memory, the allocationSize can be determined by calling vkGetAndroidHardwareBufferPropertiesANDROID. When allocating new memory for an image that can be exported to a Android hardware buffer, the memory’s allocationSize must be zero; the actual size will be determined by the dedicated image’s parameters. After the memory has been allocated, the amount of space allocated from the memory’s heap can be obtained by getting the image’s memory requirements or by calling vkGetAndroidHardwareBufferPropertiesANDROID with the Android hardware buffer exported from the memory.

The implementation guarantees certain properties about the memory requirements returned by vkGetBufferMemoryRequirements and vkGetImageMemoryRequirements:

  • The memoryTypeBits member always contains at least one bit set.

  • If buffer is a VkBuffer not created with the VK_BUFFER_CREATE_SPARSE_BINDING_BIT bit set, or if image is a VkImage that was created with a VK_IMAGE_TILING_LINEAR value in the tiling member of the VkImageCreateInfo structure passed to vkCreateImage, then the memoryTypeBits member always contains at least one bit set corresponding to a VkMemoryType with a propertyFlags that has both the VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT bit and the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT bit set. In other words, mappable coherent memory can always be attached to these objects.

  • If buffer was created with VkExternalMemoryBufferCreateInfo::handleTypes set to 0 or image was created with VkExternalMemoryImageCreateInfo::handleTypes set to 0, the memoryTypeBits member always contains at least one bit set corresponding to a VkMemoryType with a propertyFlags that has the VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT bit set.

  • The memoryTypeBits member is identical for all VkBuffer objects created with the same value for the flags and usage members in the VkBufferCreateInfo structure and the handleTypes member of the VkExternalMemoryBufferCreateInfo structure passed to vkCreateBuffer. Further, if usage1 and usage2 of type VkBufferUsageFlags are such that the bits set in usage2 are a subset of the bits set in usage1, and they have the same flags and VkExternalMemoryBufferCreateInfo::handleTypes, then the bits set in memoryTypeBits returned for usage1 must be a subset of the bits set in memoryTypeBits returned for usage2, for all values of flags.

  • The alignment member is a power of two.

  • The alignment member is identical for all VkBuffer objects created with the same combination of values for the usage and flags members in the VkBufferCreateInfo structure passed to vkCreateBuffer.

  • For images created with a color format, the memoryTypeBits member is identical for all VkImage objects created with the same combination of values for the tiling member, the VK_IMAGE_CREATE_SPARSE_BINDING_BIT bit of the flags member, the VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT bit of the flags member, handleTypes member of VkExternalMemoryImageCreateInfo, and the VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT of the usage member in the VkImageCreateInfo structure passed to vkCreateImage.

  • For images created with a depth/stencil format, the memoryTypeBits member is identical for all VkImage objects created with the same combination of values for the format member, the tiling member, the VK_IMAGE_CREATE_SPARSE_BINDING_BIT bit of the flags member, the VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT bit of the flags member, handleTypes member of VkExternalMemoryImageCreateInfo, and the VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT of the usage member in the VkImageCreateInfo structure passed to vkCreateImage.

  • If the memory requirements are for a VkImage, the memoryTypeBits member must not refer to a VkMemoryType with a propertyFlags that has the VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT bit set if the vkGetImageMemoryRequirements::image did not have VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT bit set in the usage member of the VkImageCreateInfo structure passed to vkCreateImage.

  • If the memory requirements are for a VkBuffer, the memoryTypeBits member must not refer to a VkMemoryType with a propertyFlags that has the VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT bit set.

    Note

    The implication of this requirement is that lazily allocated memory is disallowed for buffers in all cases.

  • The size member is identical for all VkBuffer objects created with the same combination of creation parameters specified in VkBufferCreateInfo and its pNext chain.

  • The size member is identical for all VkImage objects created with the same combination of creation parameters specified in VkImageCreateInfo and its pNext chain.

    Note

    This, however, does not imply that they interpret the contents of the bound memory identically with each other. That additional guarantee, however, can be explicitly requested using VK_IMAGE_CREATE_ALIAS_BIT.

To determine the memory requirements for a buffer resource, call:

void vkGetBufferMemoryRequirements2(
    VkDevice                                    device,
    const VkBufferMemoryRequirementsInfo2*      pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

or the equivalent command

void vkGetBufferMemoryRequirements2KHR(
    VkDevice                                    device,
    const VkBufferMemoryRequirementsInfo2*      pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);
  • device is the logical device that owns the buffer.

  • pInfo is a pointer to an instance of the VkBufferMemoryRequirementsInfo2 structure containing parameters required for the memory requirements query.

  • pMemoryRequirements points to an instance of the VkMemoryRequirements2 structure in which the memory requirements of the buffer object are returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pInfo must be a valid pointer to a valid VkBufferMemoryRequirementsInfo2 structure

  • pMemoryRequirements must be a valid pointer to a VkMemoryRequirements2 structure

The VkBufferMemoryRequirementsInfo2 structure is defined as:

typedef struct VkBufferMemoryRequirementsInfo2 {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           buffer;
} VkBufferMemoryRequirementsInfo2;

or the equivalent

typedef VkBufferMemoryRequirementsInfo2 VkBufferMemoryRequirementsInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • buffer is the buffer to query.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2

  • pNext must be NULL

  • buffer must be a valid VkBuffer handle

To determine the memory requirements for an image resource, call:

void vkGetImageMemoryRequirements2(
    VkDevice                                    device,
    const VkImageMemoryRequirementsInfo2*       pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

or the equivalent command

void vkGetImageMemoryRequirements2KHR(
    VkDevice                                    device,
    const VkImageMemoryRequirementsInfo2*       pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);
  • device is the logical device that owns the image.

  • pInfo is a pointer to an instance of the VkImageMemoryRequirementsInfo2 structure containing parameters required for the memory requirements query.

  • pMemoryRequirements points to an instance of the VkMemoryRequirements2 structure in which the memory requirements of the image object are returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pInfo must be a valid pointer to a valid VkImageMemoryRequirementsInfo2 structure

  • pMemoryRequirements must be a valid pointer to a VkMemoryRequirements2 structure

The VkImageMemoryRequirementsInfo2 structure is defined as:

typedef struct VkImageMemoryRequirementsInfo2 {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
} VkImageMemoryRequirementsInfo2;

or the equivalent

typedef VkImageMemoryRequirementsInfo2 VkImageMemoryRequirementsInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • image is the image to query.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2

  • pNext must be NULL or a pointer to a valid instance of VkImagePlaneMemoryRequirementsInfo

  • image must be a valid VkImage handle

To determine the memory requirements for a plane of a disjoint image, add a VkImagePlaneMemoryRequirementsInfo to the pNext chain of the VkImageMemoryRequirementsInfo2 structure.

The VkImagePlaneMemoryRequirementsInfo structure is defined as:

typedef struct VkImagePlaneMemoryRequirementsInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkImageAspectFlagBits    planeAspect;
} VkImagePlaneMemoryRequirementsInfo;

or the equivalent

typedef VkImagePlaneMemoryRequirementsInfo VkImagePlaneMemoryRequirementsInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • planeAspect is the aspect corresponding to the image plane to query.

Valid Usage
  • planeAspect must be an aspect that exists in the format; that is, for a two-plane image planeAspect must be VK_IMAGE_ASPECT_PLANE_0_BIT or VK_IMAGE_ASPECT_PLANE_1_BIT, and for a three-plane image planeAspect must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT or VK_IMAGE_ASPECT_PLANE_2_BIT

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO

  • planeAspect must be a valid VkImageAspectFlagBits value

The VkMemoryRequirements2 structure is defined as:

typedef struct VkMemoryRequirements2 {
    VkStructureType         sType;
    void*                   pNext;
    VkMemoryRequirements    memoryRequirements;
} VkMemoryRequirements2;

or the equivalent

typedef VkMemoryRequirements2 VkMemoryRequirements2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memoryRequirements is a structure of type VkMemoryRequirements describing the memory requirements of the resource.

Valid Usage (Implicit)

To determine the dedicated allocation requirements of a buffer or image resource, add a VkMemoryDedicatedRequirements structure to the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter of vkGetBufferMemoryRequirements2 or vkGetImageMemoryRequirements2.

The VkMemoryDedicatedRequirements structure is defined as:

typedef struct VkMemoryDedicatedRequirements {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           prefersDedicatedAllocation;
    VkBool32           requiresDedicatedAllocation;
} VkMemoryDedicatedRequirements;

or the equivalent

typedef VkMemoryDedicatedRequirements VkMemoryDedicatedRequirementsKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • prefersDedicatedAllocation specifies that the implementation would prefer a dedicated allocation for this resource. The application is still free to suballocate the resource but it may get better performance if a dedicated allocation is used.

  • requiresDedicatedAllocation specifies that a dedicated allocation is required for this resource.

When the implementation sets requiresDedicatedAllocation to VK_TRUE, it must also set prefersDedicatedAllocation to VK_TRUE.

If the VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter of a vkGetBufferMemoryRequirements2 call, requiresDedicatedAllocation may be VK_TRUE under one of the following conditions:

  • The pNext chain of VkBufferCreateInfo for the call to vkCreateBuffer used to create the buffer being queried contained an instance of VkExternalMemoryBufferCreateInfo, and any of the handle types specified in VkExternalMemoryBufferCreateInfo::handleTypes requires dedicated allocation, as reported by vkGetPhysicalDeviceExternalBufferProperties in VkExternalBufferProperties::externalMemoryProperties::externalMemoryFeatures, the requiresDedicatedAllocation field will be set to VK_TRUE.

In all other cases, requiresDedicatedAllocation must be set to VK_FALSE by the implementation whenever a VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed to a call to vkGetBufferMemoryRequirements2.

If the VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter of a vkGetBufferMemoryRequirements2 call and VK_BUFFER_CREATE_SPARSE_BINDING_BIT was set in VkBufferCreateInfo::flags when buffer was created then the implementation must set both prefersDedicatedAllocation and requiresDedicatedAllocation to VK_FALSE.

If the VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter of a vkGetImageMemoryRequirements2 call, requiresDedicatedAllocation may be VK_TRUE under one of the following conditions:

  • The pNext chain of VkImageCreateInfo for the call to vkCreateImage used to create the image being queried contained an instance of VkExternalMemoryImageCreateInfo, and any of the handle types specified in VkExternalMemoryImageCreateInfo::handleTypes requires dedicated allocation, as reported by vkGetPhysicalDeviceImageFormatProperties2 in VkExternalImageFormatProperties::externalMemoryProperties::externalMemoryFeatures, the requiresDedicatedAllocation field will be set to VK_TRUE.

In all other cases, requiresDedicatedAllocation must be set to VK_FALSE by the implementation whenever a VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed to a call to vkGetImageMemoryRequirements2.

If the VkMemoryDedicatedRequirements structure is included in the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter of a vkGetImageMemoryRequirements2 call and VK_IMAGE_CREATE_SPARSE_BINDING_BIT was set in VkImageCreateInfo::flags when image was created then the implementation must set both prefersDedicatedAllocation and requiresDedicatedAllocation to VK_FALSE.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS

To attach memory to a buffer object, call:

VkResult vkBindBufferMemory(
    VkDevice                                    device,
    VkBuffer                                    buffer,
    VkDeviceMemory                              memory,
    VkDeviceSize                                memoryOffset);
  • device is the logical device that owns the buffer and memory.

  • buffer is the buffer to be attached to memory.

  • memory is a VkDeviceMemory object describing the device memory to attach.

  • memoryOffset is the start offset of the region of memory which is to be bound to the buffer. The number of bytes returned in the VkMemoryRequirements::size member in memory, starting from memoryOffset bytes, will be bound to the specified buffer.

vkBindBufferMemory is equivalent to passing the same parameters through VkBindBufferMemoryInfo to vkBindBufferMemory2.

Valid Usage
  • buffer must not already be backed by a memory object

  • buffer must not have been created with any sparse memory binding flags

  • memoryOffset must be less than the size of memory

  • If buffer was created with the VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minTexelBufferOffsetAlignment

  • If buffer was created with the VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minUniformBufferOffsetAlignment

  • If buffer was created with the VK_BUFFER_USAGE_STORAGE_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minStorageBufferOffsetAlignment

  • memory must have been allocated using one of the memory types allowed in the memoryTypeBits member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer

  • memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer

  • The size member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer must be less than or equal to the size of memory minus memoryOffset

  • If buffer requires a dedicated allocation(as reported by vkGetBufferMemoryRequirements2 in VkMemoryDedicatedRequirements::requiresDedicatedAllocation for buffer), memory must have been created with VkMemoryDedicatedAllocateInfo::buffer equal to buffer

  • If the VkMemoryAllocateInfo provided when memory was allocated included an instance of VkMemoryDedicatedAllocateInfo in its pNext chain, and VkMemoryDedicatedAllocateInfo::buffer was not VK_NULL_HANDLE, then buffer must equal VkMemoryDedicatedAllocateInfo::buffer, and memoryOffset must be zero.

  • If buffer was created with the VK_BUFFER_CREATE_PROTECTED_BIT bit set, the buffer must be bound to a memory object allocated with a memory type that reports VK_MEMORY_PROPERTY_PROTECTED_BIT

  • If buffer was created with the VK_BUFFER_CREATE_PROTECTED_BIT bit not set, the buffer must not be bound to a memory object created with a memory type that reports VK_MEMORY_PROPERTY_PROTECTED_BIT

  • If buffer was created with VkDedicatedAllocationBufferCreateInfoNV::dedicatedAllocation equal to VK_TRUE, memory must have been created with VkDedicatedAllocationMemoryAllocateInfoNV::buffer equal to a buffer handle created with identical creation parameters to buffer and memoryOffset must be zero

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • buffer must be a valid VkBuffer handle

  • memory must be a valid VkDeviceMemory handle

  • buffer must have been created, allocated, or retrieved from device

  • memory must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to buffer must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

To attach memory to buffer objects for one or more buffers at a time, call:

VkResult vkBindBufferMemory2(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindBufferMemoryInfo*               pBindInfos);

or the equivalent command

VkResult vkBindBufferMemory2KHR(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindBufferMemoryInfo*               pBindInfos);
  • device is the logical device that owns the buffers and memory.

  • bindInfoCount is the number of elements in pBindInfos.

  • pBindInfos is a pointer to an array of structures of type VkBindBufferMemoryInfo, describing buffers and memory to bind.

On some implementations, it may be more efficient to batch memory bindings into a single command.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pBindInfos must be a valid pointer to an array of bindInfoCount valid VkBindBufferMemoryInfo structures

  • bindInfoCount must be greater than 0

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

VkBindBufferMemoryInfo contains members corresponding to the parameters of vkBindBufferMemory.

The VkBindBufferMemoryInfo structure is defined as:

typedef struct VkBindBufferMemoryInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           buffer;
    VkDeviceMemory     memory;
    VkDeviceSize       memoryOffset;
} VkBindBufferMemoryInfo;

or the equivalent

typedef VkBindBufferMemoryInfo VkBindBufferMemoryInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • buffer is the buffer to be attached to memory.

  • memory is a VkDeviceMemory object describing the device memory to attach.

  • memoryOffset is the start offset of the region of memory which is to be bound to the buffer. The number of bytes returned in the VkMemoryRequirements::size member in memory, starting from memoryOffset bytes, will be bound to the specified buffer.

Valid Usage
  • buffer must not already be backed by a memory object

  • buffer must not have been created with any sparse memory binding flags

  • memoryOffset must be less than the size of memory

  • If buffer was created with the VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minTexelBufferOffsetAlignment

  • If buffer was created with the VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minUniformBufferOffsetAlignment

  • If buffer was created with the VK_BUFFER_USAGE_STORAGE_BUFFER_BIT, memoryOffset must be a multiple of VkPhysicalDeviceLimits::minStorageBufferOffsetAlignment

  • memory must have been allocated using one of the memory types allowed in the memoryTypeBits member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer

  • memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer

  • The size member of the VkMemoryRequirements structure returned from a call to vkGetBufferMemoryRequirements with buffer must be less than or equal to the size of memory minus memoryOffset

  • If buffer requires a dedicated allocation(as reported by vkGetBufferMemoryRequirements2 in VkMemoryDedicatedRequirements::requiresDedicatedAllocation for buffer), memory must have been created with VkMemoryDedicatedAllocateInfo::buffer equal to buffer and memoryOffset must be zero

  • If the VkMemoryAllocateInfo provided when memory was allocated included an instance of VkMemoryDedicatedAllocateInfo in its pNext chain, and VkMemoryDedicatedAllocateInfo::buffer was not VK_NULL_HANDLE, then buffer must equal VkMemoryDedicatedAllocateInfo::buffer and memoryOffset must be zero.

  • If buffer was created with VkDedicatedAllocationBufferCreateInfoNV::dedicatedAllocation equal to VK_TRUE, memory must have been created with VkDedicatedAllocationMemoryAllocateInfoNV::buffer equal to buffer and memoryOffset must be zero

  • If the pNext chain includes VkBindBufferMemoryDeviceGroupInfo, all instances of memory specified by VkBindBufferMemoryDeviceGroupInfo::pDeviceIndices must have been allocated

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO

  • pNext must be NULL or a pointer to a valid instance of VkBindBufferMemoryDeviceGroupInfo

  • buffer must be a valid VkBuffer handle

  • memory must be a valid VkDeviceMemory handle

  • Both of buffer, and memory must have been created, allocated, or retrieved from the same VkDevice

typedef struct VkBindBufferMemoryDeviceGroupInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceIndexCount;
    const uint32_t*    pDeviceIndices;
} VkBindBufferMemoryDeviceGroupInfo;

or the equivalent

typedef VkBindBufferMemoryDeviceGroupInfo VkBindBufferMemoryDeviceGroupInfoKHR;

If the pNext list of VkBindBufferMemoryInfo includes a VkBindBufferMemoryDeviceGroupInfo structure, then that structure determines how memory is bound to buffers across multiple devices in a device group.

The VkBindBufferMemoryDeviceGroupInfo structure is defined as:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • deviceIndexCount is the number of elements in pDeviceIndices.

  • pDeviceIndices is a pointer to an array of device indices.

If deviceIndexCount is greater than zero, then on device index i the buffer is attached to the instance of memory on the physical device with device index pDeviceIndices[i].

If deviceIndexCount is zero and memory comes from a memory heap with the VK_MEMORY_HEAP_MULTI_INSTANCE_BIT bit set, then it is as if pDeviceIndices contains consecutive indices from zero to the number of physical devices in the logical device, minus one. In other words, by default each physical device attaches to its own instance of memory.

If deviceIndexCount is zero and memory comes from a memory heap without the VK_MEMORY_HEAP_MULTI_INSTANCE_BIT bit set, then it is as if pDeviceIndices contains an array of zeros. In other words, by default each physical device attaches to instance zero.

Valid Usage
  • deviceIndexCount must either be zero or equal to the number of physical devices in the logical device

  • All elements of pDeviceIndices must be valid device indices

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO

  • If deviceIndexCount is not 0, pDeviceIndices must be a valid pointer to an array of deviceIndexCount uint32_t values

To attach memory to a VkImage object created without the VK_IMAGE_CREATE_DISJOINT_BIT set, call:

VkResult vkBindImageMemory(
    VkDevice                                    device,
    VkImage                                     image,
    VkDeviceMemory                              memory,
    VkDeviceSize                                memoryOffset);
  • device is the logical device that owns the image and memory.

  • image is the image.

  • memory is the VkDeviceMemory object describing the device memory to attach.

  • memoryOffset is the start offset of the region of memory which is to be bound to the image. The number of bytes returned in the VkMemoryRequirements::size member in memory, starting from memoryOffset bytes, will be bound to the specified image.

vkBindImageMemory is equivalent to passing the same parameters through VkBindImageMemoryInfo to vkBindImageMemory2.

Valid Usage
  • image must not have been created with the VK_IMAGE_CREATE_DISJOINT_BIT set.

  • image must not already be backed by a memory object

  • image must not have been created with any sparse memory binding flags

  • memoryOffset must be less than the size of memory

  • memory must have been allocated using one of the memory types allowed in the memoryTypeBits member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements with image

  • memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements with image

  • The size member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements with image must be less than or equal to the size of memory minus memoryOffset

  • If image requires a dedicated allocation (as reported by vkGetImageMemoryRequirements2 in VkMemoryDedicatedRequirements::requiresDedicatedAllocation for image), memory must have been created with VkMemoryDedicatedAllocateInfo::image equal to image

  • If the VkMemoryAllocateInfo provided when memory was allocated included an instance of VkMemoryDedicatedAllocateInfo in its pNext chain, and VkMemoryDedicatedAllocateInfo::image was not VK_NULL_HANDLE, then image must equal VkMemoryDedicatedAllocateInfo::image and memoryOffset must be zero.

  • If image was created with the VK_IMAGE_CREATE_PROTECTED_BIT bit set, the image must be bound to a memory object allocated with a memory type that reports VK_MEMORY_PROPERTY_PROTECTED_BIT

  • If image was created with the VK_IMAGE_CREATE_PROTECTED_BIT bit not set, the image must not be bound to a memory object created with a memory type that reports VK_MEMORY_PROPERTY_PROTECTED_BIT

  • If image was created with VkDedicatedAllocationImageCreateInfoNV::dedicatedAllocation equal to VK_TRUE, memory must have been created with VkDedicatedAllocationMemoryAllocateInfoNV::image equal to an image handle created with identical creation parameters to image and memoryOffset must be zero

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • image must be a valid VkImage handle

  • memory must be a valid VkDeviceMemory handle

  • image must have been created, allocated, or retrieved from device

  • memory must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to image must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

To attach memory to image objects for one or more images at a time, call:

VkResult vkBindImageMemory2(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindImageMemoryInfo*                pBindInfos);

or the equivalent command

VkResult vkBindImageMemory2KHR(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindImageMemoryInfo*                pBindInfos);
  • device is the logical device that owns the images and memory.

  • bindInfoCount is the number of elements in pBindInfos.

  • pBindInfos is a pointer to an array of structures of type VkBindImageMemoryInfo, describing images and memory to bind.

On some implementations, it may be more efficient to batch memory bindings into a single command.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pBindInfos must be a valid pointer to an array of bindInfoCount valid VkBindImageMemoryInfo structures

  • bindInfoCount must be greater than 0

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

VkBindImageMemoryInfo contains members corresponding to the parameters of vkBindImageMemory.

The VkBindImageMemoryInfo structure is defined as:

typedef struct VkBindImageMemoryInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
    VkDeviceMemory     memory;
    VkDeviceSize       memoryOffset;
} VkBindImageMemoryInfo;

or the equivalent

typedef VkBindImageMemoryInfo VkBindImageMemoryInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • image is the image to be attached to memory.

  • memory is a VkDeviceMemory object describing the device memory to attach.

  • memoryOffset is the start offset of the region of memory which is to be bound to the image. The number of bytes returned in the VkMemoryRequirements::size member in memory, starting from memoryOffset bytes, will be bound to the specified image.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkBindImageMemoryDeviceGroupInfo, VkBindImageMemorySwapchainInfoKHR, or VkBindImagePlaneMemoryInfo

  • Each sType member in the pNext chain must be unique

  • image must be a valid VkImage handle

  • Both of image, and memory that are valid handles must have been created, allocated, or retrieved from the same VkDevice

typedef struct VkBindImageMemoryDeviceGroupInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceIndexCount;
    const uint32_t*    pDeviceIndices;
    uint32_t           splitInstanceBindRegionCount;
    const VkRect2D*    pSplitInstanceBindRegions;
} VkBindImageMemoryDeviceGroupInfo;

or the equivalent

typedef VkBindImageMemoryDeviceGroupInfo VkBindImageMemoryDeviceGroupInfoKHR;

If the pNext list of VkBindImageMemoryInfo includes a VkBindImageMemoryDeviceGroupInfo structure, then that structure determines how memory is bound to images across multiple devices in a device group.

The VkBindImageMemoryDeviceGroupInfo structure is defined as:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • deviceIndexCount is the number of elements in pDeviceIndices.

  • pDeviceIndices is a pointer to an array of device indices.

  • splitInstanceBindRegionCount is the number of elements in pSplitInstanceBindRegions.

  • pSplitInstanceBindRegions is a pointer to an array of rectangles describing which regions of the image are attached to each instance of memory.

If deviceIndexCount is greater than zero, then on device index i image is attached to the instance of the memory on the physical device with device index pDeviceIndices[i].

Let N be the number of physical devices in the logical device. If splitInstanceBindRegionCount is greater than zero, then pSplitInstanceBindRegions is an array of N2 rectangles, where the image region specified by the rectangle at element i*N+j in resource instance i is bound to the memory instance j. The blocks of the memory that are bound to each sparse image block region use an offset in memory, relative to memoryOffset, computed as if the whole image were being bound to a contiguous range of memory. In other words, horizontally adjacent image blocks use consecutive blocks of memory, vertically adjacent image blocks are separated by the number of bytes per block multiplied by the width in blocks of image, and the block at (0,0) corresponds to memory starting at memoryOffset.

If splitInstanceBindRegionCount and deviceIndexCount are zero and the memory comes from a memory heap with the VK_MEMORY_HEAP_MULTI_INSTANCE_BIT bit set, then it is as if pDeviceIndices contains consecutive indices from zero to the number of physical devices in the logical device, minus one. In other words, by default each physical device attaches to its own instance of the memory.

If splitInstanceBindRegionCount and deviceIndexCount are zero and the memory comes from a memory heap without the VK_MEMORY_HEAP_MULTI_INSTANCE_BIT bit set, then it is as if pDeviceIndices contains an array of zeros. In other words, by default each physical device attaches to instance zero.

Valid Usage
  • At least one of deviceIndexCount and splitInstanceBindRegionCount must be zero.

  • deviceIndexCount must either be zero or equal to the number of physical devices in the logical device

  • All elements of pDeviceIndices must be valid device indices.

  • splitInstanceBindRegionCount must either be zero or equal to the number of physical devices in the logical device squared

  • Elements of pSplitInstanceBindRegions that correspond to the same instance of an image must not overlap.

  • The offset.x member of any element of pSplitInstanceBindRegions must be a multiple of the sparse image block width (VkSparseImageFormatProperties::imageGranularity.width) of all non-metadata aspects of the image

  • The offset.y member of any element of pSplitInstanceBindRegions must be a multiple of the sparse image block height (VkSparseImageFormatProperties::imageGranularity.height) of all non-metadata aspects of the image

  • The extent.width member of any element of pSplitInstanceBindRegions must either be a multiple of the sparse image block width of all non-metadata aspects of the image, or else extent.width + offset.x must equal the width of the image subresource

  • The extent.height member of any element of pSplitInstanceBindRegions must either be a multiple of the sparse image block height of all non-metadata aspects of the image, or else extent.height
    offset.y must equal the width of the image subresource

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO

  • If deviceIndexCount is not 0, pDeviceIndices must be a valid pointer to an array of deviceIndexCount uint32_t values

  • If splitInstanceBindRegionCount is not 0, pSplitInstanceBindRegions must be a valid pointer to an array of splitInstanceBindRegionCount VkRect2D structures

If the pNext chain of VkBindImageMemoryInfo includes a VkBindImageMemorySwapchainInfoKHR structure, then that structure includes a swapchain handle and image index indicating that the image will be bound to memory from that swapchain.

The VkBindImageMemorySwapchainInfoKHR structure is defined as:

typedef struct VkBindImageMemorySwapchainInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkSwapchainKHR     swapchain;
    uint32_t           imageIndex;
} VkBindImageMemorySwapchainInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchain is VK_NULL_HANDLE or a swapchain handle.

  • imageIndex is an image index within swapchain.

If swapchain is not NULL, the swapchain and imageIndex are used to determine the memory that the image is bound to, instead of memory and memoryOffset.

Memory can be bound to a swapchain and use the pDeviceIndices or pSplitInstanceBindRegions members of VkBindImageMemoryDeviceGroupInfo.

Valid Usage
  • imageIndex must be less than the number of images in swapchain

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_SWAPCHAIN_INFO_KHR

  • swapchain must be a valid VkSwapchainKHR handle

Host Synchronization
  • Host access to swapchain must be externally synchronized

In order to bind planes of a disjoint image, include a VkBindImagePlaneMemoryInfo structure in the pNext chain of VkBindImageMemoryInfo.

The VkBindImagePlaneMemoryInfo structure is defined as:

typedef struct VkBindImagePlaneMemoryInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkImageAspectFlagBits    planeAspect;
} VkBindImagePlaneMemoryInfo;

or the equivalent

typedef VkBindImagePlaneMemoryInfo VkBindImagePlaneMemoryInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • planeAspect is the aspect of the disjoint image plane to bind.

Valid Usage
  • planeAspect must be a single valid plane aspect for the image format (that is, planeAspect must be VK_IMAGE_ASPECT_PLANE_0_BIT or VK_IMAGE_ASPECT_PLANE_1_BIT for “_2PLANE” formats and planeAspect must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT for “_3PLANE” formats)

  • A single call to vkBindImageMemory2 must bind all or none of the planes of an image (i.e. bindings to all planes of an image must be made in a single vkBindImageMemory2 call), as separate bindings

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO

  • planeAspect must be a valid VkImageAspectFlagBits value

Buffer-Image Granularity

There is an implementation-dependent limit, bufferImageGranularity, which specifies a page-like granularity at which linear and non-linear resources must be placed in adjacent memory locations to avoid aliasing. Two resources which do not satisfy this granularity requirement are said to alias. bufferImageGranularity is specified in bytes, and must be a power of two. Implementations which do not impose a granularity restriction may report a bufferImageGranularity value of one.

Note

Despite its name, bufferImageGranularity is really a granularity between “linear” and “non-linear” resources.

Given resourceA at the lower memory offset and resourceB at the higher memory offset in the same VkDeviceMemory object, where one resource is linear and the other is non-linear (as defined in the Glossary), and the following:

resourceA.end       = resourceA.memoryOffset + resourceA.size - 1
resourceA.endPage   = resourceA.end & ~(bufferImageGranularity-1)
resourceB.start     = resourceB.memoryOffset
resourceB.startPage = resourceB.start & ~(bufferImageGranularity-1)

The following property must hold:

resourceA.endPage < resourceB.startPage

That is, the end of the first resource (A) and the beginning of the second resource (B) must be on separate “pages” of size bufferImageGranularity. bufferImageGranularity may be different than the physical page size of the memory heap. This restriction is only needed when a linear resource and a non-linear resource are adjacent in memory and will be used simultaneously. The memory ranges of adjacent resources can be closer than bufferImageGranularity, provided they meet the alignment requirement for the objects in question.

Sparse block size in bytes and sparse image and buffer memory alignments must all be multiples of the bufferImageGranularity. Therefore, memory bound to sparse resources naturally satisfies the bufferImageGranularity.

11.7. Resource Sharing Mode

Buffer and image objects are created with a sharing mode controlling how they can be accessed from queues. The supported sharing modes are:

typedef enum VkSharingMode {
    VK_SHARING_MODE_EXCLUSIVE = 0,
    VK_SHARING_MODE_CONCURRENT = 1,
} VkSharingMode;
  • VK_SHARING_MODE_EXCLUSIVE specifies that access to any range or image subresource of the object will be exclusive to a single queue family at a time.

  • VK_SHARING_MODE_CONCURRENT specifies that concurrent access to any range or image subresource of the object from multiple queue families is supported.

Note

VK_SHARING_MODE_CONCURRENT may result in lower performance access to the buffer or image than VK_SHARING_MODE_EXCLUSIVE.

Ranges of buffers and image subresources of image objects created using VK_SHARING_MODE_EXCLUSIVE must only be accessed by queues in the queue family that has ownership of the resource. Upon creation, such resources are not owned by any queue family; ownership is implicitly acquired upon first use within a queue. Once a resource using VK_SHARING_MODE_EXCLUSIVE is owned by some queue family, the application must perform a queue family ownership transfer to make the memory contents of a range or image subresource accessible to a different queue family.

Note

Images still require a layout transition from VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED before being used on the first queue.

A queue family can take ownership of an image subresource or buffer range of a resource created with VK_SHARING_MODE_EXCLUSIVE, without an ownership transfer, in the same way as for a resource that was just created; however, taking ownership in this way has the effect that the contents of the image subresource or buffer range are undefined.

Ranges of buffers and image subresources of image objects created using VK_SHARING_MODE_CONCURRENT must only be accessed by queues from the queue families specified through the queueFamilyIndexCount and pQueueFamilyIndices members of the corresponding create info structures.

11.7.1. External Resource Sharing

Resources should only be accessed in the Vulkan instance that has exclusive ownership of their underlying memory. Only one Vulkan instance has exclusive ownership of a resource’s underlying memory at a given time, regardless of whether the resource was created using VK_SHARING_MODE_EXCLUSIVE or VK_SHARING_MODE_CONCURRENT. Applications can transfer ownership of a resource’s underlying memory only if the memory has been imported from or exported to another instance or external API using external memory handles. The semantics for transferring ownership outside of the instance are similar to those used for transferring ownership of VK_SHARING_MODE_EXCLUSIVE resources between queues, and is also accomplished using VkBufferMemoryBarrier or VkImageMemoryBarrier operations. Applications must

  1. Release exclusive ownership from the source instance or API.

  2. Ensure the release operation has completed using semaphores or fences.

  3. Acquire exclusive ownership in the destination instance or API

Unlike queue ownership transfers, the destination instance or API is not specified explicitly when releasing ownership, nor is the source instance or API specified when acquiring ownership. Instead, the image or memory barrier’s dstQueueFamilyIndex or srcQueueFamilyIndex parameters are set to the reserved queue family index VK_QUEUE_FAMILY_EXTERNAL or VK_QUEUE_FAMILY_FOREIGN_EXT to represent the external destination or source respectively.

Binding a resource to a memory object shared between multiple Vulkan instances or other APIs does not change the ownership of the underlying memory. The first entity to access the resource implicitly acquires ownership. Accessing a resource backed by memory that is owned by a particular instance or API has the same semantics as accessing a VK_SHARING_MODE_EXCLUSIVE resource, with one exception: Implementations must ensure layout transitions performed on one member of a set of identical subresources of identical images that alias the same range of an underlying memory object affect the layout of all the subresources in the set.

As a corollary, writes to any image subresources in such a set must not make the contents of memory used by other subresources in the set undefined. An application can define the content of a subresource of one image by performing device writes to an identical subresource of another image provided both images are bound to the same region of external memory. Applications may also add resources to such a set after the content of the existing set members has been defined without making the content undefined by creating a new image with the initial layout VK_IMAGE_LAYOUT_UNDEFINED and binding it to the same region of external memory as the existing images.

Note

Because layout transitions apply to all identical images aliasing the same region of external memory, the actual layout of the memory backing a new image as well as an existing image with defined content will not be undefined. Such an image is not usable until it acquires ownership of its memory from the existing owner. Therefore, the layout specified as part of this transition will be the true initial layout of the image. The undefined layout specified when creating it is a placeholder to simplify valid usage requirements.

11.8. Memory Aliasing

A range of a VkDeviceMemory allocation is aliased if it is bound to multiple resources simultaneously, as described below, via vkBindImageMemory, vkBindBufferMemory, via sparse memory bindings, or by binding the memory to resources in multiple Vulkan instances or external APIs using external memory handle export and import mechanisms.

Consider two resources, resourceA and resourceB, bound respectively to memory rangeA and rangeB. Let paddedRangeA and paddedRangeB be, respectively, rangeA and rangeB aligned to bufferImageGranularity. If the resources are both linear or both non-linear (as defined in the Glossary), then the resources alias the memory in the intersection of rangeA and rangeB. If one resource is linear and the other is non-linear, then the resources alias the memory in the intersection of paddedRangeA and paddedRangeB.

Applications can alias memory, but use of multiple aliases is subject to several constraints.

Note

Memory aliasing can be useful to reduce the total device memory footprint of an application, if some large resources are used for disjoint periods of time.

When an opaque, non-VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT image is bound to an aliased range, all image subresources of the image overlap the range. When a linear image is bound to an aliased range, the image subresources that (according to the image’s advertised layout) include bytes from the aliased range overlap the range. When a VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT image has sparse image blocks bound to an aliased range, only image subresources including those sparse image blocks overlap the range, and when the memory bound to the image’s mip tail overlaps an aliased range all image subresources in the mip tail overlap the range.

Buffers, and linear image subresources in either the VK_IMAGE_LAYOUT_PREINITIALIZED or VK_IMAGE_LAYOUT_GENERAL layouts, are host-accessible subresources. That is, the host has a well-defined addressing scheme to interpret the contents, and thus the layout of the data in memory can be consistently interpreted across aliases if each of those aliases is a host-accessible subresource. Non-linear images, and linear image subresources in other layouts, are not host-accessible.

If two aliases are both host-accessible, then they interpret the contents of the memory in consistent ways, and data written to one alias can be read by the other alias.

If two aliases are both images that were created with identical creation parameters, both were created with the VK_IMAGE_CREATE_ALIAS_BIT flag set, and both are bound identically to memory except for VkBindImageMemoryDeviceGroupInfo::pDeviceIndices and VkBindImageMemoryDeviceGroupInfo::pSplitInstanceBindRegions, then they interpret the contents of the memory in consistent ways, and data written to one alias can be read by the other alias.

Additionally, if an invididual plane of a multi-planar image and a single-plane image alias the same memory, then they also interpret the contents of the memory in consistent ways under the same conditions, but with the following modifications:

  • Both must have been created with the VK_IMAGE_CREATE_DISJOINT_BIT flag.

  • The single-plane image must have an VkFormat that is equivalent to that of the multi-planar image’s individual plane.

  • The single-plane image and the individual plane of the multi-planar image must be bound identically to memory except for VkBindImageMemoryDeviceGroupInfo::pDeviceIndices and VkBindImageMemoryDeviceGroupInfo::pSplitInstanceBindRegions.

  • The width and height of the single-plane image are derived from the multi-planar image’s dimensions in the manner listed for plane compatibility for the aliased plane.

  • All other creation parameters must be identical

Aliases created by binding the same memory to resources in multiple Vulkan instances or external APIs using external memory handle export and import mechanisms interpret the contents of the memory in consistent ways, and data written to one alias can be read by the other alias.

Otherwise, the aliases interpret the contents of the memory differently, and writes via one alias make the contents of memory partially or completely undefined to the other alias. If the first alias is a host-accessible subresource, then the bytes affected are those written by the memory operations according to its addressing scheme. If the first alias is not host-accessible, then the bytes affected are those overlapped by the image subresources that were written. If the second alias is a host-accessible subresource, the affected bytes become undefined. If the second alias is a not host-accessible, all sparse image blocks (for sparse partially-resident images) or all image subresources (for non-sparse image and fully resident sparse images) that overlap the affected bytes become undefined.

If any image subresources are made undefined due to writes to an alias, then each of those image subresources must have its layout transitioned from VK_IMAGE_LAYOUT_UNDEFINED to a valid layout before it is used, or from VK_IMAGE_LAYOUT_PREINITIALIZED if the memory has been written by the host. If any sparse blocks of a sparse image have been made undefined, then only the image subresources containing them must be transitioned.

Use of an overlapping range by two aliases must be separated by a memory dependency using the appropriate access types if at least one of those uses performs writes, whether the aliases interpret memory consistently or not. If buffer or image memory barriers are used, the scope of the barrier must contain the entire range and/or set of image subresources that overlap.

If two aliasing image views are used in the same framebuffer, then the render pass must declare the attachments using the VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT, and follow the other rules listed in that section.

Access to resources which alias memory from shaders using variables decorated with Coherent are not automatically coherent with each other.

Note

Memory recycled via an application suballocator (i.e. without freeing and reallocating the memory objects) is not substantially different from memory aliasing. However, a suballocator usually waits on a fence before recycling a region of memory, and signaling a fence involves sufficient implicit dependencies to satisfy all the above requirements.

12. Samplers

VkSampler objects represent the state of an image sampler which is used by the implementation to read image data and apply filtering and other transformations for the shader.

Samplers are represented by VkSampler handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSampler)

To create a sampler object, call:

VkResult vkCreateSampler(
    VkDevice                                    device,
    const VkSamplerCreateInfo*                  pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSampler*                                  pSampler);
  • device is the logical device that creates the sampler.

  • pCreateInfo is a pointer to an instance of the VkSamplerCreateInfo structure specifying the state of the sampler object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pSampler points to a VkSampler handle in which the resulting sampler object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkSamplerCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSampler must be a valid pointer to a VkSampler handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_TOO_MANY_OBJECTS

The VkSamplerCreateInfo structure is defined as:

typedef struct VkSamplerCreateInfo {
    VkStructureType         sType;
    const void*             pNext;
    VkSamplerCreateFlags    flags;
    VkFilter                magFilter;
    VkFilter                minFilter;
    VkSamplerMipmapMode     mipmapMode;
    VkSamplerAddressMode    addressModeU;
    VkSamplerAddressMode    addressModeV;
    VkSamplerAddressMode    addressModeW;
    float                   mipLodBias;
    VkBool32                anisotropyEnable;
    float                   maxAnisotropy;
    VkBool32                compareEnable;
    VkCompareOp             compareOp;
    float                   minLod;
    float                   maxLod;
    VkBorderColor           borderColor;
    VkBool32                unnormalizedCoordinates;
} VkSamplerCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • magFilter is a VkFilter value specifying the magnification filter to apply to lookups.

  • minFilter is a VkFilter value specifying the minification filter to apply to lookups.

  • mipmapMode is a VkSamplerMipmapMode value specifying the mipmap filter to apply to lookups.

  • addressModeU is a VkSamplerAddressMode value specifying the addressing mode for outside [0..1] range for U coordinate.

  • addressModeV is a VkSamplerAddressMode value specifying the addressing mode for outside [0..1] range for V coordinate.

  • addressModeW is a VkSamplerAddressMode value specifying the addressing mode for outside [0..1] range for W coordinate.

  • mipLodBias is the bias to be added to mipmap LOD (level-of-detail) calculation and bias provided by image sampling functions in SPIR-V, as described in the Level-of-Detail Operation section.

  • anisotropyEnable is VK_TRUE to enable anisotropic filtering, as described in the Texel Anisotropic Filtering section, or VK_FALSE otherwise.

  • maxAnisotropy is the anisotropy value clamp used by the sampler when anisotropyEnable is VK_TRUE. If anisotropyEnable is VK_FALSE, maxAnisotropy is ignored.

  • compareEnable is VK_TRUE to enable comparison against a reference value during lookups, or VK_FALSE otherwise.

    • Note: Some implementations will default to shader state if this member does not match.

  • compareOp is a VkCompareOp value specifying the comparison function to apply to fetched data before filtering as described in the Depth Compare Operation section.

  • minLod and maxLod are the values used to clamp the computed LOD value, as described in the Level-of-Detail Operation section. maxLod must be greater than or equal to minLod.

  • borderColor is a VkBorderColor value specifying the predefined border color to use.

  • unnormalizedCoordinates controls whether to use unnormalized or normalized texel coordinates to address texels of the image. When set to VK_TRUE, the range of the image coordinates used to lookup the texel is in the range of zero to the image dimensions for x, y and z. When set to VK_FALSE the range of image coordinates is zero to one. When unnormalizedCoordinates is VK_TRUE, samplers have the following requirements:

    • minFilter and magFilter must be equal.

    • mipmapMode must be VK_SAMPLER_MIPMAP_MODE_NEAREST.

    • minLod and maxLod must be zero.

    • addressModeU and addressModeV must each be either VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE or VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER.

    • anisotropyEnable must be VK_FALSE.

    • compareEnable must be VK_FALSE.

    • The sampler must not enable sampler Y’CBCR conversion.

  • When unnormalizedCoordinates is VK_TRUE, images the sampler is used with in the shader have the following requirements:

    • The viewType must be either VK_IMAGE_VIEW_TYPE_1D or VK_IMAGE_VIEW_TYPE_2D.

    • The image view must have a single layer and a single mip level.

  • When unnormalizedCoordinates is VK_TRUE, image built-in functions in the shader that use the sampler have the following requirements:

    • The functions must not use projection.

    • The functions must not use offsets.

Mapping of OpenGL to Vulkan filter modes

magFilter values of VK_FILTER_NEAREST and VK_FILTER_LINEAR directly correspond to GL_NEAREST and GL_LINEAR magnification filters. minFilter and mipmapMode combine to correspond to the similarly named OpenGL minification filter of GL_minFilter_MIPMAP_mipmapMode (e.g. minFilter of VK_FILTER_LINEAR and mipmapMode of VK_SAMPLER_MIPMAP_MODE_NEAREST correspond to GL_LINEAR_MIPMAP_NEAREST).

There are no Vulkan filter modes that directly correspond to OpenGL minification filters of GL_LINEAR or GL_NEAREST, but they can be emulated using VK_SAMPLER_MIPMAP_MODE_NEAREST, minLod = 0, and maxLod = 0.25, and using minFilter = VK_FILTER_LINEAR or minFilter = VK_FILTER_NEAREST, respectively.

Note that using a maxLod of zero would cause magnification to always be performed, and the magFilter to always be used. This is valid, just not an exact match for OpenGL behavior. Clamping the maximum LOD to 0.25 allows the λ value to be non-zero and minification to be performed, while still always rounding down to the base level. If the minFilter and magFilter are equal, then using a maxLod of zero also works.

The maximum number of sampler objects which can be simultaneously created on a device is implementation-dependent and specified by the maxSamplerAllocationCount member of the VkPhysicalDeviceLimits structure. If maxSamplerAllocationCount is exceeded, vkCreateSampler will return VK_ERROR_TOO_MANY_OBJECTS.

Since VkSampler is a non-dispatchable handle type, implementations may return the same handle for sampler state vectors that are identical. In such cases, all such objects would only count once against the maxSamplerAllocationCount limit.

Valid Usage
  • The absolute value of mipLodBias must be less than or equal to VkPhysicalDeviceLimits::maxSamplerLodBias

  • If the anisotropic sampling feature is not enabled, anisotropyEnable must be VK_FALSE

  • If anisotropyEnable is VK_TRUE, maxAnisotropy must be between 1.0 and VkPhysicalDeviceLimits::maxSamplerAnisotropy, inclusive

  • If sampler Y’CBCR conversion is enabled and VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT is not set for the format, minFilter and magFilter must be equal to the sampler Y’CBCR conversion’s chromaFilter

  • If unnormalizedCoordinates is VK_TRUE, minFilter and magFilter must be equal

  • If unnormalizedCoordinates is VK_TRUE, mipmapMode must be VK_SAMPLER_MIPMAP_MODE_NEAREST

  • If unnormalizedCoordinates is VK_TRUE, minLod and maxLod must be zero

  • If unnormalizedCoordinates is VK_TRUE, addressModeU and addressModeV must each be either VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE or VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER

  • If unnormalizedCoordinates is VK_TRUE, anisotropyEnable must be VK_FALSE

  • If unnormalizedCoordinates is VK_TRUE, compareEnable must be VK_FALSE

  • If any of addressModeU, addressModeV or addressModeW are VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER, borderColor must be a valid VkBorderColor value

  • If sampler Y’CBCR conversion is enabled, addressModeU, addressModeV, and addressModeW must be VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE, anisotropyEnable must be VK_FALSE, and unnormalizedCoordinates must be VK_FALSE

  • The sampler reduction mode must be set to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT if sampler Y’CBCR conversion is enabled

  • If the VK_KHR_sampler_mirror_clamp_to_edge extension is not enabled, addressModeU, addressModeV and addressModeW must not be VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE

  • If compareEnable is VK_TRUE, compareOp must be a valid VkCompareOp value

  • If either magFilter or minFilter is VK_FILTER_CUBIC_IMG, anisotropyEnable must be VK_FALSE

  • If either magFilter or minFilter is VK_FILTER_CUBIC_IMG, the reductionMode member of VkSamplerReductionModeCreateInfoEXT must be VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT

  • If compareEnable is VK_TRUE, the reductionMode member of VkSamplerReductionModeCreateInfoEXT must be VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT

Valid Usage (Implicit)
typedef VkFlags VkSamplerCreateFlags;

VkSamplerCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

If the pNext chain of VkSamplerCreateInfo includes a VkSamplerReductionModeCreateInfoEXT structure, then that structure includes a mode that controls how texture filtering combines texel values.

The VkSamplerReductionModeCreateInfoEXT structure is defined as:

typedef struct VkSamplerReductionModeCreateInfoEXT {
    VkStructureType              sType;
    const void*                  pNext;
    VkSamplerReductionModeEXT    reductionMode;
} VkSamplerReductionModeCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • reductionMode is an enum of type VkSamplerReductionModeEXT that controls how texture filtering combines texel values.

If this structure is not present, reductionMode is considered to be VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SAMPLER_REDUCTION_MODE_CREATE_INFO_EXT

  • reductionMode must be a valid VkSamplerReductionModeEXT value

Reduction modes are specified by VkSamplerReductionModeEXT, which takes values:

typedef enum VkSamplerReductionModeEXT {
    VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT = 0,
    VK_SAMPLER_REDUCTION_MODE_MIN_EXT = 1,
    VK_SAMPLER_REDUCTION_MODE_MAX_EXT = 2,
} VkSamplerReductionModeEXT;
  • VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT specifies that texel values are combined by computing a weighted average of values in the footprint, using weights as specified in the image operations chapter.

  • VK_SAMPLER_REDUCTION_MODE_MIN_EXT specifies that texel values are combined by taking the component-wise minimum of values in the footprint with non-zero weights.

  • VK_SAMPLER_REDUCTION_MODE_MAX_EXT specifies that texel values are combined by taking the component-wise maximum of values in the footprint with non-zero weights.

Possible values of the VkSamplerCreateInfo::magFilter and minFilter parameters, specifying filters used for texture lookups, are:

typedef enum VkFilter {
    VK_FILTER_NEAREST = 0,
    VK_FILTER_LINEAR = 1,
    VK_FILTER_CUBIC_IMG = 1000015000,
} VkFilter;
  • VK_FILTER_NEAREST specifies nearest filtering.

  • VK_FILTER_LINEAR specifies linear filtering.

  • VK_FILTER_CUBIC_IMG specifies cubic filtering.

These filters are described in detail in Texel Filtering.

Possible values of the VkSamplerCreateInfo::mipmapMode, specifying the mipmap mode used for texture lookups, are:

typedef enum VkSamplerMipmapMode {
    VK_SAMPLER_MIPMAP_MODE_NEAREST = 0,
    VK_SAMPLER_MIPMAP_MODE_LINEAR = 1,
} VkSamplerMipmapMode;
  • VK_SAMPLER_MIPMAP_MODE_NEAREST specifies nearest filtering.

  • VK_SAMPLER_MIPMAP_MODE_LINEAR specifies linear filtering.

These modes are described in detail in Texel Filtering.

Possible values of the VkSamplerCreateInfo::addressMode* parameters, specifying the behavior of sampling with coordinates outside the range [0,1] for the respective u, v, or w coordinate as defined in the Wrapping Operation section, are:

typedef enum VkSamplerAddressMode {
    VK_SAMPLER_ADDRESS_MODE_REPEAT = 0,
    VK_SAMPLER_ADDRESS_MODE_MIRRORED_REPEAT = 1,
    VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE = 2,
    VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER = 3,
    VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE = 4,
} VkSamplerAddressMode;
  • VK_SAMPLER_ADDRESS_MODE_REPEAT specifies that the repeat wrap mode will be used.

  • VK_SAMPLER_ADDRESS_MODE_MIRRORED_REPEAT specifies that the mirrored repeat wrap mode will be used.

  • VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE specifies that the clamp to edge wrap mode will be used.

  • VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER specifies that the clamp to border wrap mode will be used.

  • VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE specifies that the mirror clamp to edge wrap mode will be used. This is only valid if the VK_KHR_sampler_mirror_clamp_to_edge extension is enabled.

Possible values of VkSamplerCreateInfo::borderColor, specifying the border color used for texture lookups, are:

typedef enum VkBorderColor {
    VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK = 0,
    VK_BORDER_COLOR_INT_TRANSPARENT_BLACK = 1,
    VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK = 2,
    VK_BORDER_COLOR_INT_OPAQUE_BLACK = 3,
    VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE = 4,
    VK_BORDER_COLOR_INT_OPAQUE_WHITE = 5,
} VkBorderColor;
  • VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK specifies a transparent, floating-point format, black color.

  • VK_BORDER_COLOR_INT_TRANSPARENT_BLACK specifies a transparent, integer format, black color.

  • VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK specifies an opaque, floating-point format, black color.

  • VK_BORDER_COLOR_INT_OPAQUE_BLACK specifies an opaque, integer format, black color.

  • VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE specifies an opaque, floating-point format, white color.

  • VK_BORDER_COLOR_INT_OPAQUE_WHITE specifies an opaque, integer format, white color.

These colors are described in detail in Texel Replacement.

To destroy a sampler, call:

void vkDestroySampler(
    VkDevice                                    device,
    VkSampler                                   sampler,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the sampler.

  • sampler is the sampler to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to sampler must have completed execution

  • If VkAllocationCallbacks were provided when sampler was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when sampler was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If sampler is not VK_NULL_HANDLE, sampler must be a valid VkSampler handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If sampler is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to sampler must be externally synchronized

12.1. Sampler Y’CBCR conversion

To create a sampler with Y’CBCR conversion enabled, add a VkSamplerYcbcrConversionInfo to the pNext chain of the VkSamplerCreateInfo structure. To create a sampler Y’CBCR conversion, the samplerYcbcrConversion feature must be enabled. Conversion must be fixed at pipeline creation time, through use of a combined image sampler with an immutable sampler in VkDescriptorSetLayoutBinding.

A VkSamplerYcbcrConversionInfo must be provided for samplers to be used with image views that access VK_IMAGE_ASPECT_COLOR_BIT if the format appears in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views , or if the image view has an external format .

The VkSamplerYcbcrConversionInfo structure is defined as:

typedef struct VkSamplerYcbcrConversionInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkSamplerYcbcrConversion    conversion;
} VkSamplerYcbcrConversionInfo;

or the equivalent

typedef VkSamplerYcbcrConversionInfo VkSamplerYcbcrConversionInfoKHR;
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO

  • conversion must be a valid VkSamplerYcbcrConversion handle

A sampler Y’CBCR conversion is an opaque representation of a device-specific sampler Y’CBCR conversion description, represented as a VkSamplerYcbcrConversion handle:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSamplerYcbcrConversion)

or the equivalent

typedef VkSamplerYcbcrConversion VkSamplerYcbcrConversionKHR;

To create a VkSamplerYcbcrConversion, call:

VkResult vkCreateSamplerYcbcrConversion(
    VkDevice                                    device,
    const VkSamplerYcbcrConversionCreateInfo*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSamplerYcbcrConversion*                   pYcbcrConversion);

or the equivalent command

VkResult vkCreateSamplerYcbcrConversionKHR(
    VkDevice                                    device,
    const VkSamplerYcbcrConversionCreateInfo*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSamplerYcbcrConversion*                   pYcbcrConversion);
  • device is the logical device that creates the sampler Y’CBCR conversion.

  • pCreateInfo is a pointer to an instance of the VkSamplerYcbcrConversionCreateInfo specifying the requested sampler Y’CBCR conversion.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pYcbcrConversion points to a VkSamplerYcbcrConversion handle in which the resulting sampler Y’CBCR conversion is returned.

The interpretation of the configured sampler Y’CBCR conversion is described in more detail in the description of sampler Y’CBCR conversion in the Image Operations chapter.

Valid Usage
Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkSamplerYcbcrConversionCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pYcbcrConversion must be a valid pointer to a VkSamplerYcbcrConversion handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkSamplerYcbcrConversionCreateInfo structure is defined as:

typedef struct VkSamplerYcbcrConversionCreateInfo {
    VkStructureType                  sType;
    const void*                      pNext;
    VkFormat                         format;
    VkSamplerYcbcrModelConversion    ycbcrModel;
    VkSamplerYcbcrRange              ycbcrRange;
    VkComponentMapping               components;
    VkChromaLocation                 xChromaOffset;
    VkChromaLocation                 yChromaOffset;
    VkFilter                         chromaFilter;
    VkBool32                         forceExplicitReconstruction;
} VkSamplerYcbcrConversionCreateInfo;

or the equivalent

typedef VkSamplerYcbcrConversionCreateInfo VkSamplerYcbcrConversionCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • format is the format of the image from which color information will be retrieved.

  • ycbcrModel describes the color matrix for conversion between color models.

  • ycbcrRange describes whether the encoded values have headroom and foot room, or whether the encoding uses the full numerical range.

  • components applies a swizzle based on VkComponentSwizzle enums prior to range expansion and color model conversion.

  • xChromaOffset describes the sample location associated with downsampled chroma channels in the x dimension. xChromaOffset has no effect for formats in which chroma channels are the same resolution as the luma channel.

  • yChromaOffset describes the sample location associated with downsampled chroma channels in the y dimension. yChromaOffset has no effect for formats in which the chroma channels are not downsampled vertically.

  • chromaFilter is the filter for chroma reconstruction.

  • forceExplicitReconstruction can be used to ensure that reconstruction is done explicitly, if supported.

Note

Setting forceExplicitReconstruction to VK_TRUE may have a performance penalty on implementations where explicit reconstruction is not the default mode of operation.

If the pNext chain has an instance of VkExternalFormatANDROID with non-zero externalFormat member, the sampler Y’CBCR conversion object represents an external format conversion, and format must be VK_FORMAT_UNDEFINED. Such conversions must only be used to sample image views with a matching external format. When creating an external format conversion, the value of components is ignored.

Valid Usage
  • If an external format conversion is being created, format must be VK_FORMAT_UNDEFINED, otherwise it must not be VK_FORMAT_UNDEFINED.

  • format must support VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT or VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT

  • If the format does not support VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT, xChromaOffset and yChromaOffset must not be VK_CHROMA_LOCATION_COSITED_EVEN

  • If the format does not support VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT, xChromaOffset and yChromaOffset must not be VK_CHROMA_LOCATION_MIDPOINT

  • format must represent unsigned normalized values (i.e. the format must be a UNORM format)

  • If the format has a _422 or _420 suffix:

    • components.g must be VK_COMPONENT_SWIZZLE_IDENTITY

    • components.a must be VK_COMPONENT_SWIZZLE_IDENTITY, VK_COMPONENT_SWIZZLE_ONE, or VK_COMPONENT_SWIZZLE_ZERO

    • components.r must be VK_COMPONENT_SWIZZLE_IDENTITY or VK_COMPONENT_SWIZZLE_B

    • components.b must be VK_COMPONENT_SWIZZLE_IDENTITY or VK_COMPONENT_SWIZZLE_R

    • If either components.r or components.b is VK_COMPONENT_SWIZZLE_IDENTITY, both values must be VK_COMPONENT_SWIZZLE_IDENTITY

  • If ycbcrModel is not VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY, then components.r, components.g, and components.b must correspond to channels of the format; that is, components.r, components.g, and components.b must not be VK_COMPONENT_SWIZZLE_ZERO or VK_COMPONENT_SWIZZLE_ONE, and must not correspond to a channel which contains zero or one as a consequence of conversion to RGBA

  • If the format does not support VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT, forceExplicitReconstruction must be FALSE

  • If the format does not support VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT, chromaFilter must be VK_FILTER_NEAREST

Valid Usage (Implicit)

If chromaFilter is VK_FILTER_NEAREST, chroma samples are reconstructed to luma channel resolution using nearest-neighbour sampling. Otherwise, chroma samples are reconstructed using interpolation. More details can be found in the description of sampler Y’CBCR conversion in the Image Operations chapter.

VkSamplerYcbcrModelConversion defines the conversion from the source color model to the shader color model. Possible values are:

typedef enum VkSamplerYcbcrModelConversion {
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY = 0,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY = 1,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709 = 2,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601 = 3,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020 = 4,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601,
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020,
} VkSamplerYcbcrModelConversion;

or the equivalent

typedef VkSamplerYcbcrModelConversion VkSamplerYcbcrModelConversionKHR;
  • VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY specifies that the input values to the conversion are unmodified.

  • VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY specifies no model conversion but the inputs are range expanded as for Y’CBCR.

  • VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709 specifies the color model conversion from Y’CBCR to R’G’B' defined in BT.709 and described in the “BT.709 Y’CBCR conversion” section of the Khronos Data Format Specification.

  • VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601 specifies the color model conversion from Y’CBCR to R’G’B' defined in BT.601 and described in the “BT.601 Y’CBCR conversion” section of the Khronos Data Format Specification.

  • VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020 specifies the color model conversion from Y’CBCR to R’G’B' defined in BT.2020 and described in the “BT.2020 Y’CBCR conversion” section of the Khronos Data Format Specification.

In the VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_* color models, for the input to the sampler Y’CBCR range expansion and model conversion:

  • the Y (Y' luma) channel corresponds to the G channel of an RGB image.

  • the CB (CB or “U” blue color difference) channel corresponds to the B channel of an RGB image.

  • the CR (CR or “V” red color difference) channel corresponds to the R channel of an RGB image.

  • the alpha channel, if present, is not modified by color model conversion.

These rules reflect the mapping of channels after the channel swizzle operation (controlled by VkSamplerYcbcrConversionCreateInfo::components).

Note

For example, an “YUVA” 32-bit format comprising four 8-bit channels can be implemented as VK_FORMAT_R8G8B8A8_UNORM with a component mapping:

  • components.a = VK_COMPONENT_SWIZZLE_IDENTITY

  • components.r = VK_COMPONENT_SWIZZLE_B

  • components.g = VK_COMPONENT_SWIZZLE_R

  • components.b = VK_COMPONENT_SWIZZLE_G

The VkSamplerYcbcrRange enum describes whether color channels are encoded using the full range of numerical values or whether values are reserved for headroom and foot room. VkSamplerYcbcrRange is defined as:

typedef enum VkSamplerYcbcrRange {
    VK_SAMPLER_YCBCR_RANGE_ITU_FULL = 0,
    VK_SAMPLER_YCBCR_RANGE_ITU_NARROW = 1,
    VK_SAMPLER_YCBCR_RANGE_ITU_FULL_KHR = VK_SAMPLER_YCBCR_RANGE_ITU_FULL,
    VK_SAMPLER_YCBCR_RANGE_ITU_NARROW_KHR = VK_SAMPLER_YCBCR_RANGE_ITU_NARROW,
} VkSamplerYcbcrRange;

or the equivalent

typedef VkSamplerYcbcrRange VkSamplerYcbcrRangeKHR;
  • VK_SAMPLER_YCBCR_RANGE_ITU_FULL specifies that the full range of the encoded values are valid and interpreted according to the ITU “full range” quantization rules.

  • VK_SAMPLER_YCBCR_RANGE_ITU_NARROW specifies that headroom and foot room are reserved in the numerical range of encoded values, and the remaining values are expanded according to the ITU “narrow range” quantization rules.

The formulae for these conversions is described in the Sampler Y’CBCR Range Expansion section of the Image Operations chapter.

No range modification takes place if ycbcrModel is VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY; the ycbcrRange field of VkSamplerYcbcrConversionCreateInfo is ignored in this case.

The VkChromaLocation enum, which defines the location of downsampled chroma channel samples relative to the luma samples, is defined as:

typedef enum VkChromaLocation {
    VK_CHROMA_LOCATION_COSITED_EVEN = 0,
    VK_CHROMA_LOCATION_MIDPOINT = 1,
    VK_CHROMA_LOCATION_COSITED_EVEN_KHR = VK_CHROMA_LOCATION_COSITED_EVEN,
    VK_CHROMA_LOCATION_MIDPOINT_KHR = VK_CHROMA_LOCATION_MIDPOINT,
} VkChromaLocation;

or the equivalent

typedef VkChromaLocation VkChromaLocationKHR;
  • VK_CHROMA_LOCATION_COSITED_EVEN specifies that downsampled chroma samples are aligned with luma samples with even coordinates.

  • VK_CHROMA_LOCATION_MIDPOINT specifies that downsampled chroma samples are located half way between each even luma sample and the nearest higher odd luma sample.

To destroy a sampler Y’CBCR conversion, call:

void vkDestroySamplerYcbcrConversion(
    VkDevice                                    device,
    VkSamplerYcbcrConversion                    ycbcrConversion,
    const VkAllocationCallbacks*                pAllocator);

or the equivalent command

void vkDestroySamplerYcbcrConversionKHR(
    VkDevice                                    device,
    VkSamplerYcbcrConversion                    ycbcrConversion,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the Y’CBCR conversion.

  • ycbcrConversion is the conversion to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If ycbcrConversion is not VK_NULL_HANDLE, ycbcrConversion must be a valid VkSamplerYcbcrConversion handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If ycbcrConversion is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to ycbcrConversion must be externally synchronized

13. Resource Descriptors

A descriptor is an opaque data structure representing a shader resource such as a buffer, buffer view, image view, sampler, or combined image sampler. Descriptors are organised into descriptor sets, which are bound during command recording for use in subsequent draw commands. The arrangement of content in each descriptor set is determined by a descriptor set layout, which determines what descriptors can be stored within it. The sequence of descriptor set layouts that can be used by a pipeline is specified in a pipeline layout. Each pipeline object can use up to maxBoundDescriptorSets (see Limits) descriptor sets.

Shaders access resources via variables decorated with a descriptor set and binding number that link them to a descriptor in a descriptor set. The shader interface mapping to bound descriptor sets is described in the Shader Resource Interface section.

13.1. Descriptor Types

There are a number of different types of descriptor supported by Vulkan, corresponding to different resources or usage. The following sections describe the API definitions of each descriptor type. The mapping of each type to SPIR-V is listed in the Shader Resource and Descriptor Type Correspondence and Shader Resource and Storage Class Correspondence tables in the Shader Interfaces chapter.

13.1.1. Storage Image

A storage image (VK_DESCRIPTOR_TYPE_STORAGE_IMAGE) is a descriptor type associated with an image resource via an image view that load, store, and atomic operations can be performed on.

Storage image loads are supported in all shader stages for image formats which report support for the VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT feature bit via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::linearTilingFeatures (for images with linear tiling) or VkFormatProperties::optimalTilingFeatures (for images with optimal tiling).

Stores to storage images are supported in compute shaders for image formats which report support for the VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT feature via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::linearTilingFeatures (for images with linear tiling) or VkFormatProperties::optimalTilingFeatures (for images with optimal tiling).

Atomic operations on storage images are supported in compute shaders for image formats which report support for the VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT feature via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::linearTilingFeatures (for images with linear tiling) or VkFormatProperties::optimalTilingFeatures (for images with optimal tiling).

When the fragmentStoresAndAtomics feature is enabled, stores and atomic operations are also supported for storage images in fragment shaders with the same set of image formats as supported in compute shaders. When the vertexPipelineStoresAndAtomics feature is enabled, stores and atomic operations are also supported in vertex, tessellation, and geometry shaders with the same set of image formats as supported in compute shaders.

The image subresources for a storage image must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR or VK_IMAGE_LAYOUT_GENERAL layout in order to access its data in a shader.

13.1.2. Sampler

A sampler descriptor (VK_DESCRIPTOR_TYPE_SAMPLER) is a descriptor type associated with a sampler object, used to control the behaviour of sampling operations performed on a sampled image.

13.1.3. Sampled Image

A sampled image (VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE) is a descriptor type associated with an image resource via an image view that sampling operations can be performed on.

Shaders combine a sampled image variable and a sampler variable to perform sampling operations.

Sampled images are supported in all shader stages for image formats which report support for the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT feature bit via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::linearTilingFeatures (for images with linear tiling) or VkFormatProperties::optimalTilingFeatures (for images with optimal tiling).

The image subresources for a sampled image must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_GENERAL layout in order to access its data in a shader.

13.1.4. Combined Image Sampler

A combined image sampler (VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER) is a single descriptor type associated with both a sampler and an image resource, combining both a sampler and sampled image descriptor into a single descriptor.

If the descriptor refers to a sampler that performs Y’CBCR conversion, the sampler must only be used to sample the image in the same descriptor. Otherwise, the sampler and image in this type of descriptor can be used freely with any other samplers and images.

The image subresources for a combined image sampler must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_GENERAL layout in order to access its data in a shader.

Note

On some implementations, it may be more efficient to sample from an image using a combination of sampler and sampled image that are stored together in the descriptor set in a combined descriptor.

13.1.5. Uniform Texel Buffer

A uniform texel buffer (VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER) is a descriptor type associated with a buffer resource via a buffer view that formatted load operations can be performed on.

Uniform texel buffers define a tightly-packed 1-dimensional linear array of texels, with texels going through format conversion when read in a shader in the same way as they are for an image.

Load operations from uniform texel buffers are supported in all shader stages for image formats which report support for the VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT feature bit via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::bufferFeatures.

13.1.6. Storage Texel Buffer

A storage texel buffer (VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER) is a descriptor type associated with a buffer resource via a buffer view that formatted load, store, and atomic operations can be performed on.

Storage texel buffers define a tightly-packed 1-dimensional linear array of texels, with texels going through format conversion when read in a shader in the same way as they are for an image. Unlike uniform texel buffers, these buffers can also be written to in the same way as for storage images.

Storage texel buffer loads are supported in all shader stages for texel buffer formats which report support for the VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT feature bit via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::bufferFeatures.

Stores to storage texel buffers are supported in compute shaders for texel buffer formats which report support for the VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT feature via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::bufferFeatures.

Atomic operations on storage texel buffers are supported in compute shaders for texel buffer formats which report support for the VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT feature via vkGetPhysicalDeviceFormatProperties in VkFormatProperties::bufferFeatures.

When the fragmentStoresAndAtomics feature is enabled, stores and atomic operations are also supported for storage texel buffers in fragment shaders with the same set of texel buffer formats as supported in compute shaders. When the vertexPipelineStoresAndAtomics feature is enabled, stores and atomic operations are also supported in vertex, tessellation, and geometry shaders with the same set of texel buffer formats as supported in compute shaders.

13.1.7. Storage Buffer

A storage buffer (VK_DESCRIPTOR_TYPE_STORAGE_BUFFER) is a descriptor type associated with a buffer resource directly, described in a shader as a structure with various members that load, store, and atomic operations can be performed on.

Note

Atomic operations can only be performed on members of certain types as defined in the SPIR-V environment appendix.

13.1.8. Uniform Buffer

A uniform buffer (VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER) is a descriptor type associated with a buffer resource directly, described in a shader as a structure with various members that load operations can be performed on.

13.1.9. Dynamic Uniform Buffer

A dynamic uniform buffer (VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC) is almost identical to a uniform buffer, and differs only in how the offset into the buffer is specified. The base offset calculated by the VkDescriptorBufferInfo when initially updating the descriptor set is added to a dynamic offset when binding the descriptor set.

13.1.10. Dynamic Storage Buffer

A dynamic storage buffer (VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC) is almost identical to a storage buffer, and differs only in how the offset into the buffer is specified. The base offset calculated by the VkDescriptorBufferInfo when initially updating the descriptor set is added to a dynamic offset when binding the descriptor set.

13.1.11. Input Attachment

An input attachment (VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT) is a descriptor type associated with an image resource via an image view that can be used for framebuffer local load operations in fragment shaders.

All image formats that are supported for color attachments (VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT) or depth/stencil attachments (VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT) for a given image tiling mode are also supported for input attachments.

The image subresources for an input attachment must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_GENERAL layout in order to access its data in a shader.

13.2. Descriptor Sets

Descriptors are grouped together into descriptor set objects. A descriptor set object is an opaque object that contains storage for a set of descriptors, where the types and number of descriptors is defined by a descriptor set layout. The layout object may be used to define the association of each descriptor binding with memory or other implementation resources. The layout is used both for determining the resources that need to be associated with the descriptor set, and determining the interface between shader stages and shader resources.

13.2.1. Descriptor Set Layout

A descriptor set layout object is defined by an array of zero or more descriptor bindings. Each individual descriptor binding is specified by a descriptor type, a count (array size) of the number of descriptors in the binding, a set of shader stages that can access the binding, and (if using immutable samplers) an array of sampler descriptors.

Descriptor set layout objects are represented by VkDescriptorSetLayout handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorSetLayout)

To create descriptor set layout objects, call:

VkResult vkCreateDescriptorSetLayout(
    VkDevice                                    device,
    const VkDescriptorSetLayoutCreateInfo*      pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorSetLayout*                      pSetLayout);
  • device is the logical device that creates the descriptor set layout.

  • pCreateInfo is a pointer to an instance of the VkDescriptorSetLayoutCreateInfo structure specifying the state of the descriptor set layout object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pSetLayout points to a VkDescriptorSetLayout handle in which the resulting descriptor set layout object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkDescriptorSetLayoutCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSetLayout must be a valid pointer to a VkDescriptorSetLayout handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Information about the descriptor set layout is passed in an instance of the VkDescriptorSetLayoutCreateInfo structure:

typedef struct VkDescriptorSetLayoutCreateInfo {
    VkStructureType                        sType;
    const void*                            pNext;
    VkDescriptorSetLayoutCreateFlags       flags;
    uint32_t                               bindingCount;
    const VkDescriptorSetLayoutBinding*    pBindings;
} VkDescriptorSetLayoutCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkDescriptorSetLayoutCreateFlagBits specifying options for descriptor set layout creation.

  • bindingCount is the number of elements in pBindings.

  • pBindings is a pointer to an array of VkDescriptorSetLayoutBinding structures.

Valid Usage
  • The VkDescriptorSetLayoutBinding::binding members of the elements of the pBindings array must each have different values.

  • If flags contains VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR, then all elements of pBindings must not have a descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC

  • If flags contains VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR, then the total number of elements of all bindings must be less than or equal to VkPhysicalDevicePushDescriptorPropertiesKHR::maxPushDescriptors

  • If any binding has the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT bit set, flags must include VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT

  • If any binding has the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT bit set, then all bindings must not have descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC

Valid Usage (Implicit)

Bits which can be set in VkDescriptorSetLayoutCreateInfo::flags to specify options for descriptor set layout are:

typedef enum VkDescriptorSetLayoutCreateFlagBits {
    VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR = 0x00000001,
    VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT = 0x00000002,
} VkDescriptorSetLayoutCreateFlagBits;
  • VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR specifies that descriptor sets must not be allocated using this layout, and descriptors are instead pushed by vkCmdPushDescriptorSetKHR.

  • VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT specifies that descriptor sets using this layout must be allocated from a descriptor pool created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT bit set. Descriptor set layouts created with this bit set have alternate limits for the maximum number of descriptors per-stage and per-pipeline layout. The non-UpdateAfterBind limits only count descriptors in sets created without this flag. The UpdateAfterBind limits count all descriptors, but the limits may be higher than the non-UpdateAfterBind limits.

typedef VkFlags VkDescriptorSetLayoutCreateFlags;

VkDescriptorSetLayoutCreateFlags is a bitmask type for setting a mask of zero or more VkDescriptorSetLayoutCreateFlagBits.

The VkDescriptorSetLayoutBinding structure is defined as:

typedef struct VkDescriptorSetLayoutBinding {
    uint32_t              binding;
    VkDescriptorType      descriptorType;
    uint32_t              descriptorCount;
    VkShaderStageFlags    stageFlags;
    const VkSampler*      pImmutableSamplers;
} VkDescriptorSetLayoutBinding;
  • binding is the binding number of this entry and corresponds to a resource of the same binding number in the shader stages.

  • descriptorType is a VkDescriptorType specifying which type of resource descriptors are used for this binding.

  • descriptorCount is the number of descriptors contained in the binding, accessed in a shader as an array. If descriptorCount is zero this binding entry is reserved and the resource must not be accessed from any stage via this binding within any pipeline using the set layout.

  • stageFlags member is a bitmask of VkShaderStageFlagBits specifying which pipeline shader stages can access a resource for this binding. VK_SHADER_STAGE_ALL is a shorthand specifying that all defined shader stages, including any additional stages defined by extensions, can access the resource.

    If a shader stage is not included in stageFlags, then a resource must not be accessed from that stage via this binding within any pipeline using the set layout. Other than input attachments which are limited to the fragment shader, there are no limitations on what combinations of stages can use a descriptor binding, and in particular a binding can be used by both graphics stages and the compute stage.

  • pImmutableSamplers affects initialization of samplers. If descriptorType specifies a VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER type descriptor, then pImmutableSamplers can be used to initialize a set of immutable samplers. Immutable samplers are permanently bound into the set layout; later binding a sampler into an immutable sampler slot in a descriptor set is not allowed. If pImmutableSamplers is not NULL, then it is considered to be a pointer to an array of sampler handles that will be consumed by the set layout and used for the corresponding binding. If pImmutableSamplers is NULL, then the sampler slots are dynamic and sampler handles must be bound into descriptor sets using this layout. If descriptorType is not one of these descriptor types, then pImmutableSamplers is ignored.

The above layout definition allows the descriptor bindings to be specified sparsely such that not all binding numbers between 0 and the maximum binding number need to be specified in the pBindings array. Bindings that are not specified have a descriptorCount and stageFlags of zero, and the descriptorType is treated as undefined. However, all binding numbers between 0 and the maximum binding number in the VkDescriptorSetLayoutCreateInfo::pBindings array may consume memory in the descriptor set layout even if not all descriptor bindings are used, though it should not consume additional memory from the descriptor pool.

Note

The maximum binding number specified should be as compact as possible to avoid wasted memory.

Valid Usage
  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and descriptorCount is not 0 and pImmutableSamplers is not NULL, pImmutableSamplers must be a valid pointer to an array of descriptorCount valid VkSampler handles

  • If descriptorCount is not 0, stageFlags must be a valid combination of VkShaderStageFlagBits values

  • If descriptorType is VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT and descriptorCount is not 0, then stageFlags must be 0 or VK_SHADER_STAGE_FRAGMENT_BIT

Valid Usage (Implicit)

If the pNext chain of a VkDescriptorSetLayoutCreateInfo structure includes a VkDescriptorSetLayoutBindingFlagsCreateInfoEXT structure, then that structure includes an array of flags, one for each descriptor set layout binding.

The VkDescriptorSetLayoutBindingFlagsCreateInfoEXT structure is defined as:

typedef struct VkDescriptorSetLayoutBindingFlagsCreateInfoEXT {
    VkStructureType                       sType;
    const void*                           pNext;
    uint32_t                              bindingCount;
    const VkDescriptorBindingFlagsEXT*    pBindingFlags;
} VkDescriptorSetLayoutBindingFlagsCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • bindingCount is zero or the number of elements in pBindingFlags.

  • pBindingFlags is a pointer to an array of VkDescriptorBindingFlagsEXT bitfields, one for each descriptor set layout binding.

If bindingCount is zero or if this structure is not in the pNext chain, the VkDescriptorBindingFlagsEXT for each descriptor set layout binding is considered to be zero. Otherwise, the descriptor set layout binding at VkDescriptorSetLayoutCreateInfo::pBindings[i] uses the flags in pBindingFlags[i].

Valid Usage
  • If bindingCount is not zero, bindingCount must equal VkDescriptorSetLayoutCreateInfo::bindingCount

  • If VkDescriptorSetLayoutCreateInfo::flags includes VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR, then all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT, VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT, or VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT

  • If an element of pBindingFlags includes VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT, then all other elements of VkDescriptorSetLayoutCreateInfo::pBindings must have a smaller value of binding

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingUniformBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingSampledImageUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, or VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingStorageImageUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_STORAGE_IMAGE must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingStorageBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingUniformTexelBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingStorageTexelBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • All bindings with descriptor type VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingUpdateUnusedWhilePending is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingPartiallyBound is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT

  • If VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingVariableDescriptorCount is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT

  • If an element of pBindingFlags includes VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT, that element’s descriptorType must not be VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_BINDING_FLAGS_CREATE_INFO_EXT

  • If bindingCount is not 0, pBindingFlags must be a valid pointer to an array of bindingCount valid combinations of VkDescriptorBindingFlagBitsEXT values

  • Each element of pBindingFlags must not be 0

Bits which can be set in each element of VkDescriptorSetLayoutBindingFlagsCreateInfoEXT::pBindingFlags to specify options for the corresponding descriptor set layout binding are:

typedef enum VkDescriptorBindingFlagBitsEXT {
    VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT = 0x00000001,
    VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT = 0x00000002,
    VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT = 0x00000004,
    VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT = 0x00000008,
} VkDescriptorBindingFlagBitsEXT;
  • VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT indicates that if descriptors in this binding are updated between when the descriptor set is bound in a command buffer and when that command buffer is submitted to a queue, then the submission will use the most recently set descriptors for this binding and the updates do not invalidate the command buffer. Descriptor bindings created with this flag are also partially exempt from the external synchronization requirement in vkUpdateDescriptorSetWithTemplateKHR and vkUpdateDescriptorSets. They can be updated concurrently with the set being bound to a command buffer in another thread, but not concurrently with the set being reset or freed.

  • VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT indicates that descriptors in this binding that are not dynamically used need not contain valid descriptors at the time the descriptors are consumed. A descriptor is dynamically used if any shader invocation executes an instruction that performs any memory access using the descriptor.

  • VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT indicates that descriptors in this binding can be updated after a command buffer has bound this descriptor set, or while a command buffer that uses this descriptor set is pending execution, as long as the descriptors that are updated are not used by those command buffers. If VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT is also set, then descriptors can be updated as long as they are not dynamically used by any shader invocations. If VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT is not set, then descriptors can be updated as long as they are not statically used by any shader invocations.

  • VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT indicates that this descriptor binding has a variable size that will be specified when a descriptor set is allocated using this layout. The value of descriptorCount is treated as an upper bound on the size of the binding. This must only be used for the last binding in the descriptor set layout (i.e. the binding with the largest value of binding). For the purposes of counting against limits such as maxDescriptorSet* and maxPerStageDescriptor*, the full value of descriptorCount is counted.

Note

Note that while VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT and VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT both involve updates to descriptor sets after they are bound, VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT is a weaker requirement since it is only about descriptors that are not used, whereas VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT requires the implementation to observe updates to descriptors that are used.

To query information about whether a descriptor set layout can be created, call:

void vkGetDescriptorSetLayoutSupport(
    VkDevice                                    device,
    const VkDescriptorSetLayoutCreateInfo*      pCreateInfo,
    VkDescriptorSetLayoutSupport*               pSupport);

or the equivalent command

void vkGetDescriptorSetLayoutSupportKHR(
    VkDevice                                    device,
    const VkDescriptorSetLayoutCreateInfo*      pCreateInfo,
    VkDescriptorSetLayoutSupport*               pSupport);
  • device is the logical device that would create the descriptor set layout.

  • pCreateInfo is a pointer to an instance of the VkDescriptorSetLayoutCreateInfo structure specifying the state of the descriptor set layout object.

  • pSupport points to a VkDescriptorSetLayoutSupport structure in which information about support for the descriptor set layout object is returned.

Some implementations have limitations on what fits in a descriptor set which are not easily expressible in terms of existing limits like maxDescriptorSet*, for example if all descriptor types share a limited space in memory but each descriptor is a different size or alignment. This command returns information about whether a descriptor set satisfies this limit. If the descriptor set layout satisfies the VkPhysicalDeviceMaintenance3Properties::maxPerSetDescriptors limit, this command is guaranteed to return VK_TRUE in VkDescriptorSetLayoutSupport::supported. If the descriptor set layout exceeds the VkPhysicalDeviceMaintenance3Properties::maxPerSetDescriptors limit, whether the descriptor set layout is supported is implementation-dependent and may depend on whether the descriptor sizes and alignments cause the layout to exceed an internal limit.

This command does not consider other limits such as maxPerStageDescriptor*, and so a descriptor set layout that is supported according to this command must still satisfy the pipeline layout limits such as maxPerStageDescriptor* in order to be used in a pipeline layout.

Note

This is a VkDevice query rather than VkPhysicalDevice because the answer may depend on enabled features.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkDescriptorSetLayoutCreateInfo structure

  • pSupport must be a valid pointer to a VkDescriptorSetLayoutSupport structure

Information about support for the descriptor set layout is returned in an instance of the VkDescriptorSetLayoutSupport structure:

typedef struct VkDescriptorSetLayoutSupport {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           supported;
} VkDescriptorSetLayoutSupport;

or the equivalent

typedef VkDescriptorSetLayoutSupport VkDescriptorSetLayoutSupportKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • supported specifies whether the descriptor set layout can be created.

supported is set to VK_TRUE if the descriptor set can be created, or else is set to VK_FALSE.

Valid Usage (Implicit)

If the pNext chain of a VkDescriptorSetLayoutSupport structure includes a VkDescriptorSetVariableDescriptorCountLayoutSupportEXT structure, then that structure returns additional information about whether the descriptor set layout is supported.

typedef struct VkDescriptorSetVariableDescriptorCountLayoutSupportEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxVariableDescriptorCount;
} VkDescriptorSetVariableDescriptorCountLayoutSupportEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • maxVariableDescriptorCount indicates the maximum number of descriptors supported in the highest numbered binding of the layout, if that binding is variable-sized.

If the create info includes a variable-sized descriptor, then supported is determined assuming the requested size of the variable-sized descriptor, and maxVariableDescriptorCount is set to the maximum size of that descriptor that can be successfully created (which is greater than or equal to the requested size passed in). If the create info does not include a variable-sized descriptor or if the VkPhysicalDeviceDescriptorIndexingFeaturesEXT::descriptorBindingVariableDescriptorCount feature is not enabled, then maxVariableDescriptorCount is set to zero. For the purposes of this command, a variable-sized descriptor binding with a descriptorCount of zero is treated as if the descriptorCount is one, and thus the binding is not ignored and the maximum descriptor count will be returned. If the layout is not supported, then the value written to maxVariableDescriptorCount is undefined.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_LAYOUT_SUPPORT_EXT

The following examples show a shader snippet using two descriptor sets, and application code that creates corresponding descriptor set layouts.

GLSL example
//
// binding to a single sampled image descriptor in set 0
//
layout (set=0, binding=0) uniform texture2D mySampledImage;

//
// binding to an array of sampled image descriptors in set 0
//
layout (set=0, binding=1) uniform texture2D myArrayOfSampledImages[12];

//
// binding to a single uniform buffer descriptor in set 1
//
layout (set=1, binding=0) uniform myUniformBuffer
{
    vec4 myElement[32];
};
SPIR-V example
               ...
          %1 = OpExtInstImport "GLSL.std.450"
               ...
               OpName %9 "mySampledImage"
               OpName %14 "myArrayOfSampledImages"
               OpName %18 "myUniformBuffer"
               OpMemberName %18 0 "myElement"
               OpName %20 ""
               OpDecorate %9 DescriptorSet 0
               OpDecorate %9 Binding 0
               OpDecorate %14 DescriptorSet 0
               OpDecorate %14 Binding 1
               OpDecorate %17 ArrayStride 16
               OpMemberDecorate %18 0 Offset 0
               OpDecorate %18 Block
               OpDecorate %20 DescriptorSet 1
               OpDecorate %20 Binding 0
          %2 = OpTypeVoid
          %3 = OpTypeFunction %2
          %6 = OpTypeFloat 32
          %7 = OpTypeImage %6 2D 0 0 0 1 Unknown
          %8 = OpTypePointer UniformConstant %7
          %9 = OpVariable %8 UniformConstant
         %10 = OpTypeInt 32 0
         %11 = OpConstant %10 12
         %12 = OpTypeArray %7 %11
         %13 = OpTypePointer UniformConstant %12
         %14 = OpVariable %13 UniformConstant
         %15 = OpTypeVector %6 4
         %16 = OpConstant %10 32
         %17 = OpTypeArray %15 %16
         %18 = OpTypeStruct %17
         %19 = OpTypePointer Uniform %18
         %20 = OpVariable %19 Uniform
               ...
API example
VkResult myResult;

const VkDescriptorSetLayoutBinding myDescriptorSetLayoutBinding[] =
{
    // binding to a single image descriptor
    {
        0,                                      // binding
        VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE,       // descriptorType
        1,                                      // descriptorCount
        VK_SHADER_STAGE_FRAGMENT_BIT,           // stageFlags
        NULL                                    // pImmutableSamplers
    },

    // binding to an array of image descriptors
    {
        1,                                      // binding
        VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE,       // descriptorType
        12,                                     // descriptorCount
        VK_SHADER_STAGE_FRAGMENT_BIT,           // stageFlags
        NULL                                    // pImmutableSamplers
    },

    // binding to a single uniform buffer descriptor
    {
        0,                                      // binding
        VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER,      // descriptorType
        1,                                      // descriptorCount
        VK_SHADER_STAGE_FRAGMENT_BIT,           // stageFlags
        NULL                                    // pImmutableSamplers
    }
};

const VkDescriptorSetLayoutCreateInfo myDescriptorSetLayoutCreateInfo[] =
{
    // Create info for first descriptor set with two descriptor bindings
    {
        VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO,    // sType
        NULL,                                                   // pNext
        0,                                                      // flags
        2,                                                      // bindingCount
        &myDescriptorSetLayoutBinding[0]                        // pBindings
    },

    // Create info for second descriptor set with one descriptor binding
    {
        VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO,    // sType
        NULL,                                                   // pNext
        0,                                                      // flags
        1,                                                      // bindingCount
        &myDescriptorSetLayoutBinding[2]                        // pBindings
    }
};

VkDescriptorSetLayout myDescriptorSetLayout[2];

//
// Create first descriptor set layout
//
myResult = vkCreateDescriptorSetLayout(
    myDevice,
    &myDescriptorSetLayoutCreateInfo[0],
    NULL,
    &myDescriptorSetLayout[0]);

//
// Create second descriptor set layout
//
myResult = vkCreateDescriptorSetLayout(
    myDevice,
    &myDescriptorSetLayoutCreateInfo[1],
    NULL,
    &myDescriptorSetLayout[1]);

To destroy a descriptor set layout, call:

void vkDestroyDescriptorSetLayout(
    VkDevice                                    device,
    VkDescriptorSetLayout                       descriptorSetLayout,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the descriptor set layout.

  • descriptorSetLayout is the descriptor set layout to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when descriptorSetLayout was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when descriptorSetLayout was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If descriptorSetLayout is not VK_NULL_HANDLE, descriptorSetLayout must be a valid VkDescriptorSetLayout handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If descriptorSetLayout is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to descriptorSetLayout must be externally synchronized

13.2.2. Pipeline Layouts

Access to descriptor sets from a pipeline is accomplished through a pipeline layout. Zero or more descriptor set layouts and zero or more push constant ranges are combined to form a pipeline layout object which describes the complete set of resources that can be accessed by a pipeline. The pipeline layout represents a sequence of descriptor sets with each having a specific layout. This sequence of layouts is used to determine the interface between shader stages and shader resources. Each pipeline is created using a pipeline layout.

Pipeline layout objects are represented by VkPipelineLayout handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPipelineLayout)

To create a pipeline layout, call:

VkResult vkCreatePipelineLayout(
    VkDevice                                    device,
    const VkPipelineLayoutCreateInfo*           pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkPipelineLayout*                           pPipelineLayout);
  • device is the logical device that creates the pipeline layout.

  • pCreateInfo is a pointer to an instance of the VkPipelineLayoutCreateInfo structure specifying the state of the pipeline layout object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pPipelineLayout points to a VkPipelineLayout handle in which the resulting pipeline layout object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkPipelineLayoutCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pPipelineLayout must be a valid pointer to a VkPipelineLayout handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineLayoutCreateInfo structure is defined as:

typedef struct VkPipelineLayoutCreateInfo {
    VkStructureType                 sType;
    const void*                     pNext;
    VkPipelineLayoutCreateFlags     flags;
    uint32_t                        setLayoutCount;
    const VkDescriptorSetLayout*    pSetLayouts;
    uint32_t                        pushConstantRangeCount;
    const VkPushConstantRange*      pPushConstantRanges;
} VkPipelineLayoutCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • setLayoutCount is the number of descriptor sets included in the pipeline layout.

  • pSetLayouts is a pointer to an array of VkDescriptorSetLayout objects.

  • pushConstantRangeCount is the number of push constant ranges included in the pipeline layout.

  • pPushConstantRanges is a pointer to an array of VkPushConstantRange structures defining a set of push constant ranges for use in a single pipeline layout. In addition to descriptor set layouts, a pipeline layout also describes how many push constants can be accessed by each stage of the pipeline.

    Note

    Push constants represent a high speed path to modify constant data in pipelines that is expected to outperform memory-backed resource updates.

Valid Usage
  • setLayoutCount must be less than or equal to VkPhysicalDeviceLimits::maxBoundDescriptorSets

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_SAMPLER and VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorSamplers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER and VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorUniformBuffers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER and VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorStorageBuffers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, and VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorSampledImages

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, and VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorStorageImages

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxPerStageDescriptorInputAttachments

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_SAMPLER and VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindSamplers

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER and VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindUniformBuffers

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER and VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindStorageBuffers

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, and VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindSampledImages

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, and VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindStorageImages

  • The total number of descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxPerStageDescriptorUpdateAfterBindInputAttachments

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_SAMPLER and VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetSamplers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetUniformBuffers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetUniformBuffersDynamic

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetStorageBuffers

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetStorageBuffersDynamic

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, and VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetSampledImages

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, and VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetStorageImages

  • The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceLimits::maxDescriptorSetInputAttachments

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_SAMPLER and VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindSamplers

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindUniformBuffers

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindUniformBuffersDynamic

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindStorageBuffers

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindStorageBuffersDynamic

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, and VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindSampledImages

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, and VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindStorageImages

  • The total number of descriptors of the type VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceDescriptorIndexingPropertiesEXT::maxDescriptorSetUpdateAfterBindInputAttachments

  • Any two elements of pPushConstantRanges must not include the same stage in stageFlags

  • pSetLayouts must not contain more than one descriptor set layout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR set

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • If setLayoutCount is not 0, pSetLayouts must be a valid pointer to an array of setLayoutCount valid VkDescriptorSetLayout handles

  • If pushConstantRangeCount is not 0, pPushConstantRanges must be a valid pointer to an array of pushConstantRangeCount valid VkPushConstantRange structures

typedef VkFlags VkPipelineLayoutCreateFlags;

VkPipelineLayoutCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPushConstantRange structure is defined as:

typedef struct VkPushConstantRange {
    VkShaderStageFlags    stageFlags;
    uint32_t              offset;
    uint32_t              size;
} VkPushConstantRange;
  • stageFlags is a set of stage flags describing the shader stages that will access a range of push constants. If a particular stage is not included in the range, then accessing members of that range of push constants from the corresponding shader stage will result in undefined data being read.

  • offset and size are the start offset and size, respectively, consumed by the range. Both offset and size are in units of bytes and must be a multiple of 4. The layout of the push constant variables is specified in the shader.

Valid Usage
  • offset must be less than VkPhysicalDeviceLimits::maxPushConstantsSize

  • offset must be a multiple of 4

  • size must be greater than 0

  • size must be a multiple of 4

  • size must be less than or equal to VkPhysicalDeviceLimits::maxPushConstantsSize minus offset

Valid Usage (Implicit)

Once created, pipeline layouts are used as part of pipeline creation (see Pipelines), as part of binding descriptor sets (see Descriptor Set Binding), and as part of setting push constants (see Push Constant Updates). Pipeline creation accepts a pipeline layout as input, and the layout may be used to map (set, binding, arrayElement) tuples to implementation resources or memory locations within a descriptor set. The assignment of implementation resources depends only on the bindings defined in the descriptor sets that comprise the pipeline layout, and not on any shader source.

All resource variables statically used in all shaders in a pipeline must be declared with a (set,binding,arrayElement) that exists in the corresponding descriptor set layout and is of an appropriate descriptor type and includes the set of shader stages it is used by in stageFlags. The pipeline layout can include entries that are not used by a particular pipeline, or that are dead-code eliminated from any of the shaders. The pipeline layout allows the application to provide a consistent set of bindings across multiple pipeline compiles, which enables those pipelines to be compiled in a way that the implementation may cheaply switch pipelines without reprogramming the bindings.

Similarly, the push constant block declared in each shader (if present) must only place variables at offsets that are each included in a push constant range with stageFlags including the bit corresponding to the shader stage that uses it. The pipeline layout can include ranges or portions of ranges that are not used by a particular pipeline, or for which the variables have been dead-code eliminated from any of the shaders.

There is a limit on the total number of resources of each type that can be included in bindings in all descriptor set layouts in a pipeline layout as shown in Pipeline Layout Resource Limits. The “Total Resources Available” column gives the limit on the number of each type of resource that can be included in bindings in all descriptor sets in the pipeline layout. Some resource types count against multiple limits. Additionally, there are limits on the total number of each type of resource that can be used in any pipeline stage as described in Shader Resource Limits.

Table 17. Pipeline Layout Resource Limits
Total Resources Available Resource Types

maxDescriptorSetSamplers or maxDescriptorSetUpdateAfterBindSamplers

sampler

combined image sampler

maxDescriptorSetSampledImages or maxDescriptorSetUpdateAfterBindSampledImages

sampled image

combined image sampler

uniform texel buffer

maxDescriptorSetStorageImages or maxDescriptorSetUpdateAfterBindStorageImages

storage image

storage texel buffer

maxDescriptorSetUniformBuffers or maxDescriptorSetUpdateAfterBindUniformBuffers

uniform buffer

uniform buffer dynamic

maxDescriptorSetUniformBuffersDynamic or maxDescriptorSetUpdateAfterBindUniformBuffersDynamic

uniform buffer dynamic

maxDescriptorSetStorageBuffers or maxDescriptorSetUpdateAfterBindStorageBuffers

storage buffer

storage buffer dynamic

maxDescriptorSetStorageBuffersDynamic or maxDescriptorSetUpdateAfterBindStorageBuffersDynamic

storage buffer dynamic

maxDescriptorSetInputAttachments or maxDescriptorSetUpdateAfterBindInputAttachments

input attachment

To destroy a pipeline layout, call:

void vkDestroyPipelineLayout(
    VkDevice                                    device,
    VkPipelineLayout                            pipelineLayout,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the pipeline layout.

  • pipelineLayout is the pipeline layout to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when pipelineLayout was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when pipelineLayout was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If pipelineLayout is not VK_NULL_HANDLE, pipelineLayout must be a valid VkPipelineLayout handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If pipelineLayout is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to pipelineLayout must be externally synchronized

Pipeline Layout Compatibility

Two pipeline layouts are defined to be “compatible for push constants” if they were created with identical push constant ranges. Two pipeline layouts are defined to be “compatible for set N” if they were created with identically defined descriptor set layouts for sets zero through N, and if they were created with identical push constant ranges.

When binding a descriptor set (see Descriptor Set Binding) to set number N, if the previously bound descriptor sets for sets zero through N-1 were all bound using compatible pipeline layouts, then performing this binding does not disturb any of the lower numbered sets. If, additionally, the previous bound descriptor set for set N was bound using a pipeline layout compatible for set N, then the bindings in sets numbered greater than N are also not disturbed.

Similarly, when binding a pipeline, the pipeline can correctly access any previously bound descriptor sets which were bound with compatible pipeline layouts, as long as all lower numbered sets were also bound with compatible layouts.

Layout compatibility means that descriptor sets can be bound to a command buffer for use by any pipeline created with a compatible pipeline layout, and without having bound a particular pipeline first. It also means that descriptor sets can remain valid across a pipeline change, and the same resources will be accessible to the newly bound pipeline.

Implementor’s Note

A consequence of layout compatibility is that when the implementation compiles a pipeline layout and maps pipeline resources to implementation resources, the mechanism for set N should only be a function of sets [0..N].

Note

Place the least frequently changing descriptor sets near the start of the pipeline layout, and place the descriptor sets representing the most frequently changing resources near the end. When pipelines are switched, only the descriptor set bindings that have been invalidated will need to be updated and the remainder of the descriptor set bindings will remain in place.

The maximum number of descriptor sets that can be bound to a pipeline layout is queried from physical device properties (see maxBoundDescriptorSets in Limits).

API example
const VkDescriptorSetLayout layouts[] = { layout1, layout2 };

const VkPushConstantRange ranges[] =
{
    {
        VK_PIPELINE_STAGE_VERTEX_SHADER_BIT,    // stageFlags
        0,                                      // offset
        4                                       // size
    },

    {
        VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT,  // stageFlags
        4,                                      // offset
        4                                       // size
    },
};

const VkPipelineLayoutCreateInfo createInfo =
{
    VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO,  // sType
    NULL,                                           // pNext
    0,                                              // flags
    2,                                              // setLayoutCount
    layouts,                                        // pSetLayouts
    2,                                              // pushConstantRangeCount
    ranges                                          // pPushConstantRanges
};

VkPipelineLayout myPipelineLayout;
myResult = vkCreatePipelineLayout(
    myDevice,
    &createInfo,
    NULL,
    &myPipelineLayout);

13.2.3. Allocation of Descriptor Sets

A descriptor pool maintains a pool of descriptors, from which descriptor sets are allocated. Descriptor pools are externally synchronized, meaning that the application must not allocate and/or free descriptor sets from the same pool in multiple threads simultaneously.

Descriptor pools are represented by VkDescriptorPool handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorPool)

To create a descriptor pool object, call:

VkResult vkCreateDescriptorPool(
    VkDevice                                    device,
    const VkDescriptorPoolCreateInfo*           pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorPool*                           pDescriptorPool);
  • device is the logical device that creates the descriptor pool.

  • pCreateInfo is a pointer to an instance of the VkDescriptorPoolCreateInfo structure specifying the state of the descriptor pool object.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pDescriptorPool points to a VkDescriptorPool handle in which the resulting descriptor pool object is returned.

pAllocator controls host memory allocation as described in the Memory Allocation chapter.

The created descriptor pool is returned in pDescriptorPool.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkDescriptorPoolCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pDescriptorPool must be a valid pointer to a VkDescriptorPool handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_FRAGMENTATION_EXT

Additional information about the pool is passed in an instance of the VkDescriptorPoolCreateInfo structure:

typedef struct VkDescriptorPoolCreateInfo {
    VkStructureType                sType;
    const void*                    pNext;
    VkDescriptorPoolCreateFlags    flags;
    uint32_t                       maxSets;
    uint32_t                       poolSizeCount;
    const VkDescriptorPoolSize*    pPoolSizes;
} VkDescriptorPoolCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkDescriptorPoolCreateFlagBits specifying certain supported operations on the pool.

  • maxSets is the maximum number of descriptor sets that can be allocated from the pool.

  • poolSizeCount is the number of elements in pPoolSizes.

  • pPoolSizes is a pointer to an array of VkDescriptorPoolSize structures, each containing a descriptor type and number of descriptors of that type to be allocated in the pool.

If multiple VkDescriptorPoolSize structures appear in the pPoolSizes array then the pool will be created with enough storage for the total number of descriptors of each type.

Fragmentation of a descriptor pool is possible and may lead to descriptor set allocation failures. A failure due to fragmentation is defined as failing a descriptor set allocation despite the sum of all outstanding descriptor set allocations from the pool plus the requested allocation requiring no more than the total number of descriptors requested at pool creation. Implementations provide certain guarantees of when fragmentation must not cause allocation failure, as described below.

If a descriptor pool has not had any descriptor sets freed since it was created or most recently reset then fragmentation must not cause an allocation failure (note that this is always the case for a pool created without the VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT bit set). Additionally, if all sets allocated from the pool since it was created or most recently reset use the same number of descriptors (of each type) and the requested allocation also uses that same number of descriptors (of each type), then fragmentation must not cause an allocation failure.

If an allocation failure occurs due to fragmentation, an application can create an additional descriptor pool to perform further descriptor set allocations.

If flags has the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT bit set, descriptor pool creation may fail with the error VK_ERROR_FRAGMENTATION_EXT if the total number of descriptors across all pools (including this one) created with this bit set exceeds maxUpdateAfterBindDescriptorsInAllPools, or if fragmentation of the underlying hardware resources occurs.

Valid Usage
  • maxSets must be greater than 0

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO

  • pNext must be NULL

  • flags must be a valid combination of VkDescriptorPoolCreateFlagBits values

  • pPoolSizes must be a valid pointer to an array of poolSizeCount valid VkDescriptorPoolSize structures

  • poolSizeCount must be greater than 0

Bits which can be set in VkDescriptorPoolCreateInfo::flags to enable operations on a descriptor pool are:

typedef enum VkDescriptorPoolCreateFlagBits {
    VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT = 0x00000001,
    VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT = 0x00000002,
} VkDescriptorPoolCreateFlagBits;
  • VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT specifies that descriptor sets can return their individual allocations to the pool, i.e. all of vkAllocateDescriptorSets, vkFreeDescriptorSets, and vkResetDescriptorPool are allowed. Otherwise, descriptor sets allocated from the pool must not be individually freed back to the pool, i.e. only vkAllocateDescriptorSets and vkResetDescriptorPool are allowed.

  • VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT specifies that descriptor sets allocated from this pool can include bindings with the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT bit set. It is valid to allocate descriptor sets that have bindings that don’t set the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT bit from a pool that has VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT set.

typedef VkFlags VkDescriptorPoolCreateFlags;

VkDescriptorPoolCreateFlags is a bitmask type for setting a mask of zero or more VkDescriptorPoolCreateFlagBits.

The VkDescriptorPoolSize structure is defined as:

typedef struct VkDescriptorPoolSize {
    VkDescriptorType    type;
    uint32_t            descriptorCount;
} VkDescriptorPoolSize;
  • type is the type of descriptor.

  • descriptorCount is the number of descriptors of that type to allocate.

Valid Usage
  • descriptorCount must be greater than 0

Valid Usage (Implicit)

To destroy a descriptor pool, call:

void vkDestroyDescriptorPool(
    VkDevice                                    device,
    VkDescriptorPool                            descriptorPool,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the descriptor pool.

  • descriptorPool is the descriptor pool to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

When a pool is destroyed, all descriptor sets allocated from the pool are implicitly freed and become invalid. Descriptor sets allocated from a given pool do not need to be freed before destroying that descriptor pool.

Valid Usage
  • All submitted commands that refer to descriptorPool (via any allocated descriptor sets) must have completed execution

  • If VkAllocationCallbacks were provided when descriptorPool was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when descriptorPool was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If descriptorPool is not VK_NULL_HANDLE, descriptorPool must be a valid VkDescriptorPool handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If descriptorPool is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to descriptorPool must be externally synchronized

Descriptor sets are allocated from descriptor pool objects, and are represented by VkDescriptorSet handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorSet)

To allocate descriptor sets from a descriptor pool, call:

VkResult vkAllocateDescriptorSets(
    VkDevice                                    device,
    const VkDescriptorSetAllocateInfo*          pAllocateInfo,
    VkDescriptorSet*                            pDescriptorSets);
  • device is the logical device that owns the descriptor pool.

  • pAllocateInfo is a pointer to an instance of the VkDescriptorSetAllocateInfo structure describing parameters of the allocation.

  • pDescriptorSets is a pointer to an array of VkDescriptorSet handles in which the resulting descriptor set objects are returned.

The allocated descriptor sets are returned in pDescriptorSets.

When a descriptor set is allocated, the initial state is largely uninitialized and all descriptors are undefined. However, the descriptor set can be bound in a command buffer without causing errors or exceptions. For descriptor set bindings created with the VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT bit set, all descriptors in that binding that are dynamically used must have been populated before the descriptor set is consumed. For descriptor set bindings created without the VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT bit set, all descriptors in that binding that are statically used must have been populated before the descriptor set is consumed. Entries that are not used by a pipeline can have uninitialized descriptors or descriptors of resources that have been destroyed, and executing a draw or dispatch with such a descriptor set bound does not cause undefined behavior. This means applications need not populate unused entries with dummy descriptors.

If a call to vkAllocateDescriptorSets would cause the total number of descriptor sets allocated from the pool to exceed the value of VkDescriptorPoolCreateInfo::maxSets used to create pAllocateInfodescriptorPool, then the allocation may fail due to lack of space in the descriptor pool. Similarly, the allocation may fail due to lack of space if the call to vkAllocateDescriptorSets would cause the number of any given descriptor type to exceed the sum of all the descriptorCount members of each element of VkDescriptorPoolCreateInfo::pPoolSizes with a member equal to that type. If the allocation fails due to no more space in the descriptor pool, and not because of system or device memory exhaustion, then VK_ERROR_OUT_OF_POOL_MEMORY must be returned.

vkAllocateDescriptorSets can be used to create multiple descriptor sets. If the creation of any of those descriptor sets fails, then the implementation must destroy all successfully created descriptor set objects from this command, set all entries of the pDescriptorSets array to VK_NULL_HANDLE and return the error.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pAllocateInfo must be a valid pointer to a valid VkDescriptorSetAllocateInfo structure

  • pDescriptorSets must be a valid pointer to an array of pAllocateInfo::descriptorSetCount VkDescriptorSet handles

Host Synchronization
  • Host access to pAllocateInfo::descriptorPool must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_FRAGMENTED_POOL

  • VK_ERROR_OUT_OF_POOL_MEMORY

The VkDescriptorSetAllocateInfo structure is defined as:

typedef struct VkDescriptorSetAllocateInfo {
    VkStructureType                 sType;
    const void*                     pNext;
    VkDescriptorPool                descriptorPool;
    uint32_t                        descriptorSetCount;
    const VkDescriptorSetLayout*    pSetLayouts;
} VkDescriptorSetAllocateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • descriptorPool is the pool which the sets will be allocated from.

  • descriptorSetCount determines the number of descriptor sets to be allocated from the pool.

  • pSetLayouts is an array of descriptor set layouts, with each member specifying how the corresponding descriptor set is allocated.

Valid Usage
  • Each element of pSetLayouts must not have been created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR set

  • If any element of pSetLayouts was created with the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set, descriptorPool must have been created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT flag set

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkDescriptorSetVariableDescriptorCountAllocateInfoEXT

  • descriptorPool must be a valid VkDescriptorPool handle

  • pSetLayouts must be a valid pointer to an array of descriptorSetCount valid VkDescriptorSetLayout handles

  • descriptorSetCount must be greater than 0

  • Both of descriptorPool, and the elements of pSetLayouts must have been created, allocated, or retrieved from the same VkDevice

If the pNext chain of a VkDescriptorSetAllocateInfo structure includes a VkDescriptorSetVariableDescriptorCountAllocateInfoEXT structure, then that structure includes an array of descriptor counts for variable descriptor count bindings, one for each descriptor set being allocated.

The VkDescriptorSetVariableDescriptorCountAllocateInfoEXT structure is defined as:

typedef struct VkDescriptorSetVariableDescriptorCountAllocateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           descriptorSetCount;
    const uint32_t*    pDescriptorCounts;
} VkDescriptorSetVariableDescriptorCountAllocateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • descriptorSetCount is zero or the number of elements in pDescriptorCounts.

  • pDescriptorCounts is an array of descriptor counts, with each member specifying the number of descriptors in a variable descriptor count binding in the corresponding descriptor set being allocated.

If descriptorSetCount is zero or this structure is not included in the pNext chain, then the variable lengths are considered to be zero. Otherwise, pDescriptorCounts[i] is the number of descriptors in the variable count descriptor binding in the corresponding descriptor set layout. If VkDescriptorSetAllocateInfo::pSetLayouts[i] does not include a variable count descriptor binding, then pDescriptorCounts[i] is ignored.

Valid Usage
  • If descriptorSetCount is not zero, descriptorSetCount must equal VkDescriptorSetAllocateInfo::descriptorSetCount

  • If VkDescriptorSetAllocateInfo::pSetLayouts[i] has a variable descriptor count binding, then pDescriptorCounts[i] must be less than or equal to the descriptor count specified for that binding when the descriptor set layout was created.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_ALLOCATE_INFO_EXT

  • If descriptorSetCount is not 0, pDescriptorCounts must be a valid pointer to an array of descriptorSetCount uint32_t values

To free allocated descriptor sets, call:

VkResult vkFreeDescriptorSets(
    VkDevice                                    device,
    VkDescriptorPool                            descriptorPool,
    uint32_t                                    descriptorSetCount,
    const VkDescriptorSet*                      pDescriptorSets);
  • device is the logical device that owns the descriptor pool.

  • descriptorPool is the descriptor pool from which the descriptor sets were allocated.

  • descriptorSetCount is the number of elements in the pDescriptorSets array.

  • pDescriptorSets is an array of handles to VkDescriptorSet objects.

After a successful call to vkFreeDescriptorSets, all descriptor sets in pDescriptorSets are invalid.

Valid Usage
  • All submitted commands that refer to any element of pDescriptorSets must have completed execution

  • pDescriptorSets must be a valid pointer to an array of descriptorSetCount VkDescriptorSet handles, each element of which must either be a valid handle or VK_NULL_HANDLE

  • Each valid handle in pDescriptorSets must have been allocated from descriptorPool

  • descriptorPool must have been created with the VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT flag

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • descriptorPool must be a valid VkDescriptorPool handle

  • descriptorSetCount must be greater than 0

  • descriptorPool must have been created, allocated, or retrieved from device

  • Each element of pDescriptorSets that is a valid handle must have been created, allocated, or retrieved from descriptorPool

Host Synchronization
  • Host access to descriptorPool must be externally synchronized

  • Host access to each member of pDescriptorSets must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

To return all descriptor sets allocated from a given pool to the pool, rather than freeing individual descriptor sets, call:

VkResult vkResetDescriptorPool(
    VkDevice                                    device,
    VkDescriptorPool                            descriptorPool,
    VkDescriptorPoolResetFlags                  flags);
  • device is the logical device that owns the descriptor pool.

  • descriptorPool is the descriptor pool to be reset.

  • flags is reserved for future use.

Resetting a descriptor pool recycles all of the resources from all of the descriptor sets allocated from the descriptor pool back to the descriptor pool, and the descriptor sets are implicitly freed.

Valid Usage
  • All uses of descriptorPool (via any allocated descriptor sets) must have completed execution

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • descriptorPool must be a valid VkDescriptorPool handle

  • flags must be 0

  • descriptorPool must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to descriptorPool must be externally synchronized

  • Host access to any VkDescriptorSet objects allocated from descriptorPool must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

typedef VkFlags VkDescriptorPoolResetFlags;

VkDescriptorPoolResetFlags is a bitmask type for setting a mask, but is currently reserved for future use.

13.2.4. Descriptor Set Updates

Once allocated, descriptor sets can be updated with a combination of write and copy operations. To update descriptor sets, call:

void vkUpdateDescriptorSets(
    VkDevice                                    device,
    uint32_t                                    descriptorWriteCount,
    const VkWriteDescriptorSet*                 pDescriptorWrites,
    uint32_t                                    descriptorCopyCount,
    const VkCopyDescriptorSet*                  pDescriptorCopies);
  • device is the logical device that updates the descriptor sets.

  • descriptorWriteCount is the number of elements in the pDescriptorWrites array.

  • pDescriptorWrites is a pointer to an array of VkWriteDescriptorSet structures describing the descriptor sets to write to.

  • descriptorCopyCount is the number of elements in the pDescriptorCopies array.

  • pDescriptorCopies is a pointer to an array of VkCopyDescriptorSet structures describing the descriptor sets to copy between.

The operations described by pDescriptorWrites are performed first, followed by the operations described by pDescriptorCopies. Within each array, the operations are performed in the order they appear in the array.

Each element in the pDescriptorWrites array describes an operation updating the descriptor set using descriptors for resources specified in the structure.

Each element in the pDescriptorCopies array is a VkCopyDescriptorSet structure describing an operation copying descriptors between sets.

If the dstSet member of any element of pDescriptorWrites or pDescriptorCopies is bound, accessed, or modified by any command that was recorded to a command buffer which is currently in the recording or executable state, and any of the descriptor bindings that are updated were not created with the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT or VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT bits set, that command buffer becomes invalid.

Valid Usage
  • Descriptor bindings updated by this command which were created without the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT or VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT bits set must not be used by any command that was recorded to a command buffer which is in the pending state.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If descriptorWriteCount is not 0, pDescriptorWrites must be a valid pointer to an array of descriptorWriteCount valid VkWriteDescriptorSet structures

  • If descriptorCopyCount is not 0, pDescriptorCopies must be a valid pointer to an array of descriptorCopyCount valid VkCopyDescriptorSet structures

Host Synchronization
  • Host access to pDescriptorWrites[].dstSet must be externally synchronized

  • Host access to pDescriptorCopies[].dstSet must be externally synchronized

The VkWriteDescriptorSet structure is defined as:

typedef struct VkWriteDescriptorSet {
    VkStructureType                  sType;
    const void*                      pNext;
    VkDescriptorSet                  dstSet;
    uint32_t                         dstBinding;
    uint32_t                         dstArrayElement;
    uint32_t                         descriptorCount;
    VkDescriptorType                 descriptorType;
    const VkDescriptorImageInfo*     pImageInfo;
    const VkDescriptorBufferInfo*    pBufferInfo;
    const VkBufferView*              pTexelBufferView;
} VkWriteDescriptorSet;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • dstSet is the destination descriptor set to update.

  • dstBinding is the descriptor binding within that set.

  • dstArrayElement is the starting element in that array.

  • descriptorCount is the number of descriptors to update (the number of elements in pImageInfo, pBufferInfo, or pTexelBufferView).

  • descriptorType is a VkDescriptorType specifying the type of each descriptor in pImageInfo, pBufferInfo, or pTexelBufferView, as described below. It must be the same type as that specified in VkDescriptorSetLayoutBinding for dstSet at dstBinding. The type of the descriptor also controls which array the descriptors are taken from.

  • pImageInfo points to an array of VkDescriptorImageInfo structures or is ignored, as described below.

  • pBufferInfo points to an array of VkDescriptorBufferInfo structures or is ignored, as described below.

  • pTexelBufferView points to an array of VkBufferView handles as described in the Buffer Views section or is ignored, as described below.

Only one of pImageInfo, pBufferInfo, or pTexelBufferView members is used according to the descriptor type specified in the descriptorType member of the containing VkWriteDescriptorSet structure, as specified below.

If the dstBinding has fewer than descriptorCount array elements remaining starting from dstArrayElement, then the remainder will be used to update the subsequent binding - dstBinding+1 starting at array element zero. If a binding has a descriptorCount of zero, it is skipped. This behavior applies recursively, with the update affecting consecutive bindings as needed to update all descriptorCount descriptors.

Valid Usage
  • dstBinding must be less than or equal to the maximum value of binding of all VkDescriptorSetLayoutBinding structures specified when dstSet’s descriptor set layout was created

  • dstBinding must be a binding with a non-zero descriptorCount

  • All consecutive bindings updated via a single VkWriteDescriptorSet structure, except those with a descriptorCount of zero, must have identical descriptorType and stageFlags.

  • All consecutive bindings updated via a single VkWriteDescriptorSet structure, except those with a descriptorCount of zero, must all either use immutable samplers or must all not use immutable samplers.

  • descriptorType must match the type of dstBinding within dstSet

  • dstSet must be a valid VkDescriptorSet handle

  • The sum of dstArrayElement and descriptorCount must be less than or equal to the number of array elements in the descriptor set binding specified by dstBinding, and all applicable consecutive bindings, as described by consecutive binding updates

  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, pImageInfo must be a valid pointer to an array of descriptorCount valid VkDescriptorImageInfo structures

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER, pTexelBufferView must be a valid pointer to an array of descriptorCount valid VkBufferView handles

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER, VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, pBufferInfo must be a valid pointer to an array of descriptorCount valid VkDescriptorBufferInfo structures

  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and dstSet was not allocated with a layout that included immutable samplers for dstBinding with descriptorType, the sampler member of each element of pImageInfo must be a valid VkSampler object

  • If descriptorType is VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, the imageView and imageLayout members of each element of pImageInfo must be a valid VkImageView and VkImageLayout, respectively

  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, then the imageView member of each pImageInfo element must have been created without a VkSamplerYcbcrConversionInfo structure in its pNext chain

  • If descriptorType is VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and if any element of pImageInfo has a imageView member that was created with a VkSamplerYcbcrConversionInfo structure in its pNext chain, then dstSet must have been allocated with a layout that included immutable samplers for dstBinding

  • If descriptorType is VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and dstSet was allocated with a layout that included immutable samplers for dstBinding, then the imageView member of each element of pImageInfo which corresponds to a immutable sampler that enables sampler Y’CBCR conversion must have been created with a VkSamplerYcbcrConversionInfo structure in its pNext chain with an identically defined VkSamplerYcbcrConversionInfo to the corresponding immutable sampler

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, for each descriptor that will be accessed via load or store operations the imageLayout member for corresponding elements of pImageInfo must be VK_IMAGE_LAYOUT_GENERAL

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, the offset member of each element of pBufferInfo must be a multiple of VkPhysicalDeviceLimits::minUniformBufferOffsetAlignment

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, the offset member of each element of pBufferInfo must be a multiple of VkPhysicalDeviceLimits::minStorageBufferOffsetAlignment

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, and the buffer member of any element of pBufferInfo is the handle of a non-sparse buffer, then that buffer must be bound completely and contiguously to a single VkDeviceMemory object

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, the buffer member of each element of pBufferInfo must have been created with VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, the buffer member of each element of pBufferInfo must have been created with VK_BUFFER_USAGE_STORAGE_BUFFER_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, the range member of each element of pBufferInfo, or the effective range if range is VK_WHOLE_SIZE, must be less than or equal to VkPhysicalDeviceLimits::maxUniformBufferRange

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, the range member of each element of pBufferInfo, or the effective range if range is VK_WHOLE_SIZE, must be less than or equal to VkPhysicalDeviceLimits::maxStorageBufferRange

  • If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER, the VkBuffer that each element of pTexelBufferView was created from must have been created with VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER, the VkBuffer that each element of pTexelBufferView was created from must have been created with VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_IMAGE or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, the imageView member of each element of pImageInfo must have been created with the identity swizzle

  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, the imageView member of each element of pImageInfo must have been created with VK_IMAGE_USAGE_SAMPLED_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, the imageLayout member of each element of pImageInfo must be VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • If descriptorType is VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, the imageView member of each element of pImageInfo must have been created with VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT set

  • If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, the imageView member of each element of pImageInfo must have been created with VK_IMAGE_USAGE_STORAGE_BIT set

  • All consecutive bindings updated via a single VkWriteDescriptorSet structure, except those with a descriptorCount of zero, must have identical VkDescriptorBindingFlagBitsEXT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET

  • pNext must be NULL

  • descriptorType must be a valid VkDescriptorType value

  • descriptorCount must be greater than 0

  • Both of dstSet, and the elements of pTexelBufferView that are valid handles must have been created, allocated, or retrieved from the same VkDevice

The type of descriptors in a descriptor set is specified by VkWriteDescriptorSet::descriptorType, which must be one of the values:

typedef enum VkDescriptorType {
    VK_DESCRIPTOR_TYPE_SAMPLER = 0,
    VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER = 1,
    VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE = 2,
    VK_DESCRIPTOR_TYPE_STORAGE_IMAGE = 3,
    VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER = 4,
    VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER = 5,
    VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER = 6,
    VK_DESCRIPTOR_TYPE_STORAGE_BUFFER = 7,
    VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC = 8,
    VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC = 9,
    VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT = 10,
} VkDescriptorType;

When a descriptor set is updated via elements of VkWriteDescriptorSet, members of pImageInfo, pBufferInfo and pTexelBufferView are only accessed by the implementation when they correspond to descriptor type being defined - otherwise they are ignored. The members accessed are as follows for each descriptor type:

  • For VK_DESCRIPTOR_TYPE_SAMPLER, only the sample member of each element of VkWriteDescriptorSet::pImageInfo is accessed.

  • For VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT, only the imageView and imageLayout members of each element of VkWriteDescriptorSet::pImageInfo are accessed.

  • For VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, all members of each element of VkWriteDescriptorSet::pImageInfo are accessed.

  • For VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER, VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, all members of each element of VkWriteDescriptorSet::pBufferInfo are accessed.

  • For VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER, each element of VkWriteDescriptorSet::pTexelBufferView is accessed.

The VkDescriptorBufferInfo structure is defined as:

typedef struct VkDescriptorBufferInfo {
    VkBuffer        buffer;
    VkDeviceSize    offset;
    VkDeviceSize    range;
} VkDescriptorBufferInfo;
  • buffer is the buffer resource.

  • offset is the offset in bytes from the start of buffer. Access to buffer memory via this descriptor uses addressing that is relative to this starting offset.

  • range is the size in bytes that is used for this descriptor update, or VK_WHOLE_SIZE to use the range from offset to the end of the buffer.

Note

When setting range to VK_WHOLE_SIZE, the effective range must not be larger than the maximum range for the descriptor type (maxUniformBufferRange or maxStorageBufferRange). This means that VK_WHOLE_SIZE is not typically useful in the common case where uniform buffer descriptors are suballocated from a buffer that is much larger than maxUniformBufferRange.

For VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC and VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC descriptor types, offset is the base offset from which the dynamic offset is applied and range is the static size used for all dynamic offsets.

Valid Usage
  • offset must be less than the size of buffer

  • If range is not equal to VK_WHOLE_SIZE, range must be greater than 0

  • If range is not equal to VK_WHOLE_SIZE, range must be less than or equal to the size of buffer minus offset

Valid Usage (Implicit)
  • buffer must be a valid VkBuffer handle

The VkDescriptorImageInfo structure is defined as:

typedef struct VkDescriptorImageInfo {
    VkSampler        sampler;
    VkImageView      imageView;
    VkImageLayout    imageLayout;
} VkDescriptorImageInfo;
  • sampler is a sampler handle, and is used in descriptor updates for types VK_DESCRIPTOR_TYPE_SAMPLER and VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER if the binding being updated does not use immutable samplers.

  • imageView is an image view handle, and is used in descriptor updates for types VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT.

  • imageLayout is the layout that the image subresources accessible from imageView will be in at the time this descriptor is accessed. imageLayout is used in descriptor updates for types VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, and VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT.

Members of VkDescriptorImageInfo that are not used in an update (as described above) are ignored.

Valid Usage
  • imageView must not be 2D or 2D array image view created from a 3D image

  • imageLayout must match the actual VkImageLayout of each subresource accessible from imageView at the time this descriptor is accessed

  • If sampler is used and the VkFormat of the image is a multi-planar format, the image must have been created with VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT, and the aspectMask of the imageView must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT or (for three-plane formats only) VK_IMAGE_ASPECT_PLANE_2_BIT

Valid Usage (Implicit)
  • Both of imageView, and sampler that are valid handles must have been created, allocated, or retrieved from the same VkDevice

The VkCopyDescriptorSet structure is defined as:

typedef struct VkCopyDescriptorSet {
    VkStructureType    sType;
    const void*        pNext;
    VkDescriptorSet    srcSet;
    uint32_t           srcBinding;
    uint32_t           srcArrayElement;
    VkDescriptorSet    dstSet;
    uint32_t           dstBinding;
    uint32_t           dstArrayElement;
    uint32_t           descriptorCount;
} VkCopyDescriptorSet;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • srcSet, srcBinding, and srcArrayElement are the source set, binding, and array element, respectively.

  • dstSet, dstBinding, and dstArrayElement are the destination set, binding, and array element, respectively.

  • descriptorCount is the number of descriptors to copy from the source to destination. If descriptorCount is greater than the number of remaining array elements in the source or destination binding, those affect consecutive bindings in a manner similar to VkWriteDescriptorSet above.

Valid Usage
  • srcBinding must be a valid binding within srcSet

  • The sum of srcArrayElement and descriptorCount must be less than or equal to the number of array elements in the descriptor set binding specified by srcBinding, and all applicable consecutive bindings, as described by consecutive binding updates

  • dstBinding must be a valid binding within dstSet

  • The sum of dstArrayElement and descriptorCount must be less than or equal to the number of array elements in the descriptor set binding specified by dstBinding, and all applicable consecutive bindings, as described by consecutive binding updates

  • If srcSet is equal to dstSet, then the source and destination ranges of descriptors must not overlap, where the ranges may include array elements from consecutive bindings as described by consecutive binding updates

  • If srcSet’s layout was created with the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT flag set, then dstSet’s layout must also have been created with the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT flag set

  • If srcSet’s layout was created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT flag set, then dstSet’s layout must also have been created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT flag set

  • If the descriptor pool from which srcSet was allocated was created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT flag set, then the descriptor pool from which dstSet was allocated must also have been created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT flag set

  • If the descriptor pool from which srcSet was allocated was created without the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT flag set, then the descriptor pool from which dstSet was allocated must also have been created without the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT flag set

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_COPY_DESCRIPTOR_SET

  • pNext must be NULL

  • srcSet must be a valid VkDescriptorSet handle

  • dstSet must be a valid VkDescriptorSet handle

  • Both of dstSet, and srcSet must have been created, allocated, or retrieved from the same VkDevice

13.2.5. Descriptor Update Templates

A descriptor update template specifies a mapping from descriptor update information in host memory to descriptors in a descriptor set. It is designed to avoid passing redundant information to the driver when frequently updating the same set of descriptors in descriptor sets.

Descriptor update template objects are represented by VkDescriptorUpdateTemplate handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorUpdateTemplate)

or the equivalent

typedef VkDescriptorUpdateTemplate VkDescriptorUpdateTemplateKHR;

13.2.6. Descriptor Set Updates with Templates

Updating a large VkDescriptorSet array can be an expensive operation since an application must specify one VkWriteDescriptorSet structure for each descriptor or descriptor array to update, each of which re-specifies the same state when updating the same descriptor in multiple descriptor sets. For cases when an application wishes to update the same set of descriptors in multiple descriptor sets allocated using the same VkDescriptorSetLayout, vkUpdateDescriptorSetWithTemplate can be used as a replacement for vkUpdateDescriptorSets.

VkDescriptorUpdateTemplate allows implementations to convert a set of descriptor update operations on a single descriptor set to an internal format that, in conjunction with vkUpdateDescriptorSetWithTemplate or vkCmdPushDescriptorSetWithTemplateKHR , can be more efficient compared to calling vkUpdateDescriptorSets or vkCmdPushDescriptorSetKHR . The descriptors themselves are not specified in the VkDescriptorUpdateTemplate, rather, offsets into an application provided pointer to host memory are specified, which are combined with a pointer passed to vkUpdateDescriptorSetWithTemplate or vkCmdPushDescriptorSetWithTemplateKHR . This allows large batches of updates to be executed without having to convert application data structures into a strictly-defined Vulkan data structure.

To create a descriptor update template, call:

VkResult vkCreateDescriptorUpdateTemplate(
    VkDevice                                    device,
    const VkDescriptorUpdateTemplateCreateInfo* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorUpdateTemplate*                 pDescriptorUpdateTemplate);

or the equivalent command

VkResult vkCreateDescriptorUpdateTemplateKHR(
    VkDevice                                    device,
    const VkDescriptorUpdateTemplateCreateInfo* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorUpdateTemplate*                 pDescriptorUpdateTemplate);
Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkDescriptorUpdateTemplateCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pDescriptorUpdateTemplate must be a valid pointer to a VkDescriptorUpdateTemplate handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDescriptorUpdateTemplateCreateInfo structure is defined as:

typedef struct VkDescriptorUpdateTemplateCreateInfo {
    VkStructureType                           sType;
    void*                                     pNext;
    VkDescriptorUpdateTemplateCreateFlags     flags;
    uint32_t                                  descriptorUpdateEntryCount;
    const VkDescriptorUpdateTemplateEntry*    pDescriptorUpdateEntries;
    VkDescriptorUpdateTemplateType            templateType;
    VkDescriptorSetLayout                     descriptorSetLayout;
    VkPipelineBindPoint                       pipelineBindPoint;
    VkPipelineLayout                          pipelineLayout;
    uint32_t                                  set;
} VkDescriptorUpdateTemplateCreateInfo;

or the equivalent

typedef VkDescriptorUpdateTemplateCreateInfo VkDescriptorUpdateTemplateCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • descriptorUpdateEntryCount is the number of elements in the pDescriptorUpdateEntries array.

  • pDescriptorUpdateEntries is a pointer to an array of VkDescriptorUpdateTemplateEntry structures describing the descriptors to be updated by the descriptor update template.

  • templateType Specifies the type of the descriptor update template. If set to VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET it can only be used to update descriptor sets with a fixed descriptorSetLayout. If set to VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR it can only be used to push descriptor sets using the provided pipelineBindPoint, pipelineLayout, and set number.

  • descriptorSetLayout is the descriptor set layout the parameter update template will be used with. All descriptor sets which are going to be updated through the newly created descriptor update template must be created with this layout. descriptorSetLayout is the descriptor set layout used to build the descriptor update template. All descriptor sets which are going to be updated through the newly created descriptor update template must be created with a layout that matches (is the same as, or defined identically to) this layout. This parameter is ignored if templateType is not VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET.

  • pipelineBindPoint is a VkPipelineBindPoint indicating whether the descriptors will be used by graphics pipelines or compute pipelines. This parameter is ignored if templateType is not VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR

  • pipelineLayout is a VkPipelineLayout object used to program the bindings. This parameter is ignored if templateType is not VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR

  • set is the set number of the descriptor set in the pipeline layout that will be updated. This parameter is ignored if templateType is not VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR

Valid Usage
  • If templateType is VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET, descriptorSetLayout must be a valid VkDescriptorSetLayout handle

  • If templateType is VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR, pipelineBindPoint must be a valid VkPipelineBindPoint value

  • If templateType is VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR, pipelineLayout must be a valid VkPipelineLayout handle

  • If templateType is VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR, set must be the unique set number in the pipeline layout that uses a descriptor set layout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • pDescriptorUpdateEntries must be a valid pointer to an array of descriptorUpdateEntryCount valid VkDescriptorUpdateTemplateEntry structures

  • templateType must be a valid VkDescriptorUpdateTemplateType value

  • If descriptorSetLayout is not VK_NULL_HANDLE, descriptorSetLayout must be a valid VkDescriptorSetLayout handle

  • descriptorUpdateEntryCount must be greater than 0

  • Both of descriptorSetLayout, and pipelineLayout that are valid handles must have been created, allocated, or retrieved from the same VkDevice

typedef VkFlags VkDescriptorUpdateTemplateCreateFlags;

or the equivalent

typedef VkDescriptorUpdateTemplateCreateFlags VkDescriptorUpdateTemplateCreateFlagsKHR;

VkDescriptorUpdateTemplateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The descriptor update template type is determined by the VkDescriptorUpdateTemplateCreateInfo::templateType property, which takes the following values:

typedef enum VkDescriptorUpdateTemplateType {
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET = 0,
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR = 1,
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET_KHR = VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET,
} VkDescriptorUpdateTemplateType;

or the equivalent

typedef VkDescriptorUpdateTemplateType VkDescriptorUpdateTemplateTypeKHR;
  • VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET specifies that the descriptor update template will be used for descriptor set updates only.

  • VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR specifies that the descriptor update template will be used for push descriptor updates only.

The VkDescriptorUpdateTemplateEntry structure is defined as:

typedef struct VkDescriptorUpdateTemplateEntry {
    uint32_t            dstBinding;
    uint32_t            dstArrayElement;
    uint32_t            descriptorCount;
    VkDescriptorType    descriptorType;
    size_t              offset;
    size_t              stride;
} VkDescriptorUpdateTemplateEntry;

or the equivalent

typedef VkDescriptorUpdateTemplateEntry VkDescriptorUpdateTemplateEntryKHR;
  • dstBinding is the descriptor binding to update when using this descriptor update template.

  • dstArrayElement is the starting element in the array belonging to dstBinding.

  • descriptorCount is the number of descriptors to update. If descriptorCount is greater than the number of remaining array elements in the destination binding, those affect consecutive bindings in a manner similar to VkWriteDescriptorSet above.

  • descriptorType is a VkDescriptorType specifying the type of the descriptor.

  • offset is the offset in bytes of the first binding in the raw data structure.

  • stride is the stride in bytes between two consecutive array elements of the descriptor update informations in the raw data structure. The actual pointer ptr for each array element j of update entry i is computed using the following formula:

        const char *ptr = (const char *)pData + pDescriptorUpdateEntries[i].offset + j * pDescriptorUpdateEntries[i].stride

    The stride is useful in case the bindings are stored in structs along with other data.

Valid Usage
  • dstBinding must be a valid binding in the descriptor set layout implicitly specified when using a descriptor update template to update descriptors.

  • dstArrayElement and descriptorCount must be less than or equal to the number of array elements in the descriptor set binding implicitly specified when using a descriptor update template to update descriptors, and all applicable consecutive bindings, as described by consecutive binding updates

Valid Usage (Implicit)

To destroy a descriptor update template, call:

void vkDestroyDescriptorUpdateTemplate(
    VkDevice                                    device,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const VkAllocationCallbacks*                pAllocator);

or the equivalent command

void vkDestroyDescriptorUpdateTemplateKHR(
    VkDevice                                    device,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that has been used to create the descriptor update template

  • descriptorUpdateTemplate is the descriptor update template to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when descriptorSetLayout was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when descriptorSetLayout was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If descriptorUpdateTemplate is not VK_NULL_HANDLE, descriptorUpdateTemplate must be a valid VkDescriptorUpdateTemplate handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If descriptorUpdateTemplate is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to descriptorUpdateTemplate must be externally synchronized

Once a VkDescriptorUpdateTemplate has been created, descriptor sets can be updated by calling:

void vkUpdateDescriptorSetWithTemplate(
    VkDevice                                    device,
    VkDescriptorSet                             descriptorSet,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const void*                                 pData);

or the equivalent command

void vkUpdateDescriptorSetWithTemplateKHR(
    VkDevice                                    device,
    VkDescriptorSet                             descriptorSet,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const void*                                 pData);
  • device is the logical device that updates the descriptor sets.

  • descriptorSet is the descriptor set to update

  • descriptorUpdateTemplate is the VkDescriptorUpdateTemplate which specifies the update mapping between pData and the descriptor set to update.

  • pData is a pointer to memory which contains one or more structures of VkDescriptorImageInfo, VkDescriptorBufferInfo, or VkBufferView used to write the descriptors.

Valid Usage
Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • descriptorSet must be a valid VkDescriptorSet handle

  • descriptorUpdateTemplate must be a valid VkDescriptorUpdateTemplate handle

  • descriptorUpdateTemplate must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to descriptorSet must be externally synchronized

API example
struct AppBufferView {
    VkBufferView bufferView;
    uint32_t     applicationRelatedInformation;
};

struct AppDataStructure
{
    VkDescriptorImageInfo  imageInfo;          // a single image info
    VkDescriptorBufferInfo bufferInfoArray[3]; // 3 buffer infos in an array
    AppBufferView          bufferView[2];      // An application defined structure containing a bufferView
    // ... some more application related data
};

const VkDescriptorUpdateTemplateEntry descriptorUpdateTemplateEntries[] =
{
    // binding to a single image descriptor
    {
        0,                                           // binding
        0,                                           // dstArrayElement
        1,                                           // descriptorCount
        VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,   // descriptorType
        offsetof(AppDataStructure, imageInfo),       // offset
        0                                            // stride is not required if descriptorCount is 1.
    },

    // binding to an array of buffer descriptors
    {
        0,                                           // binding
        0,                                           // dstArrayElement
        3,                                           // descriptorCount
        VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER,           // descriptorType
        offsetof(AppDataStructure, bufferInfoArray), // offset
        sizeof(VkDescriptorBufferInfo)               // stride, descriptor buffer infos are compact
    },

    // binding to an array of buffer views
    {
        0,                                           // binding
        3,                                           // dstArrayElement
        1,                                           // descriptorCount
        VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER,     // descriptorType
        offsetof(AppDataStructure, bufferView),      // offset
        sizeof(AppBufferView)                        // stride, bufferViews do not have to be compact
    },
};

// create an descriptor update template for descriptor set updates
const VkDescriptorUpdateTemplateCreateInfo createInfo =
{
    VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO,  // sType
    NULL,                                                          // pNext
    0,                                                             // flags
    3,                                                             // descriptorUpdateEntryCount
    descriptorUpdateTemplateEntries,                               // pDescriptorUpdateEntries
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET,         // templateType
    myLayout,                                                      // descriptorSetLayout
    0,                                                             // pipelineBindPoint, ignored by given templateType
    0,                                                             // pipelineLayout, ignored by given templateType
    0,                                                             // set, ignored by given templateType
};

VkDescriptorUpdateTemplate myDescriptorUpdateTemplate;
myResult = vkCreateDescriptorUpdateTemplate(
    myDevice,
    &createInfo,
    NULL,
    &myDescriptorUpdateTemplate);
}


AppDataStructure appData;

// fill appData here or cache it in your engine
vkUpdateDescriptorSetWithTemplate(myDevice, myDescriptorSet, myDescriptorUpdateTemplate, &appData);

13.2.7. Descriptor Set Binding

To bind one or more descriptor sets to a command buffer, call:

void vkCmdBindDescriptorSets(
    VkCommandBuffer                             commandBuffer,
    VkPipelineBindPoint                         pipelineBindPoint,
    VkPipelineLayout                            layout,
    uint32_t                                    firstSet,
    uint32_t                                    descriptorSetCount,
    const VkDescriptorSet*                      pDescriptorSets,
    uint32_t                                    dynamicOffsetCount,
    const uint32_t*                             pDynamicOffsets);
  • commandBuffer is the command buffer that the descriptor sets will be bound to.

  • pipelineBindPoint is a VkPipelineBindPoint indicating whether the descriptors will be used by graphics pipelines or compute pipelines. There is a separate set of bind points for each of graphics and compute, so binding one does not disturb the other.

  • layout is a VkPipelineLayout object used to program the bindings.

  • firstSet is the set number of the first descriptor set to be bound.

  • descriptorSetCount is the number of elements in the pDescriptorSets array.

  • pDescriptorSets is an array of handles to VkDescriptorSet objects describing the descriptor sets to write to.

  • dynamicOffsetCount is the number of dynamic offsets in the pDynamicOffsets array.

  • pDynamicOffsets is a pointer to an array of uint32_t values specifying dynamic offsets.

vkCmdBindDescriptorSets causes the sets numbered [firstSet.. firstSet+descriptorSetCount-1] to use the bindings stored in pDescriptorSets[0..descriptorSetCount-1] for subsequent rendering commands (either compute or graphics, according to the pipelineBindPoint). Any bindings that were previously applied via these sets are no longer valid.

Once bound, a descriptor set affects rendering of subsequent graphics or compute commands in the command buffer until a different set is bound to the same set number, or else until the set is disturbed as described in Pipeline Layout Compatibility.

A compatible descriptor set must be bound for all set numbers that any shaders in a pipeline access, at the time that a draw or dispatch command is recorded to execute using that pipeline. However, if none of the shaders in a pipeline statically use any bindings with a particular set number, then no descriptor set need be bound for that set number, even if the pipeline layout includes a non-trivial descriptor set layout for that set number.

If any of the sets being bound include dynamic uniform or storage buffers, then pDynamicOffsets includes one element for each array element in each dynamic descriptor type binding in each set. Values are taken from pDynamicOffsets in an order such that all entries for set N come before set N+1; within a set, entries are ordered by the binding numbers in the descriptor set layouts; and within a binding array, elements are in order. dynamicOffsetCount must equal the total number of dynamic descriptors in the sets being bound.

The effective offset used for dynamic uniform and storage buffer bindings is the sum of the relative offset taken from pDynamicOffsets, and the base address of the buffer plus base offset in the descriptor set. The length of the dynamic uniform and storage buffer bindings is the buffer range as specified in the descriptor set.

Each of the pDescriptorSets must be compatible with the pipeline layout specified by layout. The layout used to program the bindings must also be compatible with the pipeline used in subsequent graphics or compute commands, as defined in the Pipeline Layout Compatibility section.

The descriptor set contents bound by a call to vkCmdBindDescriptorSets may be consumed at the following times:

  • For descriptor bindings created with the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT bit set, the contents may be consumed when the command buffer is submitted to a queue, or during shader execution of the resulting draws and dispatches, or any time in between. Otherwise,

  • during host execution of the command, or during shader execution of the resulting draws and dispatches, or any time in between.

Thus, the contents of a descriptor set binding must not be altered (overwritten by an update command, or freed) between the first point in time that it may be consumed, and when the command completes executing on the queue.

The contents of pDynamicOffsets are consumed immediately during execution of vkCmdBindDescriptorSets. Once all pending uses have completed, it is legal to update and reuse a descriptor set.

Valid Usage
  • Each element of pDescriptorSets must have been allocated with a VkDescriptorSetLayout that matches (is the same as, or identically defined as) the VkDescriptorSetLayout at set n in layout, where n is the sum of firstSet and the index into pDescriptorSets

  • dynamicOffsetCount must be equal to the total number of dynamic descriptors in pDescriptorSets

  • The sum of firstSet and descriptorSetCount must be less than or equal to VkPipelineLayoutCreateInfo::setLayoutCount provided when layout was created

  • pipelineBindPoint must be supported by the commandBuffer’s parent VkCommandPool’s queue family

  • Each element of pDynamicOffsets must satisfy the required alignment for the corresponding descriptor binding’s descriptor type

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pipelineBindPoint must be a valid VkPipelineBindPoint value

  • layout must be a valid VkPipelineLayout handle

  • pDescriptorSets must be a valid pointer to an array of descriptorSetCount valid VkDescriptorSet handles

  • If dynamicOffsetCount is not 0, pDynamicOffsets must be a valid pointer to an array of dynamicOffsetCount uint32_t values

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • descriptorSetCount must be greater than 0

  • Each of commandBuffer, layout, and the elements of pDescriptorSets must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

13.2.8. Push Descriptor Updates

In addition to allocating descriptor sets and binding them to a command buffer, an application can record descriptor updates into the command buffer.

To push descriptor updates into a command buffer, call:

void vkCmdPushDescriptorSetKHR(
    VkCommandBuffer                             commandBuffer,
    VkPipelineBindPoint                         pipelineBindPoint,
    VkPipelineLayout                            layout,
    uint32_t                                    set,
    uint32_t                                    descriptorWriteCount,
    const VkWriteDescriptorSet*                 pDescriptorWrites);
  • commandBuffer is the command buffer that the descriptors will be recorded in.

  • pipelineBindPoint is a VkPipelineBindPoint indicating whether the descriptors will be used by graphics pipelines or compute pipelines. There is a separate set of push descriptor bindings for each of graphics and compute, so binding one does not disturb the other.

  • layout is a VkPipelineLayout object used to program the bindings.

  • set is the set number of the descriptor set in the pipeline layout that will be updated.

  • descriptorWriteCount is the number of elements in the pDescriptorWrites array.

  • pDescriptorWrites is a pointer to an array of VkWriteDescriptorSet structures describing the descriptors to be updated.

Push descriptors are a small bank of descriptors whose storage is internally managed by the command buffer rather than being written into a descriptor set and later bound to a command buffer. Push descriptors allow for incremental updates of descriptors without managing the lifetime of descriptor sets.

When a command buffer begins recording, all push descriptors have undefined contents. Push descriptors can be updated incrementally and cause shaders to use the updated descriptors for subsequent rendering commands (either compute or graphics, according to the pipelineBindPoint) until the descriptor is overwritten, or else until the set is disturbed as described in Pipeline Layout Compatibility. When the set is disturbed or push descriptors with a different descriptor set layout are set, all push descriptors become invalid.

Valid descriptors must be pushed for all bindings that any shaders in a pipeline access, at the time that a draw or dispatch command is recorded to execute using that pipeline. This includes immutable sampler descriptors, which must be pushed before they are accessed by a pipeline. However, if none of the shaders in a pipeline statically use certain bindings in the push descriptor set, then those descriptors need not be valid.

Push descriptors do not use dynamic offsets. Instead, the corresponding non-dynamic descriptor types can be used and the offset member of VkDescriptorBufferInfo can be changed each time the descriptor is written.

Each element of pDescriptorWrites is interpreted as in VkWriteDescriptorSet, except the dstSet member is ignored.

To push an immutable sampler, use a VkWriteDescriptorSet with dstBinding and dstArrayElement selecting the immutable sampler’s binding. If the descriptor type is VK_DESCRIPTOR_TYPE_SAMPLER, the pImageInfo parameter is ignored and the immutable sampler is taken from the push descriptor set layout in the pipeline layout. If the descriptor type is VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, the sampler member of the pImageInfo parameter is ignored and the immutable sampler is taken from the push descriptor set layout in the pipeline layout.

Valid Usage
  • pipelineBindPoint must be supported by the commandBuffer’s parent VkCommandPool’s queue family

  • set must be less than VkPipelineLayoutCreateInfo::setLayoutCount provided when layout was created

  • set must be the unique set number in the pipeline layout that uses a descriptor set layout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pipelineBindPoint must be a valid VkPipelineBindPoint value

  • layout must be a valid VkPipelineLayout handle

  • pDescriptorWrites must be a valid pointer to an array of descriptorWriteCount valid VkWriteDescriptorSet structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • descriptorWriteCount must be greater than 0

  • Both of commandBuffer, and layout must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

13.2.9. Push Descriptor Updates with Descriptor Update Templates

It is also possible to use a descriptor update template to specify the push descriptors to update. To do so, call:

void vkCmdPushDescriptorSetWithTemplateKHR(
    VkCommandBuffer                             commandBuffer,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    VkPipelineLayout                            layout,
    uint32_t                                    set,
    const void*                                 pData);
  • commandBuffer is the command buffer that the descriptors will be recorded in.

  • descriptorUpdateTemplate A descriptor update template which defines how to interpret the descriptor information in pData.

  • layout is a VkPipelineLayout object used to program the bindings. It must be compatible with the layout used to create the descriptorUpdateTemplate handle.

  • set is the set number of the descriptor set in the pipeline layout that will be updated. This must be the same number used to create the descriptorUpdateTemplate handle.

  • pData Points to memory which contains the descriptors for the templated update.

Valid Usage
Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • descriptorUpdateTemplate must be a valid VkDescriptorUpdateTemplate handle

  • layout must be a valid VkPipelineLayout handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • Each of commandBuffer, descriptorUpdateTemplate, and layout must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

API example
struct AppBufferView {
    VkBufferView bufferView;
    uint32_t     applicationRelatedInformation;
};

struct AppDataStructure
{
    VkDescriptorImageInfo  imageInfo;          // a single image info
    // ... some more application related data
};

const VkDescriptorUpdateTemplateEntry descriptorUpdateTemplateEntries[] =
{
    // binding to a single image descriptor
    {
        0,                                           // binding
        0,                                           // dstArrayElement
        1,                                           // descriptorCount
        VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,   // descriptorType
        offsetof(AppDataStructure, imageInfo),       // offset
        0                                            // stride is not required if descriptorCount is 1.
    }

};

// create an descriptor update template for descriptor set updates
const VkDescriptorUpdateTemplateCreateInfo createInfo =
{
    VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO,  // sType
    NULL,                                                          // pNext
    0,                                                             // flags
    1,                                                             // descriptorUpdateEntryCount
    descriptorUpdateTemplateEntries,                               // pDescriptorUpdateEntries
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR,       // templateType
    0,                                                             // descriptorSetLayout, ignored by given templateType
    VK_PIPELINE_BIND_POINT_GRAPHICS,                               // pipelineBindPoint
    myPipelineLayout,                                              // pipelineLayout
    0,                                                             // set
};

VkDescriptorUpdateTemplate myDescriptorUpdateTemplate;
myResult = vkCreateDescriptorUpdateTemplate(
    myDevice,
    &createInfo,
    NULL,
    &myDescriptorUpdateTemplate);
}

AppDataStructure appData;
// fill appData here or cache it in your engine
vkCmdPushDescriptorSetWithTemplateKHR(myCmdBuffer, myDescriptorUpdateTemplate, myPipelineLayout, 0,&appData);

13.2.10. Push Constant Updates

As described above in section Pipeline Layouts, the pipeline layout defines shader push constants which are updated via Vulkan commands rather than via writes to memory or copy commands.

Note

Push constants represent a high speed path to modify constant data in pipelines that is expected to outperform memory-backed resource updates.

The values of push constants are undefined at the start of a command buffer.

To update push constants, call:

void vkCmdPushConstants(
    VkCommandBuffer                             commandBuffer,
    VkPipelineLayout                            layout,
    VkShaderStageFlags                          stageFlags,
    uint32_t                                    offset,
    uint32_t                                    size,
    const void*                                 pValues);
  • commandBuffer is the command buffer in which the push constant update will be recorded.

  • layout is the pipeline layout used to program the push constant updates.

  • stageFlags is a bitmask of VkShaderStageFlagBits specifying the shader stages that will use the push constants in the updated range.

  • offset is the start offset of the push constant range to update, in units of bytes.

  • size is the size of the push constant range to update, in units of bytes.

  • pValues is an array of size bytes containing the new push constant values.

Valid Usage
  • For each byte in the range specified by offset and size and for each shader stage in stageFlags, there must be a push constant range in layout that includes that byte and that stage

  • For each byte in the range specified by offset and size and for each push constant range that overlaps that byte, stageFlags must include all stages in that push constant range’s VkPushConstantRange::stageFlags

  • offset must be a multiple of 4

  • size must be a multiple of 4

  • offset must be less than VkPhysicalDeviceLimits::maxPushConstantsSize

  • size must be less than or equal to VkPhysicalDeviceLimits::maxPushConstantsSize minus offset

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • layout must be a valid VkPipelineLayout handle

  • stageFlags must be a valid combination of VkShaderStageFlagBits values

  • stageFlags must not be 0

  • pValues must be a valid pointer to an array of size bytes

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • size must be greater than 0

  • Both of commandBuffer, and layout must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

14. Shader Interfaces

When a pipeline is created, the set of shaders specified in the corresponding Vk*PipelineCreateInfo structure are implicitly linked at a number of different interfaces.

Interface definitions make use of the following SPIR-V decorations:

  • DescriptorSet and Binding

  • Location, Component, and Index

  • Flat, NoPerspective, Centroid, and Sample

  • Block and BufferBlock

  • InputAttachmentIndex

  • Offset, ArrayStride, and MatrixStride

  • BuiltIn

  • PassthroughNV

This specification describes valid uses for Vulkan of these decorations. Any other use of one of these decorations is invalid.

14.1. Shader Input and Output Interfaces

When multiple stages are present in a pipeline, the outputs of one stage form an interface with the inputs of the next stage. When such an interface involves a shader, shader outputs are matched against the inputs of the next stage, and shader inputs are matched against the outputs of the previous stage.

There are two classes of variables that can be matched between shader stages, built-in variables and user-defined variables. Each class has a different set of matching criteria. Generally, when non-shader stages are between shader stages, the user-defined variables, and most built-in variables, form an interface between the shader stages.

The variables forming the input or output interfaces are listed as operands to the OpEntryPoint instruction and are declared with the Input or Output storage classes, respectively, in the SPIR-V module.

Output variables of a shader stage have undefined values until the shader writes to them or uses the Initializer operand when declaring the variable.

14.1.1. Built-in Interface Block

Shader built-in variables meeting the following requirements define the built-in interface block. They must

  • be explicitly declared (there are no implicit built-ins),

  • be identified with a BuiltIn decoration,

  • form object types as described in the Built-in Variables section, and

  • be declared in a block whose top-level members are the built-ins.

Built-ins only participate in interface matching if they are declared in such a block. They must not have any Location or Component decorations.

There must be no more than one built-in interface block per shader per interface.

14.1.2. User-defined Variable Interface

The remaining variables listed by OpEntryPoint with the Input or Output storage class form the user-defined variable interface. By default such variables have a type with a width of 32 or 64. If an implementation supports storageInputOutput16, user-defined variables in the Input and Output storage classes can also have types with a width of 16. These variables must be identified with a Location decoration and can also be identified with a Component decoration.

14.1.3. Interface Matching

A user-defined output variable is considered to match an input variable in the subsequent stage if the two variables are declared with the same Location and Component decoration and match in type and decoration, except that interpolation decorations are not required to match. For the purposes of interface matching, variables declared without a Component decoration are considered to have a Component decoration of zero.

Note

Matching rules for passthrough geometry shaders are slightly different and are described in the Passthrough Interface Matching section.

Variables or block members declared as structures are considered to match in type if and only if the structure members match in type, decoration, number, and declaration order. Variables or block members declared as arrays are considered to match in type only if both declarations specify the same element type and size.

Tessellation control shader per-vertex output variables and blocks, and tessellation control, tessellation evaluation, and geometry shader per-vertex input variables and blocks are required to be declared as arrays, with each element representing input or output values for a single vertex of a multi-vertex primitive. For the purposes of interface matching, the outermost array dimension of such variables and blocks is ignored.

At an interface between two non-fragment shader stages, the built-in interface block must match exactly, as described above. At an interface involving the fragment shader inputs, the presence or absence of any built-in output does not affect the interface matching.

At an interface between two shader stages, the user-defined variable interface must match exactly, as described above.

Any input value to a shader stage is well-defined as long as the preceding stages writes to a matching output, as described above.

Additionally, scalar and vector inputs are well-defined if there is a corresponding output satisfying all of the following conditions:

  • the input and output match exactly in decoration,

  • the output is a vector with the same basic type and has at least as many components as the input, and

  • the common component type of the input and output is 16-bit integer or floating-point, or 32-bit integer or floating-point (64-bit component types are excluded).

In this case, the components of the input will be taken from the first components of the output, and any extra components of the output will be ignored.

14.1.4. Location Assignment

This section describes how many locations are consumed by a given type. As mentioned above, geometry shader inputs, tessellation control shader inputs and outputs, and tessellation evaluation inputs all have an additional level of arrayness relative to other shader inputs and outputs. This outer array level is removed from the type before considering how many locations the type consumes.

The Location value specifies an interface slot comprised of a 32-bit four-component vector conveyed between stages. The Component specifies components within these vector locations. Only types with widths of 16, 32 or 64 are supported in shader interfaces.

Inputs and outputs of the following types consume a single interface location:

  • 16-bit scalar and vector types, and

  • 32-bit scalar and vector types, and

  • 64-bit scalar and 2-component vector types.

64-bit three- and four-component vectors consume two consecutive locations.

If a declared input or output is an array of size n and each element takes m locations, it will be assigned m × n consecutive locations starting with the location specified.

If the declared input or output is an n × m 16-, 32- or 64-bit matrix, it will be assigned multiple locations starting with the location specified. The number of locations assigned for each matrix will be the same as for an n-element array of m-component vectors.

The layout of a structure type used as an Input or Output depends on whether it is also a Block (i.e. has a Block decoration).

If it is a not a Block, then the structure type must have a Location decoration. Its members are assigned consecutive locations in their declaration order, with the first member assigned to the location specified for the structure type. The members, and their nested types, must not themselves have Location decorations.

If the structure type is a Block but without a Location, then each of its members must have a Location decoration. If it is a Block with a Location decoration, then its members are assigned consecutive locations in declaration order, starting from the first member which is initially assigned the location specified for the Block. Any member with its own Location decoration is assigned that location. Each remaining member is assigned the location after the immediately preceding member in declaration order.

The locations consumed by block and structure members are determined by applying the rules above in a depth-first traversal of the instantiated members as though the structure or block member were declared as an input or output variable of the same type.

Any two inputs listed as operands on the same OpEntryPoint must not be assigned the same location, either explicitly or implicitly. Any two outputs listed as operands on the same OpEntryPoint must not be assigned the same location, either explicitly or implicitly.

The number of input and output locations available for a shader input or output interface are limited, and dependent on the shader stage as described in Shader Input and Output Locations.

Table 18. Shader Input and Output Locations
Shader Interface Locations Available

vertex input

maxVertexInputAttributes

vertex output

maxVertexOutputComponents / 4

tessellation control input

maxTessellationControlPerVertexInputComponents / 4

tessellation control output

maxTessellationControlPerVertexOutputComponents / 4

tessellation evaluation input

maxTessellationEvaluationInputComponents / 4

tessellation evaluation output

maxTessellationEvaluationOutputComponents / 4

geometry input

maxGeometryInputComponents / 4

geometry output

maxGeometryOutputComponents / 4

fragment input

maxFragmentInputComponents / 4

fragment output

maxFragmentOutputAttachments

14.1.5. Component Assignment

The Component decoration allows the Location to be more finely specified for scalars and vectors, down to the individual components within a location that are consumed. The components within a location are 0, 1, 2, and 3. A variable or block member starting at component N will consume components N, N+1, N+2, …​ up through its size. For 16-, and 32-bit types, it is invalid if this sequence of components gets larger than 3. A scalar 64-bit type will consume two of these components in sequence, and a two-component 64-bit vector type will consume all four components available within a location. A three- or four-component 64-bit vector type must not specify a Component decoration. A three-component 64-bit vector type will consume all four components of the first location and components 0 and 1 of the second location. This leaves components 2 and 3 available for other component-qualified declarations.

A scalar or two-component 64-bit data type must not specify a Component decoration of 1 or 3. A Component decoration must not be specified for any type that is not a scalar or vector.

14.2. Vertex Input Interface

When the vertex stage is present in a pipeline, the vertex shader input variables form an interface with the vertex input attributes. The vertex shader input variables are matched by the Location and Component decorations to the vertex input attributes specified in the pVertexInputState member of the VkGraphicsPipelineCreateInfo structure.

The vertex shader input variables listed by OpEntryPoint with the Input storage class form the vertex input interface. These variables must be identified with a Location decoration and can also be identified with a Component decoration.

For the purposes of interface matching: variables declared without a Component decoration are considered to have a Component decoration of zero. The number of available vertex input locations is given by the maxVertexInputAttributes member of the VkPhysicalDeviceLimits structure.

All vertex shader inputs declared as above must have a corresponding attribute and binding in the pipeline.

14.3. Fragment Output Interface

When the fragment stage is present in a pipeline, the fragment shader outputs form an interface with the output attachments of the current subpass. The fragment shader output variables are matched by the Location and Component decorations to the color attachments specified in the pColorAttachments array of the VkSubpassDescription structure that describes the subpass that the fragment shader is executed in.

The fragment shader output variables listed by OpEntryPoint with the Output storage class form the fragment output interface. These variables must be identified with a Location decoration. They can also be identified with a Component decoration and/or an Index decoration. For the purposes of interface matching: variables declared without a Component decoration are considered to have a Component decoration of zero, and variables declared without an Index decoration are considered to have an Index decoration of zero.

A fragment shader output variable identified with a Location decoration of i is directed to the color attachment indicated by pColorAttachments[i], after passing through the blending unit as described in Blending, if enabled. Locations are consumed as described in Location Assignment. The number of available fragment output locations is given by the maxFragmentOutputAttachments member of the VkPhysicalDeviceLimits structure.

Components of the output variables are assigned as described in Component Assignment. Output components identified as 0, 1, 2, and 3 will be directed to the R, G, B, and A inputs to the blending unit, respectively, or to the output attachment if blending is disabled. If two variables are placed within the same location, they must have the same underlying type (floating-point or integer). The input to blending or color attachment writes is undefined for components which do not correspond to a fragment shader output.

Fragment outputs identified with an Index of zero are directed to the first input of the blending unit associated with the corresponding Location. Outputs identified with an Index of one are directed to the second input of the corresponding blending unit.

No component aliasing of output variables is allowed, that is there must not be two output variables which have the same location, component, and index, either explicitly declared or implied.

Output values written by a fragment shader must be declared with either OpTypeFloat or OpTypeInt, and a Width of 32. If storageInputOutput16 is supported, output values written by a fragment shader can be also declared with either OpTypeFloat or OpTypeInt and a Width of 16. Composites of these types are also permitted. If the color attachment has a signed or unsigned normalized fixed-point format, color values are assumed to be floating-point and are converted to fixed-point as described in Conversion from Floating-Point to Normalized Fixed-Point; If the color attachment has an integer format, color values are assumed to be integers and converted to the bit-depth of the target. Any value that cannot be represented in the attachment’s format is undefined. For any other attachment format no conversion is performed. If the type of the values written by the fragment shader do not match the format of the corresponding color attachment, the result is undefined for those components.

14.4. Fragment Input Attachment Interface

When a fragment stage is present in a pipeline, the fragment shader subpass inputs form an interface with the input attachments of the current subpass. The fragment shader subpass input variables are matched by InputAttachmentIndex decorations to the input attachments specified in the pInputAttachments array of the VkSubpassDescription structure that describes the subpass that the fragment shader is executed in.

The fragment shader subpass input variables with the UniformConstant storage class and a decoration of InputAttachmentIndex that are statically used by OpEntryPoint form the fragment input attachment interface. These variables must be declared with a type of OpTypeImage, a Dim operand of SubpassData, and a Sampled operand of 2.

A subpass input variable identified with an InputAttachmentIndex decoration of i reads from the input attachment indicated by pInputAttachments[i] member of VkSubpassDescription. If the subpass input variable is declared as an array of size N, it consumes N consecutive input attachments, starting with the index specified. There must not be more than one input variable with the same InputAttachmentIndex whether explicitly declared or implied by an array declaration. The number of available input attachment indices is given by the maxPerStageDescriptorInputAttachments member of the VkPhysicalDeviceLimits structure.

Variables identified with the InputAttachmentIndex must only be used by a fragment stage. The basic data type (floating-point, integer, unsigned integer) of the subpass input must match the basic format of the corresponding input attachment, or the values of subpass loads from these variables are undefined.

See Input Attachment for more details.

14.5. Shader Resource Interface

When a shader stage accesses buffer or image resources, as described in the Resource Descriptors section, the shader resource variables must be matched with the pipeline layout that is provided at pipeline creation time.

The set of shader resources that form the shader resource interface for a stage are the variables statically used by OpEntryPoint with the storage class of Uniform, UniformConstant, or PushConstant. For the fragment shader, this includes the fragment input attachment interface.

The shader resource interface consists of two sub-interfaces: the push constant interface and the descriptor set interface.

14.5.1. Push Constant Interface

The shader variables defined with a storage class of PushConstant that are statically used by the shader entry points for the pipeline define the push constant interface. They must be:

  • typed as OpTypeStruct,

  • identified with a Block decoration, and

  • laid out explicitly using the Offset, ArrayStride, and MatrixStride decorations as specified in Offset and Stride Assignment.

There must be no more than one push constant block statically used per shader entry point.

Each variable in a push constant block must be placed at an Offset such that the entire constant value is entirely contained within the VkPushConstantRange for each OpEntryPoint that uses it, and the stageFlags for that range must specify the appropriate VkShaderStageFlagBits for that stage. The Offset decoration for any variable in a push constant block must not cause the space required for that variable to extend outside the range [0, maxPushConstantsSize).

Any variable in a push constant block that is declared as an array must only be accessed with dynamically uniform indices.

14.5.2. Descriptor Set Interface

The descriptor set interface is comprised of the shader variables with the storage class of Uniform or UniformConstant (including the variables in the fragment input attachment interface) that are statically used by the shader entry points for the pipeline.

These variables must have DescriptorSet and Binding decorations specified, which are assigned and matched with the VkDescriptorSetLayout objects in the pipeline layout as described in DescriptorSet and Binding Assignment.

Variables identified with the UniformConstant storage class are used only as handles to refer to opaque resources. Such variables must be typed as OpTypeImage, OpTypeSampler, OpTypeSampledImage, or an array of one of these types.

The Sampled Type of an OpTypeImage declaration must match the same basic data type as the corresponding resource, or the values obtained by reading or sampling from this image are undefined.

The Image Format of an OpTypeImage declaration must not be Unknown, for variables which are used for OpImageRead, OpImageSparseRead, or OpImageWrite operations, except under the following conditions:

  • For OpImageWrite, if the shaderStorageImageWriteWithoutFormat feature is enabled and the shader module declares the StorageImageWriteWithoutFormat capability.

  • For OpImageRead or OpImageSparseRead, if the shaderStorageImageReadWithoutFormat feature is enabled and the shader module declares the StorageImageReadWithoutFormat capability.

The Image Format of an OpTypeImage declaration must not be Unknown, for variables which are used for OpAtomic* operations.

Variables identified with the Uniform storage class are used to access transparent buffer backed resources. Such variables must be:

  • typed as OpTypeStruct, or an array of this type,

  • identified with a Block or BufferBlock decoration, and

  • laid out explicitly using the Offset, ArrayStride, and MatrixStride decorations as specified in Offset and Stride Assignment.

Variables identified with the StorageBuffer storage class are used to access transparent buffer backed resources. Such variables must be:

  • typed as OpTypeStruct, or an array of this type,

  • identified with a Block decoration, and

  • laid out explicitly using the Offset, ArrayStride, and MatrixStride decorations as specified in Offset and Stride Assignment.

The Offset decoration for any member of a Block-decorated variable in the Uniform storage class must not cause the space required for that variable to extend outside the range [0, maxUniformBufferRange). The Offset decoration for any member of a Block-decorated variable in the StorageBuffer storage class must not cause the space required for that variable to extend outside the range [0, maxStorageBufferRange).

Variables identified with a storage class of UniformConstant and a decoration of InputAttachmentIndex must be declared as described in Fragment Input Attachment Interface.

Each shader variable in the descriptor set interface must be of a type that corresponds to the descriptorType in the descriptor set layout binding that the variable is assigned to, as described in DescriptorSet and Binding Assignment. See Shader Resource and Descriptor Type Correspondence for the relationship between shader types and descriptor types.

SPIR-V variables decorated with a descriptor set and binding that identify a combined image sampler descriptor can have a type of OpTypeImage, OpTypeSampler (Sampled=1), or OpTypeSampledImage.

Arrays of any of these types can be indexed with constant integral expressions. The following features must be enabled and capabilities must be declared in order to index such arrays with dynamically uniform or non-uniform indices:

  • Storage images (except storage texel buffers and input attachments):

    • Dynamically uniform: shaderStorageImageArrayDynamicIndexing and StorageImageArrayDynamicIndexing

    • Non-uniform: shaderStorageImageArrayNonUniformIndexing and StorageImageArrayNonUniformIndexingEXT

  • Storage texel buffers:

    • Dynamically uniform: shaderStorageTexelBufferArrayDynamicIndexing and StorageTexelBufferArrayDynamicIndexingEXT

    • Non-uniform: shaderStorageTexelBufferArrayNonUniformIndexing and StorageTexelBufferArrayNonUniformIndexingEXT

  • Input attachments:

    • Dynamically uniform: shaderInputAttachmentArrayDynamicIndexing and InputAttachmentArrayDynamicIndexingEXT

    • Non-uniform: shaderInputAttachmentArrayNonUniformIndexing and InputAttachmentArrayNonUniformIndexingEXT

  • Sampled images (except uniform texel buffers):

    • Dynamically uniform: shaderSampledImageArrayDynamicIndexing and SampledImageArrayDynamicIndexing

    • Non-uniform: shaderSampledImageArrayNonUniformIndexing and SampledImageArrayNonUniformIndexingEXT

  • Uniform texel buffers:

    • Dynamically uniform: shaderUniformTexelBufferArrayDynamicIndexing and UniformTexelBufferArrayDynamicIndexingEXT

    • Non-uniform: shaderUniformTexelBufferArrayNonUniformIndexing and UniformTexelBufferArrayNonUniformIndexingEXT

  • Uniform buffers:

    • Dynamically uniform: shaderUniformBufferArrayDynamicIndexing and UniformBufferArrayDynamicIndexing

    • Non-uniform: shaderUniformBufferArrayNonUniformIndexing and UniformBufferArrayNonUniformIndexingEXT

  • Storage buffers:

    • Dynamically uniform: shaderStorageBufferArrayDynamicIndexing and StorageBufferArrayDynamicIndexing

    • Non-uniform: shaderStorageBufferArrayNonUniformIndexing and StorageBufferArrayNonUniformIndexingEXT

If an instruction loads from or stores to a resource (including atomics and image instructions) and the resource descriptor being accessed is not dynamically uniform, then the corresponding non-uniform indexing feature must be enabled and the capability must be declared. If an instruction loads from or stores to a resource (including atomics and image instructions) and the resource descriptor being accessed is not uniform, then the corresponding dynamic indexing or non-uniform feature must be enabled and the capability must be declared.

If sampler Y’CBCR conversion is enabled, the combined image sampler must be indexed only by constant integral expressions when aggregated into arrays in shader code, irrespective of the shaderSampledImageArrayDynamicIndexing feature.

Table 19. Shader Resource and Descriptor Type Correspondence
Resource type Descriptor Type

sampler

VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER

sampled image

VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER

storage image

VK_DESCRIPTOR_TYPE_STORAGE_IMAGE

combined image sampler

VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER

uniform texel buffer

VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER

storage texel buffer

VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER

uniform buffer

VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC

storage buffer

VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC

input attachment

VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT

Table 20. Shader Resource and Storage Class Correspondence
Resource type Storage Class Type Decoration(s)1

sampler

UniformConstant

OpTypeSampler

sampled image

UniformConstant

OpTypeImage (Sampled=1)

storage image

UniformConstant

OpTypeImage (Sampled=2)

combined image sampler

UniformConstant

OpTypeSampledImage
OpTypeImage (Sampled=1)
OpTypeSampler

uniform texel buffer

UniformConstant

OpTypeImage (Dim=Buffer, Sampled=1)

storage texel buffer

UniformConstant

OpTypeImage (Dim=Buffer, Sampled=2)

uniform buffer

Uniform

OpTypeStruct

Block, Offset, (ArrayStride), (MatrixStride)

storage buffer

Uniform

OpTypeStruct

BufferBlock, Offset, (ArrayStride), (MatrixStride)

StorageBuffer

Block, Offset, (ArrayStride), (MatrixStride)

input attachment

UniformConstant

OpTypeImage (Dim=SubpassData, Sampled=2)

InputAttachmentIndex

1

in addition to DescriptorSet and Binding

14.5.3. DescriptorSet and Binding Assignment

A variable decorated with a DescriptorSet decoration of s and a Binding decoration of b indicates that this variable is associated with the VkDescriptorSetLayoutBinding that has a binding equal to b in pSetLayouts[s] that was specified in VkPipelineLayoutCreateInfo.

DescriptorSet decoration values must be between zero and maxBoundDescriptorSets minus one, inclusive. Binding decoration values can be any 32-bit unsigned integer value, as described in Descriptor Set Layout. Each descriptor set has its own binding name space.

If the Binding decoration is used with an array, the entire array is assigned that binding value. The array must be a single-dimensional array and size of the array must be no larger than the number of descriptors in the binding. If the array is runtime-sized, then array elements greater than or equal to the size of that binding in the bound descriptor set must not be used. If the array is runtime-sized, the runtimeDescriptorArray feature must be enabled and the RuntimeDescriptorArrayEXT capability must be declared. The index of each element of the array is referred to as the arrayElement. For the purposes of interface matching and descriptor set operations, if a resource variable is not an array, it is treated as if it has an arrayElement of zero.

There is a limit on the number of resources of each type that can be accessed by a pipeline stage as shown in Shader Resource Limits. The “Resources Per Stage” column gives the limit on the number each type of resource that can be statically used for an entry point in any given stage in a pipeline. The “Resource Types” column lists which resource types are counted against the limit. Some resource types count against multiple limits.

The pipeline layout may include descriptor sets and bindings which are not referenced by any variables statically used by the entry points for the shader stages in the binding’s stageFlags.

However, if a variable assigned to a given DescriptorSet and Binding is statically used by the entry point for a shader stage, the pipeline layout must contain a descriptor set layout binding in that descriptor set layout and for that binding number, and that binding’s stageFlags must include the appropriate VkShaderStageFlagBits for that stage. The descriptor set layout binding must be of a corresponding descriptor type, as defined in Shader Resource and Descriptor Type Correspondence.

Note

There are no limits on the number of shader variables that can have overlapping set and binding values in a shader; but which resources are statically used has an impact. If any shader variable identifying a resource is statically used in a shader, then the underlying descriptor bound at the declared set and binding must support the declared type in the shader when the shader executes.

If multiple shader variables are declared with the same set and binding values, and with the same underlying descriptor type, they can all be statically used within the same shader. However, accesses are not automatically synchronized, and Aliased decorations should be used to avoid data hazards (see section 2.18.2 Aliasing in the SPIR-V specification).

If multiple shader variables with the same set and binding values are declared in a single shader, but with different declared types, where any of those are not supported by the relevant bound descriptor, that shader can only be executed if the variables with the unsupported type are not statically used.

A noteworthy example of using multiple statically-used shader variables sharing the same descriptor set and binding values is a descriptor of type VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER that has multiple corresponding shader variables in the UniformConstant storage class, where some could be OpTypeImage, some could be OpTypeSampler (Sampled=1), and some could be OpTypeSampledImage.

Table 21. Shader Resource Limits
Resources per Stage Resource Types

maxPerStageDescriptorSamplers or maxPerStageDescriptorUpdateAfterBindSamplers

sampler

combined image sampler

maxPerStageDescriptorSampledImages or maxPerStageDescriptorUpdateAfterBindSampledImages

sampled image

combined image sampler

uniform texel buffer

maxPerStageDescriptorStorageImages or maxPerStageDescriptorUpdateAfterBindStorageImages

storage image

storage texel buffer

maxPerStageDescriptorUniformBuffers or maxPerStageDescriptorUpdateAfterBindUniformBuffers

uniform buffer

uniform buffer dynamic

maxPerStageDescriptorStorageBuffers or maxPerStageDescriptorUpdateAfterBindStorageBuffers

storage buffer

storage buffer dynamic

maxPerStageDescriptorInputAttachments or maxPerStageDescriptorUpdateAfterBindInputAttachments

input attachment1

1

Input attachments can only be used in the fragment shader stage

14.5.4. Offset and Stride Assignment

All variables with a storage class of PushConstant or Uniform must be explicitly laid out using the Offset, ArrayStride, and MatrixStride decorations. There are two different layouts requirements depending on the specific resources.

Standard Uniform Buffer Layout

The base alignment of the type of an OpTypeStruct member of is defined recursively as follows:

  • A scalar of size N has a base alignment of N.

  • A two-component vector, with components of size N, has a base alignment of 2 N.

  • A three- or four-component vector, with components of size N, has a base alignment of 4 N.

  • An array has a base alignment equal to the base alignment of its element type, rounded up to a multiple of 16.

  • A structure has a base alignment equal to the largest base alignment of any of its members, rounded up to a multiple of 16.

  • A row-major matrix of C columns has a base alignment equal to the base alignment of a vector of C matrix components.

  • A column-major matrix has a base alignment equal to the base alignment of the matrix column type.

A member is defined to improperly straddle if either of the following are true:

  • It is a vector with total size less than or equal to 16 bytes, and has Offset decorations placing its first byte at F and its last byte at L, where floor(F / 16) != floor(L / 16).

  • It is a vector with total size greater than 16 bytes and has its Offset decorations placing its first byte at a non-integer multiple of 16.

Every member of an OpTypeStruct with storage class of Uniform and a decoration of Block (uniform buffers) must be laid out according to the following rules:

  • The Offset decoration of a scalar, an array, a structure, or a matrix must be a multiple of its base alignment.

  • The Offset decoration of a vector must be an integer multiple of the base alignment of its scalar component type, and must not improperly straddle, as defined above.

  • Any ArrayStride or MatrixStride decoration must be an integer multiple of the base alignment of the array or matrix from above.

  • The Offset decoration of a member must not place it between the end of a structure or an array and the next multiple of the base alignment of that structure or array.

  • The numeric order of Offset decorations need not follow member declaration order.

Note

The std140 layout in GLSL satisfies these rules.

Standard Storage Buffer Layout

Member variables of an OpTypeStruct with a storage class of PushConstant (push constants), or a storage class of Uniform with a decoration of BufferBlock (storage buffers) , or a storage class of StorageBuffer with a decoration of Block must be laid out as above, except for array and structure base alignment which do not need to be rounded up to a multiple of 16.

Note

The std430 layout in GLSL satisfies these rules.

14.6. Built-In Variables

Built-in variables are accessed in shaders by declaring a variable decorated with a BuiltIn decoration. The meaning of each BuiltIn decoration is as follows. In the remainder of this section, the name of a built-in is used interchangeably with a term equivalent to a variable decorated with that particular built-in. Built-ins that represent integer values can be declared as either signed or unsigned 32-bit integers.

BaryCoordNoPerspAMD

The BaryCoordNoPerspAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using linear interpolation at the pixel’s center. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordNoPerspCentroidAMD

The BaryCoordNoPerspCentroidAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using linear interpolation at the centroid. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordNoPerspSampleAMD

The BaryCoordNoPerspCentroidAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using linear interpolation at each covered sample. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordPullModelAMD

The BaryCoordPullModelAMD decoration can be used to decorate a fragment shader input variable. This variable will contain (1/W, 1/I, 1/J) evaluated at the pixel center and can be used to calculate gradients and then interpolate I, J, and W at any desired sample location.

BaryCoordSmoothAMD

The BaryCoordSmoothAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using perspective interpolation at the pixel’s center. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordSmoothCentroidAMD

The BaryCoordSmoothCentroidAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using perspective interpolation at the centroid. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordSmoothSampleAMD

The BaryCoordSmoothCentroidAMD decoration can be used to decorate a fragment shader input variable. This variable will contain the (I,J) pair of the barycentric coordinates corresponding to the fragment evaluated using perspective interpolation at each covered sample. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaseInstance

Decorating a variable with the BaseInstance built-in will make that variable contain the integer value corresponding to the first instance that was passed to the command that invoked the current vertex shader invocation. BaseInstance is the firstInstance parameter to a direct drawing command or the firstInstance member of a structure consumed by an indirect drawing command.

The BaseInstance decoration must be used only within vertex shaders.

The variable decorated with BaseInstance must be declared using the input storage class.

The variable decorated with BaseInstance must be declared as a scalar 32-bit integer.

BaseVertex

Decorating a variable with the BaseVertex built-in will make that variable contain the integer value corresponding to the first vertex or vertex offset that was passed to the command that invoked the current vertex shader invocation. For non-indexed drawing commands, this variable is the firstVertex parameter to a direct drawing command or the firstVertex member of the structure consumed by an indirect drawing command. For indexed drawing commands, this variable is the vertexOffset parameter to a direct drawing command or the vertexOffset member of the structure consumed by an indirect drawing command.

The BaseVertex decoration must be used only within vertex shaders.

The variable decorated with BaseVertex must be declared using the input storage class.

The variable decorated with codeBaseVertex must be declared as a scalar 32-bit integer.

ClipDistance

Decorating a variable with the ClipDistance built-in decoration will make that variable contain the mechanism for controlling user clipping. ClipDistance is an array such that the ith element of the array specifies the clip distance for plane i. A clip distance of 0 means the vertex is on the plane, a positive distance means the vertex is inside the clip half-space, and a negative distance means the point is outside the clip half-space.

The ClipDistance decoration must be used only within vertex, fragment, tessellation control, tessellation evaluation, and geometry shaders.

In vertex shaders, any variable decorated with ClipDistance must be declared using the Output storage class.

In fragment shaders, any variable decorated with ClipDistance must be declared using the Input storage class.

In tessellation control, tessellation evaluation, or geometry shaders, any variable decorated with ClipDistance must not be in a storage class other than Input or Output.

Any variable decorated with ClipDistance must be declared as an array of 32-bit floating-point values.

Note

The array variable decorated with ClipDistance is explicitly sized by the shader.

Note

In the last vertex processing stage, these values will be linearly interpolated across the primitive and the portion of the primitive with interpolated distances less than 0 will be considered outside the clip volume. If ClipDistance is then used by a fragment shader, ClipDistance contains these linearly interpolated values.

CullDistance

Decorating a variable with the CullDistance built-in decoration will make that variable contain the mechanism for controlling user culling. If any member of this array is assigned a negative value for all vertices belonging to a primitive, then the primitive is discarded before rasterization.

The CullDistance decoration must be used only within vertex, fragment, tessellation control, tessellation evaluation, and geometry shaders.

In vertex shaders, any variable decorated with CullDistance must be declared using the Output storage class.

In fragment shaders, any variable decorated with CullDistance must be declared using the Input storage class.

In tessellation control, tessellation evaluation, or geometry shaders, any variable decorated with CullDistance must not be declared in a storage class other than input or output.

Any variable decorated with CullDistance must be declared as an array of 32-bit floating-point values.

Note

In fragment shaders, the values of the CullDistance array are linearly interpolated across each primitive.

Note

If CullDistance decorates an input variable, that variable will contain the corresponding value from the CullDistance decorated output variable from the previous shader stage.

DeviceIndex

The DeviceIndex decoration can be applied to a shader input which will be filled with the device index of the physical device that is executing the current shader invocation. This value will be in the range \([0,max(1,physicalDeviceCount))\), where physicalDeviceCount is the physicalDeviceCount member of VkDeviceGroupDeviceCreateInfo.

The DeviceIndex decoration can be used in any shader.

The variable decorated with DeviceIndex must be declared using the Input storage class.

The variable decorated with DeviceIndex must be declared as a scalar 32-bit integer.

DrawIndex

Decorating a variable with the DrawIndex built-in will make that variable contain the integer value corresponding to the zero-based index of the drawing command that invoked the current vertex shader invocation. For indirect drawing commands, DrawIndex begins at zero and increments by one for each draw command executed. The number of draw commands is given by the drawCount parameter. For direct drawing commands, DrawIndex is always zero. DrawIndex is dynamically uniform.

The DrawIndex decoration must be used only within vertex shaders.

The variable decorated with DrawIndex must be declared using the input storage class.

The variable decorated with DrawIndex must be declared as a scalar 32-bit integer.

FragCoord

Decorating a variable with the FragCoord built-in decoration will make that variable contain the framebuffer coordinate \((x,y,z,\frac{1}{w})\) of the fragment being processed. The (x,y) coordinate (0,0) is the upper left corner of the upper left pixel in the framebuffer.

When Sample Shading is enabled, the x and y components of FragCoord reflect the location of one of the samples corresponding to the shader invocation.

When sample shading is not enabled, the x and y components of FragCoord reflect the location of the center of the pixel, (0.5,0.5).

The z component of FragCoord is the interpolated depth value of the primitive.

The w component is the interpolated \(\frac{1}{w}\).

The FragCoord decoration must be used only within fragment shaders.

The variable decorated with FragCoord must be declared using the Input storage class.

The Centroid interpolation decoration is ignored, but allowed, on FragCoord.

The variable decorated with FragCoord must be declared as a four-component vector of 32-bit floating-point values.

FragDepth

Decorating a variable with the FragDepth built-in decoration will make that variable contain the new depth value for all samples covered by the fragment. This value will be used for depth testing and, if the depth test passes, any subsequent write to the depth/stencil attachment.

To write to FragDepth, a shader must declare the DepthReplacing execution mode. If a shader declares the DepthReplacing execution mode and there is an execution path through the shader that does not set FragDepth, then the fragment’s depth value is undefined for executions of the shader that take that path.

The FragDepth decoration must be used only within fragment shaders.

The variable decorated with FragDepth must be declared using the Output storage class.

The variable decorated with FragDepth must be declared as a scalar 32-bit floating-point value.

FragStencilRefEXT

Decorating a variable with the FragStencilRefEXT built-in decoration will make that variable contain the stencil reference value for all samples covered by the fragment. This value will be used as the stencil reference value used in stencil testing.

To write to FragStencilRefEXT, a shader must declare the StencilRefReplacingEXT execution mode. If a shader declares the StencilRefReplacingEXT execution mode and there is an execution path through the shader that does not set FragStencilRefEXT, then the fragment’s stencil reference value is undefined for executions of the shader that take that path.

The FragStencilRefEXT decoration must be used only within fragment shaders.

The variable decorated with FragStencilRefEXT must be declared using the Output storage class.

The variable decorated with FragStencilRefEXT must be declared as a scalar integer value. Only the least significant s bits of the integer value of the variable decorated with FragStencilRefEXT are considered for stencil testing, where s is the number of bits in the stencil framebuffer attachment, and higher order bits are discarded.

FrontFacing

Decorating a variable with the FrontFacing built-in decoration will make that variable contain whether the fragment is front or back facing. This variable is non-zero if the current fragment is considered to be part of a front-facing polygon primitive or of a non-polygon primitive and is zero if the fragment is considered to be part of a back-facing polygon primitive.

The FrontFacing decoration must be used only within fragment shaders.

The variable decorated with FrontFacing must be declared using the Input storage class.

The variable decorated with FrontFacing must be declared as a boolean.

FullyCoveredEXT

Decorating a variable with the FullyCoveredEXT built-in decoration will make that variable indicate whether the fragment pixel square is fully covered by the generating primitive. This variable is non-zero if conservative rasterization is enabled and the current fragment pixel square is fully covered by the generating primitive, and is zero if the fragment is not covered or partially covered, or conservative rasterization is disabled.

The FullyCoveredEXT decoration must be used only within fragment shaders and the FragmentFullyCoveredEXT capability must be declared.

The variable decorated with FullyCoveredEXT must be declared using the Input storage class.

The variable decorated with FullyCoveredEXT must be declared as a boolean.

If the implementation supports VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativeRasterizationPostDepthCoverage and the PostDepthCoverage execution mode is specified the SampleMask built-in input variable will reflect the coverage after the early per-fragment depth and stencil tests are applied. If VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativeRasterizationPostDepthCoverage is not supported the PostDepthCoverage execution mode must not be specified.

GlobalInvocationId

Decorating a variable with the GlobalInvocationId built-in decoration will make that variable contain the location of the current invocation within the global workgroup. Each component is equal to the index of the local workgroup multiplied by the size of the local workgroup plus LocalInvocationId.

The GlobalInvocationId decoration must be used only within compute shaders.

The variable decorated with GlobalInvocationId must be declared using the Input storage class.

The variable decorated with GlobalInvocationId must be declared as a three-component vector of 32-bit integers.

HelperInvocation

Decorating a variable with the HelperInvocation built-in decoration will make that variable contain whether the current invocation is a helper invocation. This variable is non-zero if the current fragment being shaded is a helper invocation and zero otherwise. A helper invocation is an invocation of the shader that is produced to satisfy internal requirements such as the generation of derivatives.

The HelperInvocation decoration must be used only within fragment shaders.

The variable decorated with HelperInvocation must be declared using the Input storage class.

The variable decorated with HelperInvocation must be declared as a boolean.

Note

It is very likely that a helper invocation will have a value of SampleMask fragment shader input value that is zero.

InvocationId

Decorating a variable with the InvocationId built-in decoration will make that variable contain the index of the current shader invocation in a geometry shader, or the index of the output patch vertex in a tessellation control shader.

In a geometry shader, the index of the current shader invocation ranges from zero to the number of instances declared in the shader minus one. If the instance count of the geometry shader is one or is not specified, then InvocationId will be zero.

The InvocationId decoration must be used only within tessellation control and geometry shaders.

The variable decorated with InvocationId must be declared using the Input storage class.

The variable decorated with InvocationId must be declared as a scalar 32-bit integer.

InstanceIndex

Decorating a variable with the InstanceIndex built-in decoration will make that variable contain the index of the instance that is being processed by the current vertex shader invocation. InstanceIndex begins at the firstInstance parameter to vkCmdDraw or vkCmdDrawIndexed or at the firstInstance member of a structure consumed by vkCmdDrawIndirect or vkCmdDrawIndexedIndirect.

The InstanceIndex decoration must be used only within vertex shaders.

The variable decorated with InstanceIndex must be declared using the Input storage class.

The variable decorated with InstanceIndex must be declared as a scalar 32-bit integer.

Layer

Decorating a variable with the Layer built-in decoration will make that variable contain the select layer of a multi-layer framebuffer attachment.

In a vertex, tessellation evaluation, or geometry shader, any variable decorated with Layer can be written with the framebuffer layer index to which the primitive produced by that shader will be directed.

The last active vertex processing stage (in pipeline order) controls the Layer that is used. Outputs in previous shader stages are not used, even if the last stage fails to write the Layer.

If the last active vertex processing stage shader entry point’s interface does not include a variable decorated with Layer, then the first layer is used. If a vertex processing stage shader entry point’s interface includes a variable decorated with Layer, it must write the same value to Layer for all output vertices of a given primitive. If the Layer value is less than 0 or greater than or equal to the number of layers in the framebuffer, then primitives may still be rasterized, fragment shaders may be executed, and the framebuffer values for all layers are undefined.

The Layer decoration must be used only within vertex, tessellation evaluation, geometry, and fragment shaders.

In a vertex, tessellation evaluation, or geometry shader, any variable decorated with Layer must be declared using the Output storage class. If such a variable is also decorated with ViewportRelativeNV, then the ViewportIndex is added to the layer that is used for rendering and that is made available in the fragment shader. If the shader writes to a variable decorated ViewportMaskNV, then the layer selected has a different value for each viewport a primitive is rendered to.

In a fragment shader, a variable decorated with Layer contains the layer index of the primitive that the fragment invocation belongs to.

In a fragment shader, any variable decorated with Layer must be declared using the Input storage class.

Any variable decorated with Layer must be declared as a scalar 32-bit integer.

LocalInvocationId

Decorating a variable with the LocalInvocationId built-in decoration will make that variable contain the location of the current compute shader invocation within the local workgroup. Each component ranges from zero through to the size of the workgroup in that dimension minus one.

The LocalInvocationId decoration must be used only within compute shaders.

The variable decorated with LocalInvocationId must be declared using the Input storage class.

The variable decorated with LocalInvocationId must be declared as a three-component vector of 32-bit integers.

Note

If the size of the workgroup in a particular dimension is one, then the LocalInvocationId in that dimension will be zero. If the workgroup is effectively two-dimensional, then LocalInvocationId.z will be zero. If the workgroup is effectively one-dimensional, then both LocalInvocationId.y and LocalInvocationId.z will be zero.

LocalInvocationIndex

Decorating a variable with the LocalInvocationIndex built-in decoration will make that variable contain a one-dimensional representation of LocalInvocationId. This is computed as:

LocalInvocationIndex =
    LocalInvocationId.z * WorkgroupSize.x * WorkgroupSize.y +
    LocalInvocationId.y * WorkgroupSize.x +
    LocalInvocationId.x;

The LocalInvocationIndex decoration must be used only within compute shaders.

+ The variable decorated with LocalInvocationIndex must be declared using the Input storage class.

+ The variable decorated with LocalInvocationIndex must be declared as a scalar 32-bit integer.

NumSubgroups

Decorating a variable with the NumSubgroups built-in decoration will make that variable contain the number of subgroups in the local workgroup.

The NumSubgroups decoration must be used only within compute shaders.

The variable decorated with NumSubgroups must be declared using the Input storage class.

The object decorated with NumSubgroups must be declared as a scalar 32-bit integer.

NumWorkgroups

Decorating a variable with the NumWorkgroups built-in decoration will make that variable contain the number of local workgroups that are part of the dispatch that the invocation belongs to. Each component is equal to the values of the workgroup count parameters passed into the dispatch commands.

The NumWorkgroups decoration must be used only within compute shaders.

The variable decorated with NumWorkgroups must be declared using the Input storage class.

The variable decorated with NumWorkgroups must be declared as a three-component vector of 32-bit integers.

PatchVertices

Decorating a variable with the PatchVertices built-in decoration will make that variable contain the number of vertices in the input patch being processed by the shader. A single tessellation control or tessellation evaluation shader can read patches of differing sizes, so the value of the PatchVertices variable may differ between patches.

The PatchVertices decoration must be used only within tessellation control and tessellation evaluation shaders.

The variable decorated with PatchVertices must be declared using the Input storage class.

The variable decorated with PatchVertices must be declared as a scalar 32-bit integer.

PointCoord

Decorating a variable with the PointCoord built-in decoration will make that variable contain the coordinate of the current fragment within the point being rasterized, normalized to the size of the point with origin in the upper left corner of the point, as described in Basic Point Rasterization. If the primitive the fragment shader invocation belongs to is not a point, then the variable decorated with PointCoord contains an undefined value.

The PointCoord decoration must be used only within fragment shaders.

The variable decorated with PointCoord must be declared using the Input storage class.

The variable decorated with PointCoord must be declared as two-component vector of 32-bit floating-point values.

Note

Depending on how the point is rasterized, PointCoord may never reach (0,0) or (1,1).

PointSize

Decorating a variable with the PointSize built-in decoration will make that variable contain the size of point primitives. The value written to the variable decorated with PointSize by the last vertex processing stage in the pipeline is used as the framebuffer-space size of points produced by rasterization.

The PointSize decoration must be used only within vertex, tessellation control, tessellation evaluation, and geometry shaders.

In a vertex shader, any variable decorated with PointSize must be declared using the Output storage class.

In a tessellation control, tessellation evaluation, or geometry shader, any variable decorated with PointSize must be declared using either the Input or Output storage class.

Any variable decorated with PointSize must be declared as a scalar 32-bit floating-point value.

Note

When PointSize decorates a variable in the Input storage class, it contains the data written to the output variable decorated with PointSize from the previous shader stage.

Position

Decorating a variable with the Position built-in decoration will make that variable contain the position of the current vertex. In the last vertex processing stage, the value of the variable decorated with Position is used in subsequent primitive assembly, clipping, and rasterization operations.

The Position decoration must be used only within vertex, tessellation control, tessellation evaluation, and geometry shaders.

In a vertex shader, any variable decorated with Position must be declared using the Output storage class.

In a tessellation control, tessellation evaluation, or geometry shader, any variable decorated with Position must not be declared in a storage class other than Input or Output.

Any variable decorated with Position must be declared as a four-component vector of 32-bit floating-point values.

Note

When Position decorates a variable in the Input storage class, it contains the data written to the output variable decorated with Position from the previous shader stage.

PositionPerViewNV

Decorating a variable with the PositionPerViewNV built-in decoration will make that variable contain the position of the current vertex, for each view.

The PositionPerViewNV decoration must be used only within vertex, tessellation control, tessellation evaluation, and geometry shaders.

In a vertex shader, any variable decorated with PositionPerViewNV must be declared using the Output storage class.

In a tessellation control, tessellation evaluation, or geometry shader, any variable decorated with PositionPerViewNV must not be declared in a storage class other than input or output.

Any variable decorated with PositionPerViewNV must be declared as an array of four-component vector of 32-bit floating-point values with at least as many elements as the maximum view in the subpass’s view mask plus one. The array must be indexed by a constant or specialization constant.

Elements of the array correspond to views in a multiview subpass, and those elements corresponding to views in the view mask of the subpass the shader is compiled against will be used as the position value for those views. For the final vertex processing stage in the pipeline, values written to an output variable decorated with PositionPerViewNV are used in subsequent primitive assembly, clipping, and rasterization operations, as with Position. PositionPerViewNV output in an earlier vertex processing stage is available as an input in the subsequent vertex processing stage.

If a shader is compiled against a subpass that has the VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX bit set, then the position values for each view must not differ in any component other than the X component. If the values do differ, one will be chosen in an implementation-dependent manner.

PrimitiveId

Decorating a variable with the PrimitiveId built-in decoration will make that variable contain the index of the current primitive.

The index of the first primitive generated by a drawing command is zero, and the index is incremented after every individual point, line, or triangle primitive is processed.

For triangles drawn as points or line segments (see Polygon Mode), the primitive index is incremented only once, even if multiple points or lines are eventually drawn.

Variables decorated with PrimitiveId are reset to zero between each instance drawn.

Restarting a primitive topology using primitive restart has no effect on the value of variables decorated with PrimitiveId.

In tessellation control and tessellation evaluation shaders, it will contain the index of the patch within the current set of rendering primitives that correspond to the shader invocation.

In a geometry shader, it will contain the number of primitives presented as input to the shader since the current set of rendering primitives was started.

In a fragment shader, it will contain the primitive index written by the geometry shader if a geometry shader is present, or with the value that would have been presented as input to the geometry shader had it been present.

If a geometry shader is present and the fragment shader reads from an input variable decorated with PrimitiveId, then the geometry shader must write to an output variable decorated with PrimitiveId in all execution paths.

The PrimitiveId decoration must be used only within fragment, tessellation control, tessellation evaluation, and geometry shaders.

In a tessellation control or tessellation evaluation shader, any variable decorated with PrimitiveId must be declared using the Input storage class.

In a geometry shader, any variable decorated with PrimitiveId must be declared using either the Input or Output storage class.

In a fragment shader, any variable decorated with PrimitiveId must be declared using the Input storage class, and either the Geometry or Tessellation capability must also be declared.

Any variable decorated with PrimitiveId must be declared as a scalar 32-bit integer.

Note

When the PrimitiveId decoration is applied to an output variable in the geometry shader, the resulting value is seen through the PrimitiveId decorated input variable in the fragment shader.

SampleId

Decorating a variable with the SampleId built-in decoration will make that variable contain the zero-based index of the sample the invocation corresponds to. SampleId ranges from zero to the number of samples in the framebuffer minus one. If a fragment shader entry point’s interface includes an input variable decorated with SampleId, Sample Shading is considered enabled with a minSampleShading value of 1.0.

The SampleId decoration must be used only within fragment shaders.

The variable decorated with SampleId must be declared using the Input storage class.

The variable decorated with SampleId must be declared as a scalar 32-bit integer.

SampleMask

Decorating a variable with the SampleMask built-in decoration will make any variable contain the sample coverage mask for the current fragment shader invocation.

A variable in the Input storage class decorated with SampleMask will contain a bitmask of the set of samples covered by the primitive generating the fragment during rasterization. It has a sample bit set if and only if the sample is considered covered for this fragment shader invocation. SampleMask[] is an array of integers. Bits are mapped to samples in a manner where bit B of mask M (SampleMask[M]) corresponds to sample 32 × M + B.

When state specifies multiple fragment shader invocations for a given fragment, the sample mask for any single fragment shader invocation specifies the subset of the covered samples for the fragment that correspond to the invocation. In this case, the bit corresponding to each covered sample will be set in exactly one fragment shader invocation.

If the PostDepthCoverage execution mode is specified, the sample is considered covered if and only if the sample is covered by the primitive and the sample passes the early per-fragment tests. Otherwise the sample is considered covered if the sample is covered by the primitive, regardless of the result of the fragment tests.

A variable in the Output storage class decorated with SampleMask is an array of integers forming a bit array in a manner similar an input variable decorated with SampleMask, but where each bit represents coverage as computed by the shader. Modifying the sample mask by writing zero to a bit of SampleMask causes the sample to be considered uncovered. If this variable is also decorated with OverrideCoverageNV, the fragment coverage is replaced with the sample mask bits set in the shader otherwise the fragment coverage is ANDed with the bits of the sample mask. If the fragment shader is being evaluated at any frequency other than per-fragment, bits of the sample mask not corresponding to the current fragment shader invocation are ignored. This array must be sized in the fragment shader either implicitly or explicitly, to be no larger than the implementation-dependent maximum sample-mask (as an array of 32-bit elements), determined by the maximum number of samples. If a fragment shader entry point’s interface includes an output variable decorated with SampleMask, the sample mask will be undefined for any array elements of any fragment shader invocations that fail to assign a value. If a fragment shader entry point’s interface does not include an output variable decorated with SampleMask, the sample mask has no effect on the processing of a fragment.

The SampleMask decoration must be used only within fragment shaders.

Any variable decorated with SampleMask must be declared using either the Input or Output storage class.

Any variable decorated with SampleMask must be declared as an array of 32-bit integers.

SamplePosition

Decorating a variable with the SamplePosition built-in decoration will make that variable contain the sub-pixel position of the sample being shaded. The top left of the pixel is considered to be at coordinate (0,0) and the bottom right of the pixel is considered to be at coordinate (1,1). If a fragment shader entry point’s interface includes an input variable decorated with SamplePosition, Sample Shading is considered enabled with a minSampleShading value of 1.0.

The SamplePosition decoration must be used only within fragment shaders.

The variable decorated with SamplePosition must be declared using the Input storage class. If the current pipeline uses custom sample locations the value of any variable decorated with the SamplePosition built-in decoration is undefined.

The variable decorated with SamplePosition must be declared as a two-component vector of 32-bit floating-point values.

SubgroupId

Decorating a variable with the SubgroupId built-in decoration will make that variable contain the index of the subgroup within the local workgroup. This variable is in range [0, NumSubgroups-1].

The SubgroupId decoration must be used only within compute shaders.

The variable decorated with SubgroupId must be declared using the Input storage class.

The variable decorated with SubgroupId must be declared as a scalar 32-bit integer.

SubgroupEqMask

Decorating a variable with the SubgroupEqMask builtin decoration will make that variable contain the subgroup mask of the current subgroup invocation. The bit corresponding to the SubgroupLocalInvocationId is set in the variable decorated with SubgroupEqMask. All other bits are set to zero.

The variable decorated with SubgroupEqMask must be declared using the Input storage class.

The variable decorated with SubgroupEqMask must be declared as a four-component vector of 32-bit integer values.

SubgroupEqMaskKHR is an alias of SubgroupEqMask.

SubgroupGeMask

Decorating a variable with the SubgroupGeMask builtin decoration will make that variable contain the subgroup mask of the current subgroup invocation. The bits corresponding to the invocations greater than or equal to SubgroupLocalInvocationId through SubgroupSize-1 are set in the variable decorated with SubgroupGeMask. All other bits are set to zero.

The variable decorated with SubgroupGeMask must be declared using the Input storage class.

The variable decorated with SubgroupGeMask must be declared as a four-component vector of 32-bit integer values.

SubgroupGeMaskKHR is an alias of SubgroupGeMask.

SubgroupGtMask

Decorating a variable with the SubgroupGtMask builtin decoration will make that variable contain the subgroup mask of the current subgroup invocation. The bits corresponding to the invocations greater than SubgroupLocalInvocationId through SubgroupSize-1 are set in the variable decorated with SubgroupGtMask. All other bits are set to zero.

The variable decorated with SubgroupGtMask must be declared using the Input storage class.

The variable decorated with SubgroupGtMask must be declared as a four-component vector of 32-bit integer values.

SubgroupGtMaskKHR is an alias of SubgroupGtMask.

SubgroupLeMask

Decorating a variable with the SubgroupLeMask builtin decoration will make that variable contain the subgroup mask of the current subgroup invocation. The bits corresponding to the invocations less than or equal to SubgroupLocalInvocationId are set in the variable decorated with SubgroupLeMask. All other bits are set to zero.

The variable decorated with SubgroupLeMask must be declared using the Input storage class.

The variable decorated with SubgroupLeMask must be declared as a four-component vector of 32-bit integer values.

SubgroupLeMaskKHR is an alias of SubgroupLeMask.

SubgroupLtMask

Decorating a variable with the SubgroupLtMask builtin decoration will make that variable contain the subgroup mask of the current subgroup invocation. The bits corresponding to the invocations less than SubgroupLocalInvocationId are set in the variable decorated with SubgroupLtMask. All other bits are set to zero.

The variable decorated with SubgroupLtMask must be declared using the Input storage class.

The variable decorated with SubgroupLtMask must be declared as a four-component vector of 32-bit integer values.

SubgroupLtMaskKHR is an alias of SubgroupLtMask.

SubgroupLocalInvocationId

Decorating a variable with the SubgroupLocalInvocationId builtin decoration will make that variable contain the index of the invocation within the subgroup. This variable is in range [0,SubgroupSize-1].

The variable decorated with SubgroupLocalInvocationId must be declared using the Input storage class.

The variable decorated with SubgroupLocalInvocationId must be declared as a scalar 32-bit integer.

SubgroupSize

Decorating a variable with the SubgroupSize builtin decoration will make that variable contain the implementation-dependent maximum number of invocations in a subgroup. The maximum number of invocations that an implementation can support per subgroup is 128.

The variable decorated with SubgroupSize must be declared using the Input storage class.

The variable decorated with SubgroupSize must be declared as a scalar 32-bit integer.

TessCoord

Decorating a variable with the TessCoord built-in decoration will make that variable contain the three-dimensional (u,v,w) barycentric coordinate of the tessellated vertex within the patch. u, v, and w are in the range [0,1] and vary linearly across the primitive being subdivided. For the tessellation modes of Quads or IsoLines, the third component is always zero.

The TessCoord decoration must be used only within tessellation evaluation shaders.

The variable decorated with TessCoord must be declared using the Input storage class.

The variable decorated with TessCoord must be declared as three-component vector of 32-bit floating-point values.

TessLevelOuter

Decorating a variable with the TessLevelOuter built-in decoration will make that variable contain the outer tessellation levels for the current patch.

In tessellation control shaders, the variable decorated with TessLevelOuter can be written to which controls the tessellation factors for the resulting patch. These values are used by the tessellator to control primitive tessellation and can be read by tessellation evaluation shaders.

In tessellation evaluation shaders, the variable decorated with TessLevelOuter can read the values written by the tessellation control shader.

The TessLevelOuter decoration must be used only within tessellation control and tessellation evaluation shaders.

In a tessellation control shader, any variable decorated with TessLevelOuter must be declared using the Output storage class.

In a tessellation evaluation shader, any variable decorated with TessLevelOuter must be declared using the Input storage class.

Any variable decorated with TessLevelOuter must be declared as an array of size four, containing 32-bit floating-point values.

TessLevelInner

Decorating a variable with the TessLevelInner built-in decoration will make that variable contain the inner tessellation levels for the current patch.

In tessellation control shaders, the variable decorated with TessLevelInner can be written to, which controls the tessellation factors for the resulting patch. These values are used by the tessellator to control primitive tessellation and can be read by tessellation evaluation shaders.

In tessellation evaluation shaders, the variable decorated with TessLevelInner can read the values written by the tessellation control shader.

The TessLevelInner decoration must be used only within tessellation control and tessellation evaluation shaders.

In a tessellation control shader, any variable decorated with TessLevelInner must be declared using the Output storage class.

In a tessellation evaluation shader, any variable decorated with TessLevelInner must be declared using the Input storage class.

Any variable decorated with TessLevelInner must be declared as an array of size two, containing 32-bit floating-point values.

VertexIndex

Decorating a variable with the VertexIndex built-in decoration will make that variable contain the index of the vertex that is being processed by the current vertex shader invocation. For non-indexed draws, this variable begins at the firstVertex parameter to vkCmdDraw or the firstVertex member of a structure consumed by vkCmdDrawIndirect and increments by one for each vertex in the draw. For indexed draws, its value is the content of the index buffer for the vertex plus the vertexOffset parameter to vkCmdDrawIndexed or the vertexOffset member of the structure consumed by vkCmdDrawIndexedIndirect.

The VertexIndex decoration must be used only within vertex shaders.

The variable decorated with VertexIndex must be declared using the Input storage class.

The variable decorated with VertexIndex must be declared as a scalar 32-bit integer.

Note

VertexIndex starts at the same starting value for each instance.

ViewIndex

The ViewIndex decoration can be applied to a shader input which will be filled with the index of the view that is being processed by the current shader invocation.

If multiview is enabled in the render pass, this value will be one of the bits set in the view mask of the subpass the pipeline is compiled against. If multiview is not enabled in the render pass, this value will be zero.

The ViewIndex decoration must not be used within compute shaders.

The variable decorated with ViewIndex must be declared using the Input storage class.

The variable decorated with ViewIndex must be declared as a scalar 32-bit integer.

ViewportIndex

Decorating a variable with the ViewportIndex built-in decoration will make that variable contain the index of the viewport.

In a vertex, tessellation evaluation, or geometry shader, the variable decorated with ViewportIndex can be written to with the viewport index to which the primitive produced by that shader will be directed.

The selected viewport index is used to select the viewport transform and scissor rectangle.

The last active vertex processing stage (in pipeline order) controls the ViewportIndex that is used. Outputs in previous shader stages are not used, even if the last stage fails to write the ViewportIndex.

If the last active vertex processing stage shader entry point’s interface does not include a variable decorated with ViewportIndex, then the first viewport is used. If a vertex processing stage shader entry point’s interface includes a variable decorated with ViewportIndex, it must write the same value to ViewportIndex for all output vertices of a given primitive.

The ViewportIndex decoration must be used only within vertex, tessellation evaluation, geometry, and fragment shaders.

In a vertex, tessellation evaluation, or geometry shader, any variable decorated with ViewportIndex must be declared using the Output storage class.

In a fragment shader, the variable decorated with ViewportIndex contains the viewport index of the primitive that the fragment invocation belongs to.

In a fragment shader, any variable decorated with ViewportIndex must be declared using the Input storage class.

Any variable decorated with ViewportIndex must be declared as a scalar 32-bit integer.

ViewportMaskNV

Decorating a variable with the ViewportMaskNV built-in decoration will make that variable contain the viewport mask.

In a vertex, tessellation evaluation, or geometry shader, the variable decorated with ViewportMaskNV can be written to with the mask of which viewports the primitive produced by that shader will directed.

The ViewportMaskNV variable must be an array that has ⌈(VkPhysicalDeviceLimits::maxViewports / 32)⌉ elements. When a shader writes to this variable, bit B of element M controls whether a primitive is emitted to viewport 32 × M +B. The viewports indicated by the mask are used to select the viewport transform and scissor rectangle that a primitive will be transformed by.

The last active vertex processing stage (in pipeline order) controls the ViewportMaskNV that is used. Outputs in previous shader stages are not used, even if the last stage fails to write the ViewportMaskNV. When ViewportMaskNV is written by the final vertex processing stage, any variable decorated with ViewportIndex in the fragment shader will have the index of the viewport that was used in generating that fragment.

If a vertex processing stage shader entry point’s interface includes a variable decorated with ViewportMaskNV, it must write the same value to ViewportMaskNV for all output vertices of a given primitive.

The ViewportMaskNV decoration must be used only within vertex, tessellation evaluation, and geometry shaders.

Any variable decorated with ViewportMaskNV must be declared using the Output storage class.

Any variable decorated with ViewportMaskNV must be declared as an array of 32-bit integers.

ViewportMaskPerViewNV

Decorating a variable with the ViewportMaskPerViewNV built-in decoration will make that variable contain the mask of viewports primitives are broadcast to, for each view.

The ViewportMaskPerViewNV decoration must be used only within vertex, tessellation control, tessellation evaluation, and geometry shaders.

Any variable decorated with ViewportMaskPerViewNV must be declared using the Output storage class.

The value written to an element of ViewportMaskPerViewNV in the last vertex processing stage is a bitmask indicating which viewports the primitive will be directed to. The primitive will be broadcast to the viewport corresponding to each non-zero bit of the bitmask, and that viewport index is used to select the viewport transform and scissor rectangle, for each view. The same values must be written to all vertices in a given primitive, or else the set of viewports used for that primitive is undefined.

Any variable decorated with ViewportMaskPerViewNV must be declared as an array of scalar 32-bit integers with at least as many elements as the maximum view in the subpass’s view mask plus one. The array must be indexed by a constant or specialization constant.

Elements of the array correspond to views in a multiview subpass, and those elements corresponding to views in the view mask of the subpass the shader is compiled against will be used as the viewport mask value for those views. ViewportMaskPerViewNV output in an earlier vertex processing stage is not available as an input in the subsequent vertex processing stage.

Although ViewportMaskNV is an array, ViewportMaskPerViewNV is not a two-dimensional array. Instead, ViewportMaskPerViewNV is limited to 32 viewports.

WorkgroupId

Decorating a variable with the WorkgroupId built-in decoration will make that variable contain the global workgroup that the current invocation is a member of. Each component ranges from a base value to a base + count value, based on the parameters passed into the dispatch commands.

The WorkgroupId decoration must be used only within compute shaders.

The variable decorated with WorkgroupId must be declared using the Input storage class.

The variable decorated with WorkgroupId must be declared as a three-component vector of 32-bit integers.

WorkgroupSize

Decorating an object with the WorkgroupSize built-in decoration will make that object contain the dimensions of a local workgroup. If an object is decorated with the WorkgroupSize decoration, this must take precedence over any execution mode set for LocalSize.

The WorkgroupSize decoration must be used only within compute shaders.

The object decorated with WorkgroupSize must be a specialization constant or a constant.

The object decorated with WorkgroupSize must be declared as a three-component vector of 32-bit integers.

15. Image Operations

15.1. Image Operations Overview

Image Operations are steps performed by SPIR-V image instructions, where those instructions which take an OpTypeImage (representing a VkImageView) or OpTypeSampledImage (representing a (VkImageView, VkSampler) pair) and texel coordinates as operands, and return a value based on one or more neighboring texture elements (texels) in the image.

Note

Texel is a term which is a combination of the words texture and element. Early interactive computer graphics supported texture operations on textures, a small subset of the image operations on images described here. The discrete samples remain essentially equivalent, however, so we retain the historical term texel to refer to them.

SPIR-V Image Instructions include the following functionality:

  • OpImageSample* and OpImageSparseSample* read one or more neighboring texels of the image, and filter the texel values based on the state of the sampler.

    • Instructions with ImplicitLod in the name determine the LOD used in the sampling operation based on the coordinates used in neighboring fragments.

    • Instructions with ExplicitLod in the name determine the LOD used in the sampling operation based on additional coordinates.

    • Instructions with Proj in the name apply homogeneous projection to the coordinates.

  • OpImageFetch and OpImageSparseFetch return a single texel of the image. No sampler is used.

  • OpImage*Gather and OpImageSparse*Gather read neighboring texels and return a single component of each.

  • OpImageRead (and OpImageSparseRead) and OpImageWrite read and write, respectively, a texel in the image. No sampler is used.

  • Instructions with Dref in the name apply depth comparison on the texel values.

  • Instructions with Sparse in the name additionally return a sparse residency code.

15.1.1. Texel Coordinate Systems

Images are addressed by texel coordinates. There are three texel coordinate systems:

  • normalized texel coordinates [0.0, 1.0]

  • unnormalized texel coordinates [0.0, width / height / depth)

  • integer texel coordinates [0, width / height / depth)

SPIR-V OpImageFetch, OpImageSparseFetch, OpImageRead, OpImageSparseRead, and OpImageWrite instructions use integer texel coordinates. Other image instructions can use either normalized or unnormalized texel coordinates (selected by the unnormalizedCoordinates state of the sampler used in the instruction), but there are limitations on what operations, image state, and sampler state is supported. Normalized coordinates are logically converted to unnormalized as part of image operations, and certain steps are only performed on normalized coordinates. The array layer coordinate is always treated as unnormalized even when other coordinates are normalized.

Normalized texel coordinates are referred to as (s,t,r,q,a), with the coordinates having the following meanings:

  • s: Coordinate in the first dimension of an image.

  • t: Coordinate in the second dimension of an image.

  • r: Coordinate in the third dimension of an image.

    • (s,t,r) are interpreted as a direction vector for Cube images.

  • q: Fourth coordinate, for homogeneous (projective) coordinates.

  • a: Coordinate for array layer.

The coordinates are extracted from the SPIR-V operand based on the dimensionality of the image variable and type of instruction. For Proj instructions, the components are in order (s, [t,] [r,] q) with t and r being conditionally present based on the Dim of the image. For non-Proj instructions, the coordinates are (s [,t] [,r] [,a]), with t and r being conditionally present based on the Dim of the image and a being conditionally present based on the Arrayed property of the image. Projective image instructions are not supported on Arrayed images.

Unnormalized texel coordinates are referred to as (u,v,w,a), with the coordinates having the following meanings:

  • u: Coordinate in the first dimension of an image.

  • v: Coordinate in the second dimension of an image.

  • w: Coordinate in the third dimension of an image.

  • a: Coordinate for array layer.

Only the u and v coordinates are directly extracted from the SPIR-V operand, because only 1D and 2D (non-Arrayed) dimensionalities support unnormalized coordinates. The components are in order (u [,v]), with v being conditionally present when the dimensionality is 2D. When normalized coordinates are converted to unnormalized coordinates, all four coordinates are used.

Integer texel coordinates are referred to as (i,j,k,l,n), and the first four in that order have the same meanings as unnormalized texel coordinates. They are extracted from the SPIR-V operand in order (i, [,j], [,k], [,l]), with j and k conditionally present based on the Dim of the image, and l conditionally present based on the Arrayed property of the image. n is the sample index and is taken from the Sample image operand.

For all coordinate types, unused coordinates are assigned a value of zero.

vulkantexture0
Figure 3. Texel Coordinate Systems

The Texel Coordinate Systems - For the example shown of an 8×4 texel two dimensional image.

  • Normalized texel coordinates:

    • The s coordinate goes from 0.0 to 1.0, left to right.

    • The t coordinate goes from 0.0 to 1.0, top to bottom.

  • Unnormalized texel coordinates:

    • The u coordinate goes from -1.0 to 9.0, left to right. The u coordinate within the range 0.0 to 8.0 is within the image, otherwise it is within the border.

    • The v coordinate goes from -1.0 to 5.0, top to bottom. The v coordinate within the range 0.0 to 4.0 is within the image, otherwise it is within the border.

  • Integer texel coordinates:

    • The i coordinate goes from -1 to 8, left to right. The i coordinate within the range 0 to 7 addresses texels within the image, otherwise it addresses a border texel.

    • The j coordinate goes from -1 to 5, top to bottom. The j coordinate within the range 0 to 3 addresses texels within the image, otherwise it addresses a border texel.

  • Also shown for linear filtering:

    • Given the unnormalized coordinates (u,v), the four texels selected are i0j0, i1j0, i0j1, and i1j1.

    • The weights α and β.

    • Given the offset Δi and Δj, the four texels selected by the offset are i0j'0, i1j'0, i0j'1, and i1j'1.

Note

For formats with reduced-resolution channels, Δi and Δj are relative to the resolution of the highest-resolution channel, and therefore may be divided by two relative to the unnormalized coordinate space of the lower-resolution channels.

vulkantexture1
Figure 4. Texel Coordinate Systems

The Texel Coordinate Systems - For the example shown of an 8×4 texel two dimensional image.

  • Texel coordinates as above. Also shown for nearest filtering:

    • Given the unnormalized coordinates (u,v), the texel selected is ij.

    • Given the offset Δi and Δj, the texel selected by the offset is ij'.

15.2. Conversion Formulas

editing-note

(Bill) These Conversion Formulas will likely move to Section 2.7 Fixed-Point Data Conversions (RGB to sRGB and sRGB to RGB) and section 2.6 Numeric Representation and Computation (RGB to Shared Exponent and Shared Exponent to RGB)

15.2.1. RGB to Shared Exponent Conversion

An RGB color (red, green, blue) is transformed to a shared exponent color (redshared, greenshared, blueshared, expshared) as follows:

First, the components (red, green, blue) are clamped to (redclamped, greenclamped, blueclamped) as:

redclamped = max(0, min(sharedexpmax, red))

greenclamped = max(0, min(sharedexpmax, green))

blueclamped = max(0, min(sharedexpmax, blue))

Where:

\[\begin{aligned} N & = 9 & \text{number of mantissa bits per component} \\ B & = 15 & \text{exponent bias} \\ E_{max} & = 31 & \text{maximum possible biased exponent value} \\ sharedexp_{max} & = \frac{(2^N-1)}{2^N} \times 2^{(E_{max}-B)} \end{aligned}\]
Note

NaN, if supported, is handled as in IEEE 754-2008 minNum() and maxNum(). That is the result is a NaN is mapped to zero.

The largest clamped component, maxclamped is determined:

maxclamped = max(redclamped, greenclamped, blueclamped)

A preliminary shared exponent exp' is computed:

\[\begin{aligned} exp' = \begin{cases} \left \lfloor \log_2(max_{clamped}) \right \rfloor + (B+1) & \text{for}\ max_{clamped} > 2^{-(B+1)} \\ 0 & \text{for}\ max_{clamped} \leq 2^{-(B+1)} \end{cases} \end{aligned}\]

The shared exponent expshared is computed:

\[\begin{aligned} max_{shared} = \left \lfloor { \frac{max_{clamped}}{2^{(exp'-B-N)}} + \frac{1}{2} } \right \rfloor \end{aligned}\]
\[\begin{aligned} exp_{shared} = \begin{cases} exp' & \text{for}\ 0 \leq max_{shared} < 2^N \\ exp'+1 & \text{for}\ max_{shared} = 2^N \end{cases} \end{aligned}\]

Finally, three integer values in the range 0 to 2N are computed:

\[\begin{aligned} red_{shared} & = \left \lfloor { \frac{red_{clamped}}{2^{(exp_{shared}-B-N)}}+ \frac{1}{2} } \right \rfloor \\ green_{shared} & = \left \lfloor { \frac{green_{clamped}}{2^{(exp_{shared}-B-N)}}+ \frac{1}{2} } \right \rfloor \\ blue_{shared} & = \left \lfloor { \frac{blue_{clamped}}{2^{(exp_{shared}-B-N)}}+ \frac{1}{2} } \right \rfloor \end{aligned}\]

15.2.2. Shared Exponent to RGB

A shared exponent color (redshared, greenshared, blueshared, expshared) is transformed to an RGB color (red, green, blue) as follows:

\(red = red_{shared} \times {2^{(exp_{shared}-B-N)}}\)

\(green = green_{shared} \times {2^{(exp_{shared}-B-N)}}\)

\(blue = blue_{shared} \times {2^{(exp_{shared}-B-N)}}\)

Where:

N = 9 (number of mantissa bits per component)

B = 15 (exponent bias)

15.3. Texel Input Operations

Texel input instructions are SPIR-V image instructions that read from an image. Texel input operations are a set of steps that are performed on state, coordinates, and texel values while processing a texel input instruction, and which are common to some or all texel input instructions. They include the following steps, which are performed in the listed order:

For texel input instructions involving multiple texels (for sampling or gathering), these steps are applied for each texel that is used in the instruction. Depending on the type of image instruction, other steps are conditionally performed between these steps or involving multiple coordinate or texel values.

If Chroma Reconstruction is implicit, Texel Filtering instead takes place during chroma reconstruction, before sampler Y’CBCR conversion occurs.

15.3.1. Texel Input Validation Operations

Texel input validation operations inspect instruction/image/sampler state or coordinates, and in certain circumstances cause the texel value to be replaced or become undefined. There are a series of validations that the texel undergoes.

Instruction/Sampler/Image View Validation

There are a number of cases where a SPIR-V instruction can mismatch with the sampler, the image view, or both. There are a number of cases where the sampler can mismatch with the image view. In such cases the value of the texel returned is undefined.

These cases include:

  • The sampler borderColor is an integer type and the image view format is not one of the VkFormat integer types or a stencil component of a depth/stencil format.

  • The sampler borderColor is a float type and the image view format is not one of the VkFormat float types or a depth component of a depth/stencil format.

  • The sampler borderColor is one of the opaque black colors (VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK or VK_BORDER_COLOR_INT_OPAQUE_BLACK) and the image view VkComponentSwizzle for any of the VkComponentMapping components is not VK_COMPONENT_SWIZZLE_IDENTITY.

  • The VkImageLayout of any subresource in the image view does not match that specified in VkDescriptorImageInfo::imageLayout used to write the image descriptor.

  • If the instruction is OpImageRead or OpImageSparseRead and the shaderStorageImageReadWithoutFormat feature is not enabled, or the instruction is OpImageWrite and the shaderStorageImageWriteWithoutFormat feature is not enabled, then the SPIR-V Image Format must be compatible with the image view’s format.

  • The sampler unnormalizedCoordinates is VK_TRUE and any of the limitations of unnormalized coordinates are violated.

  • The SPIR-V instruction is one of the OpImage*Dref* instructions and the sampler compareEnable is VK_FALSE

  • The SPIR-V instruction is not one of the OpImage*Dref* instructions and the sampler compareEnable is VK_TRUE

  • The SPIR-V instruction is one of the OpImage*Dref* instructions and the image view format is not one of the depth/stencil formats with a depth component, or the image view aspect is not VK_IMAGE_ASPECT_DEPTH_BIT.

  • The SPIR-V instruction’s image variable’s properties are not compatible with the image view:

    • Rules for viewType:

      • VK_IMAGE_VIEW_TYPE_1D must have Dim = 1D, Arrayed = 0, MS = 0.

      • VK_IMAGE_VIEW_TYPE_2D must have Dim = 2D, Arrayed = 0.

      • VK_IMAGE_VIEW_TYPE_3D must have Dim = 3D, Arrayed = 0, MS = 0.

      • VK_IMAGE_VIEW_TYPE_CUBE must have Dim = Cube, Arrayed = 0, MS = 0.

      • VK_IMAGE_VIEW_TYPE_1D_ARRAY must have Dim = 1D, Arrayed = 1, MS = 0.

      • VK_IMAGE_VIEW_TYPE_2D_ARRAY must have Dim = 2D, Arrayed = 1.

      • VK_IMAGE_VIEW_TYPE_CUBE_ARRAY must have Dim = Cube, Arrayed = 1, MS = 0.

    • If the image was created with VkImageCreateInfo::samples equal to VK_SAMPLE_COUNT_1_BIT, the instruction must have MS = 0.

    • If the image was created with VkImageCreateInfo::samples not equal to VK_SAMPLE_COUNT_1_BIT, the instruction must have MS = 1.

Only OpImageSample* and OpImageSparseSample* can be used with a sampler that enables sampler Y’CBCR conversion.

OpImageFetch, OpImageSparseFetch, OpImage*Gather, and OpImageSparse*Gather must not be used with a sampler that enables sampler Y'CBCR conversion.

The ConstOffset and Offset operands must not be used with a sampler that enables sampler Y’CBCR conversion.

Integer Texel Coordinate Validation

Integer texel coordinates are validated against the size of the image level, and the number of layers and number of samples in the image. For SPIR-V instructions that use integer texel coordinates, this is performed directly on the integer coordinates. For instructions that use normalized or unnormalized texel coordinates, this is performed on the coordinates that result after conversion to integer texel coordinates.

If the integer texel coordinates do not satisfy all of the conditions

0 ≤ i < ws

0 ≤ j < hs

0 ≤ k < ds

0 ≤ l < layers

0 ≤ n < samples

where:

ws = width of the image level

hs = height of the image level

ds = depth of the image level

layers = number of layers in the image

samples = number of samples per texel in the image

then the texel fails integer texel coordinate validation.

There are four cases to consider:

  1. Valid Texel Coordinates

    • If the texel coordinates pass validation (that is, the coordinates lie within the image),

    then the texel value comes from the value in image memory.

  2. Border Texel

    • If the texel coordinates fail validation, and

    • If the read is the result of an image sample instruction or image gather instruction, and

    • If the image is not a cube image,

    then the texel is a border texel and texel replacement is performed.

  3. Invalid Texel

    • If the texel coordinates fail validation, and

    • If the read is the result of an image fetch instruction, image read instruction, or atomic instruction,

    then the texel is an invalid texel and texel replacement is performed.

  4. Cube Map Edge or Corner

    Otherwise the texel coordinates lie on the borders along the edges and corners of a cube map image, and Cube map edge handling is performed.

Cube Map Edge Handling

If the texel coordinates lie on the borders along the edges and corners of a cube map image, the following steps are performed. Note that this only occurs when using VK_FILTER_LINEAR filtering within a mip level, since VK_FILTER_NEAREST is treated as using VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE.

  • Cube Map Edge Texel

    • If the texel lies along the border in either only i or only j

    then the texel lies along an edge, so the coordinates (i,j) and the array layer l are transformed to select the adjacent texel from the appropriate neighboring face.

  • Cube Map Corner Texel

    • If the texel lies along the border in both i and j

    then the texel lies at a corner and there is no unique neighboring face from which to read that texel. The texel should be replaced by the average of the three values of the adjacent texels in each incident face. However, implementations may replace the cube map corner texel by other methods, subject to the constraint that if the three available samples have the same value, the replacement texel also has that value.

Sparse Validation

If the texel reads from an unbound region of a sparse image, the texel is a sparse unbound texel, and processing continues with texel replacement.

Layout Validation

If all planes of a disjoint multi-planar image are not in the same image layout when the image is sampled with sampler Y’CBCR conversion, the result of texel reads is undefined.

15.3.2. Format Conversion

Texels undergo a format conversion from the VkFormat of the image view to a vector of either floating point or signed or unsigned integer components, with the number of components based on the number of components present in the format.

  • Color formats have one, two, three, or four components, according to the format.

  • Depth/stencil formats are one component. The depth or stencil component is selected by the aspectMask of the image view.

Each component is converted based on its type and size (as defined in the Format Definition section for each VkFormat), using the appropriate equations in 16-Bit Floating-Point Numbers, Unsigned 11-Bit Floating-Point Numbers, Unsigned 10-Bit Floating-Point Numbers, Fixed-Point Data Conversion, and Shared Exponent to RGB. Signed integer components smaller than 32 bits are sign-extended.

If the image format is sRGB, the color components are first converted as if they are UNORM, and then sRGB to linear conversion is applied to the R, G, and B components as described in the “sRGB EOTF” section of the Khronos Data Format Specification. The A component, if present, is unchanged.

If the image view format is block-compressed, then the texel value is first decoded, then converted based on the type and number of components defined by the compressed format.

15.3.3. Texel Replacement

A texel is replaced if it is one (and only one) of:

  • a border texel,

  • an invalid texel, or

  • a sparse unbound texel.

Border texels are replaced with a value based on the image format and the borderColor of the sampler. The border color is:

Table 22. Border Color B
Sampler borderColor Corresponding Border Color

VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK

B = (0.0, 0.0, 0.0, 0.0)

VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK

B = (0.0, 0.0, 0.0, 1.0)

VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE

B = (1.0, 1.0, 1.0, 1.0)

VK_BORDER_COLOR_INT_TRANSPARENT_BLACK

B = (0, 0, 0, 0)

VK_BORDER_COLOR_INT_OPAQUE_BLACK

B = (0, 0, 0, 1)

VK_BORDER_COLOR_INT_OPAQUE_WHITE

B = (1, 1, 1, 1)

Note

The names VK_BORDER_COLOR_*_TRANSPARENT_BLACK, VK_BORDER_COLOR_*_OPAQUE_BLACK, and VK_BORDER_COLOR_*_OPAQUE_WHITE are meant to describe which components are zeros and ones in the vocabulary of compositing, and are not meant to imply that the numerical value of VK_BORDER_COLOR_INT_OPAQUE_WHITE is a saturating value for integers.

This is substituted for the texel value by replacing the number of components in the image format

Table 23. Border Texel Components After Replacement
Texel Aspect or Format Component Assignment

Depth aspect

D = Br

Stencil aspect

S = Br

One component color format

Cr = Br

Two component color format

Crg = (Br,Bg)

Three component color format

Crgb = (Br,Bg,Bb)

Four component color format

Crgba = (Br,Bg,Bb,Ba)

The value returned by a read of an invalid texel is undefined, unless that read operation is from a buffer resource and the robustBufferAccess feature is enabled. In that case, an invalid texel is replaced as described by the robustBufferAccess feature.

If the VkPhysicalDeviceSparseProperties::residencyNonResidentStrict property is VK_TRUE, a sparse unbound texel is replaced with 0 or 0.0 values for integer and floating-point components of the image format, respectively.

If residencyNonResidentStrict is VK_FALSE, the value of the sparse unbound texel is undefined.

15.3.4. Depth Compare Operation

If the image view has a depth/stencil format, the depth component is selected by the aspectMask, and the operation is a Dref instruction, a depth comparison is performed. The value of the result D is 1.0 if the result of the compare operation is true, and 0.0 otherwise. The compare operation is selected by the compareOp member of the sampler.

\[\begin{aligned} D & = 1.0 & \begin{cases} D_{\textit{ref}} \leq D & \text{for LEQUAL} \\ D_{\textit{ref}} \geq D & \text{for GEQUAL} \\ D_{\textit{ref}} < D & \text{for LESS} \\ D_{\textit{ref}} > D & \text{for GREATER} \\ D_{\textit{ref}} = D & \text{for EQUAL} \\ D_{\textit{ref}} \neq D & \text{for NOTEQUAL} \\ \textit{true} & \text{for ALWAYS} \\ \textit{false} & \text{for NEVER} \end{cases} \\ D & = 0.0 & \text{otherwise} \end{aligned}\]

where, in the depth comparison:

Dref = shaderOp.Dref (from optional SPIR-V operand)

D (texel depth value)

15.3.5. Conversion to RGBA

The texel is expanded from one, two, or three to four components based on the image base color:

Table 24. Texel Color After Conversion To RGBA
Texel Aspect or Format RGBA Color

Depth aspect

Crgba = (D,0,0,one)

Stencil aspect

Crgba = (S,0,0,one)

One component color format

Crgba = (Cr,0,0,one)

Two component color format

Crgba = (Crg,0,one)

Three component color format

Crgba = (Crgb,one)

Four component color format

Crgba = Crgba

where one = 1.0f for floating-point formats and depth aspects, and one = 1 for integer formats and stencil aspects.

15.3.6. Component Swizzle

All texel input instructions apply a swizzle based on:

The swizzle can rearrange the components of the texel, or substitute zero and one for any components. It is defined as follows for the R component, and operates similarly for the other components.

\[\begin{aligned} C'_{rgba}[R] & = \begin{cases} C_{rgba}[R] & \text{for RED swizzle} \\ C_{rgba}[G] & \text{for GREEN swizzle} \\ C_{rgba}[B] & \text{for BLUE swizzle} \\ C_{rgba}[A] & \text{for ALPHA swizzle} \\ 0 & \text{for ZERO swizzle} \\ one & \text{for ONE swizzle} \\ C_{rgba}[R] & \text{for IDENTITY swizzle} \end{cases} \end{aligned}\]

where:

\[\begin{aligned} C_{rgba}[R] & \text{is the RED component} \\ C_{rgba}[G] & \text{is the GREEN component} \\ C_{rgba}[B] & \text{is the BLUE component} \\ C_{rgba}[A] & \text{is the ALPHA component} \\ one & = 1.0\text{f} & \text{for floating point components} \\ one & = 1 & \text{for integer components} \end{aligned}\]

For each component this is applied to, the VK_COMPONENT_SWIZZLE_IDENTITY swizzle selects the corresponding component from Crgba.

If the border color is one of the VK_BORDER_COLOR_*_OPAQUE_BLACK enums and the VkComponentSwizzle is not VK_COMPONENT_SWIZZLE_IDENTITY for all components (or the equivalent identity mapping), the value of the texel after swizzle is undefined.

15.3.7. Sparse Residency

OpImageSparse* instructions return a structure which includes a residency code indicating whether any texels accessed by the instruction are sparse unbound texels. This code can be interpreted by the OpImageSparseTexelsResident instruction which converts the residency code to a boolean value.

15.3.8. Chroma Reconstruction

In some color models, the color representation is defined in terms of monochromatic light intensity (often called “luma”) and color differences relative to this intensity, often called “chroma”. It is common for color models other than RGB to represent the chroma channels at lower spatial resolution than the luma channel. This approach is used to take advantage of the eye’s lower spatial sensitivity to color compared with its sensitivity to brightness. Less commonly, the same approach is used with additive color, since the green channel dominates the eye’s sensitivity to light intensity and the spatial sensitivity to color introduced by red and blue is lower.

Lower-resolution channels are “downsampled” by resizing them to a lower spatial resolution than the channel representing luminance. The process of reconstructing a full color value for texture access involves accessing both chroma and luma values at the same location. To generate the color accurately, the values of the lower-resolution channels at the location of the luma samples must be reconstructed from the lower-resolution sample locations, an operation known here as “chroma reconstruction” irrespective of the actual color model.

The location of the chroma samples relative to the luma coordinates is determined by the xChromaOffset and yChromaOffset members of the VkSamplerYcbcrConversionCreateInfo structure used to create the sampler Y’CBCR conversion.

The following diagrams show the relationship between unnormalized (u,v) coordinates and (i,j) integer texel positions in the luma channel (shown in black, with circles showing integer sample positions) and the texel coordinates of reduced-resolution chroma channels, shown as crosses in red.

Note

If the chroma values are reconstructed at the locations of the luma samples by means of interpolation, chroma samples from outside the image bounds are needed; these are determined according to Wrapping Operation. These diagrams represent this by showing the bounds of the “chroma texel” extending beyond the image bounds, and including additional chroma sample positions where required for interpolation. The limits of a sample for NEAREST sampling is shown as a grid.

chromasamples 422 cosited
Figure 5. 422 downsampling, xChromaOffset=COSITED_EVEN
chromasamples 422 midpoint
Figure 6. 422 downsampling, xChromaOffset=MIDPOINT
chromasamples 420 xcosited ycosited
Figure 7. 420 downsampling, xChromaOffset=COSITED_EVEN, yChromaOffset=COSITED_EVEN
chromasamples 420 xmidpoint ycosited
Figure 8. 420 downsampling, xChromaOffset=MIDPOINT, yChromaOffset=COSITED_EVEN
chromasamples 420 xcosited ymidpoint
Figure 9. 420 downsampling, xChromaOffset=COSITED_EVEN, yChromaOffset=MIDPOINT
chromasamples 420 xmidpoint ymidpoint
Figure 10. 420 downsampling, xChromaOffset=MIDPOINT, yChromaOffset=MIDPOINT

Reconstruction is implemented in one of two ways:

If the format of the image that is to be sampled sets VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT, or the VkSamplerYcbcrConversionCreateInfo’s forceExplicitReconstruction is set to VK_TRUE, reconstruction is performed as an explicit step independent of filtering, described in the Explicit Reconstruction section.

If the format of the image that is to be sampled does not set VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT and if the VkSamplerYcbcrConversionCreateInfo’s forceExplicitReconstruction is set to VK_FALSE, reconstruction is performed as an implicit part of filtering prior to color model conversion, with no separate post-conversion texel filtering step, as described in the Implicit Reconstruction section.

Explicit Reconstruction
  • If the chromaFilter member of the VkSamplerYcbcrConversionCreateInfo structure is VK_FILTER_NEAREST:

    • If the format’s R and B channels are reduced in resolution in just width by a factor of two relative to the G channel (i.e. this is a “_422” format), the \(\tau_{ijk}[level]\) values accessed by texel filtering are reconstructed as follows:

      \[\begin{aligned} \tau_R'(i, j) & = \tau_R(\lfloor{i\times 0.5}\rfloor, j)[level] \\ \tau_B'(i, j) & = \tau_B(\lfloor{i\times 0.5}\rfloor, j)[level] \end{aligned}\]
    • If the format’s R and B channels are reduced in resolution in width and height by a factor of two relative to the G channel (i.e. this is a “_420” format), the \(\tau_{ijk}[level]\) values accessed by texel filtering are reconstructed as follows:

      \[\begin{aligned} \tau_R'(i, j) & = \tau_R(\lfloor{i\times 0.5}\rfloor, \lfloor{j\times 0.5}\rfloor)[level] \\ \tau_B'(i, j) & = \tau_B(\lfloor{i\times 0.5}\rfloor, \lfloor{j\times 0.5}\rfloor)[level] \end{aligned}\]
      Note

      xChromaOffset and yChromaOffset have no effect if chromaFilter is VK_FILTER_NEAREST for explicit reconstruction.

  • If the chromaFilter member of the VkSamplerYcbcrConversionCreateInfo structure is VK_FILTER_LINEAR:

    • If the format’s R and B channels are reduced in resolution in just width by a factor of two relative to the G channel (i.e. this is a “422” format):

      • If xChromaOffset is VK_CHROMA_LOCATION_COSITED_EVEN:

        \[\tau_{RB}'(i,j) = \begin{cases} \tau_{RB}(\lfloor{i\times 0.5}\rfloor,j)[level], & 0.5 \times i = \lfloor{0.5 \times i}\rfloor\\ 0.5\times\tau_{RB}(\lfloor{i\times 0.5}\rfloor,j)[level] + \\ 0.5\times\tau_{RB}(\lfloor{i\times 0.5}\rfloor + 1,j)[level], & 0.5 \times i \neq \lfloor{0.5 \times i}\rfloor \end{cases}\]
      • If xChromaOffset is VK_CHROMA_LOCATION_MIDPOINT:

        \[\tau_{RB}(i,j)' = \begin{cases} 0.25 \times \tau_{RB}(\lfloor{i\times 0.5}\rfloor - 1,j)[level] + \\ 0.75 \times \tau_{RB}(\lfloor{i\times 0.5}\rfloor,j)[level], & 0.5 \times i = \lfloor{0.5 \times i}\rfloor\\ 0.75 \times \tau_{RB}(\lfloor{i\times 0.5}\rfloor,j)[level] + \\ 0.25 \times \tau_{RB}(\lfloor{i\times 0.5}\rfloor + 1,j)[level], & 0.5 \times i \neq \lfloor{0.5 \times i}\rfloor \end{cases}\]
    • If the format’s R and B channels are reduced in resolution in width and height by a factor of two relative to the G channel (i.e. this is a “420” format), a similar relationship applies. Due to the number of options, these formulae are expressed more concisely as follows:

      xChromaOffset δi

      COSITED_EVEN

      0

      MIDPOINT

      0.5

      yChromaOffset δj

      COSITED_EVEN

      0

      MIDPOINT

      0.5

      \[\begin{aligned} \tau_{RB}'(i,j) = &\\ &\tau_{RB}(\lfloor 0.5\times(i-\delta_i)\rfloor, \lfloor 0.5\times(j-\delta_j)\rfloor)[level] && \times (1 - (0.5\times(i-\delta_i) - \lfloor 0.5\times(i-\delta_i)\rfloor)) && \times (1 - (0.5\times(j-\delta_j) - \lfloor 0.5\times(j-\delta_j)\rfloor)) +\\ &\tau_{RB}(1+\lfloor 0.5\times(i-\delta_i)\rfloor, \lfloor 0.5\times(j-\delta_j)\rfloor)[level] && \times (0.5\times(i-\delta_i) - \lfloor 0.5\times(i-\delta_i)\rfloor) && \times (1 - (0.5\times(j-\delta_j) - \lfloor 0.5\times(j-\delta_j)\rfloor)) +\\ &\tau_{RB}(\lfloor 0.5\times(i-\delta_i)\rfloor, 1+\lfloor 0.5\times(j-\delta_j)\rfloor)[level] && \times (1 - (0.5\times(i-\delta_i) - \lfloor 0.5\times(i-\delta_i)\rfloor)) && \times (0.5\times(j-\delta_j) - \lfloor 0.5\times(j-\delta_j)\rfloor) +\\ &\tau_{RB}(1+\lfloor 0.5\times(i-\delta_i)\rfloor, 1+\lfloor 0.5\times(j-\delta_j)\rfloor)[level] && \times (0.5\times(i-\delta_i) - \lfloor 0.5\times(i-\delta_i)\rfloor) && \times (0.5\times(j-\delta_j) - \lfloor 0.5\times(j-\delta_j)\rfloor) \end{aligned}\]
Note

In the case where the texture itself is bilinearly interpolated as described in Texel Filtering, thus requiring four full-color samples for the filtering operation, and where the reconstruction of these samples uses bilinear interpolation in the chroma channels due to chromaFilter=VK_FILTER_LINEAR, up to nine chroma samples may be required, depending on the sample location.

Implicit Reconstruction

Implicit reconstruction takes place by the samples being interpolated, as required by the filter settings of the sampler, except that chromaFilter takes precedence for the chroma samples.

If chromaFilter is VK_FILTER_NEAREST, an implementation may behave as if xChromaOffset and yChromaOffset were both VK_CHROMA_LOCATION_MIDPOINT, irrespective of the values set.

Note

This will not have any visible effect if the locations of the luma samples coincide with the location of the samples used for rasterization.

The sample coordinates are adjusted by the downsample factor of the channel (such that, for example, the sample coordinates are divided by two if the channel has a downsample factor of two relative to the luma channel):

\[\begin{aligned} u_{RB}' (422/420) &= \begin{cases} 0.5\times (u + 0.5), & \textrm{xChromaOffset = COSITED}\_\textrm{EVEN} \\ 0.5\times u, & \textrm{xChromaOffset = MIDPOINT} \end{cases} \\ v_{RB}' (420) &= \begin{cases} 0.5\times (v + 0.5), & \textrm{yChromaOffset = COSITED}\_\textrm{EVEN} \\ 0.5\times v, & \textrm{yChromaOffset = MIDPOINT} \end{cases} \end{aligned}\]

15.3.9. Sampler Y’CBCR Conversion

Sampler Y’CBCR conversion performs the following operations, which an implementation may combine into a single mathematical operation:

Sampler Y’CBCR Range Expansion

Sampler Y’CBCR range expansion is applied to color channel values after all texel input operations which are not specific to sampler Y’CBCR conversion. For example, the input values to this stage have been converted using the normal format conversion rules.

Sampler Y’CBCR range expansion is not applied if ycbcrModel is VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY. That is, the shader receives the vector C'rgba as output by the Component Swizzle stage without further modification.

For other values of ycbcrModel, range expansion is applied to the texel channel values output by the Component Swizzle defined by the components member of VkSamplerYcbcrConversionCreateInfo. Range expansion applies independently to each channel of the image. For the purposes of range expansion and Y’CBCR model conversion, the R and B channels contain color difference (chroma) values and the G channel contains luma. The A channel is not modified by sampler Y’CBCR range expansion.

The range expansion to be applied is defined by the ycbcrRange member of the VkSamplerYcbcrConversionCreateInfo structure:

  • If ycbcrRange is VK_SAMPLER_YCBCR_RANGE_ITU_FULL, the following transformations are applied:

    \[\begin{aligned} Y' &= C'_{rgba}[G] \\ C_B &= C'_{rgba}[B] - {{2^{(n-1)}}\over{(2^n) - 1}} \\ C_R &= C'_{rgba}[R] - {{2^{(n-1)}}\over{(2^n) - 1}} \end{aligned}\]
    Note

    These formulae correspond to the “full range” encoding in the Khronos Data Format Specification.

    Should any future amendments be made to the ITU specifications from which these equations are derived, the formulae used by Vulkan may also be updated to maintain parity.

  • If ycbcrRange is VK_SAMPLER_YCBCR_RANGE_ITU_NARROW, the following transformations are applied:

    \[\begin{aligned} Y' &= {{C'_{rgba}[G] \times (2^n-1) - 16\times 2^{n-8}}\over{219\times 2^{n-8}}} \\ C_B &= {{C'_{rgba}[B] \times \left(2^n-1\right) - 128\times 2^{n-8}}\over{224\times 2^{n-8}}} \\ C_R &= {{C'_{rgba}[R] \times \left(2^n-1\right) - 128\times 2^{n-8}}\over{224\times 2^{n-8}}} \end{aligned}\]
    Note

    These formulae correspond to the “narrow range” encoding in the Khronos Data Format Specification.

  • n is the bit-depth of the channels in the format.

The precision of the operations performed during range expansion must be at least that of the source format.

An implementation may clamp the results of these range expansion operations such that Y' falls in the range [0,1], and/or such that CB and CR fall in the range [-0.5,0.5].

Sampler Y’CBCR Model Conversion

The range-expanded values are converted between color models, according to the color model conversion specified in the ycbcrModel member:

VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY

The color channels are not modified by the color model conversion since they are assumed already to represent the desired color model in which the shader is operating; Y’CBCR range expansion is also ignored.

VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY

The color channels are not modified by the color model conversion and are assumed to be treated as though in Y’CBCR form both in memory and in the shader; Y’CBCR range expansion is applied to the channels as for other Y’CBCR models, with the vector (CR,Y',CB,A) provided to the shader.

VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709

The color channels are transformed from a Y’CBCR representation to an R’G’B' representation as described in the “BT.709 Y’CBCR conversion” section of the Khronos Data Format Specification.

VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601

The color channels are transformed from a Y’CBCR representation to an R’G’B' representation as described in the “BT.601 Y’CBCR conversion” section of the Khronos Data Format Specification.

VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020

The color channels are transformed from a Y’CBCR representation to an R’G’B' representation as described in the “BT.2020 Y’CBCR conversion” section of the Khronos Data Format Specification.

In this operation, each output channel is dependent on each input channel.

An implementation may clamp the R’G’B' results of these conversions to the range [0,1].

The precision of the operations performed during model conversion must be at least that of the source format.

The alpha channel is not modified by these model conversions.

Note

Sampling operations in a non-linear color space can introduce color and intensity shifts at sharp transition boundaries. To avoid this issue, the technically precise color correction sequence described in the “Introduction to Color Conversions” chapter of the Khronos Data Format Specification may be performed as follows:

The additional calculations and, especially, additional number of sampling operations in the VK_FILTER_LINEAR case can be expected to have a performance impact compared with using the outputs directly; since the variation from “correct” results are subtle for most content, the application author should determine whether a more costly implementation is strictly necessary. Note that if chromaFilter and minFilter/magFilter are both VK_FILTER_NEAREST, these operations are redundant and sampling using sampler Y’CBCR conversion at the desired sample coordinates will produce the “correct” results without further processing.

15.4. Texel Output Operations

Texel output instructions are SPIR-V image instructions that write to an image. Texel output operations are a set of steps that are performed on state, coordinates, and texel values while processing a texel output instruction, and which are common to some or all texel output instructions. They include the following steps, which are performed in the listed order:

15.4.1. Texel Output Validation Operations

Texel output validation operations inspect instruction/image state or coordinates, and in certain circumstances cause the write to have no effect. There are a series of validations that the texel undergoes.

Texel Format Validation

If the image format of the OpTypeImage is not compatible with the VkImageView’s format, the effect of the write on the image view’s memory is undefined, but the write must not access memory outside of the image view.

15.4.2. Integer Texel Coordinate Validation

The integer texel coordinates are validated according to the same rules as for texel input coordinate validation.

If the texel fails integer texel coordinate validation, then the write has no effect.

15.4.3. Sparse Texel Operation

If the texel attempts to write to an unbound region of a sparse image, the texel is a sparse unbound texel. In such a case, if the VkPhysicalDeviceSparseProperties::residencyNonResidentStrict property is VK_TRUE, the sparse unbound texel write has no effect. If residencyNonResidentStrict is VK_FALSE, the write may have a side effect that becomes visible to other accesses to unbound texels in any resource, but will not be visible to any device memory allocated by the application.

15.4.4. Texel Output Format Conversion

If the image format is sRGB, a linear to sRGB conversion is applied to the R, G, and B components as described in the “sRGB EOTF” section of the Khronos Data Format Specification. The A component, if present, is unchanged.

Texels then undergo a format conversion from the floating point, signed, or unsigned integer type of the texel data to the VkFormat of the image view. Any unused components are ignored.

Each component is converted based on its type and size (as defined in the Format Definition section for each VkFormat), using the appropriate equations in 16-Bit Floating-Point Numbers and Fixed-Point Data Conversion.

15.5. Derivative Operations

SPIR-V derivative instructions include OpDPdx, OpDPdy, OpDPdxFine, OpDPdyFine, OpDPdxCoarse, and OpDPdyCoarse. Derivative instructions are only available in a fragment shader.

vulkantexture2
Figure 11. Implicit Derivatives

Derivatives are computed as if there is a 2×2 neighborhood of fragments for each fragment shader invocation. These neighboring fragments are used to compute derivatives with the assumption that the values of P in the neighborhood are piecewise linear. It is further assumed that the values of P in the neighborhood are locally continuous, therefore derivatives in non-uniform control flow are undefined.

\[\begin{aligned} dPdx_{i_1,j_0} & = dPdx_{i_0,j_0} & = P_{i_1,j_0} - P_{i_0,j_0} \\ dPdx_{i_1,j_1} & = dPdx_{i_0,j_1} & = P_{i_1,j_1} - P_{i_0,j_1} \\ \\ dPdy_{i_0,j_1} & = dPdy_{i_0,j_0} & = P_{i_0,j_1} - P_{i_0,j_0} \\ dPdy_{i_1,j_1} & = dPdy_{i_1,j_0} & = P_{i_1,j_1} - P_{i_1,j_0} \end{aligned}\]

The Fine derivative instructions must return the values above, for a group of fragments in a 2×2 neighborhood. Coarse derivatives may return only two values. In this case, the values should be:

\[\begin{aligned} dPdx & = \begin{cases} dPdx_{i_0,j_0} & \text{preferred}\\ dPdx_{i_0,j_1} \end{cases} \\ dPdy & = \begin{cases} dPdy_{i_0,j_0} & \text{preferred}\\ dPdy_{i_1,j_0} \end{cases} \end{aligned}\]

OpDPdx and OpDPdy must return the same result as either OpDPdxFine or OpDPdxCoarse and either OpDPdyFine or OpDPdyCoarse, respectively. Implementations must make the same choice of either coarse or fine for both OpDPdx and OpDPdy, and implementations should make the choice that is more efficient to compute.

For multi-planar formats, the derivatives are computed based on the plane with the largest dimensions.

15.6. Normalized Texel Coordinate Operations

If the image sampler instruction provides normalized texel coordinates, some of the following operations are performed.

15.6.1. Projection Operation

For Proj image operations, the normalized texel coordinates (s,t,r,q,a) and (if present) the Dref coordinate are transformed as follows:

\[\begin{aligned} s & = \frac{s}{q}, & \text{for 1D, 2D, or 3D image} \\ \\ t & = \frac{t}{q}, & \text{for 2D or 3D image} \\ \\ r & = \frac{r}{q}, & \text{for 3D image} \\ \\ D_{\textit{ref}} & = \frac{D_{\textit{ref}}}{q}, & \text{if provided} \end{aligned}\]

15.6.2. Derivative Image Operations

Derivatives are used for LOD selection. These derivatives are either implicit (in an ImplicitLod image instruction in a fragment shader) or explicit (provided explicitly by shader to the image instruction in any shader).

For implicit derivatives image instructions, the derivatives of texel coordinates are calculated in the same manner as derivative operations above. That is:

\[\begin{aligned} \partial{s}/\partial{x} & = dPdx(s), & \partial{s}/\partial{y} & = dPdy(s), & \text{for 1D, 2D, Cube, or 3D image} \\ \partial{t}/\partial{x} & = dPdx(t), & \partial{t}/\partial{y} & = dPdy(t), & \text{for 2D, Cube, or 3D image} \\ \partial{u}/\partial{x} & = dPdx(u), & \partial{u}/\partial{y} & = dPdy(u), & \text{for Cube or 3D image} \end{aligned}\]

Partial derivatives not defined above for certain image dimensionalities are set to zero.

For explicit LOD image instructions, if the optional SPIR-V operand Grad is provided, then the operand values are used for the derivatives. The number of components present in each derivative for a given image dimensionality matches the number of partial derivatives computed above.

If the optional SPIR-V operand Lod is provided, then derivatives are set to zero, the cube map derivative transformation is skipped, and the scale factor operation is skipped. Instead, the floating point scalar coordinate is directly assigned to λbase as described in Level-of-Detail Operation.

For implicit derivative image instructions, the partial derivative values may be computed by linear approximation using a 2×2 neighborhood of shader invocations (known as a quad), as described above. If the instruction is in control flow that is not uniform across the quad, then the derivative values and hence the implicit LOD values are undefined.

If the image or sampler object used by an implicit derivative image instruction is not uniform across the quad and quadDivergentImplicitLod is not supported, then the derivative and LOD values are undefined. Implicit derivatives are well-defined when the image and sampler and control flow are uniform across the quad, even if they diverge between different quads.

If quadDivergentImplicitLod is supported, then derivatives and implicit LOD values are well-defined even if the image or sampler object are not uniform within a quad. The derivatives are computed as specified above, and the implicit LOD calculation proceeds for each shader invocation using its respective image and sampler object.

For the purposes of implicit derivatives, Flat fragment input variables are uniform within a quad.

15.6.3. Cube Map Face Selection and Transformations

For cube map image instructions, the (s,t,r) coordinates are treated as a direction vector (rx,ry,rz). The direction vector is used to select a cube map face. The direction vector is transformed to a per-face texel coordinate system (sface,tface), The direction vector is also used to transform the derivatives to per-face derivatives.

15.6.4. Cube Map Face Selection

The direction vector selects one of the cube map’s faces based on the largest magnitude coordinate direction (the major axis direction). Since two or more coordinates can have identical magnitude, the implementation must have rules to disambiguate this situation.

The rules should have as the first rule that rz wins over ry and rx, and the second rule that ry wins over rx. An implementation may choose other rules, but the rules must be deterministic and depend only on (rx,ry,rz).

The layer number (corresponding to a cube map face), the coordinate selections for sc, tc, rc, and the selection of derivatives, are determined by the major axis direction as specified in the following two tables.

Table 25. Cube map face and coordinate selection
Major Axis Direction Layer Number Cube Map Face sc tc rc

+rx

0

Positive X

-rz

-ry

rx

-rx

1

Negative X

+rz

-ry

rx

+ry

2

Positive Y

+rx

+rz

ry

-ry

3

Negative Y

+rx

-rz

ry

+rz

4

Positive Z

+rx

-ry

rz

-rz

5

Negative Z

-rx

-ry

rz

Table 26. Cube map derivative selection
Major Axis Direction ∂sc / ∂x ∂sc / ∂y ∂tc / ∂x ∂tc / ∂y ∂rc / ∂x ∂rc / ∂y

+rx

-∂rz / ∂x

-∂rz / ∂y

-∂ry / ∂x

-∂ry / ∂y

+∂rx / ∂x

+∂rx / ∂y

-rx

+∂rz / ∂x

+∂rz / ∂y

-∂ry / ∂x

-∂ry / ∂y

-∂rx / ∂x

-∂rx / ∂y

+ry

+∂rx / ∂x

+∂rx / ∂y

+∂rz / ∂x

+∂rz / ∂y

+∂ry / ∂x

+∂ry / ∂y

-ry

+∂rx / ∂x

+∂rx / ∂y

-∂rz / ∂x

-∂rz / ∂y

-∂ry / ∂x

-∂ry / ∂y

+rz

+∂rx / ∂x

+∂rx / ∂y

-∂ry / ∂x

-∂ry / ∂y

+∂rz / ∂x

+∂rz / ∂y

-rz

-∂rx / ∂x

-∂rx / ∂y

-∂ry / ∂x

-∂ry / ∂y

-∂rz / ∂x

-∂rz / ∂y

15.6.5. Cube Map Coordinate Transformation

\[\begin{aligned} s_{\textit{face}} & = \frac{1}{2} \times \frac{s_c}{|r_c|} + \frac{1}{2} \\ t_{\textit{face}} & = \frac{1}{2} \times \frac{t_c}{|r_c|} + \frac{1}{2} \\ \end{aligned}\]

15.6.6. Cube Map Derivative Transformation

\[\begin{aligned} \frac{\partial{s_{\textit{face}}}}{\partial{x}} &= \frac{\partial}{\partial{x}} \left ( \frac{1}{2} \times \frac{s_{c}}{|r_{c}|} + \frac{1}{2}\right ) \\ \frac{\partial{s_{\textit{face}}}}{\partial{x}} &= \frac{1}{2} \times \frac{\partial}{\partial{x}} \left ( \frac{s_{c}}{|r_{c}|} \right ) \\ \frac{\partial{s_{\textit{face}}}}{\partial{x}} &= \frac{1}{2} \times \left ( \frac{ |r_{c}| \times \partial{s_c}/\partial{x} -s_c \times {\partial{r_{c}}}/{\partial{x}}} {\left ( r_{c} \right )^2} \right ) \end{aligned}\]
\[\begin{aligned} \frac{\partial{s_{\textit{face}}}}{\partial{y}} &= \frac{1}{2} \times \left ( \frac{ |r_{c}| \times \partial{s_c}/\partial{y} -s_c \times {\partial{r_{c}}}/{\partial{y}}} {\left ( r_{c} \right )^2} \right )\\ \frac{\partial{t_{\textit{face}}}}{\partial{x}} &= \frac{1}{2} \times \left ( \frac{ |r_{c}| \times \partial{t_c}/\partial{x} -t_c \times {\partial{r_{c}}}/{\partial{x}}} {\left ( r_{c} \right )^2} \right ) \\ \frac{\partial{t_{\textit{face}}}}{\partial{y}} &= \frac{1}{2} \times \left ( \frac{ |r_{c}| \times \partial{t_c}/\partial{y} -t_c \times {\partial{r_{c}}}/{\partial{y}}} {\left ( r_{c} \right )^2} \right ) \end{aligned}\]
editing-note

(Bill) Note that we never revisited ARB_texture_cubemap after we introduced dependent texture fetches (ARB_fragment_program and ARB_fragment_shader).

The derivatives of sface and tface are only valid for non-dependent texture fetches (pre OpenGL 2.0).

15.6.7. Scale Factor Operation, Level-of-Detail Operation and Image Level(s) Selection

LOD selection can be either explicit (provided explicitly by the image instruction) or implicit (determined from a scale factor calculated from the derivatives). The implicit LOD selected can be queried using the SPIR-V instruction OpImageQueryLod, which gives access to the λ' and dl values, defined below.

Scale Factor Operation

The magnitude of the derivatives are calculated by:

mux = |∂s/∂x| × wbase

mvx = |∂t/∂x| × hbase

mwx = |∂r/∂x| × dbase

muy = |∂s/∂y| × wbase

mvy = |∂t/∂y| × hbase

mwy = |∂r/∂y| × dbase

where:

∂t/∂x = ∂t/∂y = 0 (for 1D images)

∂r/∂x = ∂r/∂y = 0 (for 1D, 2D or Cube images)

and

wbase = image.w

hbase = image.h

dbase = image.d

(for the baseMipLevel, from the image descriptor).

A point sampled in screen space has an elliptical footprint in texture space. The minimum and maximum scale factors min, ρmax) should be the minor and major axes of this ellipse.

The scale factors ρx and ρy, calculated from the magnitude of the derivatives in x and y, are used to compute the minimum and maximum scale factors.

ρx and ρy may be approximated with functions fx and fy, subject to the following constraints:

\[\begin{aligned} & f_x \text{\ is\ continuous\ and\ monotonically\ increasing\ in\ each\ of\ } m_{ux}, m_{vx}, \text{\ and\ } m_{wx} \\ & f_y \text{\ is\ continuous\ and\ monotonically\ increasing\ in\ each\ of\ } m_{uy}, m_{vy}, \text{\ and\ } m_{wy} \end{aligned}\]
\[\begin{aligned} \max(|m_{ux}|, |m_{vx}|, |m_{wx}|) \leq f_{x} \leq \sqrt{2} (|m_{ux}| + |m_{vx}| + |m_{wx}|) \\ \max(|m_{uy}|, |m_{vy}|, |m_{wy}|) \leq f_{y} \leq \sqrt{2} (|m_{uy}| + |m_{vy}| + |m_{wy}|) \end{aligned}\]
editing-note

(Bill) For reviewers only - anticipating questions.

We only support implicit derivatives for normalized texel coordinates.

So we are documenting the derivatives in s,t,r (normalized texel coordinates) rather than u,v,w (unnormalized texel coordinates) as in OpenGL and OpenGL ES specifications. (I know, u,v,w is the way it has been documented since OpenGL V1.0.)

Also there is no reason to have conditional application of wbase, hbase, dbase for rectangle textures either, since they do not support implicit derivatives.

The minimum and maximum scale factors minmax) are determined by:

ρmax = max(ρx, ρy)

ρmin = min(ρx, ρy)

The ratio of anisotropy is determined by:

η = min(ρmaxmin, maxAniso)

where:

sampler.maxAniso = maxAnisotropy (from sampler descriptor)

limits.maxAniso = maxSamplerAnisotropy (from physical device limits)

maxAniso = min(sampler.maxAniso, limits.maxAniso)

If ρmax = ρmin = 0, then all the partial derivatives are zero, the fragment’s footprint in texel space is a point, and N should be treated as 1. If ρmax ≠ 0 and ρmin = 0 then all partial derivatives along one axis are zero, the fragment’s footprint in texel space is a line segment, and η should be treated as maxAniso. However, anytime the footprint is small in texel space the implementation may use a smaller value of η, even when ρmin is zero or close to zero. If either VkPhysicalDeviceFeatures::samplerAnisotropy or VkSamplerCreateInfo::anisotropyEnable are VK_FALSE, maxAniso is set to 1.

If η = 1, sampling is isotropic. If η > 1, sampling is anisotropic.

The sampling rate (N) is derived as:

N = ⌈η⌉

An implementation may round N up to the nearest supported sampling rate. An implementation may use the value of N as an approximation of η.

Level-of-Detail Operation

The LOD parameter λ is computed as follows:

\[\begin{aligned} \lambda_{base}(x,y) & = \begin{cases} shaderOp.Lod & \text{(from optional SPIR-V operand)} \\ \log_2 \left ( \frac{\rho_{max}}{\eta} \right ) & \text{otherwise} \end{cases} \\ \lambda'(x,y) & = \lambda_{base} + \mathbin{clamp}(sampler.bias + shaderOp.bias,-maxSamplerLodBias,maxSamplerLodBias) \\ \lambda & = \begin{cases} lod_{max}, & \lambda' > lod_{max} \\ \lambda', & lod_{min} \leq \lambda' \leq lod_{max} \\ lod_{min}, & \lambda' < lod_{min} \\ \textit{undefined}, & lod_{min} > lod_{max} \end{cases} \end{aligned}\]

where:

\[\begin{aligned} sampler.bias & = mipLodBias & \text{(from sampler descriptor)} \\ shaderOp.bias & = \begin{cases} Bias & \text{(from optional SPIR-V operand)} \\ 0 & \text{otherwise} \end{cases} \\ sampler.lod_{min} & = minLod & \text{(from sampler descriptor)} \\ shaderOp.lod_{min} & = \begin{cases} MinLod & \text{(from optional SPIR-V operand)} \\ 0 & \text{otherwise} \end{cases} \\ \\ lod_{min} & = \max(sampler.lod_{min}, shaderOp.lod_{min}) \\ lod_{max} & = maxLod & \text{(from sampler descriptor)} \end{aligned}\]

and maxSamplerLodBias is the value of the VkPhysicalDeviceLimits feature maxSamplerLodBias.

Image Level(s) Selection

The image level(s) d, dhi, and dlo which texels are read from are determined by an image-level parameter dl, which is computed based on the LOD parameter, as follows:

\[\begin{aligned} d_{l} = \begin{cases} nearest(d'), & \text{mipmapMode is VK\_SAMPLER\_MIPMAP\_MODE\_NEAREST} \\ d', & \text{otherwise} \end{cases} \end{aligned}\]

where:

\[\begin{aligned} d' = level_{base} + \text{clamp}(\lambda, 0, q) \end{aligned}\]
\[\begin{aligned} nearest(d') & = \begin{cases} \left \lceil d' + 0.5\right \rceil - 1, & \text{preferred} \\ \left \lfloor d' + 0.5\right \rfloor, & \text{alternative} \end{cases} \end{aligned}\]

and

levelbase = baseMipLevel

q = levelCount - 1

baseMipLevel and levelCount are taken from the subresourceRange of the image view.

If the sampler’s mipmapMode is VK_SAMPLER_MIPMAP_MODE_NEAREST, then the level selected is d = dl.

If the sampler’s mipmapMode is VK_SAMPLER_MIPMAP_MODE_LINEAR, two neighboring levels are selected:

\[\begin{aligned} d_{hi} & = \lfloor d_{l} \rfloor \\ d_{lo} & = min( d_{hi} + 1, q ) \\ \delta & = d_{l} - d_{hi} \end{aligned}\]

δ is the fractional value used for linear filtering between levels.

15.6.8. (s,t,r,q,a) to (u,v,w,a) Transformation

The normalized texel coordinates are scaled by the image level dimensions and the array layer is selected. This transformation is performed once for each level (d or dhi and dlo) used in filtering.

\[\begin{aligned} u(x,y) & = s(x,y) \times width_{level} \\ v(x,y) & = \begin{cases} 0 & \text{for 1D images} \\ t(x,y) \times height_{level} & \text{otherwise} \end{cases} \\ w(x,y) & = \begin{cases} 0 & \text{for 2D or Cube images} \\ r(x,y) \times depth_{level} & \text{otherwise} \end{cases} \\ \\ a(x,y) & = \begin{cases} a(x,y) & \text{for array images} \\ 0 & \text{otherwise} \end{cases} \end{aligned}\]

Operations then proceed to Unnormalized Texel Coordinate Operations.

15.7. Unnormalized Texel Coordinate Operations

15.7.1. (u,v,w,a) to (i,j,k,l,n) Transformation And Array Layer Selection

The unnormalized texel coordinates are transformed to integer texel coordinates relative to the selected mipmap level.

The layer index l is computed as:

l = clamp(RNE(a), 0, layerCount - 1) + baseArrayLayer

where layerCount is the number of layers in the image subresource range of the image view, baseArrayLayer is the first layer from the subresource range, and where:

\[\begin{aligned} \mathbin{RNE}(a) & = \begin{cases} \mathbin{roundTiesToEven}(a) & \text{preferred, from IEEE Std 754-2008 Floating-Point Arithmetic} \\ \left \lfloor a + 0.5 \right \rfloor & \text{alternative} \end{cases} \end{aligned}\]

The sample index n is assigned the value zero.

Nearest filtering (VK_FILTER_NEAREST) computes the integer texel coordinates that the unnormalized coordinates lie within:

\[\begin{aligned} i &= \lfloor u \rfloor \\ j &= \lfloor v \rfloor \\ k &= \lfloor w \rfloor \end{aligned}\]

Linear filtering (VK_FILTER_LINEAR) computes a set of neighboring coordinates which bound the unnormalized coordinates. The integer texel coordinates are combinations of i0 or i1, j0 or j1, k0 or k1, as well as weights α, β, and γ.

\[\begin{aligned} i_0 &= \lfloor u - 0.5 \rfloor \\ i_1 &= i_0 + 1 \\ j_0 &= \lfloor v - 0.5 \rfloor \\ j_1 &= j_0 + 1 \\ k_0 &= \lfloor w - 0.5 \rfloor \\ k_1 &= k_0 + 1 \\ \alpha &= \left(u - 0.5\right) - i_0 \\ \beta &= \left(v - 0.5\right) - j_0 \\ \gamma &= \left(w - 0.5\right) - k_0 \end{aligned}\]

Cubic filtering (VK_FILTER_CUBIC_IMG) computes a set of neighboring coordinates which bound the unnormalized coordinates. The integer texel coordinates are combinations of i0, i1, i2 or i3, j0, j1, j2 or j3, as well as weights α and β.

\[\begin{aligned} i_{0} & = \left \lfloor u - \frac{3}{2} \right \rfloor & i_{1} & = i_{0} + 1 & i_{2} & = i_{1} + 1 & i_{3} & = i_{2} + 1 \\[1em] j_{0} & = \left \lfloor v - \frac{3}{2} \right \rfloor & j_{1} & = j_{0} + 1 & j_{2} & = j_{1} + 1 & j_{3} & = j_{2} + 1 \\ \\ \alpha & = \mathbin{frac} \left ( u - \frac{1}{2} \right ) \\[1em] \beta & = \mathbin{frac} \left ( v - \frac{1}{2} \right ) \end{aligned}\]

If the image instruction includes a ConstOffset operand, the constant offsets i, Δj, Δk) are added to (i,j,k) components of the integer texel coordinates.

15.8. Integer Texel Coordinate Operations

Integer texel coordinate operations may supply a LOD which texels are to be read from or written to using the optional SPIR-V operand Lod. If the Lod is provided then it must be an integer.

The image level selected is:

\[\begin{aligned} d & = level_{base} + \begin{cases} Lod & \text{(from optional SPIR-V operand)} \\ 0 & \text{otherwise} \end{cases} \\ \end{aligned}\]

If d does not lie in the range [baseMipLevel, baseMipLevel + levelCount) then any values fetched are undefined, and any writes are discarded.

15.9. Image Sample Operations

15.9.1. Wrapping Operation

Cube images ignore the wrap modes specified in the sampler. Instead, if VK_FILTER_NEAREST is used within a mip level then VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE is used, and if VK_FILTER_LINEAR is used within a mip level then sampling at the edges is performed as described earlier in the Cube map edge handling section.

The first integer texel coordinate i is transformed based on the addressModeU parameter of the sampler.

\[\begin{aligned} i &= \begin{cases} i \bmod size & \text{for repeat} \\ (size - 1) - \mathbin{mirror} ((i \bmod (2 \times size)) - size) & \text{for mirrored repeat} \\ \mathbin{clamp}(i,0,size-1) & \text{for clamp to edge} \\ \mathbin{clamp}(i,-1,size) & \text{for clamp to border} \\ \mathbin{clamp}(\mathbin{mirror}(i),0,size-1) & \text{for mirror clamp to edge} \end{cases} \end{aligned}\]

where:

\[\begin{aligned} & \mathbin{mirror}(n) = \begin{cases} n & \text{for}\ n \geq 0 \\ -(1+n) & \text{otherwise} \end{cases} \end{aligned}\]

j (for 2D and Cube image) and k (for 3D image) are similarly transformed based on the addressModeV and addressModeW parameters of the sampler, respectively.

15.9.2. Texel Gathering

SPIR-V instructions with Gather in the name return a vector derived from a 2×2 rectangular region of texels in the base level of the image view. The rules for the VK_FILTER_LINEAR minification filter are applied to identify the four selected texels. Each texel is then converted to an RGBA value according to conversion to RGBA and then swizzled. A four-component vector is then assembled by taking the component indicated by the Component value in the instruction from the swizzled color value of the four texels:

\[\begin{aligned} \tau[R] &= \tau_{i0j1}[level_{base}][comp] \\ \tau[G] &= \tau_{i1j1}[level_{base}][comp] \\ \tau[B] &= \tau_{i1j0}[level_{base}][comp] \\ \tau[A] &= \tau_{i0j0}[level_{base}][comp] \end{aligned}\]

where:

\[\begin{aligned} \tau[level_{base}][comp] &= \begin{cases} \tau[level_{base}][R], & \text{for}\ comp = 0 \\ \tau[level_{base}][G], & \text{for}\ comp = 1 \\ \tau[level_{base}][B], & \text{for}\ comp = 2 \\ \tau[level_{base}][A], & \text{for}\ comp = 3 \end{cases}\\ comp & \,\text{from SPIR-V operand Component} \end{aligned}\]

OpImage*Gather must not be used on a sampled image with sampler Y’CBCR conversion enabled.

15.9.3. Texel Filtering

If λ is less than or equal to zero, the texture is said to be magnified, and the filter mode within a mip level is selected by the magFilter in the sampler. If λ is greater than zero, the texture is said to be minified, and the filter mode within a mip level is selected by the minFilter in the sampler.

Within a mip level, VK_FILTER_NEAREST filtering selects a single value using the (i, j, k) texel coordinates, with all texels taken from layer l.

\[\begin{aligned} \tau[level] &= \begin{cases} \tau_{ijk}[level], & \text{for 3D image} \\ \tau_{ij}[level], & \text{for 2D or Cube image} \\ \tau_{i}[level], & \text{for 1D image} \end{cases} \end{aligned}\]

Within a mip level, VK_FILTER_LINEAR filtering combines 8 (for 3D), 4 (for 2D or Cube), or 2 (for 1D) texel values, using the weights computed earlier:

\[\begin{aligned} \tau_{3D}[level] & = reduce((1-\alpha)(1-\beta)(1-\gamma),\tau_{i0j0k0}[level], \\ & \, (\alpha)(1-\beta)(1-\gamma),\tau_{i1j0k0}[level], \\ & \, (1-\alpha)(\beta)(1-\gamma),\tau_{i0j1k0}[level], \\ & \, (\alpha)(\beta)(1-\gamma),\tau_{i1j1k0}[level], \\ & \, (1-\alpha)(1-\beta)(\gamma),\tau_{i0j0k1}[level], \\ & \, (\alpha)(1-\beta)(\gamma),\tau_{i1j0k1}[level], \\ & \, (1-\alpha)(\beta)(\gamma),\tau_{i0j1k1}[level], \\ & \, (\alpha)(\beta)(\gamma),\tau_{i1j1k1}[level]) \end{aligned}\]
\[\begin{aligned} \tau_{2D}[level] & = reduce((1-\alpha)(1-\beta),\tau_{i0j0}[level], \\ & \, (\alpha)(1-\beta),\tau_{i1j0}[level], \\ & \, (1-\alpha)(\beta),\tau_{i0j1}[level], \\ & \, (\alpha)(\beta),\tau_{i1j1}[level]) \end{aligned}\]
\[\begin{aligned} \tau_{1D}[level] & = reduce((1-\alpha),\tau_{i0}[level], \\ & \, (\alpha),\tau_{i1}[level]) \end{aligned}\]
\[\begin{aligned} \tau[level] &= \begin{cases} \tau_{3D}[level], & \text{for 3D image} \\ \tau_{2D}[level], & \text{for 2D or Cube image} \\ \tau_{1D}[level], & \text{for 1D image} \end{cases} \end{aligned}\]

The function reduce() is defined to operate on pairs of weights and texel values as follows. When using linear or anisotropic filtering, the values of multiple texels are combined using a weighted average to produce a filtered texture value. However, a filtered texture value can also be produced by computing per-component minimum and maximum values over the set of texels that would normally be averaged. The VkSamplerReductionModeCreateInfoEXT::reductionMode controls the process by which multiple texels are combined to produce a filtered texture value. When set to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT, a weighted average is computed. If the reduction mode is VK_SAMPLER_REDUCTION_MODE_MIN_EXT or VK_SAMPLER_REDUCTION_MODE_MAX_EXT, reduce() computes a component-wise minimum or maximum, respectively, of the components of the set of provided texels with non-zero weights.

Within a mip level, VK_FILTER_CUBIC_IMG filtering computes a weighted average of 16 (for 2D), or 4 (for 1D) texel values, using the weights computed during texel selection.

Catmull-Rom Spine interpolation of four points is defined by the equation:

\[\begin{aligned} cinterp(\tau_0, \tau_1, \tau_2, \tau_3, \omega) = \frac{1}{2} \begin{bmatrix}1 & \omega & \omega^2 & \omega^3 \end{bmatrix} \times \begin{bmatrix} 0 & 2 & 0 & 0 \\ -1 & 0 & 1 & 0 \\ 2 & -5 & 4 & 1 \\ -1 & 3 & -3 & 1 \end{bmatrix} \times \begin{bmatrix} \tau_0 \\ \tau_1 \\ \tau_2 \\ \tau_3 \end{bmatrix} \end{aligned}\]

Using the values calculated in texel selection, this equation is applied to the four points in 1D images. For 2D images, the this equation is evaluated first for each row, and the result is then fed back into the equation and interpolated again:

τ1D[level] = cinterp(τi0[level], τi1[level], τi2[level], τi3[level], α)

τj0[level] = cinterp(τi0j0[level], τi1j0[level], τi2j0[level], τi3j0[level], α)

τj1[level] = cinterp(τi0j1[level], τi1j1[level], τi2j1[level], τi3j1[level], α)

τj2[level] = cinterp(τi0j2[level], τi1j2[level], τi2j2[level], τi3j2[level], α)

τj3[level] = cinterp(τi0j3[level], τi1j3[level], τi2j3[level], τi3j3[level], α)

τ2D[level] = cinterp(τj0[level], τj1[level], τj2[level], τj3[level], β)

\[\begin{aligned} \tau[level] &= \begin{cases} \tau_{2D}[level], & \text{for 2D image} \\ \tau_{1D}[level], & \text{for 1D image} \end{cases} \end{aligned}\]

Finally, mipmap filtering either selects a value from one mip level or computes a weighted average between neighboring mip levels:

\[\begin{aligned} \tau &= \begin{cases} \tau[d], & \text{for mip mode BASE or NEAREST} \\ reduce((1-\delta),\tau[d_{hi}],\delta,\tau[d_{lo}]), & \text{for mip mode LINEAR} \end{cases} \end{aligned}\]

15.9.4. Texel Anisotropic Filtering

Anisotropic filtering is enabled by the anisotropyEnable in the sampler. When enabled, the image filtering scheme accounts for a degree of anisotropy.

The particular scheme for anisotropic texture filtering is implementation dependent. Implementations should consider the magFilter, minFilter and mipmapMode of the sampler to control the specifics of the anisotropic filtering scheme used. In addition, implementations should consider minLod and maxLod of the sampler.

The following describes one particular approach to implementing anisotropic filtering for the 2D Image case, implementations may choose other methods:

Given a magFilter, minFilter of VK_FILTER_LINEAR and a mipmapMode of VK_SAMPLER_MIPMAP_MODE_NEAREST:

Instead of a single isotropic sample, N isotropic samples are be sampled within the image footprint of the image level d to approximate an anisotropic filter. The sum τ2Daniso is defined using the single isotropic τ2D(u,v) at level d.

\[\begin{aligned} \tau_{2Daniso} & = \frac{1}{N}\sum_{i=1}^{N} {\tau_{2D}\left ( u \left ( x - \frac{1}{2} + \frac{i}{N+1} , y \right ), \left ( v \left (x-\frac{1}{2}+\frac{i}{N+1} \right ), y \right ) \right )}, & \text{when}\ \rho_{x} > \rho_{y} \\ \tau_{2Daniso} &= \frac{1}{N}\sum_{i=1}^{N} {\tau_{2D}\left ( u \left ( x, y - \frac{1}{2} + \frac{i}{N+1} \right ), \left ( v \left (x,y-\frac{1}{2}+\frac{i}{N+1} \right ) \right ) \right )}, & \text{when}\ \rho_{y} \geq \rho_{x} \end{aligned}\]

When VkSamplerReductionModeCreateInfoEXT::reductionMode is set to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT, the above summation is used. If the reduction mode is VK_SAMPLER_REDUCTION_MODE_MIN_EXT or VK_SAMPLER_REDUCTION_MODE_MAX_EXT, then the value is instead computed as \tau_{2Daniso} = reduce(\tau_1, …​, \tau_N), combining all texel values with non-zero weights.

15.10. Image Operation Steps

Each step described in this chapter is performed by a subset of the image instructions:

  • Texel Input Validation Operations, Format Conversion, Texel Replacement, Conversion to RGBA, and Component Swizzle: Performed by all instructions except OpImageWrite.

  • Depth Comparison: Performed by OpImage*Dref instructions.

  • All Texel output operations: Performed by OpImageWrite.

  • Projection: Performed by all OpImage*Proj instructions.

  • Derivative Image Operations, Cube Map Operations, Scale Factor Operation, Level-of-Detail Operation and Image Level(s) Selection, and Texel Anisotropic Filtering: Performed by all OpImageSample* and OpImageSparseSample* instructions.

  • (s,t,r,q,a) to (u,v,w,a) Transformation, Wrapping, and (u,v,w,a) to (i,j,k,l,n) Transformation And Array Layer Selection: Performed by all OpImageSample, OpImageSparseSample, and OpImage*Gather instructions.

  • Texel Gathering: Performed by OpImage*Gather instructions.

  • Texel Filtering: Performed by all OpImageSample* and OpImageSparseSample* instructions.

  • Sparse Residency: Performed by all OpImageSparse* instructions.

16. Queries

Queries provide a mechanism to return information about the processing of a sequence of Vulkan commands. Query operations are asynchronous, and as such, their results are not returned immediately. Instead, their results, and their availability status, are stored in a Query Pool. The state of these queries can be read back on the host, or copied to a buffer object on the device.

The supported query types are Occlusion Queries, Pipeline Statistics Queries, and Timestamp Queries.

16.1. Query Pools

Queries are managed using query pool objects. Each query pool is a collection of a specific number of queries of a particular type.

Query pools are represented by VkQueryPool handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkQueryPool)

To create a query pool, call:

VkResult vkCreateQueryPool(
    VkDevice                                    device,
    const VkQueryPoolCreateInfo*                pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkQueryPool*                                pQueryPool);
  • device is the logical device that creates the query pool.

  • pCreateInfo is a pointer to an instance of the VkQueryPoolCreateInfo structure containing the number and type of queries to be managed by the pool.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pQueryPool is a pointer to a VkQueryPool handle in which the resulting query pool object is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkQueryPoolCreateInfo structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pQueryPool must be a valid pointer to a VkQueryPool handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkQueryPoolCreateInfo structure is defined as:

typedef struct VkQueryPoolCreateInfo {
    VkStructureType                  sType;
    const void*                      pNext;
    VkQueryPoolCreateFlags           flags;
    VkQueryType                      queryType;
    uint32_t                         queryCount;
    VkQueryPipelineStatisticFlags    pipelineStatistics;
} VkQueryPoolCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • queryType is a VkQueryType value specifying the type of queries managed by the pool.

  • queryCount is the number of queries managed by the pool.

  • pipelineStatistics is a bitmask of VkQueryPipelineStatisticFlagBits specifying which counters will be returned in queries on the new pool, as described below in Pipeline Statistics Queries.

pipelineStatistics is ignored if queryType is not VK_QUERY_TYPE_PIPELINE_STATISTICS.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_QUERY_POOL_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • queryType must be a valid VkQueryType value

typedef VkFlags VkQueryPoolCreateFlags;

VkQueryPoolCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

To destroy a query pool, call:

void vkDestroyQueryPool(
    VkDevice                                    device,
    VkQueryPool                                 queryPool,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the query pool.

  • queryPool is the query pool to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to queryPool must have completed execution

  • If VkAllocationCallbacks were provided when queryPool was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when queryPool was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If queryPool is not VK_NULL_HANDLE, queryPool must be a valid VkQueryPool handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If queryPool is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to queryPool must be externally synchronized

Possible values of VkQueryPoolCreateInfo::queryType, specifying the type of queries managed by the pool, are:

typedef enum VkQueryType {
    VK_QUERY_TYPE_OCCLUSION = 0,
    VK_QUERY_TYPE_PIPELINE_STATISTICS = 1,
    VK_QUERY_TYPE_TIMESTAMP = 2,
} VkQueryType;

16.2. Query Operation

The operation of queries is controlled by the commands vkCmdBeginQuery, vkCmdEndQuery, vkCmdResetQueryPool, vkCmdCopyQueryPoolResults, and vkCmdWriteTimestamp.

In order for a VkCommandBuffer to record query management commands, the queue family for which its VkCommandPool was created must support the appropriate type of operations (graphics, compute) suitable for the query type of a given query pool.

Each query in a query pool has a status that is either unavailable or available, and also has state to store the numerical results of a query operation of the type requested when the query pool was created. Resetting a query via vkCmdResetQueryPool sets the status to unavailable and makes the numerical results undefined. Performing a query operation with vkCmdBeginQuery and vkCmdEndQuery changes the status to available when the query finishes, and updates the numerical results. Both the availability status and numerical results are retrieved by calling either vkGetQueryPoolResults or vkCmdCopyQueryPoolResults.

Query commands, for the same query and submitted to the same queue, execute in their entirety in submission order, relative to each other. In effect there is an implicit execution dependency from each such query command to all query command previously submitted to the same queue. There is one significant exception to this; if the flags parameter of vkCmdCopyQueryPoolResults does not include VK_QUERY_RESULT_WAIT_BIT, execution of vkCmdCopyQueryPoolResults may happen-before the results of vkCmdEndQuery are available.

After query pool creation, each query is in an undefined state and must be reset prior to use. Queries must also be reset between uses. Using a query that has not been reset will result in undefined behavior.

If a logical device includes multiple physical devices, then each command that writes a query must execute on a single physical device, and any call to vkCmdBeginQuery must execute the corresponding vkCmdEndQuery command on the same physical device.

To reset a range of queries in a query pool, call:

void vkCmdResetQueryPool(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    firstQuery,
    uint32_t                                    queryCount);
  • commandBuffer is the command buffer into which this command will be recorded.

  • queryPool is the handle of the query pool managing the queries being reset.

  • firstQuery is the initial query index to reset.

  • queryCount is the number of queries to reset.

When executed on a queue, this command sets the status of query indices [firstQuery, firstQuery + queryCount - 1] to unavailable.

Valid Usage
  • firstQuery must be less than the number of queries in queryPool

  • The sum of firstQuery and queryCount must be less than or equal to the number of queries in queryPool

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • queryPool must be a valid VkQueryPool handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • Both of commandBuffer, and queryPool must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics
Compute

Once queries are reset and ready for use, query commands can be issued to a command buffer. Occlusion queries and pipeline statistics queries count events - drawn samples and pipeline stage invocations, respectively - resulting from commands that are recorded between a vkCmdBeginQuery command and a vkCmdEndQuery command within a specified command buffer, effectively scoping a set of drawing and/or compute commands. Timestamp queries write timestamps to a query pool.

A query must begin and end in the same command buffer, although if it is a primary command buffer, and the inherited queries feature is enabled, it can execute secondary command buffers during the query operation. For a secondary command buffer to be executed while a query is active, it must set the occlusionQueryEnable, queryFlags, and/or pipelineStatistics members of VkCommandBufferInheritanceInfo to conservative values, as described in the Command Buffer Recording section. A query must either begin and end inside the same subpass of a render pass instance, or must both begin and end outside of a render pass instance (i.e. contain entire render pass instances).

If queries are used while executing a render pass instance that has multiview enabled, the query uses N consecutive query indices in the query pool (starting at query) where N is the number of bits set in the view mask in the subpass the query is used in. How the numerical results of the query are distributed among the queries is implementation-dependent. For example, some implementations may write each view’s results to a distinct query, while other implementations may write the total result to the first query and write zero to the other queries. However, the sum of the results in all the queries must accurately reflect the total result of the query summed over all views. Applications can sum the results from all the queries to compute the total result.

Queries used with multiview rendering must not span subpasses, i.e. they must begin and end in the same subpass.

To begin a query, call:

void vkCmdBeginQuery(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    query,
    VkQueryControlFlags                         flags);
  • commandBuffer is the command buffer into which this command will be recorded.

  • queryPool is the query pool that will manage the results of the query.

  • query is the query index within the query pool that will contain the results.

  • flags is a bitmask of VkQueryControlFlagBits specifying constraints on the types of queries that can be performed.

If the queryType of the pool is VK_QUERY_TYPE_OCCLUSION and flags contains VK_QUERY_CONTROL_PRECISE_BIT, an implementation must return a result that matches the actual number of samples passed. This is described in more detail in Occlusion Queries.

After beginning a query, that query is considered active within the command buffer it was called in until that same query is ended. Queries active in a primary command buffer when secondary command buffers are executed are considered active for those secondary command buffers.

Valid Usage
  • queryPool must have been created with a queryType that differs from that of any queries that are active within commandBuffer

  • All queries used by the command must be unavailable

  • If the precise occlusion queries feature is not enabled, or the queryType used to create queryPool was not VK_QUERY_TYPE_OCCLUSION, flags must not contain VK_QUERY_CONTROL_PRECISE_BIT

  • query must be less than the number of queries in queryPool

  • If the queryType used to create queryPool was VK_QUERY_TYPE_OCCLUSION, the VkCommandPool that commandBuffer was allocated from must support graphics operations

  • If the queryType used to create queryPool was VK_QUERY_TYPE_PIPELINE_STATISTICS and any of the pipelineStatistics indicate graphics operations, the VkCommandPool that commandBuffer was allocated from must support graphics operations

  • If the queryType used to create queryPool was VK_QUERY_TYPE_PIPELINE_STATISTICS and any of the pipelineStatistics indicate compute operations, the VkCommandPool that commandBuffer was allocated from must support compute operations

  • commandBuffer must not be a protected command buffer

  • If vkCmdBeginQuery is called within a render pass instance, the sum of query and the number of bits set in the current subpass’s view mask must be less than or equal to the number of queries in queryPool

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • queryPool must be a valid VkQueryPool handle

  • flags must be a valid combination of VkQueryControlFlagBits values

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • Both of commandBuffer, and queryPool must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

Bits which can be set in vkCmdBeginQuery::flags, specifying constraints on the types of queries that can be performed, are:

typedef enum VkQueryControlFlagBits {
    VK_QUERY_CONTROL_PRECISE_BIT = 0x00000001,
} VkQueryControlFlagBits;
typedef VkFlags VkQueryControlFlags;

VkQueryControlFlags is a bitmask type for setting a mask of zero or more VkQueryControlFlagBits.

To end a query after the set of desired draw or dispatch commands is executed, call:

void vkCmdEndQuery(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    query);
  • commandBuffer is the command buffer into which this command will be recorded.

  • queryPool is the query pool that is managing the results of the query.

  • query is the query index within the query pool where the result is stored.

As queries operate asynchronously, ending a query does not immediately set the query’s status to available. A query is considered finished when the final results of the query are ready to be retrieved by vkGetQueryPoolResults and vkCmdCopyQueryPoolResults, and this is when the query’s status is set to available.

Once a query is ended the query must finish in finite time, unless the state of the query is changed using other commands, e.g. by issuing a reset of the query.

Valid Usage
  • All queries used by the command must be active

  • query must be less than the number of queries in queryPool

  • commandBuffer must not be a protected command buffer

  • If vkCmdEndQuery is called within a render pass instance, the sum of query and the number of bits set in the current subpass’s view mask must be less than or equal to the number of queries in queryPool

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • queryPool must be a valid VkQueryPool handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • Both of commandBuffer, and queryPool must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

An application can retrieve results either by requesting they be written into application-provided memory, or by requesting they be copied into a VkBuffer. In either case, the layout in memory is defined as follows:

  • The first query’s result is written starting at the first byte requested by the command, and each subsequent query’s result begins stride bytes later.

  • Each query’s result is a tightly packed array of unsigned integers, either 32- or 64-bits as requested by the command, storing the numerical results and, if requested, the availability status.

  • If VK_QUERY_RESULT_WITH_AVAILABILITY_BIT is used, the final element of each query’s result is an integer indicating whether the query’s result is available, with any non-zero value indicating that it is available.

  • Occlusion queries write one integer value - the number of samples passed. Pipeline statistics queries write one integer value for each bit that is enabled in the pipelineStatistics when the pool is created, and the statistics values are written in bit order starting from the least significant bit. Timestamps write one integer value.

  • If more than one query is retrieved and stride is not at least as large as the size of the array of integers corresponding to a single query, the values written to memory are undefined.

To retrieve status and results for a set of queries, call:

VkResult vkGetQueryPoolResults(
    VkDevice                                    device,
    VkQueryPool                                 queryPool,
    uint32_t                                    firstQuery,
    uint32_t                                    queryCount,
    size_t                                      dataSize,
    void*                                       pData,
    VkDeviceSize                                stride,
    VkQueryResultFlags                          flags);
  • device is the logical device that owns the query pool.

  • queryPool is the query pool managing the queries containing the desired results.

  • firstQuery is the initial query index.

  • queryCount is the number of queries. firstQuery and queryCount together define a range of queries. For pipeline statistics queries, each query index in the pool contains one integer value for each bit that is enabled in VkQueryPoolCreateInfo::pipelineStatistics when the pool is created.

  • dataSize is the size in bytes of the buffer pointed to by pData.

  • pData is a pointer to a user-allocated buffer where the results will be written

  • stride is the stride in bytes between results for individual queries within pData.

  • flags is a bitmask of VkQueryResultFlagBits specifying how and when results are returned.

If no bits are set in flags, and all requested queries are in the available state, results are written as an array of 32-bit unsigned integer values. The behavior when not all queries are available, is described below.

If VK_QUERY_RESULT_64_BIT is not set and the result overflows a 32-bit value, the value may either wrap or saturate. Similarly, if VK_QUERY_RESULT_64_BIT is set and the result overflows a 64-bit value, the value may either wrap or saturate.

If VK_QUERY_RESULT_WAIT_BIT is set, Vulkan will wait for each query to be in the available state before retrieving the numerical results for that query. In this case, vkGetQueryPoolResults is guaranteed to succeed and return VK_SUCCESS if the queries become available in a finite time (i.e. if they have been issued and not reset). If queries will never finish (e.g. due to being reset but not issued), then vkGetQueryPoolResults may not return in finite time.

If VK_QUERY_RESULT_WAIT_BIT and VK_QUERY_RESULT_PARTIAL_BIT are both not set then no result values are written to pData for queries that are in the unavailable state at the time of the call, and vkGetQueryPoolResults returns VK_NOT_READY. However, availability state is still written to pData for those queries if VK_QUERY_RESULT_WITH_AVAILABILITY_BIT is set.

Note

Applications must take care to ensure that use of the VK_QUERY_RESULT_WAIT_BIT bit has the desired effect.

For example, if a query has been used previously and a command buffer records the commands vkCmdResetQueryPool, vkCmdBeginQuery, and vkCmdEndQuery for that query, then the query will remain in the available state until the vkCmdResetQueryPool command executes on a queue. Applications can use fences or events to ensure that a query has already been reset before checking for its results or availability status. Otherwise, a stale value could be returned from a previous use of the query.

The above also applies when VK_QUERY_RESULT_WAIT_BIT is used in combination with VK_QUERY_RESULT_WITH_AVAILABILITY_BIT. In this case, the returned availability status may reflect the result of a previous use of the query unless the vkCmdResetQueryPool command has been executed since the last use of the query.

Note

Applications can double-buffer query pool usage, with a pool per frame, and reset queries at the end of the frame in which they are read.

If VK_QUERY_RESULT_PARTIAL_BIT is set, VK_QUERY_RESULT_WAIT_BIT is not set, and the query’s status is unavailable, an intermediate result value between zero and the final result value is written to pData for that query.

VK_QUERY_RESULT_PARTIAL_BIT must not be used if the pool’s queryType is VK_QUERY_TYPE_TIMESTAMP.

If VK_QUERY_RESULT_WITH_AVAILABILITY_BIT is set, the final integer value written for each query is non-zero if the query’s status was available or zero if the status was unavailable. When VK_QUERY_RESULT_WITH_AVAILABILITY_BIT is used, implementations must guarantee that if they return a non-zero availability value then the numerical results must be valid, assuming the results are not reset by a subsequent command.

Note

Satisfying this guarantee may require careful ordering by the application, e.g. to read the availability status before reading the results.

Valid Usage
  • firstQuery must be less than the number of queries in queryPool

  • If VK_QUERY_RESULT_64_BIT is not set in flags then pData and stride must be multiples of 4

  • If VK_QUERY_RESULT_64_BIT is set in flags then pData and stride must be multiples of 8

  • The sum of firstQuery and queryCount must be less than or equal to the number of queries in queryPool

  • dataSize must be large enough to contain the result of each query, as described here

  • If the queryType used to create queryPool was VK_QUERY_TYPE_TIMESTAMP, flags must not contain VK_QUERY_RESULT_PARTIAL_BIT

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • queryPool must be a valid VkQueryPool handle

  • pData must be a valid pointer to an array of dataSize bytes

  • flags must be a valid combination of VkQueryResultFlagBits values

  • dataSize must be greater than 0

  • queryPool must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_NOT_READY

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

Bits which can be set in vkGetQueryPoolResults::flags and vkCmdCopyQueryPoolResults::flags, specifying how and when results are returned, are:

typedef enum VkQueryResultFlagBits {
    VK_QUERY_RESULT_64_BIT = 0x00000001,
    VK_QUERY_RESULT_WAIT_BIT = 0x00000002,
    VK_QUERY_RESULT_WITH_AVAILABILITY_BIT = 0x00000004,
    VK_QUERY_RESULT_PARTIAL_BIT = 0x00000008,
} VkQueryResultFlagBits;
  • VK_QUERY_RESULT_64_BIT specifies the results will be written as an array of 64-bit unsigned integer values. If this bit is not set, the results will be written as an array of 32-bit unsigned integer values.

  • VK_QUERY_RESULT_WAIT_BIT specifies that Vulkan will wait for each query’s status to become available before retrieving its results.

  • VK_QUERY_RESULT_WITH_AVAILABILITY_BIT specifies that the availability status accompanies the results.

  • VK_QUERY_RESULT_PARTIAL_BIT specifies that returning partial results is acceptable.

typedef VkFlags VkQueryResultFlags;

VkQueryResultFlags is a bitmask type for setting a mask of zero or more VkQueryResultFlagBits.

To copy query statuses and numerical results directly to buffer memory, call:

void vkCmdCopyQueryPoolResults(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    firstQuery,
    uint32_t                                    queryCount,
    VkBuffer                                    dstBuffer,
    VkDeviceSize                                dstOffset,
    VkDeviceSize                                stride,
    VkQueryResultFlags                          flags);
  • commandBuffer is the command buffer into which this command will be recorded.

  • queryPool is the query pool managing the queries containing the desired results.

  • firstQuery is the initial query index.

  • queryCount is the number of queries. firstQuery and queryCount together define a range of queries.

  • dstBuffer is a VkBuffer object that will receive the results of the copy command.

  • dstOffset is an offset into dstBuffer.

  • stride is the stride in bytes between results for individual queries within dstBuffer. The required size of the backing memory for dstBuffer is determined as described above for vkGetQueryPoolResults.

  • flags is a bitmask of VkQueryResultFlagBits specifying how and when results are returned.

vkCmdCopyQueryPoolResults is guaranteed to see the effect of previous uses of vkCmdResetQueryPool in the same queue, without any additional synchronization. Thus, the results will always reflect the most recent use of the query.

flags has the same possible values described above for the flags parameter of vkGetQueryPoolResults, but the different style of execution causes some subtle behavioral differences. Because vkCmdCopyQueryPoolResults executes in order with respect to other query commands, there is less ambiguity about which use of a query is being requested.

If no bits are set in flags, results for all requested queries in the available state are written as 32-bit unsigned integer values, and nothing is written for queries in the unavailable state.

If VK_QUERY_RESULT_64_BIT is set, the results are written as an array of 64-bit unsigned integer values as described for vkGetQueryPoolResults.

If VK_QUERY_RESULT_WAIT_BIT is set, the implementation will wait for each query’s status to be in the available state before retrieving the numerical results for that query. This is guaranteed to reflect the most recent use of the query on the same queue, assuming that the query is not being simultaneously used by other queues. If the query does not become available in a finite amount of time (e.g. due to not issuing a query since the last reset), a VK_ERROR_DEVICE_LOST error may occur.

Similarly, if VK_QUERY_RESULT_WITH_AVAILABILITY_BIT is set and VK_QUERY_RESULT_WAIT_BIT is not set, the availability is guaranteed to reflect the most recent use of the query on the same queue, assuming that the query is not being simultaneously used by other queues. As with vkGetQueryPoolResults, implementations must guarantee that if they return a non-zero availability value, then the numerical results are valid.

If VK_QUERY_RESULT_PARTIAL_BIT is set, VK_QUERY_RESULT_WAIT_BIT is not set, and the query’s status is unavailable, an intermediate result value between zero and the final result value is written for that query.

VK_QUERY_RESULT_PARTIAL_BIT must not be used if the pool’s queryType is VK_QUERY_TYPE_TIMESTAMP.

vkCmdCopyQueryPoolResults is considered to be a transfer operation, and its writes to buffer memory must be synchronized using VK_PIPELINE_STAGE_TRANSFER_BIT and VK_ACCESS_TRANSFER_WRITE_BIT before using the results.

Valid Usage
  • dstOffset must be less than the size of dstBuffer

  • firstQuery must be less than the number of queries in queryPool

  • The sum of firstQuery and queryCount must be less than or equal to the number of queries in queryPool

  • If VK_QUERY_RESULT_64_BIT is not set in flags then dstOffset and stride must be multiples of 4

  • If VK_QUERY_RESULT_64_BIT is set in flags then dstOffset and stride must be multiples of 8

  • dstBuffer must have enough storage, from dstOffset, to contain the result of each query, as described here

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • If the queryType used to create queryPool was VK_QUERY_TYPE_TIMESTAMP, flags must not contain VK_QUERY_RESULT_PARTIAL_BIT

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • queryPool must be a valid VkQueryPool handle

  • dstBuffer must be a valid VkBuffer handle

  • flags must be a valid combination of VkQueryResultFlagBits values

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • Each of commandBuffer, dstBuffer, and queryPool must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics
Compute

Transfer

Rendering operations such as clears, MSAA resolves, attachment load/store operations, and blits may count towards the results of queries. This behavior is implementation-dependent and may vary depending on the path used within an implementation. For example, some implementations have several types of clears, some of which may include vertices and some not.

16.3. Occlusion Queries

Occlusion queries track the number of samples that pass the per-fragment tests for a set of drawing commands. As such, occlusion queries are only available on queue families supporting graphics operations. The application can then use these results to inform future rendering decisions. An occlusion query is begun and ended by calling vkCmdBeginQuery and vkCmdEndQuery, respectively. When an occlusion query begins, the count of passing samples always starts at zero. For each drawing command, the count is incremented as described in Sample Counting. If flags does not contain VK_QUERY_CONTROL_PRECISE_BIT an implementation may generate any non-zero result value for the query if the count of passing samples is non-zero.

Note

Not setting VK_QUERY_CONTROL_PRECISE_BIT mode may be more efficient on some implementations, and should be used where it is sufficient to know a boolean result on whether any samples passed the per-fragment tests. In this case, some implementations may only return zero or one, indifferent to the actual number of samples passing the per-fragment tests.

When an occlusion query finishes, the result for that query is marked as available. The application can then either copy the result to a buffer (via vkCmdCopyQueryPoolResults) or request it be put into host memory (via vkGetQueryPoolResults).

Note

If occluding geometry is not drawn first, samples can pass the depth test, but still not be visible in a final image.

16.4. Pipeline Statistics Queries

Pipeline statistics queries allow the application to sample a specified set of VkPipeline counters. These counters are accumulated by Vulkan for a set of either draw or dispatch commands while a pipeline statistics query is active. As such, pipeline statistics queries are available on queue families supporting either graphics or compute operations. Further, the availability of pipeline statistics queries is indicated by the pipelineStatisticsQuery member of the VkPhysicalDeviceFeatures object (see vkGetPhysicalDeviceFeatures and vkCreateDevice for detecting and requesting this query type on a VkDevice).

A pipeline statistics query is begun and ended by calling vkCmdBeginQuery and vkCmdEndQuery, respectively. When a pipeline statistics query begins, all statistics counters are set to zero. While the query is active, the pipeline type determines which set of statistics are available, but these must be configured on the query pool when it is created. If a statistic counter is issued on a command buffer that does not support the corresponding operation, that counter is undefined after the query has finished. At least one statistic counter relevant to the operations supported on the recording command buffer must be enabled.

Bits which can be set to individually enable pipeline statistics counters for query pools with VkQueryPoolCreateInfo::pipelineStatistics, and for secondary command buffers with VkCommandBufferInheritanceInfo::pipelineStatistics, are:

typedef enum VkQueryPipelineStatisticFlagBits {
    VK_QUERY_PIPELINE_STATISTIC_INPUT_ASSEMBLY_VERTICES_BIT = 0x00000001,
    VK_QUERY_PIPELINE_STATISTIC_INPUT_ASSEMBLY_PRIMITIVES_BIT = 0x00000002,
    VK_QUERY_PIPELINE_STATISTIC_VERTEX_SHADER_INVOCATIONS_BIT = 0x00000004,
    VK_QUERY_PIPELINE_STATISTIC_GEOMETRY_SHADER_INVOCATIONS_BIT = 0x00000008,
    VK_QUERY_PIPELINE_STATISTIC_GEOMETRY_SHADER_PRIMITIVES_BIT = 0x00000010,
    VK_QUERY_PIPELINE_STATISTIC_CLIPPING_INVOCATIONS_BIT = 0x00000020,
    VK_QUERY_PIPELINE_STATISTIC_CLIPPING_PRIMITIVES_BIT = 0x00000040,
    VK_QUERY_PIPELINE_STATISTIC_FRAGMENT_SHADER_INVOCATIONS_BIT = 0x00000080,
    VK_QUERY_PIPELINE_STATISTIC_TESSELLATION_CONTROL_SHADER_PATCHES_BIT = 0x00000100,
    VK_QUERY_PIPELINE_STATISTIC_TESSELLATION_EVALUATION_SHADER_INVOCATIONS_BIT = 0x00000200,
    VK_QUERY_PIPELINE_STATISTIC_COMPUTE_SHADER_INVOCATIONS_BIT = 0x00000400,
} VkQueryPipelineStatisticFlagBits;
  • VK_QUERY_PIPELINE_STATISTIC_INPUT_ASSEMBLY_VERTICES_BIT specifies that queries managed by the pool will count the number of vertices processed by the input assembly stage. Vertices corresponding to incomplete primitives may contribute to the count.

  • VK_QUERY_PIPELINE_STATISTIC_INPUT_ASSEMBLY_PRIMITIVES_BIT specifies that queries managed by the pool will count the number of primitives processed by the input assembly stage. If primitive restart is enabled, restarting the primitive topology has no effect on the count. Incomplete primitives may be counted.

  • VK_QUERY_PIPELINE_STATISTIC_VERTEX_SHADER_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of vertex shader invocations. This counter’s value is incremented each time a vertex shader is invoked.

  • VK_QUERY_PIPELINE_STATISTIC_GEOMETRY_SHADER_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of geometry shader invocations. This counter’s value is incremented each time a geometry shader is invoked. In the case of instanced geometry shaders, the geometry shader invocations count is incremented for each separate instanced invocation.

  • VK_QUERY_PIPELINE_STATISTIC_GEOMETRY_SHADER_PRIMITIVES_BIT specifies that queries managed by the pool will count the number of primitives generated by geometry shader invocations. The counter’s value is incremented each time the geometry shader emits a primitive. Restarting primitive topology using the SPIR-V instructions OpEndPrimitive or OpEndStreamPrimitive has no effect on the geometry shader output primitives count.

  • VK_QUERY_PIPELINE_STATISTIC_CLIPPING_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of primitives processed by the Primitive Clipping stage of the pipeline. The counter’s value is incremented each time a primitive reaches the primitive clipping stage.

  • VK_QUERY_PIPELINE_STATISTIC_CLIPPING_PRIMITIVES_BIT specifies that queries managed by the pool will count the number of primitives output by the Primitive Clipping stage of the pipeline. The counter’s value is incremented each time a primitive passes the primitive clipping stage. The actual number of primitives output by the primitive clipping stage for a particular input primitive is implementation-dependent but must satisfy the following conditions:

    • If at least one vertex of the input primitive lies inside the clipping volume, the counter is incremented by one or more.

    • Otherwise, the counter is incremented by zero or more.

  • VK_QUERY_PIPELINE_STATISTIC_FRAGMENT_SHADER_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of fragment shader invocations. The counter’s value is incremented each time the fragment shader is invoked.

  • VK_QUERY_PIPELINE_STATISTIC_TESSELLATION_CONTROL_SHADER_PATCHES_BIT specifies that queries managed by the pool will count the number of patches processed by the tessellation control shader. The counter’s value is incremented once for each patch for which a tessellation control shader is invoked.

  • VK_QUERY_PIPELINE_STATISTIC_TESSELLATION_EVALUATION_SHADER_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of invocations of the tessellation evaluation shader. The counter’s value is incremented each time the tessellation evaluation shader is invoked.

  • VK_QUERY_PIPELINE_STATISTIC_COMPUTE_SHADER_INVOCATIONS_BIT specifies that queries managed by the pool will count the number of compute shader invocations. The counter’s value is incremented every time the compute shader is invoked. Implementations may skip the execution of certain compute shader invocations or execute additional compute shader invocations for implementation-dependent reasons as long as the results of rendering otherwise remain unchanged.

These values are intended to measure relative statistics on one implementation. Various device architectures will count these values differently. Any or all counters may be affected by the issues described in Query Operation.

Note

For example, tile-based rendering devices may need to replay the scene multiple times, affecting some of the counts.

If a pipeline has rasterizerDiscardEnable enabled, implementations may discard primitives after the final vertex processing stage. As a result, if rasterizerDiscardEnable is enabled, the clipping input and output primitives counters may not be incremented.

When a pipeline statistics query finishes, the result for that query is marked as available. The application can copy the result to a buffer (via vkCmdCopyQueryPoolResults), or request it be put into host memory (via vkGetQueryPoolResults).

typedef VkFlags VkQueryPipelineStatisticFlags;

VkQueryPipelineStatisticFlags is a bitmask type for setting a mask of zero or more VkQueryPipelineStatisticFlagBits.

16.5. Timestamp Queries

Timestamps provide applications with a mechanism for timing the execution of commands. A timestamp is an integer value generated by the VkPhysicalDevice. Unlike other queries, timestamps do not operate over a range, and so do not use vkCmdBeginQuery or vkCmdEndQuery. The mechanism is built around a set of commands that allow the application to tell the VkPhysicalDevice to write timestamp values to a query pool and then either read timestamp values on the host (using vkGetQueryPoolResults) or copy timestamp values to a VkBuffer (using vkCmdCopyQueryPoolResults). The application can then compute differences between timestamps to determine execution time.

The number of valid bits in a timestamp value is determined by the VkQueueFamilyProperties::timestampValidBits property of the queue on which the timestamp is written. Timestamps are supported on any queue which reports a non-zero value for timestampValidBits via vkGetPhysicalDeviceQueueFamilyProperties. If the timestampComputeAndGraphics limit is VK_TRUE, timestamps are supported by every queue family that supports either graphics or compute operations (see VkQueueFamilyProperties).

The number of nanoseconds it takes for a timestamp value to be incremented by 1 can be obtained from VkPhysicalDeviceLimits::timestampPeriod after a call to vkGetPhysicalDeviceProperties.

To request a timestamp, call:

void vkCmdWriteTimestamp(
    VkCommandBuffer                             commandBuffer,
    VkPipelineStageFlagBits                     pipelineStage,
    VkQueryPool                                 queryPool,
    uint32_t                                    query);
  • commandBuffer is the command buffer into which the command will be recorded.

  • pipelineStage is one of the VkPipelineStageFlagBits, specifying a stage of the pipeline.

  • queryPool is the query pool that will manage the timestamp.

  • query is the query within the query pool that will contain the timestamp.

vkCmdWriteTimestamp latches the value of the timer when all previous commands have completed executing as far as the specified pipeline stage, and writes the timestamp value to memory. When the timestamp value is written, the availability status of the query is set to available.

Note

If an implementation is unable to detect completion and latch the timer at any specific stage of the pipeline, it may instead do so at any logically later stage.

vkCmdCopyQueryPoolResults can then be called to copy the timestamp value from the query pool into buffer memory, with ordering and synchronization behavior equivalent to how other queries operate. Timestamp values can also be retrieved from the query pool using vkGetQueryPoolResults. As with other queries, the query must be reset using vkCmdResetQueryPool before requesting the timestamp value be written to it.

While vkCmdWriteTimestamp can be called inside or outside of a render pass instance, vkCmdCopyQueryPoolResults must only be called outside of a render pass instance.

Timestamps may only be meaningfully compared if they are written by commands submitted to the same queue.

Note

An example of such a comparison is determining the execution time of a sequence of commands.

If vkCmdWriteTimestamp is called while executing a render pass instance that has multiview enabled, the timestamp uses N consecutive query indices in the query pool (starting at query) where N is the number of bits set in the view mask of the subpass the command is executed in. The resulting query values are determined by an implementation-dependent choice of one of the following behaviors:

  • The first query is a timestamp value and (if more than one bit is set in the view mask) zero is written to the remaining queries. If two timestamps are written in the same subpass, the sum of the execution time of all views between those commands is the difference between the first query written by each command.

  • All N queries are timestamp values. If two timestamps are written in the same subpass, the sum of the execution time of all views between those commands is the sum of the difference between corresponding queries written by each command. The difference between corresponding queries may be the execution time of a single view.

In either case, the application can sum the differences between all N queries to determine the total execution time.

Valid Usage
  • queryPool must have been created with a queryType of VK_QUERY_TYPE_TIMESTAMP

  • The query identified by queryPool and query must be unavailable

  • The command pool’s queue family must support a non-zero timestampValidBits

  • All queries used by the command must be unavailable

  • If vkCmdWriteTimestamp is called within a render pass instance, the sum of query and the number of bits set in the current subpass’s view mask must be less than or equal to the number of queries in queryPool

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pipelineStage must be a valid VkPipelineStageFlagBits value

  • queryPool must be a valid VkQueryPool handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • Both of commandBuffer, and queryPool must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Transfer
Graphics
Compute

Transfer

17. Clear Commands

17.1. Clearing Images Outside A Render Pass Instance

Color and depth/stencil images can be cleared outside a render pass instance using vkCmdClearColorImage or vkCmdClearDepthStencilImage, respectively. These commands are only allowed outside of a render pass instance.

To clear one or more subranges of a color image, call:

void vkCmdClearColorImage(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     image,
    VkImageLayout                               imageLayout,
    const VkClearColorValue*                    pColor,
    uint32_t                                    rangeCount,
    const VkImageSubresourceRange*              pRanges);
  • commandBuffer is the command buffer into which the command will be recorded.

  • image is the image to be cleared.

  • imageLayout specifies the current layout of the image subresource ranges to be cleared, and must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_GENERAL or VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL.

  • pColor is a pointer to a VkClearColorValue structure that contains the values the image subresource ranges will be cleared to (see Clear Values below).

  • rangeCount is the number of image subresource range structures in pRanges.

  • pRanges points to an array of VkImageSubresourceRange structures that describe a range of mipmap levels, array layers, and aspects to be cleared, as described in Image Views. The aspectMask of all image subresource ranges must only include VK_IMAGE_ASPECT_COLOR_BIT.

Each specified range in pRanges is cleared to the value specified by pColor.

Valid Usage
  • image must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT usage flag

  • image must not use a format listed in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views

  • If image is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • imageLayout must specify the layout of the image subresource ranges of image specified in pRanges at the time this command is executed on a VkDevice

  • imageLayout must be VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_GENERAL, or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR

  • The VkImageSubresourceRange::baseMipLevel members of the elements of the pRanges array must each be less than the mipLevels specified in VkImageCreateInfo when image was created

  • For each VkImageSubresourceRange element of pRanges, if the levelCount member is not VK_REMAINING_MIP_LEVELS, then baseMipLevel + levelCount must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • The VkImageSubresourceRange::baseArrayLayer members of the elements of the pRanges array must each be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • For each VkImageSubresourceRange element of pRanges, if the layerCount member is not VK_REMAINING_ARRAY_LAYERS, then baseArrayLayer + layerCount must be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • image must not have a compressed or depth/stencil format

  • If commandBuffer is an unprotected command buffer, then image must not be a protected image

  • If commandBuffer is a protected command buffer, then image must not be an unprotected image

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • image must be a valid VkImage handle

  • imageLayout must be a valid VkImageLayout value

  • pColor must be a valid pointer to a valid VkClearColorValue union

  • pRanges must be a valid pointer to an array of rangeCount valid VkImageSubresourceRange structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • rangeCount must be greater than 0

  • Both of commandBuffer, and image must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics
Compute

Transfer

To clear one or more subranges of a depth/stencil image, call:

void vkCmdClearDepthStencilImage(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     image,
    VkImageLayout                               imageLayout,
    const VkClearDepthStencilValue*             pDepthStencil,
    uint32_t                                    rangeCount,
    const VkImageSubresourceRange*              pRanges);
  • commandBuffer is the command buffer into which the command will be recorded.

  • image is the image to be cleared.

  • imageLayout specifies the current layout of the image subresource ranges to be cleared, and must be VK_IMAGE_LAYOUT_GENERAL or VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL.

  • pDepthStencil is a pointer to a VkClearDepthStencilValue structure that contains the values the depth and stencil image subresource ranges will be cleared to (see Clear Values below).

  • rangeCount is the number of image subresource range structures in pRanges.

  • pRanges points to an array of VkImageSubresourceRange structures that describe a range of mipmap levels, array layers, and aspects to be cleared, as described in Image Views. The aspectMask of each image subresource range in pRanges can include VK_IMAGE_ASPECT_DEPTH_BIT if the image format has a depth component, and VK_IMAGE_ASPECT_STENCIL_BIT if the image format has a stencil component. pDepthStencil is a pointer to a VkClearDepthStencilValue structure that contains the values the image subresource ranges will be cleared to (see Clear Values below).

Valid Usage
  • image must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT usage flag

  • If image is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • imageLayout must specify the layout of the image subresource ranges of image specified in pRanges at the time this command is executed on a VkDevice

  • imageLayout must be either of VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • The VkImageSubresourceRange::baseMipLevel members of the elements of the pRanges array must each be less than the mipLevels specified in VkImageCreateInfo when image was created

  • For each VkImageSubresourceRange element of pRanges, if the levelCount member is not VK_REMAINING_MIP_LEVELS, then baseMipLevel + levelCount must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • The VkImageSubresourceRange::baseArrayLayer members of the elements of the pRanges array must each be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • For each VkImageSubresourceRange element of pRanges, if the layerCount member is not VK_REMAINING_ARRAY_LAYERS, then baseArrayLayer + layerCount must be less than the arrayLayers specified in VkImageCreateInfo when image was created

  • image must have a depth/stencil format

  • If commandBuffer is an unprotected command buffer, then image must not be a protected image

  • If commandBuffer is a protected command buffer, then image must not be an unprotected image

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • image must be a valid VkImage handle

  • imageLayout must be a valid VkImageLayout value

  • pDepthStencil must be a valid pointer to a valid VkClearDepthStencilValue structure

  • pRanges must be a valid pointer to an array of rangeCount valid VkImageSubresourceRange structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called outside of a render pass instance

  • rangeCount must be greater than 0

  • Both of commandBuffer, and image must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics

Transfer

Clears outside render pass instances are treated as transfer operations for the purposes of memory barriers.

17.2. Clearing Images Inside A Render Pass Instance

To clear one or more regions of color and depth/stencil attachments inside a render pass instance, call:

void vkCmdClearAttachments(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    attachmentCount,
    const VkClearAttachment*                    pAttachments,
    uint32_t                                    rectCount,
    const VkClearRect*                          pRects);
  • commandBuffer is the command buffer into which the command will be recorded.

  • attachmentCount is the number of entries in the pAttachments array.

  • pAttachments is a pointer to an array of VkClearAttachment structures defining the attachments to clear and the clear values to use.

  • rectCount is the number of entries in the pRects array.

  • pRects points to an array of VkClearRect structures defining regions within each selected attachment to clear.

vkCmdClearAttachments can clear multiple regions of each attachment used in the current subpass of a render pass instance. This command must be called only inside a render pass instance, and implicitly selects the images to clear based on the current framebuffer attachments and the command parameters.

Valid Usage
  • If the aspectMask member of any element of pAttachments contains VK_IMAGE_ASPECT_COLOR_BIT, the colorAttachment member of that element must refer to a valid color attachment in the current subpass

  • The rectangular region specified by each element of pRects must be contained within the render area of the current render pass instance

  • The layers specified by each element of pRects must be contained within every attachment that pAttachments refers to

  • The layerCount member of each element of pRects must not be 0

  • If the render pass instance this is recorded in uses multiview, then baseArrayLayer must be zero and layerCount must be one.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pAttachments must be a valid pointer to an array of attachmentCount valid VkClearAttachment structures

  • pRects must be a valid pointer to an array of rectCount VkClearRect structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • attachmentCount must be greater than 0

  • rectCount must be greater than 0

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

The VkClearRect structure is defined as:

typedef struct VkClearRect {
    VkRect2D    rect;
    uint32_t    baseArrayLayer;
    uint32_t    layerCount;
} VkClearRect;
  • rect is the two-dimensional region to be cleared.

  • baseArrayLayer is the first layer to be cleared.

  • layerCount is the number of layers to clear.

The layers [baseArrayLayer, baseArrayLayer + layerCount) counting from the base layer of the attachment image view are cleared.

The VkClearAttachment structure is defined as:

typedef struct VkClearAttachment {
    VkImageAspectFlags    aspectMask;
    uint32_t              colorAttachment;
    VkClearValue          clearValue;
} VkClearAttachment;
  • aspectMask is a mask selecting the color, depth and/or stencil aspects of the attachment to be cleared. aspectMask can include VK_IMAGE_ASPECT_COLOR_BIT for color attachments, VK_IMAGE_ASPECT_DEPTH_BIT for depth/stencil attachments with a depth component, and VK_IMAGE_ASPECT_STENCIL_BIT for depth/stencil attachments with a stencil component. If the subpass’s depth/stencil attachment is VK_ATTACHMENT_UNUSED, then the clear has no effect.

  • colorAttachment is only meaningful if VK_IMAGE_ASPECT_COLOR_BIT is set in aspectMask, in which case it is an index to the pColorAttachments array in the VkSubpassDescription structure of the current subpass which selects the color attachment to clear. If colorAttachment is VK_ATTACHMENT_UNUSED then the clear has no effect.

  • clearValue is the color or depth/stencil value to clear the attachment to, as described in Clear Values below.

No memory barriers are needed between vkCmdClearAttachments and preceding or subsequent draw or attachment clear commands in the same subpass.

The vkCmdClearAttachments command is not affected by the bound pipeline state.

Attachments can also be cleared at the beginning of a render pass instance by setting loadOp (or stencilLoadOp) of VkAttachmentDescription to VK_ATTACHMENT_LOAD_OP_CLEAR, as described for vkCreateRenderPass.

Valid Usage
  • If aspectMask includes VK_IMAGE_ASPECT_COLOR_BIT, it must not include VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT

  • aspectMask must not include VK_IMAGE_ASPECT_METADATA_BIT

  • clearValue must be a valid VkClearValue union

  • If commandBuffer is an unprotected command buffer, then the attachment to be cleared must not be a protected image.

  • If commandBuffer is a protected command buffer, then the attachment to be cleared must not be an unprotected image.

Valid Usage (Implicit)

17.3. Clear Values

The VkClearColorValue structure is defined as:

typedef union VkClearColorValue {
    float       float32[4];
    int32_t     int32[4];
    uint32_t    uint32[4];
} VkClearColorValue;
  • float32 are the color clear values when the format of the image or attachment is one of the formats in the Interpretation of Numeric Format table other than signed integer (SINT) or unsigned integer (UINT). Floating point values are automatically converted to the format of the image, with the clear value being treated as linear if the image is sRGB.

  • int32 are the color clear values when the format of the image or attachment is signed integer (SINT). Signed integer values are converted to the format of the image by casting to the smaller type (with negative 32-bit values mapping to negative values in the smaller type). If the integer clear value is not representable in the target type (e.g. would overflow in conversion to that type), the clear value is undefined.

  • uint32 are the color clear values when the format of the image or attachment is unsigned integer (UINT). Unsigned integer values are converted to the format of the image by casting to the integer type with fewer bits.

The four array elements of the clear color map to R, G, B, and A components of image formats, in order.

If the image has more than one sample, the same value is written to all samples for any pixels being cleared.

The VkClearDepthStencilValue structure is defined as:

typedef struct VkClearDepthStencilValue {
    float       depth;
    uint32_t    stencil;
} VkClearDepthStencilValue;
  • depth is the clear value for the depth aspect of the depth/stencil attachment. It is a floating-point value which is automatically converted to the attachment’s format.

  • stencil is the clear value for the stencil aspect of the depth/stencil attachment. It is a 32-bit integer value which is converted to the attachment’s format by taking the appropriate number of LSBs.

Valid Usage

The VkClearValue union is defined as:

typedef union VkClearValue {
    VkClearColorValue           color;
    VkClearDepthStencilValue    depthStencil;
} VkClearValue;
  • color specifies the color image clear values to use when clearing a color image or attachment.

  • depthStencil specifies the depth and stencil clear values to use when clearing a depth/stencil image or attachment.

This union is used where part of the API requires either color or depth/stencil clear values, depending on the attachment, and defines the initial clear values in the VkRenderPassBeginInfo structure.

Valid Usage
  • depthStencil must be a valid VkClearDepthStencilValue structure

17.4. Filling Buffers

To clear buffer data, call:

void vkCmdFillBuffer(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    dstBuffer,
    VkDeviceSize                                dstOffset,
    VkDeviceSize                                size,
    uint32_t                                    data);
  • commandBuffer is the command buffer into which the command will be recorded.

  • dstBuffer is the buffer to be filled.

  • dstOffset is the byte offset into the buffer at which to start filling, and must be a multiple of 4.

  • size is the number of bytes to fill, and must be either a multiple of 4, or VK_WHOLE_SIZE to fill the range from offset to the end of the buffer. If VK_WHOLE_SIZE is used and the remaining size of the buffer is not a multiple of 4, then the nearest smaller multiple is used.

  • data is the 4-byte word written repeatedly to the buffer to fill size bytes of data. The data word is written to memory according to the host endianness.

vkCmdFillBuffer is treated as “transfer” operation for the purposes of synchronization barriers. The VK_BUFFER_USAGE_TRANSFER_DST_BIT must be specified in usage of VkBufferCreateInfo in order for the buffer to be compatible with vkCmdFillBuffer.

Valid Usage
  • dstOffset must be less than the size of dstBuffer

  • dstOffset must be a multiple of 4

  • If size is not equal to VK_WHOLE_SIZE, size must be greater than 0

  • If size is not equal to VK_WHOLE_SIZE, size must be less than or equal to the size of dstBuffer minus dstOffset

  • If size is not equal to VK_WHOLE_SIZE, size must be a multiple of 4

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • If commandBuffer is an unprotected command buffer, then dstBuffer must not be a protected buffer

  • If commandBuffer is a protected command buffer, then dstBuffer must not be an unprotected buffer

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • dstBuffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics or compute operations

  • This command must only be called outside of a render pass instance

  • Both of commandBuffer, and dstBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

17.5. Updating Buffers

To update buffer data inline in a command buffer, call:

void vkCmdUpdateBuffer(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    dstBuffer,
    VkDeviceSize                                dstOffset,
    VkDeviceSize                                dataSize,
    const void*                                 pData);
  • commandBuffer is the command buffer into which the command will be recorded.

  • dstBuffer is a handle to the buffer to be updated.

  • dstOffset is the byte offset into the buffer to start updating, and must be a multiple of 4.

  • dataSize is the number of bytes to update, and must be a multiple of 4.

  • pData is a pointer to the source data for the buffer update, and must be at least dataSize bytes in size.

dataSize must be less than or equal to 65536 bytes. For larger updates, applications can use buffer to buffer copies.

Note

Buffer updates performed with vkCmdUpdateBuffer first copy the data into command buffer memory when the command is recorded (which requires additional storage and may incur an additional allocation), and then copy the data from the command buffer into dstBuffer when the command is executed on a device.

The additional cost of this functionality compared to buffer to buffer copies means it is only recommended for very small amounts of data, and is why it is limited to only 65536 bytes.

Applications can work around this by issuing multiple vkCmdUpdateBuffer commands to different ranges of the same buffer, but it is strongly recommended that they should not.

The source data is copied from the user pointer to the command buffer when the command is called.

vkCmdUpdateBuffer is only allowed outside of a render pass. This command is treated as “transfer” operation, for the purposes of synchronization barriers. The VK_BUFFER_USAGE_TRANSFER_DST_BIT must be specified in usage of VkBufferCreateInfo in order for the buffer to be compatible with vkCmdUpdateBuffer.

Valid Usage
  • dstOffset must be less than the size of dstBuffer

  • dataSize must be less than or equal to the size of dstBuffer minus dstOffset

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstOffset must be a multiple of 4

  • dataSize must be less than or equal to 65536

  • dataSize must be a multiple of 4

  • If commandBuffer is an unprotected command buffer, then dstBuffer must not be a protected buffer

  • If commandBuffer is a protected command buffer, then dstBuffer must not be an unprotected buffer

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • dstBuffer must be a valid VkBuffer handle

  • pData must be a valid pointer to an array of dataSize bytes

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • dataSize must be greater than 0

  • Both of commandBuffer, and dstBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

Note

The pData parameter was of type uint32_t* instead of void* prior to revision 1.0.19 of the Specification and VK_HEADER_VERSION 19 of the Vulkan Header Files. This was a historical anomaly, as the source data may be of other types.

18. Copy Commands

An application can copy buffer and image data using several methods depending on the type of data transfer. Data can be copied between buffer objects with vkCmdCopyBuffer and a portion of an image can be copied to another image with vkCmdCopyImage. Image data can also be copied to and from buffer memory using vkCmdCopyImageToBuffer and vkCmdCopyBufferToImage. Image data can be blitted (with or without scaling and filtering) with vkCmdBlitImage. Multisampled images can be resolved to a non-multisampled image with vkCmdResolveImage.

18.1. Common Operation

The following valid usage rules apply to all copy commands:

  • Copy commands must be recorded outside of a render pass instance.

  • The set of all bytes bound to all the source regions must not overlap the set of all bytes bound to the destination regions.

  • The set of all bytes bound to each destination region must not overlap the set of all bytes bound to another destination region.

  • Copy regions must be non-empty.

  • Regions must not extend outside the bounds of the buffer or image level, except that regions of compressed images can extend as far as the dimension of the image level rounded up to a complete compressed texel block.

  • Source image subresources must be in either the VK_IMAGE_LAYOUT_GENERAL or VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL layout. Destination image subresources must be in the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_GENERAL or VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL layout. As a consequence, if an image subresource is used as both source and destination of a copy, it must be in the VK_IMAGE_LAYOUT_GENERAL layout.

  • Source images must use a format that supports VK_FORMAT_FEATURE_TRANSFER_SRC_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Destination images must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Source images must have been created with the VK_IMAGE_USAGE_TRANSFER_SRC_BIT usage bit enabled and destination images must have been created with the VK_IMAGE_USAGE_TRANSFER_DST_BIT usage bit enabled.

  • Source buffers must have been created with the VK_BUFFER_USAGE_TRANSFER_SRC_BIT usage bit enabled and destination buffers must have been created with the VK_BUFFER_USAGE_TRANSFER_DST_BIT usage bit enabled.

All copy commands are treated as “transfer” operations for the purposes of synchronization barriers.

18.2. Copying Data Between Buffers

To copy data between buffer objects, call:

void vkCmdCopyBuffer(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    srcBuffer,
    VkBuffer                                    dstBuffer,
    uint32_t                                    regionCount,
    const VkBufferCopy*                         pRegions);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcBuffer is the source buffer.

  • dstBuffer is the destination buffer.

  • regionCount is the number of regions to copy.

  • pRegions is a pointer to an array of VkBufferCopy structures specifying the regions to copy.

Each region in pRegions is copied from the source buffer to the same region of the destination buffer. srcBuffer and dstBuffer can be the same buffer or alias the same memory, but the result is undefined if the copy regions overlap in memory.

Valid Usage
  • The size member of each element of pRegions must be greater than 0

  • The srcOffset member of each element of pRegions must be less than the size of srcBuffer

  • The dstOffset member of each element of pRegions must be less than the size of dstBuffer

  • The size member of each element of pRegions must be less than or equal to the size of srcBuffer minus srcOffset

  • The size member of each element of pRegions must be less than or equal to the size of dstBuffer minus dstOffset

  • The union of the source regions, and the union of the destination regions, specified by the elements of pRegions, must not overlap in memory

  • srcBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_SRC_BIT usage flag

  • If srcBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • If commandBuffer is an unprotected command buffer, then srcBuffer must not be a protected buffer

  • If commandBuffer is an unprotected command buffer, then dstBuffer must not be a protected buffer

  • If commandBuffer is a protected command buffer, then dstBuffer must not be an unprotected buffer

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcBuffer must be a valid VkBuffer handle

  • dstBuffer must be a valid VkBuffer handle

  • pRegions must be a valid pointer to an array of regionCount VkBufferCopy structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstBuffer, and srcBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

The VkBufferCopy structure is defined as:

typedef struct VkBufferCopy {
    VkDeviceSize    srcOffset;
    VkDeviceSize    dstOffset;
    VkDeviceSize    size;
} VkBufferCopy;
  • srcOffset is the starting offset in bytes from the start of srcBuffer.

  • dstOffset is the starting offset in bytes from the start of dstBuffer.

  • size is the number of bytes to copy.

18.3. Copying Data Between Images

vkCmdCopyImage performs image copies in a similar manner to a host memcpy. It does not perform general-purpose conversions such as scaling, resizing, blending, color-space conversion, or format conversions. Rather, it simply copies raw image data. vkCmdCopyImage can copy between images with different formats, provided the formats are compatible as defined below.

To copy data between image objects, call:

void vkCmdCopyImage(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     srcImage,
    VkImageLayout                               srcImageLayout,
    VkImage                                     dstImage,
    VkImageLayout                               dstImageLayout,
    uint32_t                                    regionCount,
    const VkImageCopy*                          pRegions);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcImage is the source image.

  • srcImageLayout is the current layout of the source image subresource.

  • dstImage is the destination image.

  • dstImageLayout is the current layout of the destination image subresource.

  • regionCount is the number of regions to copy.

  • pRegions is a pointer to an array of VkImageCopy structures specifying the regions to copy.

Each region in pRegions is copied from the source image to the same region of the destination image. srcImage and dstImage can be the same image or alias the same memory.

The formats of srcImage and dstImage must be compatible. Formats are considered compatible if their element size is the same between both formats. For example, VK_FORMAT_R8G8B8A8_UNORM is compatible with VK_FORMAT_R32_UINT because both texels are 4 bytes in size. Depth/stencil formats must match exactly.

If the format of srcImage or dstImage is a multi-planar image format, regions of each plane to be copied must be specified separately using the srcSubresource and dstSubresource members of the VkImageCopy structure. In this case, the aspectMask of the srcSubresource or dstSubresource that refers to the multi-planar image must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT. For the purposes of vkCmdCopyImage, each plane of a multi-planar image is treated as having the format listed in Compatible formats of planes of multi-planar formats for the plane identified by the aspectMask of the corresponding subresource. This applies both to VkFormat and to coordinates used in the copy, which correspond to texels in the plane rather than how these texels map to coordinates in the image as a whole.

Note

For example, the VK_IMAGE_ASPECT_PLANE_1_BIT plane of a VK_FORMAT_G8_B8R8_2PLANE_420_UNORM image is compatible with an image of format VK_FORMAT_R8G8_UNORM and (less usefully) with the VK_IMAGE_ASPECT_PLANE_0_BIT plane of an image of format VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16, as each texel is 2 bytes in size.

vkCmdCopyImage allows copying between size-compatible compressed and uncompressed internal formats. Formats are size-compatible if the element size of the uncompressed format is equal to the element size (compressed texel block size) of the compressed format. Such a copy does not perform on-the-fly compression or decompression. When copying from an uncompressed format to a compressed format, each texel of uncompressed data of the source image is copied as a raw value to the corresponding compressed texel block of the destination image. When copying from a compressed format to an uncompressed format, each compressed texel block of the source image is copied as a raw value to the corresponding texel of uncompressed data in the destination image. Thus, for example, it is legal to copy between a 128-bit uncompressed format and a compressed format which has a 128-bit sized compressed texel block representing 4×4 texels (using 8 bits per texel), or between a 64-bit uncompressed format and a compressed format which has a 64-bit sized compressed texel block representing 4×4 texels (using 4 bits per texel).

When copying between compressed and uncompressed formats the extent members represent the texel dimensions of the source image and not the destination. When copying from a compressed image to an uncompressed image the image texel dimensions written to the uncompressed image will be source extent divided by the compressed texel block dimensions. When copying from an uncompressed image to a compressed image the image texel dimensions written to the compressed image will be the source extent multiplied by the compressed texel block dimensions. In both cases the number of bytes read and the number of bytes written will be identical.

Copying to or from block-compressed images is typically done in multiples of the compressed texel block size. For this reason the extent must be a multiple of the compressed texel block dimension. There is one exception to this rule which is required to handle compressed images created with dimensions that are not a multiple of the compressed texel block dimensions: if the srcImage is compressed, then:

  • If extent.width is not a multiple of the compressed texel block width, then (extent.width + srcOffset.x) must equal the image subresource width.

  • If extent.height is not a multiple of the compressed texel block height, then (extent.height + srcOffset.y) must equal the image subresource height.

  • If extent.depth is not a multiple of the compressed texel block depth, then (extent.depth + srcOffset.z) must equal the image subresource depth.

Similarly, if the dstImage is compressed, then:

  • If extent.width is not a multiple of the compressed texel block width, then (extent.width + dstOffset.x) must equal the image subresource width.

  • If extent.height is not a multiple of the compressed texel block height, then (extent.height + dstOffset.y) must equal the image subresource height.

  • If extent.depth is not a multiple of the compressed texel block depth, then (extent.depth + dstOffset.z) must equal the image subresource depth.

This allows the last compressed texel block of the image in each non-multiple dimension to be included as a source or destination of the copy.

_422” image formats that are not multi-planar are treated as having a 2×1 compressed texel block for the purposes of these rules.

vkCmdCopyImage can be used to copy image data between multisample images, but both images must have the same number of samples.

Valid Usage
  • The source region specified by each element of pRegions must be a region that is contained within srcImage if the srcImage’s VkFormat is not a multi-planar format, and must be a region that is contained within the plane being copied if the srcImage’s VkFormat is a multi-planar format

  • The destination region specified by each element of pRegions must be a region that is contained within dstImage if the dstImage’s VkFormat is not a multi-planar format, and must be a region that is contained within the plane being copied to if the dstImage’s VkFormat is a multi-planar format

  • The union of all source regions, and the union of all destination regions, specified by the elements of pRegions, must not overlap in memory

  • srcImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_SRC_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • srcImage must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT usage flag

  • If srcImage is non-sparse then the image or disjoint plane to be copied must be bound completely and contiguously to a single VkDeviceMemory object

  • srcImageLayout must specify the layout of the image subresources of srcImage specified in pRegions at the time this command is executed on a VkDevice

  • srcImageLayout must be VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL, VK_IMAGE_LAYOUT_GENERAL, or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR

  • dstImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • dstImage must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT usage flag

  • If dstImage is non-sparse then the image or disjoint plane that is the destination of the copy must be bound completely and contiguously to a single VkDeviceMemory object

  • dstImageLayout must specify the layout of the image subresources of dstImage specified in pRegions at the time this command is executed on a VkDevice

  • dstImageLayout must be VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_GENERAL, or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR

  • If the VkFormat of each of srcImage and dstImage is not a multi-planar format, the VkFormat of each of srcImage and dstImage must be compatible, as defined below

  • In a copy to or from a plane of a multi-planar image, the VkFormat of the image and plane must be compatible according to the description of compatible planes for the plane being copied

  • When a copy is performed to or from an image with a multi-planar format, the aspectMask of the srcSubresource and/or dstSubresource that refers to the multi-planar image must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT (with VK_IMAGE_ASPECT_PLANE_2_BIT valid only for a VkFormat with three planes)

  • The sample count of srcImage and dstImage must match

  • If commandBuffer is an unprotected command buffer, then srcImage must not be a protected image

  • If commandBuffer is an unprotected command buffer, then dstImage must not be a protected image

  • If commandBuffer is a protected command buffer, then dstImage must not be an unprotected image

  • The srcSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when dstImage was created

  • The srcSubresource.baseArrayLayer + srcSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.baseArrayLayer + dstSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when dstImage was created

  • The srcOffset and extent members of each element of pRegions must respect the image transfer granularity requirements of commandBuffer’s command pool’s queue family, as described in VkQueueFamilyProperties

  • The dstOffset and extent members of each element of pRegions must respect the image transfer granularity requirements of commandBuffer’s command pool’s queue family, as described in VkQueueFamilyProperties

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcImage must be a valid VkImage handle

  • srcImageLayout must be a valid VkImageLayout value

  • dstImage must be a valid VkImage handle

  • dstImageLayout must be a valid VkImageLayout value

  • pRegions must be a valid pointer to an array of regionCount valid VkImageCopy structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstImage, and srcImage must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

The VkImageCopy structure is defined as:

typedef struct VkImageCopy {
    VkImageSubresourceLayers    srcSubresource;
    VkOffset3D                  srcOffset;
    VkImageSubresourceLayers    dstSubresource;
    VkOffset3D                  dstOffset;
    VkExtent3D                  extent;
} VkImageCopy;
  • srcSubresource and dstSubresource are VkImageSubresourceLayers structures specifying the image subresources of the images used for the source and destination image data, respectively.

  • srcOffset and dstOffset select the initial x, y, and z offsets in texels of the sub-regions of the source and destination image data.

  • extent is the size in texels of the image to copy in width, height and depth.

For VK_IMAGE_TYPE_3D images, copies are performed slice by slice starting with the z member of the srcOffset or dstOffset, and copying depth slices. For images with multiple layers, copies are performed layer by layer starting with the baseArrayLayer member of the srcSubresource or dstSubresource and copying layerCount layers. Image data can be copied between images with different image types. If one image is VK_IMAGE_TYPE_3D and the other image is VK_IMAGE_TYPE_2D with multiple layers, then each slice is copied to or from a different layer.

Copies involving a multi-planar image format specify the region to be copied in terms of the plane to be copied, not the coordinates of the multi-planar image. This means that copies accessing the R/B planes of “_422” format images must fit the copied region within half the width of the parent image, and that copies accessing the R/B planes of “_420” format images must fit the copied region within half the width and height of the parent image.

Valid Usage
  • If neither the calling command’s srcImage nor the calling command’s dstImage has a multi-planar image format then the aspectMask member of srcSubresource and dstSubresource must match

  • If the calling command’s srcImage has a VkFormat with two planes then the srcSubresource aspectMask must be VK_IMAGE_ASPECT_PLANE_0_BIT or VK_IMAGE_ASPECT_PLANE_1_BIT

  • If the calling command’s srcImage has a VkFormat with three planes then the srcSubresource aspectMask must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT

  • If the calling command’s dstImage has a VkFormat with two planes then the dstSubresource aspectMask must be VK_IMAGE_ASPECT_PLANE_0_BIT or VK_IMAGE_ASPECT_PLANE_1_BIT

  • If the calling command’s dstImage has a VkFormat with three planes then the dstSubresource aspectMask must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT

  • If the calling command’s srcImage has a multi-planar image format and the dstImage does not have a multi-planar image format, the dstSubresource aspectMask must be VK_IMAGE_ASPECT_COLOR_BIT

  • If the calling command’s dstImage has a multi-planar image format and the srcImage does not have a multi-planar image format, the srcSubresource aspectMask must be VK_IMAGE_ASPECT_COLOR_BIT

  • The number of slices of the extent (for 3D) or layers of the srcSubresource (for non-3D) must match the number of slices of the extent (for 3D) or layers of the dstSubresource (for non-3D)

  • If either of the calling command’s srcImage or dstImage parameters are of VkImageType VK_IMAGE_TYPE_3D, the baseArrayLayer and layerCount members of the corresponding subresource must be 0 and 1, respectively

  • The aspectMask member of srcSubresource must specify aspects present in the calling command’s srcImage

  • The aspectMask member of dstSubresource must specify aspects present in the calling command’s dstImage

  • srcOffset.x and (extent.width + srcOffset.x) must both be greater than or equal to 0 and less than or equal to the source image subresource width

  • srcOffset.y and (extent.height + srcOffset.y) must both be greater than or equal to 0 and less than or equal to the source image subresource height

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D, then srcOffset.y must be 0 and extent.height must be 1.

  • srcOffset.z and (extent.depth + srcOffset.z) must both be greater than or equal to 0 and less than or equal to the source image subresource depth

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D, then srcOffset.z must be 0 and extent.depth must be 1.

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D, then dstOffset.z must be 0 and extent.depth must be 1.

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_2D, then srcOffset.z must be 0.

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_2D, then dstOffset.z must be 0.

  • If both srcImage and dstImage are of type VK_IMAGE_TYPE_2D then then extent.depth must be 1.

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_2D, and the dstImage is of type VK_IMAGE_TYPE_3D, then extent.depth must equal to the layerCount member of srcSubresource.

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_2D, and the srcImage is of type VK_IMAGE_TYPE_3D, then extent.depth must equal to the layerCount member of dstSubresource.

  • dstOffset.x and (extent.width + dstOffset.x) must both be greater than or equal to 0 and less than or equal to the destination image subresource width

  • dstOffset.y and (extent.height + dstOffset.y) must both be greater than or equal to 0 and less than or equal to the destination image subresource height

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D, then dstOffset.y must be 0 and extent.height must be 1.

  • dstOffset.z and (extent.depth + dstOffset.z) must both be greater than or equal to 0 and less than or equal to the destination image subresource depth

  • If the calling command’s srcImage is a compressed image, or a single-plane, “_422” image format, all members of srcOffset must be a multiple of the corresponding dimensions of the compressed texel block

  • If the calling command’s srcImage is a compressed image, or a single-plane, “_422” image format, extent.width must be a multiple of the compressed texel block width or (extent.width + srcOffset.x) must equal the source image subresource width

  • If the calling command’s srcImage is a compressed image, or a single-plane, “_422” image format, extent.height must be a multiple of the compressed texel block height or (extent.height + srcOffset.y) must equal the source image subresource height

  • If the calling command’s srcImage is a compressed image, or a single-plane, “_422” image format, extent.depth must be a multiple of the compressed texel block depth or (extent.depth + srcOffset.z) must equal the source image subresource depth

  • If the calling command’s dstImage is a compressed format image, or a single-plane, “_422” image format, all members of dstOffset must be a multiple of the corresponding dimensions of the compressed texel block

  • If the calling command’s dstImage is a compressed format image, or a single-plane, “_422” image format, extent.width must be a multiple of the compressed texel block width or (extent.width + dstOffset.x) must equal the destination image subresource width

  • If the calling command’s dstImage is a compressed format image, or a single-plane, “_422” image format, extent.height must be a multiple of the compressed texel block height or (extent.height + dstOffset.y) must equal the destination image subresource height

  • If the calling command’s dstImage is a compressed format image, or a single-plane, “_422” image format, extent.depth must be a multiple of the compressed texel block depth or (extent.depth + dstOffset.z) must equal the destination image subresource depth

Valid Usage (Implicit)
  • srcSubresource must be a valid VkImageSubresourceLayers structure

  • dstSubresource must be a valid VkImageSubresourceLayers structure

The VkImageSubresourceLayers structure is defined as:

typedef struct VkImageSubresourceLayers {
    VkImageAspectFlags    aspectMask;
    uint32_t              mipLevel;
    uint32_t              baseArrayLayer;
    uint32_t              layerCount;
} VkImageSubresourceLayers;
  • aspectMask is a combination of VkImageAspectFlagBits, selecting the color, depth and/or stencil aspects to be copied.

  • mipLevel is the mipmap level to copy from.

  • baseArrayLayer and layerCount are the starting layer and number of layers to copy.

Valid Usage
  • If aspectMask contains VK_IMAGE_ASPECT_COLOR_BIT, it must not contain either of VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT

  • aspectMask must not contain VK_IMAGE_ASPECT_METADATA_BIT

  • layerCount must be greater than 0

Valid Usage (Implicit)

18.4. Copying Data Between Buffers and Images

To copy data from a buffer object to an image object, call:

void vkCmdCopyBufferToImage(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    srcBuffer,
    VkImage                                     dstImage,
    VkImageLayout                               dstImageLayout,
    uint32_t                                    regionCount,
    const VkBufferImageCopy*                    pRegions);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcBuffer is the source buffer.

  • dstImage is the destination image.

  • dstImageLayout is the layout of the destination image subresources for the copy.

  • regionCount is the number of regions to copy.

  • pRegions is a pointer to an array of VkBufferImageCopy structures specifying the regions to copy.

Each region in pRegions is copied from the specified region of the source buffer to the specified region of the destination image.

If the format of dstImage is a multi-planar image format), regions of each plane to be a target of a copy must be specified separately using the pRegions member of the VkBufferImageCopy structure. In this case, the aspectMask of imageSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT. For the purposes of vkCmdCopyBufferToImage, each plane of a multi-planar image is treated as having the format listed in Compatible formats of planes of multi-planar formats for the plane identified by the aspectMask of the corresponding subresource. This applies both to VkFormat and to coordinates used in the copy, which correspond to texels in the plane rather than how these texels map to coordinates in the image as a whole.

Valid Usage
  • srcBuffer must be large enough to contain all buffer locations that are accessed according to Buffer and Image Addressing, for each element of pRegions

  • The image region specified by each element of pRegions must be a region that is contained within dstImage if the dstImage’s VkFormat is not a multi-planar format, and must be a region that is contained within the plane being copied to if the dstImage’s VkFormat is a multi-planar format

  • The union of all source regions, and the union of all destination regions, specified by the elements of pRegions, must not overlap in memory

  • srcBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_SRC_BIT usage flag

  • dstImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • If srcBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstImage must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT usage flag

  • If dstImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstImage must have a sample count equal to VK_SAMPLE_COUNT_1_BIT

  • dstImageLayout must specify the layout of the image subresources of dstImage specified in pRegions at the time this command is executed on a VkDevice

  • dstImageLayout must be VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_GENERAL, or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR

  • If commandBuffer is an unprotected command buffer, then srcBuffer must not be a protected buffer

  • If commandBuffer is an unprotected command buffer, then dstImage must not be a protected image

  • If commandBuffer is a protected command buffer, then dstImage must not be an unprotected image

  • The imageSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when dstImage was created

  • The imageSubresource.baseArrayLayer + imageSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when dstImage was created

  • The imageOffset and imageExtent members of each element of pRegions must respect the image transfer granularity requirements of commandBuffer’s command pool’s queue family, as described in VkQueueFamilyProperties

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcBuffer must be a valid VkBuffer handle

  • dstImage must be a valid VkImage handle

  • dstImageLayout must be a valid VkImageLayout value

  • pRegions must be a valid pointer to an array of regionCount valid VkBufferImageCopy structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstImage, and srcBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

To copy data from an image object to a buffer object, call:

void vkCmdCopyImageToBuffer(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     srcImage,
    VkImageLayout                               srcImageLayout,
    VkBuffer                                    dstBuffer,
    uint32_t                                    regionCount,
    const VkBufferImageCopy*                    pRegions);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcImage is the source image.

  • srcImageLayout is the layout of the source image subresources for the copy.

  • dstBuffer is the destination buffer.

  • regionCount is the number of regions to copy.

  • pRegions is a pointer to an array of VkBufferImageCopy structures specifying the regions to copy.

Each region in pRegions is copied from the specified region of the source image to the specified region of the destination buffer.

If the VkFormat of srcImage is a multi-planar image format, regions of each plane to be a source of a copy must be specified separately using the pRegions member of the VkBufferImageCopy structure. In this case, the aspectMask of imageSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT. For the purposes of vkCmdCopyBufferToImage, each plane of a multi-planar image is treated as having the format listed in Compatible formats of planes of multi-planar formats for the plane identified by the aspectMask of the corresponding subresource. This applies both to VkFormat and to coordinates used in the copy, which correspond to texels in the plane rather than how these texels map to coordinates in the image as a whole.

Valid Usage
  • The image region specified by each element of pRegions must be a region that is contained within srcImage if the srcImage’s VkFormat is not a multi-planar format, and must be a region that is contained within the plane being copied if the srcImage’s VkFormat is a multi-planar format

  • dstBuffer must be large enough to contain all buffer locations that are accessed according to Buffer and Image Addressing, for each element of pRegions

  • The union of all source regions, and the union of all destination regions, specified by the elements of pRegions, must not overlap in memory

  • srcImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_SRC_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • srcImage must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT usage flag

  • If srcImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • srcImage must have a sample count equal to VK_SAMPLE_COUNT_1_BIT

  • srcImageLayout must specify the layout of the image subresources of srcImage specified in pRegions at the time this command is executed on a VkDevice

  • srcImageLayout must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • If commandBuffer is an unprotected command buffer, then srcImage must not be a protected image

  • If commandBuffer is an unprotected command buffer, then dstBuffer must not be a protected buffer

  • If commandBuffer is a protected command buffer, then dstBuffer must not be an unprotected buffer

  • The imageSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when srcImage was created

  • The imageSubresource.baseArrayLayer + imageSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when srcImage was created

  • The imageOffset and imageExtent members of each element of pRegions must respect the image transfer granularity requirements of commandBuffer’s command pool’s queue family, as described in VkQueueFamilyProperties

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcImage must be a valid VkImage handle

  • srcImageLayout must be a valid VkImageLayout value

  • dstBuffer must be a valid VkBuffer handle

  • pRegions must be a valid pointer to an array of regionCount valid VkBufferImageCopy structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstBuffer, and srcImage must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

For both vkCmdCopyBufferToImage and vkCmdCopyImageToBuffer, each element of pRegions is a structure defined as:

typedef struct VkBufferImageCopy {
    VkDeviceSize                bufferOffset;
    uint32_t                    bufferRowLength;
    uint32_t                    bufferImageHeight;
    VkImageSubresourceLayers    imageSubresource;
    VkOffset3D                  imageOffset;
    VkExtent3D                  imageExtent;
} VkBufferImageCopy;
  • bufferOffset is the offset in bytes from the start of the buffer object where the image data is copied from or to.

  • bufferRowLength and bufferImageHeight specify the data in buffer memory as a subregion of a larger two- or three-dimensional image, and control the addressing calculations of data in buffer memory. If either of these values is zero, that aspect of the buffer memory is considered to be tightly packed according to the imageExtent.

  • imageSubresource is a VkImageSubresourceLayers used to specify the specific image subresources of the image used for the source or destination image data.

  • imageOffset selects the initial x, y, z offsets in texels of the sub-region of the source or destination image data.

  • imageExtent is the size in texels of the image to copy in width, height and depth.

When copying to or from a depth or stencil aspect, the data in buffer memory uses a layout that is a (mostly) tightly packed representation of the depth or stencil data. Specifically:

  • data copied to or from the stencil aspect of any depth/stencil format is tightly packed with one VK_FORMAT_S8_UINT value per texel.

  • data copied to or from the depth aspect of a VK_FORMAT_D16_UNORM or VK_FORMAT_D16_UNORM_S8_UINT format is tightly packed with one VK_FORMAT_D16_UNORM value per texel.

  • data copied to or from the depth aspect of a VK_FORMAT_D32_SFLOAT or VK_FORMAT_D32_SFLOAT_S8_UINT format is tightly packed with one VK_FORMAT_D32_SFLOAT value per texel.

  • data copied to or from the depth aspect of a VK_FORMAT_X8_D24_UNORM_PACK32 or VK_FORMAT_D24_UNORM_S8_UINT format is packed with one 32-bit word per texel with the D24 value in the LSBs of the word, and undefined values in the eight MSBs.

Note

To copy both the depth and stencil aspects of a depth/stencil format, two entries in pRegions can be used, where one specifies the depth aspect in imageSubresource, and the other specifies the stencil aspect.

Because depth or stencil aspect buffer to image copies may require format conversions on some implementations, they are not supported on queues that do not support graphics. When copying to a depth aspect, the data in buffer memory must be in the the range [0,1] or undefined results occur.

Copies are done layer by layer starting with image layer baseArrayLayer member of imageSubresource. layerCount layers are copied from the source image or to the destination image.

Valid Usage
  • If the calling command’s VkImage parameter’s format is not a depth/stencil format or a multi-planar format, then bufferOffset must be a multiple of the format’s element size

  • If the calling command’s VkImage parameter’s format is a multi-planar format, then bufferOffset must be a multiple of the element size of the compatible format for the format and the aspectMask of the imageSubresource as defined in Compatible formats of planes of multi-planar formats

  • bufferOffset must be a multiple of 4

  • bufferRowLength must be 0, or greater than or equal to the width member of imageExtent

  • bufferImageHeight must be 0, or greater than or equal to the height member of imageExtent

  • imageOffset.x and (imageExtent.width + imageOffset.x) must both be greater than or equal to 0 and less than or equal to the image subresource width where this refers to the width of the plane of the image involved in the copy in the case of a multi-planar format

  • imageOffset.y and (imageExtent.height + imageOffset.y) must both be greater than or equal to 0 and less than or equal to the image subresource height where this refers to the height of the plane of the image involved in the copy in the case of a multi-planar format

  • If the calling command’s srcImage (vkCmdCopyImageToBuffer) or dstImage (vkCmdCopyBufferToImage) is of type VK_IMAGE_TYPE_1D, then imageOffset.y must be 0 and imageExtent.height must be 1.

  • imageOffset.z and (imageExtent.depth + imageOffset.z) must both be greater than or equal to 0 and less than or equal to the image subresource depth

  • If the calling command’s srcImage (vkCmdCopyImageToBuffer) or dstImage (vkCmdCopyBufferToImage) is of type VK_IMAGE_TYPE_1D or VK_IMAGE_TYPE_2D, then imageOffset.z must be 0 and imageExtent.depth must be 1

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, bufferRowLength must be a multiple of the compressed texel block width

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, bufferImageHeight must be a multiple of the compressed texel block height

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, all members of imageOffset must be a multiple of the corresponding dimensions of the compressed texel block

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, bufferOffset must be a multiple of the compressed texel block size in bytes

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, imageExtent.width must be a multiple of the compressed texel block width or (imageExtent.width + imageOffset.x) must equal the image subresource width

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, imageExtent.height must be a multiple of the compressed texel block height or (imageExtent.height + imageOffset.y) must equal the image subresource height

  • If the calling command’s VkImage parameter is a compressed image, or a single-plane, “_422” image format, imageExtent.depth must be a multiple of the compressed texel block depth or (imageExtent.depth + imageOffset.z) must equal the image subresource depth

  • The aspectMask member of imageSubresource must specify aspects present in the calling command’s VkImage parameter

  • If the calling command’s VkImage parameter’s format is a multi-planar format, then the aspectMask member of imageSubresource must be VK_IMAGE_ASPECT_PLANE_0_BIT, VK_IMAGE_ASPECT_PLANE_1_BIT, or VK_IMAGE_ASPECT_PLANE_2_BIT (with VK_IMAGE_ASPECT_PLANE_2_BIT valid only for image formats with three planes)

  • The aspectMask member of imageSubresource must only have a single bit set

  • If the calling command’s VkImage parameter is of VkImageType VK_IMAGE_TYPE_3D, the baseArrayLayer and layerCount members of imageSubresource must be 0 and 1, respectively

  • When copying to the depth aspect of an image subresource, the data in the source buffer must be in the range [0,1]

Valid Usage (Implicit)
  • imageSubresource must be a valid VkImageSubresourceLayers structure

18.4.1. Buffer and Image Addressing

Pseudocode for image/buffer addressing is:

rowLength = region->bufferRowLength;
if (rowLength == 0)
    rowLength = region->imageExtent.width;

imageHeight = region->bufferImageHeight;
if (imageHeight == 0)
    imageHeight = region->imageExtent.height;

elementSize = <element size of the format of the src/dstImage>;

address of (x,y,z) = region->bufferOffset + (((z * imageHeight) + y) * rowLength + x) * elementSize;

where x,y,z range from (0,0,0) to region->imageExtent.{width,height,depth}.

Note that imageOffset does not affect addressing calculations for buffer memory. Instead, bufferOffset can be used to select the starting address in buffer memory.

For block-compression formats, all parameters are still specified in texels rather than compressed texel blocks, but the addressing math operates on whole compressed texel blocks. Pseudocode for compressed copy addressing is:

rowLength = region->bufferRowLength;
if (rowLength == 0)
    rowLength = region->imageExtent.width;

imageHeight = region->bufferImageHeight;
if (imageHeight == 0)
    imageHeight = region->imageExtent.height;

compressedTexelBlockSizeInBytes = <compressed texel block size taken from the src/dstImage>;
rowLength /= compressedTexelBlockWidth;
imageHeight /= compressedTexelBlockHeight;

address of (x,y,z) = region->bufferOffset + (((z * imageHeight) + y) * rowLength + x) * compressedTexelBlockSizeInBytes;

where x,y,z range from (0,0,0) to region->imageExtent.{width/compressedTexelBlockWidth,height/compressedTexelBlockHeight,depth/compressedTexelBlockDepth}.

Copying to or from block-compressed images is typically done in multiples of the compressed texel block size. For this reason the imageExtent must be a multiple of the compressed texel block dimension. There is one exception to this rule which is required to handle compressed images created with dimensions that are not a multiple of the compressed texel block dimensions:

  • If imageExtent.width is not a multiple of the compressed texel block width, then (imageExtent.width + imageOffset.x) must equal the image subresource width.

  • If imageExtent.height is not a multiple of the compressed texel block height, then (imageExtent.height + imageOffset.y) must equal the image subresource height.

  • If imageExtent.depth is not a multiple of the compressed texel block depth, then (imageExtent.depth + imageOffset.z) must equal the image subresource depth.

This allows the last compressed texel block of the image in each non-multiple dimension to be included as a source or destination of the copy.

18.5. Image Copies with Scaling

To copy regions of a source image into a destination image, potentially performing format conversion, arbitrary scaling, and filtering, call:

void vkCmdBlitImage(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     srcImage,
    VkImageLayout                               srcImageLayout,
    VkImage                                     dstImage,
    VkImageLayout                               dstImageLayout,
    uint32_t                                    regionCount,
    const VkImageBlit*                          pRegions,
    VkFilter                                    filter);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcImage is the source image.

  • srcImageLayout is the layout of the source image subresources for the blit.

  • dstImage is the destination image.

  • dstImageLayout is the layout of the destination image subresources for the blit.

  • regionCount is the number of regions to blit.

  • pRegions is a pointer to an array of VkImageBlit structures specifying the regions to blit.

  • filter is a VkFilter specifying the filter to apply if the blits require scaling.

vkCmdBlitImage must not be used for multisampled source or destination images. Use vkCmdResolveImage for this purpose.

As the sizes of the source and destination extents can differ in any dimension, texels in the source extent are scaled and filtered to the destination extent. Scaling occurs via the following operations:

  • For each destination texel, the integer coordinate of that texel is converted to an unnormalized texture coordinate, using the effective inverse of the equations described in unnormalized to integer conversion:

    ubase = i + ½

    vbase = j + ½

    wbase = k + ½

  • These base coordinates are then offset by the first destination offset:

    uoffset = ubase - xdst0

    voffset = vbase - ydst0

    woffset = wbase - zdst0

    aoffset = a - baseArrayCountdst

  • The scale is determined from the source and destination regions, and applied to the offset coordinates:

    scale_u = (xsrc1 - xsrc0) / (xdst1 - xdst0)

    scale_v = (ysrc1 - ysrc0) / (ydst1 - ydst0)

    scale_w = (zsrc1 - zsrc0) / (zdst1 - zdst0)

    uscaled = uoffset * scaleu

    vscaled = voffset * scalev

    wscaled = woffset * scalew

  • Finally the source offset is added to the scaled coordinates, to determine the final unnormalized coordinates used to sample from srcImage:

    u = uscaled + xsrc0

    v = vscaled + ysrc0

    w = wscaled + zsrc0

    q = mipLevel

    a = aoffset + baseArrayCountsrc

These coordinates are used to sample from the source image, as described in Image Operations chapter, with the filter mode equal to that of filter, a mipmap mode of VK_SAMPLER_MIPMAP_MODE_NEAREST and an address mode of VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE. Implementations must clamp at the edge of the source image, and may additionally clamp to the edge of the source region.

Note

Due to allowable rounding errors in the generation of the source texture coordinates, it is not always possible to guarantee exactly which source texels will be sampled for a given blit. As rounding errors are implementation dependent, the exact results of a blitting operation are also implementation dependent.

Blits are done layer by layer starting with the baseArrayLayer member of srcSubresource for the source and dstSubresource for the destination. layerCount layers are blitted to the destination image.

3D textures are blitted slice by slice. Slices in the source region bounded by srcOffsets[0].z and srcOffsets[1].z are copied to slices in the destination region bounded by dstOffsets[0].z and dstOffsets[1].z. For each destination slice, a source z coordinate is linearly interpolated between srcOffsets[0].z and srcOffsets[1].z. If the filter parameter is VK_FILTER_LINEAR then the value sampled from the source image is taken by doing linear filtering using the interpolated z coordinate. If filter parameter is VK_FILTER_NEAREST then value sampled from the source image is taken from the single nearest slice (with undefined rounding mode).

The following filtering and conversion rules apply:

  • Integer formats can only be converted to other integer formats with the same signedness.

  • No format conversion is supported between depth/stencil images. The formats must match.

  • Format conversions on unorm, snorm, unscaled and packed float formats of the copied aspect of the image are performed by first converting the pixels to float values.

  • For sRGB source formats, nonlinear RGB values are converted to linear representation prior to filtering.

  • After filtering, the float values are first clamped and then cast to the destination image format. In case of sRGB destination format, linear RGB values are converted to nonlinear representation before writing the pixel to the image.

Signed and unsigned integers are converted by first clamping to the representable range of the destination format, then casting the value.

Valid Usage
  • The source region specified by each element of pRegions must be a region that is contained within srcImage

  • The destination region specified by each element of pRegions must be a region that is contained within dstImage

  • The union of all destination regions, specified by the elements of pRegions, must not overlap in memory with any texel that may be sampled during the blit operation

  • srcImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_SRC_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • srcImage must not use a format listed in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views

  • srcImage must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT usage flag

  • If srcImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • srcImageLayout must specify the layout of the image subresources of srcImage specified in pRegions at the time this command is executed on a VkDevice

  • srcImageLayout must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • dstImage must use a format that supports VK_FORMAT_FEATURE_TRANSFER_DST_BIT, which is indicated by VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • dstImage must not use a format listed in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views

  • dstImage must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT usage flag

  • If dstImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstImageLayout must specify the layout of the image subresources of dstImage specified in pRegions at the time this command is executed on a VkDevice

  • dstImageLayout must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • The sample count of srcImage and dstImage must both be equal to VK_SAMPLE_COUNT_1_BIT

  • If either of srcImage or dstImage was created with a signed integer VkFormat, the other must also have been created with a signed integer VkFormat

  • If either of srcImage or dstImage was created with an unsigned integer VkFormat, the other must also have been created with an unsigned integer VkFormat

  • If either of srcImage or dstImage was created with a depth/stencil format, the other must have exactly the same format

  • If srcImage was created with a depth/stencil format, filter must be VK_FILTER_NEAREST

  • srcImage must have been created with a samples value of VK_SAMPLE_COUNT_1_BIT

  • dstImage must have been created with a samples value of VK_SAMPLE_COUNT_1_BIT

  • If filter is VK_FILTER_LINEAR, srcImage must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • If filter is VK_FILTER_CUBIC_IMG, srcImage must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • If filter is VK_FILTER_CUBIC_IMG, srcImage must have a VkImageType of VK_IMAGE_TYPE_2D

  • If commandBuffer is an unprotected command buffer, then srcImage must not be a protected image

  • If commandBuffer is an unprotected command buffer, then dstImage must not be a protected image

  • If commandBuffer is a protected command buffer, then dstImage must not be an unprotected image

  • The srcSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when dstImage was created

  • The srcSubresource.baseArrayLayer + srcSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.baseArrayLayer + dstSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when dstImage was created

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcImage must be a valid VkImage handle

  • srcImageLayout must be a valid VkImageLayout value

  • dstImage must be a valid VkImage handle

  • dstImageLayout must be a valid VkImageLayout value

  • pRegions must be a valid pointer to an array of regionCount valid VkImageBlit structures

  • filter must be a valid VkFilter value

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstImage, and srcImage must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics

Transfer

The VkImageBlit structure is defined as:

typedef struct VkImageBlit {
    VkImageSubresourceLayers    srcSubresource;
    VkOffset3D                  srcOffsets[2];
    VkImageSubresourceLayers    dstSubresource;
    VkOffset3D                  dstOffsets[2];
} VkImageBlit;
  • srcSubresource is the subresource to blit from.

  • srcOffsets is an array of two VkOffset3D structures specifying the bounds of the source region within srcSubresource.

  • dstSubresource is the subresource to blit into.

  • dstOffsets is an array of two VkOffset3D structures specifying the bounds of the destination region within dstSubresource.

For each element of the pRegions array, a blit operation is performed the specified source and destination regions.

Valid Usage
  • The aspectMask member of srcSubresource and dstSubresource must match

  • The layerCount member of srcSubresource and dstSubresource must match

  • If either of the calling command’s srcImage or dstImage parameters are of VkImageType VK_IMAGE_TYPE_3D, the baseArrayLayer and layerCount members of both srcSubresource and dstSubresource must be 0 and 1, respectively

  • The aspectMask member of srcSubresource must specify aspects present in the calling command’s srcImage

  • The aspectMask member of dstSubresource must specify aspects present in the calling command’s dstImage

  • srcOffset[0].x and srcOffset[1].x must both be greater than or equal to 0 and less than or equal to the source image subresource width

  • srcOffset[0].y and srcOffset[1].y must both be greater than or equal to 0 and less than or equal to the source image subresource height

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D, then srcOffset[0].y must be 0 and srcOffset[1].y must be 1.

  • srcOffset[0].z and srcOffset[1].z must both be greater than or equal to 0 and less than or equal to the source image subresource depth

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D or VK_IMAGE_TYPE_2D, then srcOffset[0].z must be 0 and srcOffset[1].z must be 1.

  • dstOffset[0].x and dstOffset[1].x must both be greater than or equal to 0 and less than or equal to the destination image subresource width

  • dstOffset[0].y and dstOffset[1].y must both be greater than or equal to 0 and less than or equal to the destination image subresource height

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D, then dstOffset[0].y must be 0 and dstOffset[1].y must be 1.

  • dstOffset[0].z and dstOffset[1].z must both be greater than or equal to 0 and less than or equal to the destination image subresource depth

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D or VK_IMAGE_TYPE_2D, then dstOffset[0].z must be 0 and dstOffset[1].z must be 1.

Valid Usage (Implicit)
  • srcSubresource must be a valid VkImageSubresourceLayers structure

  • dstSubresource must be a valid VkImageSubresourceLayers structure

18.6. Resolving Multisample Images

To resolve a multisample image to a non-multisample image, call:

void vkCmdResolveImage(
    VkCommandBuffer                             commandBuffer,
    VkImage                                     srcImage,
    VkImageLayout                               srcImageLayout,
    VkImage                                     dstImage,
    VkImageLayout                               dstImageLayout,
    uint32_t                                    regionCount,
    const VkImageResolve*                       pRegions);
  • commandBuffer is the command buffer into which the command will be recorded.

  • srcImage is the source image.

  • srcImageLayout is the layout of the source image subresources for the resolve.

  • dstImage is the destination image.

  • dstImageLayout is the layout of the destination image subresources for the resolve.

  • regionCount is the number of regions to resolve.

  • pRegions is a pointer to an array of VkImageResolve structures specifying the regions to resolve.

During the resolve the samples corresponding to each pixel location in the source are converted to a single sample before being written to the destination. If the source formats are floating-point or normalized types, the sample values for each pixel are resolved in an implementation-dependent manner. If the source formats are integer types, a single sample’s value is selected for each pixel.

srcOffset and dstOffset select the initial x, y, and z offsets in texels of the sub-regions of the source and destination image data. extent is the size in texels of the source image to resolve in width, height and depth.

Resolves are done layer by layer starting with baseArrayLayer member of srcSubresource for the source and dstSubresource for the destination. layerCount layers are resolved to the destination image.

Valid Usage
  • The source region specified by each element of pRegions must be a region that is contained within srcImage

  • The destination region specified by each element of pRegions must be a region that is contained within dstImage

  • The union of all source regions, and the union of all destination regions, specified by the elements of pRegions, must not overlap in memory

  • If srcImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • srcImage must have a sample count equal to any valid sample count value other than VK_SAMPLE_COUNT_1_BIT

  • If dstImage is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstImage must have a sample count equal to VK_SAMPLE_COUNT_1_BIT

  • srcImageLayout must specify the layout of the image subresources of srcImage specified in pRegions at the time this command is executed on a VkDevice

  • srcImageLayout must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • dstImageLayout must specify the layout of the image subresources of dstImage specified in pRegions at the time this command is executed on a VkDevice

  • dstImageLayout must be VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL or VK_IMAGE_LAYOUT_GENERAL

  • If dstImage was created with tiling equal to VK_IMAGE_TILING_LINEAR, dstImage must have been created with a format that supports being a color attachment, as specified by the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT flag in VkFormatProperties::linearTilingFeatures returned by vkGetPhysicalDeviceFormatProperties

  • If dstImage was created with tiling equal to VK_IMAGE_TILING_OPTIMAL, dstImage must have been created with a format that supports being a color attachment, as specified by the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties

  • srcImage and dstImage must have been created with the same image format

  • If commandBuffer is an unprotected command buffer, then srcImage must not be a protected image

  • If commandBuffer is an unprotected command buffer, then dstImage must not be a protected image

  • If commandBuffer is a protected command buffer, then dstImage must not be an unprotected image

  • The srcSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.mipLevel member of each element of pRegions must be less than the mipLevels specified in VkImageCreateInfo when dstImage was created

  • The srcSubresource.baseArrayLayer + srcSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when srcImage was created

  • The dstSubresource.baseArrayLayer + dstSubresource.layerCount of each element of pRegions must be less than or equal to the arrayLayers specified in VkImageCreateInfo when dstImage was created

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • srcImage must be a valid VkImage handle

  • srcImageLayout must be a valid VkImageLayout value

  • dstImage must be a valid VkImage handle

  • dstImageLayout must be a valid VkImageLayout value

  • pRegions must be a valid pointer to an array of regionCount valid VkImageResolve structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called outside of a render pass instance

  • regionCount must be greater than 0

  • Each of commandBuffer, dstImage, and srcImage must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Graphics

Transfer

The VkImageResolve structure is defined as:

typedef struct VkImageResolve {
    VkImageSubresourceLayers    srcSubresource;
    VkOffset3D                  srcOffset;
    VkImageSubresourceLayers    dstSubresource;
    VkOffset3D                  dstOffset;
    VkExtent3D                  extent;
} VkImageResolve;
  • srcSubresource and dstSubresource are VkImageSubresourceLayers structures specifying the image subresources of the images used for the source and destination image data, respectively. Resolve of depth/stencil images is not supported.

  • srcOffset and dstOffset select the initial x, y, and z offsets in texels of the sub-regions of the source and destination image data.

  • extent is the size in texels of the source image to resolve in width, height and depth.

Valid Usage
  • The aspectMask member of srcSubresource and dstSubresource must only contain VK_IMAGE_ASPECT_COLOR_BIT

  • The layerCount member of srcSubresource and dstSubresource must match

  • If either of the calling command’s srcImage or dstImage parameters are of VkImageType VK_IMAGE_TYPE_3D, the baseArrayLayer and layerCount members of both srcSubresource and dstSubresource must be 0 and 1, respectively

  • srcOffset.x and (extent.width + srcOffset.x) must both be greater than or equal to 0 and less than or equal to the source image subresource width

  • srcOffset.y and (extent.height + srcOffset.y) must both be greater than or equal to 0 and less than or equal to the source image subresource height

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D, then srcOffset.y must be 0 and extent.height must be 1.

  • srcOffset.z and (extent.depth + srcOffset.z) must both be greater than or equal to 0 and less than or equal to the source image subresource depth

  • If the calling command’s srcImage is of type VK_IMAGE_TYPE_1D or VK_IMAGE_TYPE_2D, then srcOffset.z must be 0 and extent.depth must be 1.

  • dstOffset.x and (extent.width + dstOffset.x) must both be greater than or equal to 0 and less than or equal to the destination image subresource width

  • dstOffset.y and (extent.height + dstOffset.y) must both be greater than or equal to 0 and less than or equal to the destination image subresource height

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D, then dstOffset.y must be 0 and extent.height must be 1.

  • dstOffset.z and (extent.depth + dstOffset.z) must both be greater than or equal to 0 and less than or equal to the destination image subresource depth

  • If the calling command’s dstImage is of type VK_IMAGE_TYPE_1D or VK_IMAGE_TYPE_2D, then dstOffset.z must be 0 and extent.depth must be 1.

Valid Usage (Implicit)
  • srcSubresource must be a valid VkImageSubresourceLayers structure

  • dstSubresource must be a valid VkImageSubresourceLayers structure

18.7. Buffer Markers

To write a 32-bit marker value into a buffer as a pipelined operation, call:

void vkCmdWriteBufferMarkerAMD(
    VkCommandBuffer                             commandBuffer,
    VkPipelineStageFlagBits                     pipelineStage,
    VkBuffer                                    dstBuffer,
    VkDeviceSize                                dstOffset,
    uint32_t                                    marker);
  • commandBuffer is the command buffer into which the command will be recorded.

  • pipelineStage is one of the VkPipelineStageFlagBits values, specifying the pipeline stage whose completion triggers the marker write.

  • dstBuffer is the buffer where the marker will be written to.

  • dstOffset is the byte offset into the buffer where the marker will be written to.

  • marker is the 32-bit value of the marker.

The command will write the 32-bit marker value into the buffer only after all preceding commands have finished executing up to at least the specified pipeline stage. This includes the completion of other preceding vkCmdWriteBufferMarkerAMD commands so long as their specified pipeline stages occur either at the same time or earlier than this command’s specified pipelineStage.

While consecutive buffer marker writes with the same pipelineStage parameter are implicitly complete in submission order, memory and execution dependencies between buffer marker writes and other operations must still be explicitly ordered using synchronization commands. The access scope for buffer marker writes falls under the VK_ACCESS_TRANSFER_WRITE_BIT, and the pipeline stages for identifying the synchronization scope must include both pipelineStage and VK_PIPELINE_STAGE_TRANSFER_BIT.

Note

Similar to vkCmdWriteTimestamp, if an implementation is unable to write a marker at any specific pipeline stage, it may instead do so at any logically later stage.

Note

Implementations may only support a limited number of pipelined marker write operations in flight at a given time, thus excessive number of marker write operations may degrade command execution performance.

Valid Usage
  • dstOffset must be less than or equal to the size of dstBuffer minus 4.

  • dstBuffer must have been created with VK_BUFFER_USAGE_TRANSFER_DST_BIT usage flag

  • If dstBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • dstOffset must be a multiple of 4

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pipelineStage must be a valid VkPipelineStageFlagBits value

  • dstBuffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support transfer, graphics, or compute operations

  • Both of commandBuffer, and dstBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Transfer
Graphics
Compute

Transfer

19. Drawing Commands

Drawing commands (commands with Draw in the name) provoke work in a graphics pipeline. Drawing commands are recorded into a command buffer and when executed by a queue, will produce work which executes according to the bound graphics pipeline. A graphics pipeline must be bound to a command buffer before any drawing commands are recorded in that command buffer.

Each draw is made up of zero or more vertices and zero or more instances, which are processed by the device and result in the assembly of primitives. Primitives are assembled according to the pInputAssemblyState member of the VkGraphicsPipelineCreateInfo structure, which is of type VkPipelineInputAssemblyStateCreateInfo:

typedef struct VkPipelineInputAssemblyStateCreateInfo {
    VkStructureType                            sType;
    const void*                                pNext;
    VkPipelineInputAssemblyStateCreateFlags    flags;
    VkPrimitiveTopology                        topology;
    VkBool32                                   primitiveRestartEnable;
} VkPipelineInputAssemblyStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • topology is a VkPrimitiveTopology defining the primitive topology, as described below.

  • primitiveRestartEnable controls whether a special vertex index value is treated as restarting the assembly of primitives. This enable only applies to indexed draws (vkCmdDrawIndexed and vkCmdDrawIndexedIndirect), and the special index value is either 0xFFFFFFFF when the indexType parameter of vkCmdBindIndexBuffer is equal to VK_INDEX_TYPE_UINT32, or 0xFFFF when indexType is equal to VK_INDEX_TYPE_UINT16. Primitive restart is not allowed for “list” topologies.

Restarting the assembly of primitives discards the most recent index values if those elements formed an incomplete primitive, and restarts the primitive assembly using the subsequent indices, but only assembling the immediately following element through the end of the originally specified elements. The primitive restart index value comparison is performed before adding the vertexOffset value to the index value.

Valid Usage
  • If topology is VK_PRIMITIVE_TOPOLOGY_POINT_LIST, VK_PRIMITIVE_TOPOLOGY_LINE_LIST, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST, VK_PRIMITIVE_TOPOLOGY_LINE_LIST_WITH_ADJACENCY, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY or VK_PRIMITIVE_TOPOLOGY_PATCH_LIST, primitiveRestartEnable must be VK_FALSE

  • If the geometry shaders feature is not enabled, topology must not be any of VK_PRIMITIVE_TOPOLOGY_LINE_LIST_WITH_ADJACENCY, VK_PRIMITIVE_TOPOLOGY_LINE_STRIP_WITH_ADJACENCY, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY or VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY

  • If the tessellation shaders feature is not enabled, topology must not be VK_PRIMITIVE_TOPOLOGY_PATCH_LIST

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_INPUT_ASSEMBLY_STATE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • topology must be a valid VkPrimitiveTopology value

typedef VkFlags VkPipelineInputAssemblyStateCreateFlags;

VkPipelineInputAssemblyStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

19.1. Primitive Topologies

Primitive topology determines how consecutive vertices are organized into primitives, and determines the type of primitive that is used at the beginning of the graphics pipeline. The effective topology for later stages of the pipeline is altered by tessellation or geometry shading (if either is in use) and depends on the execution modes of those shaders. Supported topologies are defined by VkPrimitiveTopology and include:

typedef enum VkPrimitiveTopology {
    VK_PRIMITIVE_TOPOLOGY_POINT_LIST = 0,
    VK_PRIMITIVE_TOPOLOGY_LINE_LIST = 1,
    VK_PRIMITIVE_TOPOLOGY_LINE_STRIP = 2,
    VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST = 3,
    VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP = 4,
    VK_PRIMITIVE_TOPOLOGY_TRIANGLE_FAN = 5,
    VK_PRIMITIVE_TOPOLOGY_LINE_LIST_WITH_ADJACENCY = 6,
    VK_PRIMITIVE_TOPOLOGY_LINE_STRIP_WITH_ADJACENCY = 7,
    VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY = 8,
    VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY = 9,
    VK_PRIMITIVE_TOPOLOGY_PATCH_LIST = 10,
} VkPrimitiveTopology;

Each primitive topology, and its construction from a list of vertices, is summarized below with a supporting diagram. In each diagram, the numbered points show the sequencing of vertices in order within the vertex arrays; however the positions chosen are arbitrary and for illustration only. Vertices connected with solid lines belong to the main primitives. In the primitive types with adjacency, the vertices connected by dashed lines are the adjacent vertices that are accessible in a geometry shader.

Note

The terminology “vertex i ” means “the vertex with index i in the ordered list of vertices defining this primitive”.

Note

Depending on the polygon mode, a polygon primitive generated from a drawing command with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_FAN, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY, or VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY is rendered in one of several ways, such as outlining its border or filling its interior. The order of vertices in such a primitive is significant during polygon rasterization and fragment shading.

19.1.1. Point Lists

A series of individual points are specified with topology VK_PRIMITIVE_TOPOLOGY_POINT_LIST. Each vertex defines a separate point.

primitive topology point list
Figure 12. Point Lists

19.1.2. Line Lists

Lists of line segments, with each segment defined by a pair of vertices, are specified with topology VK_PRIMITIVE_TOPOLOGY_LINE_LIST. The first two vertices define the first segment, with subsequent pairs of vertices each defining one more segment. If the number of vertices is odd, then the last vertex is ignored.

primitive topology line list
Figure 13. Line Lists

19.1.3. Line Strips

A series of one or more connected line segments are specified with topology VK_PRIMITIVE_TOPOLOGY_LINE_STRIP. In this case, the first vertex specifies the first segment’s start point while the second vertex specifies the first segment’s endpoint and the second segment’s start point. In general, vertex i (for i > 0) specifies the beginning of the ith segment and the end of the previous segment. The last vertex specifies the end of the last segment. If only one vertex is specified, then no primitive is generated.

primitive topology line strip
Figure 14. Line Strips

19.1.4. Triangle Lists

Lists of separate triangles are specified with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST. In this case, vertices 3 i, 3 i + 1, and 3 i + 2 (in that order) determine a triangle for each i = 0, 1, …​, n-1, where there are 3 n + k vertices drawn. k is either 0, 1, or 2; if k is not zero, the final k vertices are ignored.

primitive topology triangle list
Figure 15. Triangle Lists

19.1.5. Triangle Strips

A triangle strip is a series of triangles connected along shared edges, and is specified with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP. In this case, the first three vertices define the first triangle, and their order is significant. Each subsequent vertex defines a new triangle using that point along with the last two vertices from the previous triangle. If fewer than three vertices are specified, no primitive is produced. The order of vertices in successive triangles changes as shown in the figure below, so that all triangle faces have the same orientation.

primitive topology triangle strip
Figure 16. Triangle Strips

19.1.6. Triangle Fans

A triangle fan is specified with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_FAN. It is similar to a triangle strip, but changes the vertex replaced from the previous triangle so that all triangles in the fan share a common vertex.

primitive topology triangle fan
Figure 17. Triangle Fans

19.1.7. Line Lists With Adjacency

Lines with adjacency are specified with topology VK_PRIMITIVE_TOPOLOGY_LINE_LIST_WITH_ADJACENCY, and are independent line segments where each endpoint has a corresponding adjacent vertex that is accessible in a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored.

A line segment is drawn from vertex 4 i + 1 to vertex 4 i + 2 for each i = 0, 1, …​, n-1, where there are 4 n + k vertices. k is either 0, 1, 2, or 3; if k is not zero, the final k vertices are ignored. For line segment i, vertices 4 i and 4 i + 3 vertices are considered adjacent to vertices 4 i + 1 and 4 i + 2, respectively.

primitive topology line list with adjacency
Figure 18. Line Lists With Adjacency

19.1.8. Line Strips With Adjacency

Line strips with adjacency are specified with topology VK_PRIMITIVE_TOPOLOGY_LINE_STRIP_WITH_ADJACENCY and are similar to line strips, except that each line segment has a pair of adjacent vertices that are accessible in a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored.

A line segment is drawn from vertex i + 1 vertex to vertex i + 2 for each i = 0, 1, …​, n-1, where there are n + 3 vertices. If there are fewer than four vertices, all vertices are ignored. For line segment i, vertices i and i + 3 are considered adjacent to vertices i + 1 and i + 2, respectively.

primitive topology line strip with adjacency
Figure 19. Line Strips With Adjacency

19.1.9. Triangle Lists With Adjacency

Triangles with adjacency are specified with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY, and are similar to separate triangles except that each triangle edge has an adjacent vertex that is accessible in a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored.

Vertices 6 i, 6 i + 2, and 6 i + 4 (in that order) determine a triangle for each i = 0, 1, …​, n-1, where there are 6 n+k vertices. k is either 0, 1, 2, 3, 4, or 5; if k is non-zero, the final k vertices are ignored. For triangle i, vertices 6 i + 1, 6 i + 3, and 6 i + 5 vertices are considered adjacent to edges from vertex 6 i to 6 i + 2, from 6 i + 2 to 6 i + 4, and from 6 i + 4 to 6 i vertices, respectively.

primitive topology triangle list with adjacency
Figure 20. Triangle Lists With Adjacency

19.1.10. Triangle Strips With Adjacency

Triangle strips with adjacency are specified with topology VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY, and are similar to triangle strips except that each triangle edge has an adjacent vertex that is accessible in a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored.

In triangle strips with adjacency, n triangles are drawn where there are 2 (n + 2) + k vertices. k is either 0 or 1; if k is 1, the final vertex is ignored. If there are fewer than 6 vertices, the entire primitive is ignored.

primitive topology triangle strip with adjacency
Figure 21. Triangle Strips With Adjacency

The table below illustrates the vertices and order used to draw each triangle, and which vertices are considered adjacent to each edge of those triangles. Each triangle is drawn using the vertices whose numbers are in the 1st, 2nd, and 3rd columns under Primitive Vertices, in that order. The vertices in the 1/2, 2/3, and 3/1 columns under Adjacent Vertices are considered adjacent to the edges from the first to the second, from the second to the third, and from the third to the first vertex of the triangle, respectively. The six rows correspond to six cases: the first and only triangle (i = 0, n = 1), the first triangle of several (i = 0, n > 0), odd middle triangles (i = 1, 3, 5 …​), even middle triangles (i = 2, 4, 6, …​), and special cases for the last triangle, when i is either even or odd. For the purposes of this table, both the first vertex and first triangle are numbered 0.

Table 27. Triangles generated by triangle strips with adjacency
Primitive Vertices Adjacent Vertices

Primitive

1st

2nd

3rd

1/2

2/3

3/1

only (i = 0, n = 1)

0

2

4

1

5

3

first (i = 0)

0

2

4

1

6

3

middle (i odd)

2 i + 2

2 i

2 i + 4

2 i-2

2 i + 3

2 i + 6

middle (i even)

2 i

2 i + 2

2 i + 4

2 i-2

2 i + 6

2 i + 3

last (i=n-1, i odd)

2 i + 2

2 i

2 i + 4

2 i-2

2 i + 3

2 i + 5

last (i=n-1, i even)

2 i

2 i + 2

2 i + 4

2 i-2

2 i + 5

2 i + 3

19.1.11. Separate Patches

Separate patches are specified with topology VK_PRIMITIVE_TOPOLOGY_PATCH_LIST. A patch is an ordered collection of vertices used for primitive tessellation. The vertices comprising a patch have no implied geometric ordering, and are used by tessellation shaders and the fixed-function tessellator to generate new point, line, or triangle primitives.

Each patch in the series has a fixed number of vertices, specified by the patchControlPoints member of the VkPipelineTessellationStateCreateInfo structure passed to vkCreateGraphicsPipelines. Once assembled and vertex shaded, these patches are provided as input to the tessellation control shader stage.

If the number of vertices in a patch is given by v, vertices v × i through v × i + v - 1 (in that order) determine a patch for each i = 0, 1, …​, n-1, where there are v × n + k vertices. k is in the range [0, v - 1]; if k is not zero, the final k vertices are ignored.

19.2. Primitive Order

Primitives generated by drawing commands progress through the stages of the graphics pipeline in primitive order. Primitive order is initially determined in the following way:

  1. Submission order determines the initial ordering

  2. For indirect draw commands, the order in which accessed instances of the VkDrawIndirectCommand are stored in buffer, from lower indirect buffer addresses to higher addresses.

  3. If a draw command includes multiple instances, the order in which instances are executed, from lower numbered instances to higher.

  4. The order in which primitives are specified by a draw command:

    • For non-indexed draws, from vertices with a lower numbered vertexIndex to a higher numbered vertexIndex.

    • For indexed draws, vertices sourced from a lower index buffer addresses to higher addresses.

Within this order implementations further sort primitives:

  1. If tessellation shading is active, by an implementation-dependent order of new primitives generated by tessellation.

  2. If geometry shading is active, by the order new primitives are generated by geometry shading.

  3. If the polygon mode is not VK_POLYGON_MODE_FILL, or VK_POLYGON_MODE_FILL_RECTANGLE_NV, by an implementation-dependent ordering of the new primitives generated within the original primitive.

Primitive order is later used to define rasterization order, which determines the order in which fragments output results to a framebuffer.

19.3. Programmable Primitive Shading

Once primitives are assembled, they proceed to the vertex shading stage of the pipeline. If the draw includes multiple instances, then the set of primitives is sent to the vertex shading stage multiple times, once for each instance.

It is undefined whether vertex shading occurs on vertices that are discarded as part of incomplete primitives, but if it does occur then it operates as if they were vertices in complete primitives and such invocations can have side effects.

Vertex shading receives two per-vertex inputs from the primitive assembly stage - the vertexIndex and the instanceIndex. How these values are generated is defined below, with each command.

Drawing commands fall roughly into two categories:

To bind an index buffer to a command buffer, call:

void vkCmdBindIndexBuffer(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset,
    VkIndexType                                 indexType);
  • commandBuffer is the command buffer into which the command is recorded.

  • buffer is the buffer being bound.

  • offset is the starting offset in bytes within buffer used in index buffer address calculations.

  • indexType is a VkIndexType value specifying whether indices are treated as 16 bits or 32 bits.

Valid Usage
  • offset must be less than the size of buffer

  • The sum of offset and the address of the range of VkDeviceMemory object that is backing buffer, must be a multiple of the type indicated by indexType

  • buffer must have been created with the VK_BUFFER_USAGE_INDEX_BUFFER_BIT flag

  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • indexType must be a valid VkIndexType value

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • Both of buffer, and commandBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

Possible values of vkCmdBindIndexBuffer::indexType, specifying the size of indices, are:

typedef enum VkIndexType {
    VK_INDEX_TYPE_UINT16 = 0,
    VK_INDEX_TYPE_UINT32 = 1,
} VkIndexType;
  • VK_INDEX_TYPE_UINT16 specifies that indices are 16-bit unsigned integer values.

  • VK_INDEX_TYPE_UINT32 specifies that indices are 32-bit unsigned integer values.

The parameters for each drawing command are specified directly in the command or read from buffer memory, depending on the command. Drawing commands that source their parameters from buffer memory are known as indirect drawing commands.

All drawing commands interact with the Robust Buffer Access feature.

To record a non-indexed draw, call:

void vkCmdDraw(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    vertexCount,
    uint32_t                                    instanceCount,
    uint32_t                                    firstVertex,
    uint32_t                                    firstInstance);
  • commandBuffer is the command buffer into which the command is recorded.

  • vertexCount is the number of vertices to draw.

  • instanceCount is the number of instances to draw.

  • firstVertex is the index of the first vertex to draw.

  • firstInstance is the instance ID of the first instance to draw.

When the command is executed, primitives are assembled using the current primitive topology and vertexCount consecutive vertex indices with the first vertexIndex value equal to firstVertex. The primitives are drawn instanceCount times with instanceIndex starting with firstInstance and increasing sequentially for each instance. The assembled primitives execute the bound graphics pipeline.

Valid Usage
  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • For a given vertex buffer binding, any attribute data fetched must be entirely contained within the corresponding vertex buffer binding, as described in Vertex Input Description

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

To record an indexed draw, call:

void vkCmdDrawIndexed(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    indexCount,
    uint32_t                                    instanceCount,
    uint32_t                                    firstIndex,
    int32_t                                     vertexOffset,
    uint32_t                                    firstInstance);
  • commandBuffer is the command buffer into which the command is recorded.

  • indexCount is the number of vertices to draw.

  • instanceCount is the number of instances to draw.

  • firstIndex is the base index within the index buffer.

  • vertexOffset is the value added to the vertex index before indexing into the vertex buffer.

  • firstInstance is the instance ID of the first instance to draw.

When the command is executed, primitives are assembled using the current primitive topology and indexCount vertices whose indices are retrieved from the index buffer. The index buffer is treated as an array of tightly packed unsigned integers of size defined by the vkCmdBindIndexBuffer::indexType parameter with which the buffer was bound.

The first vertex index is at an offset of firstIndex * indexSize + offset within the bound index buffer, where offset is the offset specified by vkCmdBindIndexBuffer and indexSize is the byte size of the type specified by indexType. Subsequent index values are retrieved from consecutive locations in the index buffer. Indices are first compared to the primitive restart value, then zero extended to 32 bits (if the indexType is VK_INDEX_TYPE_UINT16) and have vertexOffset added to them, before being supplied as the vertexIndex value.

The primitives are drawn instanceCount times with instanceIndex starting with firstInstance and increasing sequentially for each instance. The assembled primitives execute the bound graphics pipeline.

Valid Usage
  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • For a given vertex buffer binding, any attribute data fetched must be entirely contained within the corresponding vertex buffer binding, as described in Vertex Input Description

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • (indexSize * (firstIndex + indexCount) + offset) must be less than or equal to the size of the bound index buffer, with indexSize being based on the type specified by indexType, where the index buffer, indexType, and offset are specified via vkCmdBindIndexBuffer

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

To record a non-indexed indirect draw, call:

void vkCmdDrawIndirect(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset,
    uint32_t                                    drawCount,
    uint32_t                                    stride);
  • commandBuffer is the command buffer into which the command is recorded.

  • buffer is the buffer containing draw parameters.

  • offset is the byte offset into buffer where parameters begin.

  • drawCount is the number of draws to execute, and can be zero.

  • stride is the byte stride between successive sets of draw parameters.

vkCmdDrawIndirect behaves similarly to vkCmdDraw except that the parameters are read by the device from a buffer during execution. drawCount draws are executed by the command, with parameters taken from buffer starting at offset and increasing by stride bytes for each successive draw. The parameters of each draw are encoded in an array of VkDrawIndirectCommand structures. If drawCount is less than or equal to one, stride is ignored.

Valid Usage
  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • buffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • offset must be a multiple of 4

  • If drawCount is greater than 1, stride must be a multiple of 4 and must be greater than or equal to sizeof(VkDrawIndirectCommand)

  • If the multi-draw indirect feature is not enabled, drawCount must be 0 or 1

  • If the drawIndirectFirstInstance feature is not enabled, all the firstInstance members of the VkDrawIndirectCommand structures accessed by this command must be 0

  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • If drawCount is equal to 1, (offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • If drawCount is greater than 1, (stride × (drawCount - 1) + offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • drawCount must be less than or equal to VkPhysicalDeviceLimits::maxDrawIndirectCount

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • Both of buffer, and commandBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

The VkDrawIndirectCommand structure is defined as:

typedef struct VkDrawIndirectCommand {
    uint32_t    vertexCount;
    uint32_t    instanceCount;
    uint32_t    firstVertex;
    uint32_t    firstInstance;
} VkDrawIndirectCommand;
  • vertexCount is the number of vertices to draw.

  • instanceCount is the number of instances to draw.

  • firstVertex is the index of the first vertex to draw.

  • firstInstance is the instance ID of the first instance to draw.

The members of VkDrawIndirectCommand have the same meaning as the similarly named parameters of vkCmdDraw.

Valid Usage
  • For a given vertex buffer binding, any attribute data fetched must be entirely contained within the corresponding vertex buffer binding, as described in Vertex Input Description

  • If the drawIndirectFirstInstance feature is not enabled, firstInstance must be 0

To record a non-indexed draw call with a draw call count sourced from a buffer, call:

void vkCmdDrawIndirectCountAMD(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset,
    VkBuffer                                    countBuffer,
    VkDeviceSize                                countBufferOffset,
    uint32_t                                    maxDrawCount,
    uint32_t                                    stride);
  • commandBuffer is the command buffer into which the command is recorded.

  • buffer is the buffer containing draw parameters.

  • offset is the byte offset into buffer where parameters begin.

  • countBuffer is the buffer containing the draw count.

  • countBufferOffset is the byte offset into countBuffer where the draw count begins.

  • maxDrawCount specifies the maximum number of draws that will be executed. The actual number of executed draw calls is the minimum of the count specified in countBuffer and maxDrawCount.

  • stride is the byte stride between successive sets of draw parameters.

vkCmdDrawIndirectCountAMD behaves similarly to vkCmdDrawIndirect except that the draw count is read by the device from a buffer during execution. The command will read an unsigned 32-bit integer from countBuffer located at countBufferOffset and use this as the draw count.

Valid Usage
  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • buffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • If countBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • countBuffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • offset must be a multiple of 4

  • countBufferOffset must be a multiple of 4

  • stride must be a multiple of 4 and must be greater than or equal to sizeof(VkDrawIndirectCommand)

  • If maxDrawCount is greater than or equal to 1, (stride × (maxDrawCount - 1) + offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • If the drawIndirectFirstInstance feature is not enabled, all the firstInstance members of the VkDrawIndirectCommand structures accessed by this command must be 0

  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • If the count stored in countBuffer is equal to 1, (offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • If the count stored in countBuffer is greater than 1, (stride × (drawCount - 1) + offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • The count stored in countBuffer must be less than or equal to VkPhysicalDeviceLimits::maxDrawIndirectCount

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • countBuffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • Each of buffer, commandBuffer, and countBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

To record an indexed indirect draw, call:

void vkCmdDrawIndexedIndirect(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset,
    uint32_t                                    drawCount,
    uint32_t                                    stride);
  • commandBuffer is the command buffer into which the command is recorded.

  • buffer is the buffer containing draw parameters.

  • offset is the byte offset into buffer where parameters begin.

  • drawCount is the number of draws to execute, and can be zero.

  • stride is the byte stride between successive sets of draw parameters.

vkCmdDrawIndexedIndirect behaves similarly to vkCmdDrawIndexed except that the parameters are read by the device from a buffer during execution. drawCount draws are executed by the command, with parameters taken from buffer starting at offset and increasing by stride bytes for each successive draw. The parameters of each draw are encoded in an array of VkDrawIndexedIndirectCommand structures. If drawCount is less than or equal to one, stride is ignored.

Valid Usage
  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • buffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • offset must be a multiple of 4

  • If drawCount is greater than 1, stride must be a multiple of 4 and must be greater than or equal to sizeof(VkDrawIndexedIndirectCommand)

  • If the multi-draw indirect feature is not enabled, drawCount must be 0 or 1

  • If the drawIndirectFirstInstance feature is not enabled, all the firstInstance members of the VkDrawIndexedIndirectCommand structures accessed by this command must be 0

  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • If drawCount is equal to 1, (offset + sizeof(VkDrawIndexedIndirectCommand)) must be less than or equal to the size of buffer

  • If drawCount is greater than 1, (stride × (drawCount - 1) + offset + sizeof(VkDrawIndexedIndirectCommand)) must be less than or equal to the size of buffer

  • drawCount must be less than or equal to VkPhysicalDeviceLimits::maxDrawIndirectCount

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • Both of buffer, and commandBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

The VkDrawIndexedIndirectCommand structure is defined as:

typedef struct VkDrawIndexedIndirectCommand {
    uint32_t    indexCount;
    uint32_t    instanceCount;
    uint32_t    firstIndex;
    int32_t     vertexOffset;
    uint32_t    firstInstance;
} VkDrawIndexedIndirectCommand;
  • indexCount is the number of vertices to draw.

  • instanceCount is the number of instances to draw.

  • firstIndex is the base index within the index buffer.

  • vertexOffset is the value added to the vertex index before indexing into the vertex buffer.

  • firstInstance is the instance ID of the first instance to draw.

The members of VkDrawIndexedIndirectCommand have the same meaning as the similarly named parameters of vkCmdDrawIndexed.

Valid Usage
  • For a given vertex buffer binding, any attribute data fetched must be entirely contained within the corresponding vertex buffer binding, as described in Vertex Input Description

  • (indexSize * (firstIndex + indexCount) + offset) must be less than or equal to the size of the bound index buffer, with indexSize being based on the type specified by indexType, where the index buffer, indexType, and offset are specified via vkCmdBindIndexBuffer

  • If the drawIndirectFirstInstance feature is not enabled, firstInstance must be 0

To record an indexed draw call with a draw call count sourced from a buffer, call:

void vkCmdDrawIndexedIndirectCountAMD(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset,
    VkBuffer                                    countBuffer,
    VkDeviceSize                                countBufferOffset,
    uint32_t                                    maxDrawCount,
    uint32_t                                    stride);
  • commandBuffer is the command buffer into which the command is recorded.

  • buffer is the buffer containing draw parameters.

  • offset is the byte offset into buffer where parameters begin.

  • countBuffer is the buffer containing the draw count.

  • countBufferOffset is the byte offset into countBuffer where the draw count begins.

  • maxDrawCount specifies the maximum number of draws that will be executed. The actual number of executed draw calls is the minimum of the count specified in countBuffer and maxDrawCount.

  • stride is the byte stride between successive sets of draw parameters.

vkCmdDrawIndexedIndirectCountAMD behaves similarly to vkCmdDrawIndexedIndirect except that the draw count is read by the device from a buffer during execution. The command will read an unsigned 32-bit integer from countBuffer located at countBufferOffset and use this as the draw count.

Valid Usage
  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • buffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • If countBuffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • countBuffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • offset must be a multiple of 4

  • countBufferOffset must be a multiple of 4

  • stride must be a multiple of 4 and must be greater than or equal to sizeof(VkDrawIndirectCommand)

  • If maxDrawCount is greater than or equal to 1, (stride × (maxDrawCount - 1) + offset + sizeof(VkDrawIndirectCommand)) must be less than or equal to the size of buffer

  • If the drawIndirectFirstInstance feature is not enabled, all the firstInstance members of the VkDrawIndexedIndirectCommand structures accessed by this command must be 0

  • The current render pass must be compatible with the renderPass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • The subpass index of the current render pass must be equal to the subpass member of the VkGraphicsPipelineCreateInfo structure specified when creating the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS.

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS, a push constant value must have been set for VK_PIPELINE_BIND_POINT_GRAPHICS, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • All vertex input bindings accessed via vertex input variables declared in the vertex shader entry point’s interface must have valid buffers bound

  • A valid graphics pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_GRAPHICS

  • If the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS requires any dynamic state, that state must have been set on the current command buffer

  • If count stored in countBuffer is equal to 1, (offset + sizeof(VkDrawIndexedIndirectCommand)) must be less than or equal to the size of buffer

  • If count stored in countBuffer is greater than 1, (stride × (drawCount - 1) + offset + sizeof(VkDrawIndexedIndirectCommand)) must be less than or equal to the size of buffer

  • drawCount must be less than or equal to VkPhysicalDeviceLimits::maxDrawIndirectCount

  • Every input attachment used by the current subpass must be bound to the pipeline via a descriptor set

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_GRAPHICS uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Image subresources used as attachments in the current render pass must not be accessed in any way other than as an attachment by this command.

  • If the draw is recorded in a render pass instance with multiview enabled, the maximum instance index must be less than or equal to VkPhysicalDeviceMultiviewProperties::maxMultiviewInstanceIndex.

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the framebuffer-space pipeline stages in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_GRAPHICS reads from or writes to any image or buffer, the image or buffer must not be a protected image or protected buffer.

  • If the bound graphics pipeline was created with VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable set to VK_TRUE and the current subpass has a depth/stencil attachment, then that attachment must have been created with the VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT bit set

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • countBuffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • This command must only be called inside of a render pass instance

  • Each of buffer, commandBuffer, and countBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics

Graphics

20. Fixed-Function Vertex Processing

Vertex fetching is controlled via configurable state, as a logically distinct graphics pipeline stage.

20.1. Vertex Attributes

Vertex shaders can define input variables, which receive vertex attribute data transferred from one or more VkBuffer(s) by drawing commands. Vertex shader input variables are bound to buffers via an indirect binding where the vertex shader associates a vertex input attribute number with each variable, vertex input attributes are associated to vertex input bindings on a per-pipeline basis, and vertex input bindings are associated with specific buffers on a per-draw basis via the vkCmdBindVertexBuffers command. Vertex input attribute and vertex input binding descriptions also contain format information controlling how data is extracted from buffer memory and converted to the format expected by the vertex shader.

There are VkPhysicalDeviceLimits::maxVertexInputAttributes number of vertex input attributes and VkPhysicalDeviceLimits::maxVertexInputBindings number of vertex input bindings (each referred to by zero-based indices), where there are at least as many vertex input attributes as there are vertex input bindings. Applications can store multiple vertex input attributes interleaved in a single buffer, and use a single vertex input binding to access those attributes.

In GLSL, vertex shaders associate input variables with a vertex input attribute number using the location layout qualifier. The component layout qualifier associates components of a vertex shader input variable with components of a vertex input attribute.

GLSL example
// Assign location M to variableName
layout (location=M, component=2) in vec2 variableName;

// Assign locations [N,N+L) to the array elements of variableNameArray
layout (location=N) in vec4 variableNameArray[L];

In SPIR-V, vertex shaders associate input variables with a vertex input attribute number using the Location decoration. The Component decoration associates components of a vertex shader input variable with components of a vertex input attribute. The Location and Component decorations are specified via the OpDecorate instruction.

SPIR-V example
               ...
          %1 = OpExtInstImport "GLSL.std.450"
               ...
               OpName %9 "variableName"
               OpName %15 "variableNameArray"
               OpDecorate %18 BuiltIn VertexIndex
               OpDecorate %19 BuiltIn InstanceIndex
               OpDecorate %9 Location M
               OpDecorate %9 Component 2
               OpDecorate %15 Location N
               ...
          %2 = OpTypeVoid
          %3 = OpTypeFunction %2
          %6 = OpTypeFloat 32
          %7 = OpTypeVector %6 2
          %8 = OpTypePointer Input %7
          %9 = OpVariable %8 Input
         %10 = OpTypeVector %6 4
         %11 = OpTypeInt 32 0
         %12 = OpConstant %11 L
         %13 = OpTypeArray %10 %12
         %14 = OpTypePointer Input %13
         %15 = OpVariable %14 Input
               ...

20.1.1. Attribute Location and Component Assignment

Vertex shaders allow Location and Component decorations on input variable declarations. The Location decoration specifies which vertex input attribute is used to read and interpret the data that a variable will consume. The Component decoration allows the location to be more finely specified for scalars and vectors, down to the individual components within a location that are consumed. The components within a location are 0, 1, 2, and 3. A variable starting at component N will consume components N, N+1, N+2, …​ up through its size. For single precision types, it is invalid if the sequence of components gets larger than 3.

When a vertex shader input variable declared using a scalar or vector 32-bit data type is assigned a location, its value(s) are taken from the components of the input attribute specified with the corresponding VkVertexInputAttributeDescription::location. The components used depend on the type of variable and the Component decoration specified in the variable declaration, as identified in Input attribute components accessed by 32-bit input variables. Any 32-bit scalar or vector input will consume a single location. For 32-bit data types, missing components are filled in with default values as described below.

Table 28. Input attribute components accessed by 32-bit input variables
32-bit data type Component decoration Components consumed

scalar

0 or unspecified

(x, o, o, o)

scalar

1

(o, y, o, o)

scalar

2

(o, o, z, o)

scalar

3

(o, o, o, w)

two-component vector

0 or unspecified

(x, y, o, o)

two-component vector

1

(o, y, z, o)

two-component vector

2

(o, o, z, w)

three-component vector

0 or unspecified

(x, y, z, o)

three-component vector

1

(o, y, z, w)

four-component vector

0 or unspecified

(x, y, z, w)

Components indicated by “o” are available for use by other input variables which are sourced from the same attribute, and if used, are either filled with the corresponding component from the input format (if present), or the default value.

When a vertex shader input variable declared using a 32-bit floating point matrix type is assigned a location i, its values are taken from consecutive input attributes starting with the corresponding VkVertexInputAttributeDescription::location. Such matrices are treated as an array of column vectors with values taken from the input attributes identified in Input attributes accessed by 32-bit input matrix variables. The VkVertexInputAttributeDescription::format must be specified with a VkFormat that corresponds to the appropriate type of column vector. The Component decoration must not be used with matrix types.

Table 29. Input attributes accessed by 32-bit input matrix variables
Data type Column vector type Locations consumed Components consumed

mat2

two-component vector

i, i+1

(x, y, o, o), (x, y, o, o)

mat2x3

three-component vector

i, i+1

(x, y, z, o), (x, y, z, o)

mat2x4

four-component vector

i, i+1

(x, y, z, w), (x, y, z, w)

mat3x2

two-component vector

i, i+1, i+2

(x, y, o, o), (x, y, o, o), (x, y, o, o)

mat3

three-component vector

i, i+1, i+2

(x, y, z, o), (x, y, z, o), (x, y, z, o)

mat3x4

four-component vector

i, i+1, i+2

(x, y, z, w), (x, y, z, w), (x, y, z, w)

mat4x2

two-component vector

i, i+1, i+2, i+3

(x, y, o, o), (x, y, o, o), (x, y, o, o), (x, y, o, o)

mat4x3

three-component vector

i, i+1, i+2, i+3

(x, y, z, o), (x, y, z, o), (x, y, z, o), (x, y, z, o)

mat4

four-component vector

i, i+1, i+2, i+3

(x, y, z, w), (x, y, z, w), (x, y, z, w), (x, y, z, w)

Components indicated by “o” are available for use by other input variables which are sourced from the same attribute, and if used, are either filled with the corresponding component from the input (if present), or the default value.

When a vertex shader input variable declared using a scalar or vector 64-bit data type is assigned a location i, its values are taken from consecutive input attributes starting with the corresponding VkVertexInputAttributeDescription::location. The locations and components used depend on the type of variable and the Component decoration specified in the variable declaration, as identified in Input attribute locations and components accessed by 64-bit input variables. For 64-bit data types, no default attribute values are provided. Input variables must not use more components than provided by the attribute. Input attributes which have one- or two-component 64-bit formats will consume a single location. Input attributes which have three- or four-component 64-bit formats will consume two consecutive locations. A 64-bit scalar data type will consume two components, and a 64-bit two-component vector data type will consume all four components available within a location. A three- or four-component 64-bit data type must not specify a component. A three-component 64-bit data type will consume all four components of the first location and components 0 and 1 of the second location. This leaves components 2 and 3 available for other component-qualified declarations. A four-component 64-bit data type will consume all four components of the first location and all four components of the second location. It is invalid for a scalar or two-component 64-bit data type to specify a component of 1 or 3.

Table 30. Input attribute locations and components accessed by 64-bit input variables
Input format Locations consumed 64-bit data type Location decoration Component decoration 32-bit components consumed

R64

i

scalar

i

0 or unspecified

(x, y, -, -)

R64G64

i

scalar

i

0 or unspecified

(x, y, o, o)

scalar

i

2

(o, o, z, w)

two-component vector

i

0 or unspecified

(x, y, z, w)

R64G64B64

i, i+1

scalar

i

0 or unspecified

(x, y, o, o), (o, o, -, -)

scalar

i

2

(o, o, z, w), (o, o, -, -)

scalar

i+1

0 or unspecified

(o, o, o, o), (x, y, -, -)

two-component vector

i

0 or unspecified

(x, y, z, w), (o, o, -, -)

three-component vector

i

unspecified

(x, y, z, w), (x, y, -, -)

R64G64B64A64

i, i+1

scalar

i

0 or unspecified

(x, y, o, o), (o, o, o, o)

scalar

i

2

(o, o, z, w), (o, o, o, o)

scalar

i+1

0 or unspecified

(o, o, o, o), (x, y, o, o)

scalar

i+1

2

(o, o, o, o), (o, o, z, w)

two-component vector

i

0 or unspecified

(x, y, z, w), (o, o, o, o)

two-component vector

i+1

0 or unspecified

(o, o, o, o), (x, y, z, w)

three-component vector

i

unspecified

(x, y, z, w), (x, y, o, o)

four-component vector

i

unspecified

(x, y, z, w), (x, y, z, w)

Components indicated by “o” are available for use by other input variables which are sourced from the same attribute. Components indicated by “-” are not available for input variables as there are no default values provided for 64-bit data types, and there is no data provided by the input format.

When a vertex shader input variable declared using a 64-bit floating-point matrix type is assigned a location i, its values are taken from consecutive input attribute locations. Such matrices are treated as an array of column vectors with values taken from the input attributes as shown in Input attribute locations and components accessed by 64-bit input variables. Each column vector starts at the location immediately following the last location of the previous column vector. The number of attributes and components assigned to each matrix is determined by the matrix dimensions and ranges from two to eight locations.

When a vertex shader input variable declared using an array type is assigned a location, its values are taken from consecutive input attributes starting with the corresponding VkVertexInputAttributeDescription::location. The number of attributes and components assigned to each element are determined according to the data type of the array elements and Component decoration (if any) specified in the declaration of the array, as described above. Each element of the array, in order, is assigned to consecutive locations, but all at the same specified component within each location.

Only input variables declared with the data types and component decorations as specified above are supported. Location aliasing is causing two variables to have the same location number. Component aliasing is assigning the same (or overlapping) component number for two location aliases. Location aliasing is allowed only if it does not cause component aliasing. Further, when location aliasing, the aliases sharing the location must all have the same SPIR-V floating-point component type or all have the same width integer-type components.

20.2. Vertex Input Description

Applications specify vertex input attribute and vertex input binding descriptions as part of graphics pipeline creation. The VkGraphicsPipelineCreateInfo::pVertexInputState points to a structure of type VkPipelineVertexInputStateCreateInfo.

The VkPipelineVertexInputStateCreateInfo structure is defined as:

typedef struct VkPipelineVertexInputStateCreateInfo {
    VkStructureType                             sType;
    const void*                                 pNext;
    VkPipelineVertexInputStateCreateFlags       flags;
    uint32_t                                    vertexBindingDescriptionCount;
    const VkVertexInputBindingDescription*      pVertexBindingDescriptions;
    uint32_t                                    vertexAttributeDescriptionCount;
    const VkVertexInputAttributeDescription*    pVertexAttributeDescriptions;
} VkPipelineVertexInputStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • vertexBindingDescriptionCount is the number of vertex binding descriptions provided in pVertexBindingDescriptions.

  • pVertexBindingDescriptions is a pointer to an array of VkVertexInputBindingDescription structures.

  • vertexAttributeDescriptionCount is the number of vertex attribute descriptions provided in pVertexAttributeDescriptions.

  • pVertexAttributeDescriptions is a pointer to an array of VkVertexInputAttributeDescription structures.

Valid Usage
  • vertexBindingDescriptionCount must be less than or equal to VkPhysicalDeviceLimits::maxVertexInputBindings

  • vertexAttributeDescriptionCount must be less than or equal to VkPhysicalDeviceLimits::maxVertexInputAttributes

  • For every binding specified by each element of pVertexAttributeDescriptions, a VkVertexInputBindingDescription must exist in pVertexBindingDescriptions with the same value of binding

  • All elements of pVertexBindingDescriptions must describe distinct binding numbers

  • All elements of pVertexAttributeDescriptions must describe distinct attribute locations

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkPipelineVertexInputDivisorStateCreateInfoEXT

  • flags must be 0

  • If vertexBindingDescriptionCount is not 0, pVertexBindingDescriptions must be a valid pointer to an array of vertexBindingDescriptionCount valid VkVertexInputBindingDescription structures

  • If vertexAttributeDescriptionCount is not 0, pVertexAttributeDescriptions must be a valid pointer to an array of vertexAttributeDescriptionCount valid VkVertexInputAttributeDescription structures

typedef VkFlags VkPipelineVertexInputStateCreateFlags;

VkPipelineVertexInputStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

Each vertex input binding is specified by an instance of the VkVertexInputBindingDescription structure.

The VkVertexInputBindingDescription structure is defined as:

typedef struct VkVertexInputBindingDescription {
    uint32_t             binding;
    uint32_t             stride;
    VkVertexInputRate    inputRate;
} VkVertexInputBindingDescription;
  • binding is the binding number that this structure describes.

  • stride is the distance in bytes between two consecutive elements within the buffer.

  • inputRate is a VkVertexInputRate value specifying whether vertex attribute addressing is a function of the vertex index or of the instance index.

Valid Usage
  • binding must be less than VkPhysicalDeviceLimits::maxVertexInputBindings

  • stride must be less than or equal to VkPhysicalDeviceLimits::maxVertexInputBindingStride

Valid Usage (Implicit)

Possible values of VkVertexInputBindingDescription::inputRate, specifying the rate at which vertex attributes are pulled from buffers, are:

typedef enum VkVertexInputRate {
    VK_VERTEX_INPUT_RATE_VERTEX = 0,
    VK_VERTEX_INPUT_RATE_INSTANCE = 1,
} VkVertexInputRate;
  • VK_VERTEX_INPUT_RATE_VERTEX specifies that vertex attribute addressing is a function of the vertex index.

  • VK_VERTEX_INPUT_RATE_INSTANCE specifies that vertex attribute addressing is a function of the instance index.

Each vertex input attribute is specified by an instance of the VkVertexInputAttributeDescription structure.

The VkVertexInputAttributeDescription structure is defined as:

typedef struct VkVertexInputAttributeDescription {
    uint32_t    location;
    uint32_t    binding;
    VkFormat    format;
    uint32_t    offset;
} VkVertexInputAttributeDescription;
  • location is the shader binding location number for this attribute.

  • binding is the binding number which this attribute takes its data from.

  • format is the size and type of the vertex attribute data.

  • offset is a byte offset of this attribute relative to the start of an element in the vertex input binding.

Valid Usage
  • location must be less than VkPhysicalDeviceLimits::maxVertexInputAttributes

  • binding must be less than VkPhysicalDeviceLimits::maxVertexInputBindings

  • offset must be less than or equal to VkPhysicalDeviceLimits::maxVertexInputAttributeOffset

  • format must be allowed as a vertex buffer format, as specified by the VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT flag in VkFormatProperties::bufferFeatures returned by vkGetPhysicalDeviceFormatProperties

Valid Usage (Implicit)

To bind vertex buffers to a command buffer for use in subsequent draw commands, call:

void vkCmdBindVertexBuffers(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    firstBinding,
    uint32_t                                    bindingCount,
    const VkBuffer*                             pBuffers,
    const VkDeviceSize*                         pOffsets);
  • commandBuffer is the command buffer into which the command is recorded.

  • firstBinding is the index of the first vertex input binding whose state is updated by the command.

  • bindingCount is the number of vertex input bindings whose state is updated by the command.

  • pBuffers is a pointer to an array of buffer handles.

  • pOffsets is a pointer to an array of buffer offsets.

The values taken from elements i of pBuffers and pOffsets replace the current state for the vertex input binding firstBinding + i, for i in [0, bindingCount). The vertex input binding is updated to start at the offset indicated by pOffsets[i] from the start of the buffer pBuffers[i]. All vertex input attributes that use each of these bindings will use these updated addresses in their address calculations for subsequent draw commands.

Valid Usage
  • firstBinding must be less than VkPhysicalDeviceLimits::maxVertexInputBindings

  • The sum of firstBinding and bindingCount must be less than or equal to VkPhysicalDeviceLimits::maxVertexInputBindings

  • All elements of pOffsets must be less than the size of the corresponding element in pBuffers

  • All elements of pBuffers must have been created with the VK_BUFFER_USAGE_VERTEX_BUFFER_BIT flag

  • Each element of pBuffers that is non-sparse must be bound completely and contiguously to a single VkDeviceMemory object

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pBuffers must be a valid pointer to an array of bindingCount valid VkBuffer handles

  • pOffsets must be a valid pointer to an array of bindingCount VkDeviceSize values

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • bindingCount must be greater than 0

  • Both of commandBuffer, and the elements of pBuffers must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

20.3. Vertex Attribute Divisor in Instanced Rendering

If the pNext chain of VkPipelineVertexInputStateCreateInfo includes a VkPipelineVertexInputDivisorStateCreateInfoEXT structure, then that structure controls how vertex attributes are assigned to an instance when instanced rendering is enabled.

The VkPipelineVertexInputDivisorStateCreateInfoEXT structure is defined as:

typedef struct VkPipelineVertexInputDivisorStateCreateInfoEXT {
    VkStructureType                                     sType;
    const void*                                         pNext;
    uint32_t                                            vertexBindingDivisorCount;
    const VkVertexInputBindingDivisorDescriptionEXT*    pVertexBindingDivisors;
} VkPipelineVertexInputDivisorStateCreateInfoEXT;
  • sType is the type of this structure

  • pNext is NULL or a pointer to an extension-specific structure

  • vertexBindingDivisorCount is the number of elements in the pVertexBindingDivisors array.

  • pVertexBindingDivisors is a pointer to an array of VkVertexInputBindingDivisorDescriptionEXT structures, which specifies the divisor value for each binding.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_DIVISOR_STATE_CREATE_INFO_EXT

  • pVertexBindingDivisors must be a valid pointer to an array of vertexBindingDivisorCount VkVertexInputBindingDivisorDescriptionEXT structures

  • vertexBindingDivisorCount must be greater than 0

The individual divisor values per binding are specified using the VkVertexInputBindingDivisorDescriptionEXT structure which is defined as:

typedef struct VkVertexInputBindingDivisorDescriptionEXT {
    uint32_t    binding;
    uint32_t    divisor;
} VkVertexInputBindingDivisorDescriptionEXT;
  • binding is the binding number for which the divisor is specified.

  • divisor is the the number of successive instances that will use the same value of the vertex attribute when instanced rendering is enabled. For example, if the divisor is N, the same vertex attribute will applied to N successive instances before moving on to the next vertex attribute. If a value of 0 is used for the divisor, then the first vertex attribute will be applied to all instances. The maximum value of divisor is implementation dependent and can be queried using VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT::maxVertexAttribDivisor.

If this structure is not used to define a divisor value for an attribute then the divisor has a logical default value of 1.

Valid Usage
  • binding must be less than VkPhysicalDeviceLimits::maxVertexInputBindings

  • divisor must be a value between 0 and VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT::maxVertexAttribDivisor, inclusive.

  • VkVertexInputBindingDescription::inputRate must be of type VK_VERTEX_INPUT_RATE_INSTANCE for this binding.

The address of each attribute for each vertexIndex and instanceIndex is calculated as follows:

  • Let attribDesc be the member of VkPipelineVertexInputStateCreateInfo::pVertexAttributeDescriptions with VkVertexInputAttributeDescription::location equal to the vertex input attribute number.

  • Let bindingDesc be the member of VkPipelineVertexInputStateCreateInfo::pVertexBindingDescriptions with VkVertexInputAttributeDescription::binding equal to attribDesc.binding.

  • Let vertexIndex be the index of the vertex within the draw (a value between firstVertex and firstVertex+vertexCount for vkCmdDraw, or a value taken from the index buffer for vkCmdDrawIndexed), and let instanceIndex be the instance number of the draw (a value between firstInstance and firstInstance+instanceCount).

  • Let divisor be the member of VkPipelineVertexInputDivisorStateCreateInfoEXT::pVertexBindingDivisors with VkVertexInputBindingDivisorDescriptionEXT::binding equal to attribDesc.binding.

bufferBindingAddress = buffer[binding].baseAddress + offset[binding];

if (bindingDesc.inputRate == VK_VERTEX_INPUT_RATE_VERTEX)
    vertexOffset = vertexIndex * bindingDesc.stride;
else
    if (divisor == 0)
        vertexOffset = 0;
    else
        vertexOffset = (instanceIndex / divisor) * bindingDesc.stride;

attribAddress = bufferBindingAddress + vertexOffset + attribDesc.offset;

For each attribute, raw data is extracted starting at attribAddress and is converted from the VkVertexInputAttributeDescription’s format to either to floating-point, unsigned integer, or signed integer based on the base type of the format; the base type of the format must match the base type of the input variable in the shader. If format is a packed format, attribAddress must be a multiple of the size in bytes of the whole attribute data type as described in Packed Formats. Otherwise, attribAddress must be a multiple of the size in bytes of the component type indicated by format (see Formats). If the format does not include G, B, or A components, then those are filled with (0,0,1) as needed (using either 1.0f or integer 1 based on the format) for attributes that are not 64-bit data types. The number of components in the vertex shader input variable need not exactly match the number of components in the format. If the vertex shader has fewer components, the extra components are discarded.

20.4. Example

To create a graphics pipeline that uses the following vertex description:

struct Vertex
{
    float   x, y, z, w;
    uint8_t u, v;
};

The application could use the following set of structures:

const VkVertexInputBindingDescription binding =
{
    0,                                          // binding
    sizeof(Vertex),                             // stride
    VK_VERTEX_INPUT_RATE_VERTEX                 // inputRate
};

const VkVertexInputAttributeDescription attributes[] =
{
    {
        0,                                      // location
        binding.binding,                        // binding
        VK_FORMAT_R32G32B32A32_SFLOAT,          // format
        0                                       // offset
    },
    {
        1,                                      // location
        binding.binding,                        // binding
        VK_FORMAT_R8G8_UNORM,                   // format
        4 * sizeof(float)                       // offset
    }
};

const VkPipelineVertexInputStateCreateInfo viInfo =
{
    VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_CREATE_INFO,    // sType
    NULL,                         // pNext
    0,                            // flags
    1,                            // vertexBindingDescriptionCount
    &binding,                     // pVertexBindingDescriptions
    2,                            // vertexAttributeDescriptionCount
    &attributes[0]                // pVertexAttributeDescriptions
};

21. Tessellation

Tessellation involves three pipeline stages. First, a tessellation control shader transforms control points of a patch and can produce per-patch data. Second, a fixed-function tessellator generates multiple primitives corresponding to a tessellation of the patch in (u,v) or (u,v,w) parameter space. Third, a tessellation evaluation shader transforms the vertices of the tessellated patch, for example to compute their positions and attributes as part of the tessellated surface. The tessellator is enabled when the pipeline contains both a tessellation control shader and a tessellation evaluation shader.

21.1. Tessellator

If a pipeline includes both tessellation shaders (control and evaluation), the tessellator consumes each input patch (after vertex shading) and produces a new set of independent primitives (points, lines, or triangles). These primitives are logically produced by subdividing a geometric primitive (rectangle or triangle) according to the per-patch outer and inner tessellation levels written by the tessellation control shader. These levels are specified using the built-in variables TessLevelOuter and TessLevelInner, respectively. This subdivision is performed in an implementation-dependent manner. If no tessellation shaders are present in the pipeline, the tessellator is disabled and incoming primitives are passed through without modification.

The type of subdivision performed by the tessellator is specified by an OpExecutionMode instruction in the tessellation evaluation or tessellation control shader using one of execution modes Triangles, Quads, and IsoLines. Other tessellation-related execution modes can also be specified in either the tessellation control or tessellation evaluation shaders, and if they are specified in both then the modes must be the same.

Tessellation execution modes include:

  • Triangles, Quads, and IsoLines. These control the type of subdivision and topology of the output primitives. One mode must be set in at least one of the tessellation shader stages.

  • VertexOrderCw and VertexOrderCcw. These control the orientation of triangles generated by the tessellator. One mode must be set in at least one of the tessellation shader stages.

  • PointMode. Controls generation of points rather than triangles or lines. This functionality defaults to disabled, and is enabled if either shader stage includes the execution mode.

  • SpacingEqual, SpacingFractionalEven, and SpacingFractionalOdd. Controls the spacing of segments on the edges of tessellated primitives. One mode must be set in at least one of the tessellation shader stages.

  • OutputVertices. Controls the size of the output patch of the tessellation control shader. One value must be set in at least one of the tessellation shader stages.

For triangles, the tessellator subdivides a triangle primitive into smaller triangles. For quads, the tessellator subdivides a rectangle primitive into smaller triangles. For isolines, the tessellator subdivides a rectangle primitive into a collection of line segments arranged in strips stretching across the rectangle in the u dimension (i.e. the coordinates in TessCoord are of the form (0,x) through (1,x) for all tessellation evaluation shader invocations that share a line).

Each vertex produced by the tessellator has an associated (u,v,w) or (u,v) position in a normalized parameter space, with parameter values in the range [0,1], as illustrated in figures Domain parameterization for tessellation primitive modes (upper-left origin) and Domain parameterization for tessellation primitive modes (lower-left origin). The domain space can have either an upper-left or lower-left origin, selected by the domainOrigin member of VkPipelineTessellationDomainOriginStateCreateInfo.

tessparamUL
Figure 22. Domain parameterization for tessellation primitive modes (upper-left origin)
tessparam
Figure 23. Domain parameterization for tessellation primitive modes (lower-left origin)
Caption

In the domain parameterization diagrams, the coordinates illustrate the value of TessCoord at the corners of the domain. The labels on the edges indicate the inner (IL0 and IL1) and outer (OL0 through OL3) tessellation level values used to control the number of subdivisions along each edge of the domain.

For triangles, the vertex’s position is a barycentric coordinate (u,v,w), where u + v + w = 1.0, and indicates the relative influence of the three vertices of the triangle on the position of the vertex. For quads and isolines, the position is a (u,v) coordinate indicating the relative horizontal and vertical position of the vertex relative to the subdivided rectangle. The subdivision process is explained in more detail in subsequent sections.

21.2. Tessellator Patch Discard

A patch is discarded by the tessellator if any relevant outer tessellation level is less than or equal to zero.

Patches will also be discarded if any relevant outer tessellation level corresponds to a floating-point NaN (not a number) in implementations supporting NaN.

No new primitives are generated and the tessellation evaluation shader is not executed for patches that are discarded. For Quads, all four outer levels are relevant. For Triangles and IsoLines, only the first three or two outer levels, respectively, are relevant. Negative inner levels will not cause a patch to be discarded; they will be clamped as described below.

21.3. Tessellator Spacing

Each of the tessellation levels is used to determine the number and spacing of segments used to subdivide a corresponding edge. The method used to derive the number and spacing of segments is specified by an OpExecutionMode in the tessellation control or tessellation evaluation shader using one of the identifiers SpacingEqual, SpacingFractionalEven, or SpacingFractionalOdd.

If SpacingEqual is used, the floating-point tessellation level is first clamped to [1, maxLevel], where maxLevel is the implementation-dependent maximum tessellation level (VkPhysicalDeviceLimits::maxTessellationGenerationLevel). The result is rounded up to the nearest integer n, and the corresponding edge is divided into n segments of equal length in (u,v) space.

If SpacingFractionalEven is used, the tessellation level is first clamped to [2, maxLevel] and then rounded up to the nearest even integer n. If SpacingFractionalOdd is used, the tessellation level is clamped to [1, maxLevel - 1] and then rounded up to the nearest odd integer n. If n is one, the edge will not be subdivided. Otherwise, the corresponding edge will be divided into n - 2 segments of equal length, and two additional segments of equal length that are typically shorter than the other segments. The length of the two additional segments relative to the others will decrease monotonically with n - f, where f is the clamped floating-point tessellation level. When n - f is zero, the additional segments will have equal length to the other segments. As n - f approaches 2.0, the relative length of the additional segments approaches zero. The two additional segments must be placed symmetrically on opposite sides of the subdivided edge. The relative location of these two segments is implementation-dependent, but must be identical for any pair of subdivided edges with identical values of f.

When the tessellator produces triangles (in the Triangles or Quads modes), the orientation of all triangles is specified with an OpExecutionMode of VertexOrderCw or VertexOrderCcw in the tessellation control or tessellation evaluation shaders. If the order is VertexOrderCw, the vertices of all generated triangles will have clockwise ordering in (u,v) or (u,v,w) space. If the order is VertexOrderCcw, the vertices will have counter-clockwise ordering.

If the tessellation domain has an upper-left origin, the vertices of a triangle have counter-clockwise ordering if

a = u0 v1 - u1 v0 + u1 v2 - u2 v1 + u2 v0 - u0 v2

is negative, and clockwise ordering if a is positive. ui and vi are the u and v coordinates in normalized parameter space of the ith vertex of the triangle. If the tessellation domain has a lower-left origin, the vertices of a triangle have counter-clockwise ordering if a is positive, and clockwise ordering if a is negative.

Note

The value a is proportional (with a positive factor) to the signed area of the triangle.

In Triangles mode, even though the vertex coordinates have a w value, it does not participate directly in the computation of a, being an affine combination of u and v.

For all primitive modes, the tessellator is capable of generating points instead of lines or triangles. If the tessellation control or tessellation evaluation shader specifies the OpExecutionMode PointMode, the primitive generator will generate one point for each distinct vertex produced by tessellation. Otherwise, the tessellator will produce a collection of line segments or triangles according to the primitive mode. When tessellating triangles or quads in point mode with fractional odd spacing, the tessellator may produce interior vertices that are positioned on the edge of the patch if an inner tessellation level is less than or equal to one. Such vertices are considered distinct from vertices produced by subdividing the outer edge of the patch, even if there are pairs of vertices with identical coordinates.

21.4. Tessellation Primitive Ordering

Few guarantees are provided for the relative ordering of primitives produced by tessellation, as they pertain to primitive order.

  • The output primitives generated from each input primitive are passed to subsequent pipeline stages in an implementation-dependent order.

  • All output primitives generated from a given input primitive are passed to subsequent pipeline stages before any output primitives generated from subsequent input primitives.

21.5. Triangle Tessellation

If the tessellation primitive mode is Triangles, an equilateral triangle is subdivided into a collection of triangles covering the area of the original triangle. First, the original triangle is subdivided into a collection of concentric equilateral triangles. The edges of each of these triangles are subdivided, and the area between each triangle pair is filled by triangles produced by joining the vertices on the subdivided edges. The number of concentric triangles and the number of subdivisions along each triangle except the outermost is derived from the first inner tessellation level. The edges of the outermost triangle are subdivided independently, using the first, second, and third outer tessellation levels to control the number of subdivisions of the u = 0 (left), v = 0 (bottom), and w = 0 (right) edges, respectively. The second inner tessellation level and the fourth outer tessellation level have no effect in this mode.

If the first inner tessellation level and all three outer tessellation levels are exactly one after clamping and rounding, only a single triangle with (u,v,w) coordinates of (0,0,1), (1,0,0), and (0,1,0) is generated. If the inner tessellation level is one and any of the outer tessellation levels is greater than one, the inner tessellation level is treated as though it were originally specified as 1 + ε and will result in a two- or three-segment subdivision depending on the tessellation spacing. When used with fractional odd spacing, the three-segment subdivision may produce inner vertices positioned on the edge of the triangle.

If any tessellation level is greater than one, tessellation begins by producing a set of concentric inner triangles and subdividing their edges. First, the three outer edges are temporarily subdivided using the clamped and rounded first inner tessellation level and the specified tessellation spacing, generating n segments. For the outermost inner triangle, the inner triangle is degenerate — a single point at the center of the triangle — if n is two. Otherwise, for each corner of the outer triangle, an inner triangle corner is produced at the intersection of two lines extended perpendicular to the corner’s two adjacent edges running through the vertex of the subdivided outer edge nearest that corner. If n is three, the edges of the inner triangle are not subdivided and is the final triangle in the set of concentric triangles. Otherwise, each edge of the inner triangle is divided into n - 2 segments, with the n - 1 vertices of this subdivision produced by intersecting the inner edge with lines perpendicular to the edge running through the n - 1 innermost vertices of the subdivision of the outer edge. Once the outermost inner triangle is subdivided, the previous subdivision process repeats itself, using the generated triangle as an outer triangle. This subdivision process is illustrated in Inner Triangle Tessellation.

innertri
Figure 24. Inner Triangle Tessellation
Caption

In the Inner Triangle Tessellation diagram, inner tessellation levels of (a) five and (b) four are shown (not to scale). Solid black circles depict vertices along the edges of the concentric triangles. The edges of inner triangles are subdivided by intersecting the edge with segments perpendicular to the edge passing through each inner vertex of the subdivided outer edge. Dotted lines depict edges connecting corresponding vertices on the inner and outer triangle edges.

Once all the concentric triangles are produced and their edges are subdivided, the area between each pair of adjacent inner triangles is filled completely with a set of non-overlapping triangles. In this subdivision, two of the three vertices of each triangle are taken from adjacent vertices on a subdivided edge of one triangle; the third is one of the vertices on the corresponding edge of the other triangle. If the innermost triangle is degenerate (i.e., a point), the triangle containing it is subdivided into six triangles by connecting each of the six vertices on that triangle with the center point. If the innermost triangle is not degenerate, that triangle is added to the set of generated triangles as-is.

After the area corresponding to any inner triangles is filled, the tessellator generates triangles to cover the area between the outermost triangle and the outermost inner triangle. To do this, the temporary subdivision of the outer triangle edge above is discarded. Instead, the u = 0, v = 0, and w = 0 edges are subdivided according to the first, second, and third outer tessellation levels, respectively, and the tessellation spacing. The original subdivision of the first inner triangle is retained. The area between the outer and first inner triangles is completely filled by non-overlapping triangles as described above. If the first (and only) inner triangle is degenerate, a set of triangles is produced by connecting each vertex on the outer triangle edges with the center point.

After all triangles are generated, each vertex in the subdivided triangle is assigned a barycentric (u,v,w) coordinate based on its location relative to the three vertices of the outer triangle.

The algorithm used to subdivide the triangular domain in (u,v,w) space into individual triangles is implementation-dependent. However, the set of triangles produced will completely cover the domain, and no portion of the domain will be covered by multiple triangles.

The order in which the vertices for a given output triangle is generated is implementation-dependent. However, when depicted in a manner similar to Inner Triangle Tessellation, the order of the vertices in each generated triangle will be either all clockwise or all counter-clockwise, according to the vertex order layout declaration.

21.6. Quad Tessellation

If the tessellation primitive mode is Quads, a rectangle is subdivided into a collection of triangles covering the area of the original rectangle. First, the original rectangle is subdivided into a regular mesh of rectangles, where the number of rectangles along the u = 0 and u = 1 (vertical) and v = 0 and v = 1 (horizontal) edges are derived from the first and second inner tessellation levels, respectively. All rectangles, except those adjacent to one of the outer rectangle edges, are decomposed into triangle pairs. The outermost rectangle edges are subdivided independently, using the first, second, third, and fourth outer tessellation levels to control the number of subdivisions of the u = 0 (left), v = 0 (bottom), u = 1 (right), and v = 1 (top) edges, respectively. The area between the inner rectangles of the mesh and the outer rectangle edges are filled by triangles produced by joining the vertices on the subdivided outer edges to the vertices on the edge of the inner rectangle mesh.

If both clamped inner tessellation levels and all four clamped outer tessellation levels are exactly one, only a single triangle pair covering the outer rectangle is generated. Otherwise, if either clamped inner tessellation level is one, that tessellation level is treated as though it were originally specified as 1 + ε and will result in a two- or three-segment subdivision depending on the tessellation spacing. When used with fractional odd spacing, the three-segment subdivision may produce inner vertices positioned on the edge of the rectangle.

If any tessellation level is greater than one, tessellation begins by subdividing the u = 0 and u = 1 edges of the outer rectangle into m segments using the clamped and rounded first inner tessellation level and the tessellation spacing. The v = 0 and v = 1 edges are subdivided into n segments using the second inner tessellation level. Each vertex on the u = 0 and v = 0 edges are joined with the corresponding vertex on the u = 1 and v = 1 edges to produce a set of vertical and horizontal lines that divide the rectangle into a grid of smaller rectangles. The primitive generator emits a pair of non-overlapping triangles covering each such rectangle not adjacent to an edge of the outer rectangle. The boundary of the region covered by these triangles forms an inner rectangle, the edges of which are subdivided by the grid vertices that lie on the edge. If either m or n is two, the inner rectangle is degenerate, and one or both of the rectangle’s edges consist of a single point. This subdivision is illustrated in Figure Inner Quad Tessellation.

innerquad
Figure 25. Inner Quad Tessellation
Caption

In the Inner Quad Tessellation diagram, inner quad tessellation levels of (a) (4,2) and (b) (7,4) are shown. The regions highlighted in red in figure (b) depict the 10 inner rectangles, each of which will be subdivided into two triangles. Solid black circles depict vertices on the boundary of the outer and inner rectangles, where the inner rectangle on the top figure is degenerate (a single line segment). Dotted lines depict the horizontal and vertical edges connecting corresponding vertices on the inner and outer rectangle edges.

After the area corresponding to the inner rectangle is filled, the tessellator must produce triangles to cover the area between the inner and outer rectangles. To do this, the subdivision of the outer rectangle edge above is discarded. Instead, the u = 0, v = 0, u = 1, and v = 1 edges are subdivided according to the first, second, third, and fourth outer tessellation levels, respectively, and the tessellation spacing. The original subdivision of the inner rectangle is retained. The area between the outer and inner rectangles is completely filled by non-overlapping triangles. Two of the three vertices of each triangle are adjacent vertices on a subdivided edge of one rectangle; the third is one of the vertices on the corresponding edge of the other triangle. If either edge of the innermost rectangle is degenerate, the area near the corresponding outer edges is filled by connecting each vertex on the outer edge with the single vertex making up the inner edge.

The algorithm used to subdivide the rectangular domain in (u,v) space into individual triangles is implementation-dependent. However, the set of triangles produced will completely cover the domain, and no portion of the domain will be covered by multiple triangles.

The order in which the vertices for a given output triangle is generated is implementation-dependent. However, when depicted in a manner similar to Inner Quad Tessellation, the order of the vertices in each generated triangle will be either all clockwise or all counter-clockwise, according to the vertex order layout declaration.

21.7. Isoline Tessellation

If the tessellation primitive mode is IsoLines, a set of independent horizontal line segments is drawn. The segments are arranged into connected strips called isolines, where the vertices of each isoline have a constant v coordinate and u coordinates covering the full range [0,1]. The number of isolines generated is derived from the first outer tessellation level; the number of segments in each isoline is derived from the second outer tessellation level. Both inner tessellation levels and the third and fourth outer tessellation levels have no effect in this mode.

As with quad tessellation above, isoline tessellation begins with a rectangle. The u = 0 and u = 1 edges of the rectangle are subdivided according to the first outer tessellation level. For the purposes of this subdivision, the tessellation spacing mode is ignored and treated as equal_spacing. An isoline is drawn connecting each vertex on the u = 0 rectangle edge to the corresponding vertex on the u = 1 rectangle edge, except that no line is drawn between (0,1) and (1,1). If the number of isolines on the subdivided u = 0 and u = 1 edges is n, this process will result in n equally spaced lines with constant v coordinates of 0, \(\frac{1}{n}, \frac{2}{n}, \ldots, \frac{n-1}{n}\).

Each of the n isolines is then subdivided according to the second outer tessellation level and the tessellation spacing, resulting in m line segments. Each segment of each line is emitted by the tessellator.

The order in which the vertices for a given output line is generated is implementation-dependent.

21.8. Tessellation Pipeline State

The pTessellationState member of VkGraphicsPipelineCreateInfo points to a structure of type VkPipelineTessellationStateCreateInfo.

The VkPipelineTessellationStateCreateInfo structure is defined as:

typedef struct VkPipelineTessellationStateCreateInfo {
    VkStructureType                           sType;
    const void*                               pNext;
    VkPipelineTessellationStateCreateFlags    flags;
    uint32_t                                  patchControlPoints;
} VkPipelineTessellationStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • patchControlPoints number of control points per patch.

Valid Usage
  • patchControlPoints must be greater than zero and less than or equal to VkPhysicalDeviceLimits::maxTessellationPatchSize

Valid Usage (Implicit)
typedef VkFlags VkPipelineTessellationStateCreateFlags;

VkPipelineTessellationStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPipelineTessellationDomainOriginStateCreateInfo structure is defined as:

typedef struct VkPipelineTessellationDomainOriginStateCreateInfo {
    VkStructureType               sType;
    const void*                   pNext;
    VkTessellationDomainOrigin    domainOrigin;
} VkPipelineTessellationDomainOriginStateCreateInfo;

or the equivalent

typedef VkPipelineTessellationDomainOriginStateCreateInfo VkPipelineTessellationDomainOriginStateCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • domainOrigin controls the origin of the tessellation domain space, and is of type VkTessellationDomainOrigin.

If the VkPipelineTessellationDomainOriginStateCreateInfo structure is included in the pNext chain of VkPipelineTessellationStateCreateInfo, it controls the origin of the tessellation domain. If this structure is not present, it is as if domainOrigin were VK_TESSELLATION_DOMAIN_ORIGIN_UPPER_LEFT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO

  • domainOrigin must be a valid VkTessellationDomainOrigin value

The possible tessellation domain origins are specified by the VkTessellationDomainOrigin enumeration:

typedef enum VkTessellationDomainOrigin {
    VK_TESSELLATION_DOMAIN_ORIGIN_UPPER_LEFT = 0,
    VK_TESSELLATION_DOMAIN_ORIGIN_LOWER_LEFT = 1,
    VK_TESSELLATION_DOMAIN_ORIGIN_UPPER_LEFT_KHR = VK_TESSELLATION_DOMAIN_ORIGIN_UPPER_LEFT,
    VK_TESSELLATION_DOMAIN_ORIGIN_LOWER_LEFT_KHR = VK_TESSELLATION_DOMAIN_ORIGIN_LOWER_LEFT,
} VkTessellationDomainOrigin;

or the equivalent

typedef VkTessellationDomainOrigin VkTessellationDomainOriginKHR;

This enum affects how the VertexOrderCw and VertexOrderCcw tessellation execution modes are interpreted, since the winding is defined relative to the orientation of the domain.

22. Geometry Shading

The geometry shader operates on a group of vertices and their associated data assembled from a single input primitive, and emits zero or more output primitives and the group of vertices and their associated data required for each output primitive. Geometry shading is enabled when a geometry shader is included in the pipeline.

22.1. Geometry Shader Input Primitives

Each geometry shader invocation has access to all vertices in the primitive (and their associated data), which are presented to the shader as an array of inputs. The input primitive type expected by the geometry shader is specified with an OpExecutionMode instruction in the geometry shader, and must be compatible with the primitive topology used by primitive assembly (if tessellation is not in use) or must match the type of primitive generated by the tessellation primitive generator (if tessellation is in use). Compatibility is defined below, with each input primitive type. The input primitive types accepted by a geometry shader are:

Points

Geometry shaders that operate on points use an OpExecutionMode instruction specifying the InputPoints input mode. Such a shader is valid only when the pipeline primitive topology is VK_PRIMITIVE_TOPOLOGY_POINT_LIST (if tessellation is not in use) or if tessellation is in use and the tessellation evaluation shader uses PointMode. There is only a single input vertex available for each geometry shader invocation. However, inputs to the geometry shader are still presented as an array, but this array has a length of one.

Lines

Geometry shaders that operate on line segments are generated by including an OpExecutionMode instruction with the InputLines mode. Such a shader is valid only for the VK_PRIMITIVE_TOPOLOGY_LINE_LIST, and VK_PRIMITIVE_TOPOLOGY_LINE_STRIP primitive topologies (if tessellation is not in use) or if tessellation is in use and the tessellation mode is Isolines. There are two input vertices available for each geometry shader invocation. The first vertex refers to the vertex at the beginning of the line segment and the second vertex refers to the vertex at the end of the line segment.

Lines with Adjacency

Geometry shaders that operate on line segments with adjacent vertices are generated by including an OpExecutionMode instruction with the InputLinesAdjacency mode. Such a shader is valid only for the VK_PRIMITIVE_TOPOLOGY_LINES_WITH_ADJACENCY and VK_PRIMITIVE_TOPOLOGY_LINE_STRIP_WITH_ADJACENCY primitive topologies and must not be used when tessellation is in use.

In this mode, there are four vertices available for each geometry shader invocation. The second vertex refers to attributes of the vertex at the beginning of the line segment and the third vertex refers to the vertex at the end of the line segment. The first and fourth vertices refer to the vertices adjacent to the beginning and end of the line segment, respectively.

Triangles

Geometry shaders that operate on triangles are created by including an OpExecutionMode instruction with the Triangles mode. Such a shader is valid when the pipeline topology is VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST, VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP, or VK_PRIMITIVE_TOPOLOGY_TRIANGLE_FAN (if tessellation is not in use) or when tessellation is in use and the tessellation mode is Triangles or Quads.

In this mode, there are three vertices available for each geometry shader invocation. The first, second, and third vertices refer to attributes of the first, second, and third vertex of the triangle, respectively.

Triangles with Adjacency

Geometry shaders that operate on triangles with adjacent vertices are created by including an OpExecutionMode instruction with the InputTrianglesAdjacency mode. Such a shader is valid when the pipeline topology is VK_PRIMITIVE_TOPOLOGY_TRIANGLES_WITH_ADJACENCY or VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY, and must not be used when tessellation is in use.

In this mode, there are six vertices available for each geometry shader invocation. The first, third and fifth vertices refer to attributes of the first, second and third vertex of the triangle, respectively. The second, fourth and sixth vertices refer to attributes of the vertices adjacent to the edges from the first to the second vertex, from the second to the third vertex, and from the third to the first vertex, respectively.

22.2. Geometry Shader Output Primitives

A geometry shader generates primitives in one of three output modes: points, line strips, or triangle strips. The primitive mode is specified in the shader using an OpExecutionMode instruction with the OutputPoints, OutputLineStrip or OutputTriangleStrip modes, respectively. Each geometry shader must include exactly one output primitive mode.

The vertices output by the geometry shader are assembled into points, lines, or triangles based on the output primitive type and the resulting primitives are then further processed as described in Rasterization. If the number of vertices emitted by the geometry shader is not sufficient to produce a single primitive, vertices corresponding to incomplete primitives are not processed by subsequent pipeline stages. The number of vertices output by the geometry shader is limited to a maximum count specified in the shader.

The maximum output vertex count is specified in the shader using an OpExecutionMode instruction with the mode set to OutputVertices and the maximum number of vertices that will be produced by the geometry shader specified as a literal. Each geometry shader must specify a maximum output vertex count.

22.3. Multiple Invocations of Geometry Shaders

Geometry shaders can be invoked more than one time for each input primitive. This is known as geometry shader instancing and is requested by including an OpExecutionMode instruction with mode specified as Invocations and the number of invocations specified as an integer literal.

In this mode, the geometry shader will execute n times for each input primitive, where n is the number of invocations specified in the OpExecutionMode instruction. The instance number is available to each invocation as a built-in input using InvocationId.

22.4. Geometry Shader Primitive Ordering

Limited guarantees are provided for the relative ordering of primitives produced by a geometry shader, as they pertain to primitive order.

  • For instanced geometry shaders, the output primitives generated from each input primitive are passed to subsequent pipeline stages using the invocation number to order the primitives, from least to greatest.

  • All output primitives generated from a given input primitive are passed to subsequent pipeline stages before any output primitives generated from subsequent input primitives.

22.5. Geometry Shader Passthrough

A geometry shader that uses the PassthroughNV decoration on a variable in its input interface is considered a passthrough geometry shader. Output primitives in a passthrough geometry shader must have the same topology as the input primitive and are not produced by emitting vertices. The vertices of the output primitive have two different types of attributes, per-vertex and per-primitive. Geometry shader input variables with PassthroughNV decoration are considered to produce per-vertex outputs, where values for each output vertex are copied from the corresponding input vertex. Any built-in or user-defined geometry shader outputs are considered per-primitive in a passthrough geometry shader, where a single output value is copied to all output vertices.

The remainder of this section details the usage of the PassthroughNV decoration and modifications to the interface matching rules when using passthrough geometry shaders.

22.5.1. PassthroughNV Decoration

Decorating a geometry shader input variable with the PassthroughNV decoration indicates that values of this input are copied through to the corresponding vertex of the output primitive. Input variables and block members which do not have the PassthroughNV decoration are consumed by the geometry shader without being passed through to subsequent stages.

The PassthroughNV decoration must only be used within a geometry shader.

Any variable decorated with PassthroughNV must be declared using the Input storage class.

The PassthroughNV decoration must not be used with any of:

  • an input primitive type other than InputPoints, InputLines, or Triangles, as specified by the mode for OpExecutionMode.

  • an invocation count other than one, as specified by the Invocations mode for OpExecutionMode.

  • an OpEntryPoint which statically uses the OpEmitVertex or OpEndPrimitive instructions.

  • a variable decorated with the InvocationId built-in decoration.

  • a variable decorated with the PrimitiveId built-in decoration that is declared using the Input storage class.

22.5.2. Passthrough Interface Matching

When a passthrough geometry shader is in use, the Interface Matching rules involving the geometry shader input and output interfaces operate as described in this section.

For the purposes of matching passthrough geometry shader inputs with outputs of the previous pipeline stages, the PassthroughNV decoration is ignored.

For the purposes of matching the outputs of the geometry shader with subsequent pipeline stages, each input variable with the PassthroughNV decoration is considered to add an equivalent output variable with the same type, decoration (other than PassthroughNV), number, and declaration order on the output interface. The output variable declaration corresponding to an input variable decorated with PassthroughNV will be identical to the input declaration, except that the outermost array dimension of such variables is removed. The output block declaration corresponding to an input block decorated with PassthroughNV or having members decorated with PassthroughNV will be identical to the input declaration, except that the outermost array dimension of such declaration is removed.

If an input block is decorated with PassthroughNV, the equivalent output block contains all the members of the input block. Otherwise, the equivalent output block contains only those input block members decorated with PassthroughNV. All members of the corresponding output block are assigned Location and Component decorations identical to those assigned to the corresponding input block members.

Output variables and blocks generated from inputs decorated with PassthroughNV will only exist for the purposes of interface matching; these declarations are not available to geometry shader code or listed in the module interface.

For the purposes of component counting, passthrough geometry shaders count all statically used input variable components declared with the PassthroughNV decoration as output components as well, since their values will be copied to the output primitive produced by the geometry shader.

23. Fixed-Function Vertex Post-Processing

After programmable vertex processing, the following fixed-function operations are applied to vertices of the resulting primitives:

editing-note

TODO:Odd that this one link to a different chapter is in this list.

Next, rasterization is performed on primitives as described in chapter Rasterization.

23.1. Viewport Swizzle

Each primitive sent to a given viewport has a swizzle and optional negation applied to its clip coordinates. The swizzle that is applied depends on the viewport index, and is controlled by the VkPipelineViewportSwizzleStateCreateInfoNV pipeline state:

typedef struct VkPipelineViewportSwizzleStateCreateInfoNV {
    VkStructureType                                sType;
    const void*                                    pNext;
    VkPipelineViewportSwizzleStateCreateFlagsNV    flags;
    uint32_t                                       viewportCount;
    const VkViewportSwizzleNV*                     pViewportSwizzles;
} VkPipelineViewportSwizzleStateCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • viewportCount is the number of viewport swizzles used by the pipeline.

  • pViewportSwizzles is a pointer to an array of VkViewportSwizzleNV structures, defining the viewport swizzles.

Valid Usage
  • viewportCount must match the viewportCount set in VkPipelineViewportStateCreateInfo

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_SWIZZLE_STATE_CREATE_INFO_NV

  • flags must be 0

  • viewportCount must be greater than 0

typedef VkFlags VkPipelineViewportSwizzleStateCreateFlagsNV;

VkPipelineViewportSwizzleStateCreateFlagsNV is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPipelineViewportSwizzleStateCreateInfoNV state is set by adding an instance of this structure to the pNext chain of an instance of the VkPipelineViewportStateCreateInfo structure and setting the graphics pipeline state with vkCreateGraphicsPipelines.

Each viewport specified from 0 to viewportCount - 1 has its x,y,z,w swizzle state set to the corresponding x, y, z and w in the VkViewportSwizzleNV structure. Each component is of type VkViewportCoordinateSwizzleNV, which determines the type of swizzle for that component. The value of x computes the new x component of the position as:

if (x == VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_X_NV) x' = x; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_X_NV) x' = -x; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Y_NV) x' = y; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Y_NV) x' = -y; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Z_NV) x' = z; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Z_NV) x' = -z; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_W_NV) x' = w; if (x ==
VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_W_NV) x' = -w;

Similar selections are performed for the y, z, and w coordinates. This swizzling is applied before clipping and perspective divide. If the swizzle for an active viewport index is not specified, the swizzle for x is VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_X_NV, y is VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Y_NV, z is VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Z_NV and w is VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_W_NV.

Viewport swizzle parameters are specified by setting the pNext pointer of VkGraphicsPipelineCreateInfo to point to an instance of VkPipelineViewportSwizzleStateCreateInfoNV. VkPipelineViewportSwizzleStateCreateInfoNV uses VkViewportSwizzleNV to set the viewport swizzle parameters.

The VkViewportSwizzleNV structure is defined as:

typedef struct VkViewportSwizzleNV {
    VkViewportCoordinateSwizzleNV    x;
    VkViewportCoordinateSwizzleNV    y;
    VkViewportCoordinateSwizzleNV    z;
    VkViewportCoordinateSwizzleNV    w;
} VkViewportSwizzleNV;
Valid Usage (Implicit)

Possible values of the VkViewportSwizzleNV::x, y, z, and w members, specifying swizzling of the corresponding components of primitives, are:

typedef enum VkViewportCoordinateSwizzleNV {
    VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_X_NV = 0,
    VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_X_NV = 1,
    VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Y_NV = 2,
    VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Y_NV = 3,
    VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Z_NV = 4,
    VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Z_NV = 5,
    VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_W_NV = 6,
    VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_W_NV = 7,
} VkViewportCoordinateSwizzleNV;

These values are described in detail in Viewport Swizzle.

23.2. Flat Shading

Flat shading a vertex output attribute means to assign all vertices of the primitive the same value for that output.

The output values assigned are those of the provoking vertex of the primitive. The provoking vertex depends on the primitive topology, and is generally the “first” vertex of the primitive. For primitives not processed by tessellation or geometry shaders, the provoking vertex is selected from the input vertices according to the following table.

Table 31. Provoking vertex selection

Primitive type of primitive i

Provoking vertex number

VK_PRIMITIVE_TOPOLOGY_POINT_LIST

i

VK_PRIMITIVE_TOPOLOGY_LINE_LIST

2 i

VK_PRIMITIVE_TOPOLOGY_LINE_STRIP

i

VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST

3 i

VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP

i

VK_PRIMITIVE_TOPOLOGY_TRIANGLE_FAN

i + 1

VK_PRIMITIVE_TOPOLOGY_LINE_LIST_WITH_ADJACENCY

4 i + 1

VK_PRIMITIVE_TOPOLOGY_LINE_STRIP_WITH_ADJACENCY

i + 1

VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST_WITH_ADJACENCY

6 i

VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP_WITH_ADJACENCY

2 i

Caption

The Provoking vertex selection table defines the output values used for flat shading the ith primitive generated by drawing commands with the indicated primitive type, derived from the corresponding values of the vertex whose index is shown in the table. Primitives and vertices are numbered starting from zero.

Flat shading is applied to those vertex attributes that match fragment input attributes which are decorated as Flat.

If a geometry shader is active, the output primitive topology is either points, line strips, or triangle strips, and the selection of the provoking vertex behaves according to the corresponding row of the table. If a tessellation evaluation shader is active and a geometry shader is not active, the provoking vertex is undefined but must be one of the vertices of the primitive.

23.3. Primitive Clipping

Primitives are culled against the cull volume and then clipped to the clip volume. In clip coordinates, the view volume is defined by:

\[\begin{array}{c} -w_c \leq x_c \leq w_c \\ -w_c \leq y_c \leq w_c \\ 0 \leq z_c \leq w_c \end{array}\]

This view volume can be further restricted by as many as VkPhysicalDeviceLimits::maxClipDistances client-defined half-spaces.

The cull volume is the intersection of up to VkPhysicalDeviceLimits::maxCullDistances client-defined half-spaces (if no client-defined cull half-spaces are enabled, culling against the cull volume is skipped).

A shader must write a single cull distance for each enabled cull half-space to elements of the CullDistance array. If the cull distance for any enabled cull half-space is negative for all of the vertices of the primitive under consideration, the primitive is discarded. Otherwise the primitive is clipped against the clip volume as defined below.

The clip volume is the intersection of up to VkPhysicalDeviceLimits::maxClipDistances client-defined half-spaces with the view volume (if no client-defined clip half-spaces are enabled, the clip volume is the view volume).

A shader must write a single clip distance for each enabled clip half-space to elements of the ClipDistance array. Clip half-space i is then given by the set of points satisfying the inequality

ci(P) ≥ 0

where ci(P) is the clip distance i at point P. For point primitives, ci(P) is simply the clip distance for the vertex in question. For line and triangle primitives, per-vertex clip distances are interpolated using a weighted mean, with weights derived according to the algorithms described in sections Basic Line Segment Rasterization and Basic Polygon Rasterization, using the perspective interpolation equations.

The number of client-defined clip and cull half-spaces that are enabled is determined by the explicit size of the built-in arrays ClipDistance and CullDistance, respectively, declared as an output in the interface of the entry point of the final shader stage before clipping.

Depth clamping is enabled or disabled via the depthClampEnable enable of the VkPipelineRasterizationStateCreateInfo structure. If depth clamping is enabled, the plane equation

0 ≤ zc ≤ wc

(see the clip volume definition above) is ignored by view volume clipping (effectively, there is no near or far plane clipping).

If the primitive under consideration is a point or line segment, then clipping passes it unchanged if its vertices lie entirely within the clip volume.

Possible values of VkPhysicalDevicePointClippingProperties::pointClippingBehavior, specifying clipping behavior of a point primitive whose vertex lies outside the clip volume, are:

typedef enum VkPointClippingBehavior {
    VK_POINT_CLIPPING_BEHAVIOR_ALL_CLIP_PLANES = 0,
    VK_POINT_CLIPPING_BEHAVIOR_USER_CLIP_PLANES_ONLY = 1,
    VK_POINT_CLIPPING_BEHAVIOR_ALL_CLIP_PLANES_KHR = VK_POINT_CLIPPING_BEHAVIOR_ALL_CLIP_PLANES,
    VK_POINT_CLIPPING_BEHAVIOR_USER_CLIP_PLANES_ONLY_KHR = VK_POINT_CLIPPING_BEHAVIOR_USER_CLIP_PLANES_ONLY,
} VkPointClippingBehavior;

or the equivalent

typedef VkPointClippingBehavior VkPointClippingBehaviorKHR;
  • VK_POINT_CLIPPING_BEHAVIOR_ALL_CLIP_PLANES specifies that the primitive is discarded if the vertex lies outside any clip plane, including the planes bounding the view volume.

  • VK_POINT_CLIPPING_BEHAVIOR_USER_CLIP_PLANES_ONLY specifies that the primitive is discarded only if the vertex lies outside any user clip plane.

If either of a line segment’s vertices lie outside of the clip volume, the line segment may be clipped, with new vertex coordinates computed for each vertex that lies outside the clip volume. A clipped line segment endpoint lies on both the original line segment and the boundary of the clip volume.

This clipping produces a value, 0 ≤ t ≤ 1, for each clipped vertex. If the coordinates of a clipped vertex are P and the original vertices’ coordinates are P1 and P2, then t is given by

P = t P1 + (1-t) P2.

editing-note

This is weird - it gives P, not t.

t is used to clip vertex output attributes as described in Clipping Shader Outputs.

If the primitive is a polygon, it passes unchanged if every one of its edges lie entirely inside the clip volume, and it is discarded if every one of its edges lie entirely outside the clip volume. If the edges of the polygon intersect the boundary of the clip volume, the intersecting edges are reconnected by new edges that lie along the boundary of the clip volume - in some cases requiring the introduction of new vertices into a polygon.

If a polygon intersects an edge of the clip volume’s boundary, the clipped polygon must include a point on this boundary edge.

Primitives rendered with user-defined half-spaces must satisfy a complementarity criterion. Suppose a series of primitives is drawn where each vertex i has a single specified clip distance di (or a number of similarly specified clip distances, if multiple half-spaces are enabled). Next, suppose that the same series of primitives are drawn again with each such clip distance replaced by -di (and the graphics pipeline is otherwise the same). In this case, primitives must not be missing any pixels, and pixels must not be drawn twice in regions where those primitives are cut by the clip planes.

23.4. Clipping Shader Outputs

Next, vertex output attributes are clipped. The output values associated with a vertex that lies within the clip volume are unaffected by clipping. If a primitive is clipped, however, the output values assigned to vertices produced by clipping are clipped.

Let the output values assigned to the two vertices P1 and P2 of an unclipped edge be c1 and c2. The value of t (see Primitive Clipping) for a clipped point P is used to obtain the output value associated with P as

c = t c1 + (1-t) c2.

(Multiplying an output value by a scalar means multiplying each of x, y, z, and w by the scalar.)

Since this computation is performed in clip space before division by wc, clipped output values are perspective-correct.

Polygon clipping creates a clipped vertex along an edge of the clip volume’s boundary. This situation is handled by noting that polygon clipping proceeds by clipping against one half-space at a time. Output value clipping is done in the same way, so that clipped points always occur at the intersection of polygon edges (possibly already clipped) with the clip volume’s boundary.

For vertex output attributes whose matching fragment input attributes are decorated with NoPerspective, the value of t used to obtain the output value associated with P will be adjusted to produce results that vary linearly in framebuffer space.

Output attributes of integer or unsigned integer type must always be flat shaded. Flat shaded attributes are constant over the primitive being rasterized (see Basic Line Segment Rasterization and Basic Polygon Rasterization), and no interpolation is performed. The output value c is taken from either c1 or c2, since flat shading has already occurred and the two values are identical.

23.5. Controlling Viewport W Scaling

If viewport W scaling is enabled, the W component of the clip coordinate is modified by the provided coefficients from the corresponding viewport as follows.

wc' = xcoeff xc + ycoeff yc + wc

The VkPipelineViewportWScalingStateCreateInfoNV structure is defined as:

typedef struct VkPipelineViewportWScalingStateCreateInfoNV {
    VkStructureType                sType;
    const void*                    pNext;
    VkBool32                       viewportWScalingEnable;
    uint32_t                       viewportCount;
    const VkViewportWScalingNV*    pViewportWScalings;
} VkPipelineViewportWScalingStateCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • viewportWScalingEnable controls whether viewport W scaling is enabled.

  • viewportCount is the number of viewports used by W scaling, and must match the number of viewports in the pipeline if viewport W scaling is enabled.

  • pViewportWScalings is a pointer to an array of VkViewportWScalingNV structures, which define the W scaling parameters for the corresponding viewport. If the viewport W scaling state is dynamic, this member is ignored.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_W_SCALING_STATE_CREATE_INFO_NV

  • viewportCount must be greater than 0

The VkPipelineViewportWScalingStateCreateInfoNV state is set by adding an instance of this structure to the pNext chain of an instance of the VkPipelineViewportStateCreateInfo structure and setting the graphics pipeline state with vkCreateGraphicsPipelines.

If the bound pipeline state object was not created with the VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV dynamic state enabled, viewport W scaling parameters are specified using the pViewportWScalings member of VkPipelineViewportWScalingStateCreateInfoNV in the pipeline state object. If the pipeline state object was created with the VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV dynamic state enabled, the viewport transformation parameters are dynamically set and changed with the command:

void vkCmdSetViewportWScalingNV(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    firstViewport,
    uint32_t                                    viewportCount,
    const VkViewportWScalingNV*                 pViewportWScalings);
  • commandBuffer is the command buffer into which the command will be recorded.

  • firstViewport is the index of the first viewport whose parameters are updated by the command.

  • viewportCount is the number of viewports whose parameters are updated by the command.

  • pViewportWScalings is a pointer to an array of VkViewportWScalingNV structures specifying viewport parameters.

The viewport parameters taken from element i of pViewportWScalings replace the current state for the viewport index firstViewport + i, for i in [0, viewportCount).

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV dynamic state enabled

  • firstViewport must be less than VkPhysicalDeviceLimits::maxViewports

  • The sum of firstViewport and viewportCount must be between 1 and VkPhysicalDeviceLimits::maxViewports, inclusive

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pViewportWScalings must be a valid pointer to an array of viewportCount VkViewportWScalingNV structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • viewportCount must be greater than 0

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

Both VkPipelineViewportWScalingStateCreateInfoNV and vkCmdSetViewportWScalingNV use VkViewportWScalingNV to set the viewport transformation parameters.

The VkViewportWScalingNV structure is defined as:

typedef struct VkViewportWScalingNV {
    float    xcoeff;
    float    ycoeff;
} VkViewportWScalingNV;
  • xcoeff and ycoeff are the viewport’s W scaling factor for x and y respectively.

23.6. Coordinate Transformations

Clip coordinates for a vertex result from shader execution, which yields a vertex coordinate Position.

Perspective division on clip coordinates yields normalized device coordinates, followed by a viewport transformation (see Controlling the Viewport) to convert these coordinates into framebuffer coordinates.

If a vertex in clip coordinates has a position given by

\[\left(\begin{array}{c} x_c \\ y_c \\ z_c \\ w_c \end{array}\right)\]

then the vertex’s normalized device coordinates are

\[\left( \begin{array}{c} x_d \\ y_d \\ z_d \end{array} \right) = \left( \begin{array}{c} \frac{x_c}{w_c} \\ \frac{y_c}{w_c} \\ \frac{z_c}{w_c} \end{array} \right)\]

23.7. Controlling the Viewport

The viewport transformation is determined by the selected viewport’s width and height in pixels, px and py, respectively, and its center (ox, oy) (also in pixels), as well as its depth range min and max determining a depth range scale value pz and a depth range bias value oz (defined below). The vertex’s framebuffer coordinates (xf, yf, zf) are given by

xf = (px / 2) xd + ox

yf = (py / 2) yd + oy

zf = pz × zd + oz

Multiple viewports are available, numbered zero up to VkPhysicalDeviceLimits::maxViewports minus one. The number of viewports used by a pipeline is controlled by the viewportCount member of the VkPipelineViewportStateCreateInfo structure used in pipeline creation.

The VkPipelineViewportStateCreateInfo structure is defined as:

typedef struct VkPipelineViewportStateCreateInfo {
    VkStructureType                       sType;
    const void*                           pNext;
    VkPipelineViewportStateCreateFlags    flags;
    uint32_t                              viewportCount;
    const VkViewport*                     pViewports;
    uint32_t                              scissorCount;
    const VkRect2D*                       pScissors;
} VkPipelineViewportStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • viewportCount is the number of viewports used by the pipeline.

  • pViewports is a pointer to an array of VkViewport structures, defining the viewport transforms. If the viewport state is dynamic, this member is ignored.

  • scissorCount is the number of scissors and must match the number of viewports.

  • pScissors is a pointer to an array of VkRect2D structures which define the rectangular bounds of the scissor for the corresponding viewport. If the scissor state is dynamic, this member is ignored.

Valid Usage
  • If the multiple viewports feature is not enabled, viewportCount must be 1

  • If the multiple viewports feature is not enabled, scissorCount must be 1

  • viewportCount must be between 1 and VkPhysicalDeviceLimits::maxViewports, inclusive

  • scissorCount must be between 1 and VkPhysicalDeviceLimits::maxViewports, inclusive

  • scissorCount and viewportCount must be identical

  • If the viewportWScalingEnable member of a VkPipelineViewportWScalingStateCreateInfoNV structure chained to the pNext chain is VK_TRUE, the viewportCount member of the VkPipelineViewportWScalingStateCreateInfoNV structure must be equal to viewportCount

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkPipelineViewportSwizzleStateCreateInfoNV or VkPipelineViewportWScalingStateCreateInfoNV

  • Each sType member in the pNext chain must be unique

  • flags must be 0

  • viewportCount must be greater than 0

  • scissorCount must be greater than 0

typedef VkFlags VkPipelineViewportStateCreateFlags;

VkPipelineViewportStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

A vertex processing stage may direct each primitive to zero or more viewports. The destination viewports for a primitive are selected by the last active vertex processing stage that has an output variable decorated with ViewportIndex (selecting a single viewport) or ViewportMaskNV (selecting multiple viewports). The viewport transform uses the viewport corresponding to either the value assigned to ViewportIndex or one of the bits set in ViewportMaskNV, and taken from an implementation-dependent vertex of each primitive. If ViewportIndex or any of the bits in ViewportMaskNV are outside the range zero to viewportCount minus one for a primitive, or if the last active vertex processing stage did not assign a value to either ViewportIndex or ViewportMaskNV for all vertices of a primitive due to flow control, the results of the viewport transformation of the vertices of such primitives are undefined. If the last vertex processing stage does not have an output decorated with ViewportIndex or ViewportMaskNV, the viewport numbered zero is used by the viewport transformation.

A single vertex can be used in more than one individual primitive, in primitives such as VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP. In this case, the viewport transformation is applied separately for each primitive.

If the bound pipeline state object was not created with the VK_DYNAMIC_STATE_VIEWPORT dynamic state enabled, viewport transformation parameters are specified using the pViewports member of VkPipelineViewportStateCreateInfo in the pipeline state object. If the pipeline state object was created with the VK_DYNAMIC_STATE_VIEWPORT dynamic state enabled, the viewport transformation parameters are dynamically set and changed with the command:

void vkCmdSetViewport(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    firstViewport,
    uint32_t                                    viewportCount,
    const VkViewport*                           pViewports);
  • commandBuffer is the command buffer into which the command will be recorded.

  • firstViewport is the index of the first viewport whose parameters are updated by the command.

  • viewportCount is the number of viewports whose parameters are updated by the command.

  • pViewports is a pointer to an array of VkViewport structures specifying viewport parameters.

The viewport parameters taken from element i of pViewports replace the current state for the viewport index firstViewport + i, for i in [0, viewportCount).

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_VIEWPORT dynamic state enabled

  • firstViewport must be less than VkPhysicalDeviceLimits::maxViewports

  • The sum of firstViewport and viewportCount must be between 1 and VkPhysicalDeviceLimits::maxViewports, inclusive

  • If the multiple viewports feature is not enabled, firstViewport must be 0

  • If the multiple viewports feature is not enabled, viewportCount must be 1

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pViewports must be a valid pointer to an array of viewportCount VkViewport structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • viewportCount must be greater than 0

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

Both VkPipelineViewportStateCreateInfo and vkCmdSetViewport use VkViewport to set the viewport transformation parameters.

The VkViewport structure is defined as:

typedef struct VkViewport {
    float    x;
    float    y;
    float    width;
    float    height;
    float    minDepth;
    float    maxDepth;
} VkViewport;
  • x and y are the viewport’s upper left corner (x,y).

  • width and height are the viewport’s width and height, respectively.

  • minDepth and maxDepth are the depth range for the viewport. It is valid for minDepth to be greater than or equal to maxDepth.

The framebuffer depth coordinate zf may be represented using either a fixed-point or floating-point representation. However, a floating-point representation must be used if the depth/stencil attachment has a floating-point depth component. If an m-bit fixed-point representation is used, we assume that it represents each value \(\frac{k}{2^m - 1}\), where k ∈ { 0, 1, …​, 2m-1 }, as k (e.g. 1.0 is represented in binary as a string of all ones).

The viewport parameters shown in the above equations are found from these values as

ox = x + width / 2

oy = y + height / 2

oz = minDepth

px = width

py = height

pz = maxDepth - minDepth.

The application can specify a negative term for height, which has the effect of negating the y coordinate in clip space before performing the transform. When using a negative height, the application should also adjust the y value to point to the lower left corner of the viewport instead of the upper left corner. Using the negative height allows the application to avoid having to negate the y component of the Position output from the last vertex processing stage in shaders that also target other graphics APIs.

The width and height of the implementation-dependent maximum viewport dimensions must be greater than or equal to the width and height of the largest image which can be created and attached to a framebuffer.

The floating-point viewport bounds are represented with an implementation-dependent precision.

Valid Usage
  • width must be greater than 0.0

  • width must be less than or equal to VkPhysicalDeviceLimits::maxViewportDimensions[0]

  • The absolute value of height must be less than or equal to VkPhysicalDeviceLimits::maxViewportDimensions[1]

  • x must be greater than or equal to viewportBoundsRange[0]

  • (x + width) must be less than or equal to viewportBoundsRange[1]

  • y must be greater than or equal to viewportBoundsRange[0]

  • y must be less than or equal to viewportBoundsRange[1]

  • (y + height) must be greater than or equal to viewportBoundsRange[0]

  • (y + height) must be less than or equal to viewportBoundsRange[1]

  • Unless VK_EXT_depth_range_unrestricted extension is enabled minDepth must be between 0.0 and 1.0, inclusive

  • Unless VK_EXT_depth_range_unrestricted extension is enabled maxDepth must be between 0.0 and 1.0, inclusive

24. Rasterization

Rasterization is the process by which a primitive is converted to a two-dimensional image. Each point of this image contains associated data such as depth, color, or other attributes.

Rasterizing a primitive begins by determining which squares of an integer grid in framebuffer coordinates are occupied by the primitive, and assigning one or more depth values to each such square. This process is described below for points, lines, and polygons.

A grid square, including its (x,y) framebuffer coordinates, z (depth), and associated data added by fragment shaders, is called a fragment. A fragment is located by its upper left corner, which lies on integer grid coordinates.

Rasterization operations also refer to a fragment’s sample locations, which are offset by subpixel fractional values from its upper left corner. The rasterization rules for points, lines, and triangles involve testing whether each sample location is inside the primitive. Fragments need not actually be square, and rasterization rules are not affected by the aspect ratio of fragments. Display of non-square grids, however, will cause rasterized points and line segments to appear fatter in one direction than the other.

We assume that fragments are square, since it simplifies antialiasing and texturing. After rasterization, fragments are processed by the early per-fragment tests, if enabled.

Several factors affect rasterization, including the members of VkPipelineRasterizationStateCreateInfo and VkPipelineMultisampleStateCreateInfo.

The VkPipelineRasterizationStateCreateInfo structure is defined as:

typedef struct VkPipelineRasterizationStateCreateInfo {
    VkStructureType                            sType;
    const void*                                pNext;
    VkPipelineRasterizationStateCreateFlags    flags;
    VkBool32                                   depthClampEnable;
    VkBool32                                   rasterizerDiscardEnable;
    VkPolygonMode                              polygonMode;
    VkCullModeFlags                            cullMode;
    VkFrontFace                                frontFace;
    VkBool32                                   depthBiasEnable;
    float                                      depthBiasConstantFactor;
    float                                      depthBiasClamp;
    float                                      depthBiasSlopeFactor;
    float                                      lineWidth;
} VkPipelineRasterizationStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • depthClampEnable controls whether to clamp the fragment’s depth values instead of clipping primitives to the z planes of the frustum, as described in Primitive Clipping.

  • rasterizerDiscardEnable controls whether primitives are discarded immediately before the rasterization stage.

  • polygonMode is the triangle rendering mode. See VkPolygonMode.

  • cullMode is the triangle facing direction used for primitive culling. See VkCullModeFlagBits.

  • frontFace is a VkFrontFace value specifying the front-facing triangle orientation to be used for culling.

  • depthBiasEnable controls whether to bias fragment depth values.

  • depthBiasConstantFactor is a scalar factor controlling the constant depth value added to each fragment.

  • depthBiasClamp is the maximum (or minimum) depth bias of a fragment.

  • depthBiasSlopeFactor is a scalar factor applied to a fragment’s slope in depth bias calculations.

  • lineWidth is the width of rasterized line segments.

The application can also add a VkPipelineRasterizationStateRasterizationOrderAMD structure to the pNext chain of a VkPipelineRasterizationStateCreateInfo structure. This structure enables selecting the rasterization order to use when rendering with the corresponding graphics pipeline as described in Rasterization Order.

Valid Usage
  • If the depth clamping feature is not enabled, depthClampEnable must be VK_FALSE

  • If the non-solid fill modes feature is not enabled, polygonMode must be VK_POLYGON_MODE_FILL or VK_POLYGON_MODE_FILL_RECTANGLE_NV

  • If the VK_NV_fill_rectangle extension is not enabled, polygonMode must not be VK_POLYGON_MODE_FILL_RECTANGLE_NV

Valid Usage (Implicit)
typedef VkFlags VkPipelineRasterizationStateCreateFlags;

VkPipelineRasterizationStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPipelineMultisampleStateCreateInfo structure is defined as:

typedef struct VkPipelineMultisampleStateCreateInfo {
    VkStructureType                          sType;
    const void*                              pNext;
    VkPipelineMultisampleStateCreateFlags    flags;
    VkSampleCountFlagBits                    rasterizationSamples;
    VkBool32                                 sampleShadingEnable;
    float                                    minSampleShading;
    const VkSampleMask*                      pSampleMask;
    VkBool32                                 alphaToCoverageEnable;
    VkBool32                                 alphaToOneEnable;
} VkPipelineMultisampleStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • rasterizationSamples is a VkSampleCountFlagBits specifying the number of samples per pixel used in rasterization.

  • sampleShadingEnable can be used to enable Sample Shading.

  • minSampleShading specifies a minimum fraction of sample shading if sampleShadingEnable is set to VK_TRUE.

  • pSampleMask is a bitmask of static coverage information that is ANDed with the coverage information generated during rasterization, as described in Sample Mask.

  • alphaToCoverageEnable controls whether a temporary coverage value is generated based on the alpha component of the fragment’s first color output as specified in the Multisample Coverage section.

  • alphaToOneEnable controls whether the alpha component of the fragment’s first color output is replaced with one as described in Multisample Coverage.

Valid Usage
  • If the sample rate shading feature is not enabled, sampleShadingEnable must be VK_FALSE

  • If the alpha to one feature is not enabled, alphaToOneEnable must be VK_FALSE

  • minSampleShading must be in the range [0,1]

  • If the subpass has any color attachments and rasterizationSamples is greater than the number of color samples, then sampleShadingEnable must be VK_FALSE

Valid Usage (Implicit)
typedef VkFlags VkPipelineMultisampleStateCreateFlags;

VkPipelineMultisampleStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

Rasterization only produces fragments corresponding to pixels in the framebuffer. Fragments which would be produced by application of any of the primitive rasterization rules described below but which lie outside the framebuffer are not produced, nor are they processed by any later stage of the pipeline, including any of the early per-fragment tests described in Early Per-Fragment Tests.

Surviving fragments are processed by fragment shaders. Fragment shaders determine associated data for fragments, and can also modify or replace their assigned depth values.

If the subpass for which this pipeline is being created uses color and/or depth/stencil attachments, then rasterizationSamples must be the same as the maximum of the sample counts of those subpass attachments.

If the subpass for which this pipeline is being created does not use color or depth/stencil attachments, rasterizationSamples must follow the rules for a zero-attachment subpass.

24.1. Discarding Primitives Before Rasterization

Primitives are discarded before rasterization if the rasterizerDiscardEnable member of VkPipelineRasterizationStateCreateInfo is enabled. When enabled, primitives are discarded after they are processed by the last active shader stage in the pipeline before rasterization.

24.2. Rasterization Order

Within a subpass of a render pass instance, for a given (x,y,layer,sample) sample location, the following operations are guaranteed to execute in rasterization order, for each separate primitive that includes that sample location:

Each of these operations is atomically executed for each primitive and sample location.

Execution of these operations for each primitive in a subpass occurs in an order determined by the application.

The rasterization order to use for a graphics pipeline is specified by adding a VkPipelineRasterizationStateRasterizationOrderAMD structure to the pNext chain of a VkPipelineRasterizationStateCreateInfo structure.

The VkPipelineRasterizationStateRasterizationOrderAMD structure is defined as:

typedef struct VkPipelineRasterizationStateRasterizationOrderAMD {
    VkStructureType            sType;
    const void*                pNext;
    VkRasterizationOrderAMD    rasterizationOrder;
} VkPipelineRasterizationStateRasterizationOrderAMD;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • rasterizationOrder is a VkRasterizationOrderAMD value specifying the primitive rasterization order to use.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_RASTERIZATION_ORDER_AMD

  • rasterizationOrder must be a valid VkRasterizationOrderAMD value

If the VK_AMD_rasterization_order device extension is not enabled or the application does not request a particular rasterization order through specifying a VkPipelineRasterizationStateRasterizationOrderAMD structure then the rasterization order used by the graphics pipeline defaults to VK_RASTERIZATION_ORDER_STRICT_AMD.

Possible values of VkPipelineRasterizationStateRasterizationOrderAMD::rasterizationOrder, specifying the primitive rasterization order, are:

typedef enum VkRasterizationOrderAMD {
    VK_RASTERIZATION_ORDER_STRICT_AMD = 0,
    VK_RASTERIZATION_ORDER_RELAXED_AMD = 1,
} VkRasterizationOrderAMD;
  • VK_RASTERIZATION_ORDER_STRICT_AMD specifies that operations for each primitive in a subpass must occur in primitive order.

  • VK_RASTERIZATION_ORDER_RELAXED_AMD specifies that operations for each primitive in a subpass may not occur in primitive order.

24.3. Multisampling

Multisampling is a mechanism to antialias all Vulkan primitives: points, lines, and polygons. The technique is to sample all primitives multiple times at each pixel. Each sample in each framebuffer attachment has storage for a color, depth, and/or stencil value, such that per-fragment operations apply to each sample independently. The color sample values can be later resolved to a single color (see Resolving Multisample Images and the Render Pass chapter for more details on how to resolve multisample images to non-multisample images).

Vulkan defines rasterization rules for single-sample modes in a way that is equivalent to a multisample mode with a single sample in the center of each pixel.

Each fragment includes a coverage value with rasterizationSamples bits (see Sample Mask). Each fragment includes rasterizationSamples depth values and sets of associated data. An implementation may choose to assign the same associated data to more than one sample. The location for evaluating such associated data may be anywhere within the pixel including the pixel center or any of the sample locations. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the pixel center must be used. The different associated data values need not all be evaluated at the same location. Each pixel fragment thus consists of integer x and y grid coordinates, rasterizationSamples depth values and sets of associated data, and a coverage value with rasterizationSamples bits.

It is understood that each pixel has rasterizationSamples locations associated with it. These locations are exact positions, rather than regions or areas, and each is referred to as a sample point. The sample points associated with a pixel must be located inside or on the boundary of the unit square that is considered to bound the pixel. Furthermore, the relative locations of sample points may be identical for each pixel in the framebuffer, or they may differ. If the current pipeline includes a fragment shader with one or more variables in its interface decorated with Sample and Input, the data associated with those variables will be assigned independently for each sample. The values for each sample must be evaluated at the location of the sample. The data associated with any other variables not decorated with Sample and Input need not be evaluated independently for each sample.

If the standardSampleLocations member of VkPhysicalDeviceLimits is VK_TRUE, then the sample counts VK_SAMPLE_COUNT_1_BIT, VK_SAMPLE_COUNT_2_BIT, VK_SAMPLE_COUNT_4_BIT, VK_SAMPLE_COUNT_8_BIT, and VK_SAMPLE_COUNT_16_BIT have sample locations as listed in the following table, with the ith entry in the table corresponding to bit i in the sample masks. VK_SAMPLE_COUNT_32_BIT and VK_SAMPLE_COUNT_64_BIT do not have standard sample locations. Locations are defined relative to an origin in the upper left corner of the pixel.

Table 32. Standard sample locations

VK_SAMPLE_COUNT_1_BIT

VK_SAMPLE_COUNT_2_BIT

VK_SAMPLE_COUNT_4_BIT

VK_SAMPLE_COUNT_8_BIT

VK_SAMPLE_COUNT_16_BIT

(0.5,0.5)

(0.25,0.25)
(0.75,0.75)

(0.375, 0.125)
(0.875, 0.375)
(0.125, 0.625)
(0.625, 0.875)

(0.5625, 0.3125)
(0.4375, 0.6875)
(0.8125, 0.5625)
(0.3125, 0.1875)
(0.1875, 0.8125)
(0.0625, 0.4375)
(0.6875, 0.9375)
(0.9375, 0.0625)

(0.5625, 0.5625)
(0.4375, 0.3125)
(0.3125, 0.625)
(0.75, 0.4375)
(0.1875, 0.375)
(0.625, 0.8125)
(0.8125, 0.6875)
(0.6875, 0.1875)
(0.375, 0.875)
(0.5, 0.0625)
(0.25, 0.125)
(0.125, 0.75)
(0.0, 0.5)
(0.9375, 0.25)
(0.875, 0.9375)
(0.0625, 0.0)

Color images created with multiple samples per pixel use a compression technique where there are two arrays of data associated with each pixel. The first array contains one element per sample where each element stores an index to the second array defining the fragment mask of the pixel. The second array contains one element per color fragment and each element stores a unique color value in the format of the image. With this compression technique it’s not always necessary to actually use unique storage locations for each color sample: when multiple samples share the same color value the fragment mask may have two samples referring to the same color fragment. The number of color fragments is determined by the samples member of the VkImageCreateInfo structure used to create the image. The VK_AMD_shader_fragment_mask device extension provides shader instructions enabling the application to get direct access to the fragment mask and the individual color fragment values.

fragment mask
Figure 26. Fragment Mask

24.4. Custom Sample Locations

Applications can also control the sample locations used for rasterization.

If the pNext chain of the VkPipelineMultisampleStateCreateInfo structure specified at pipeline creation time includes an instance of the VkPipelineSampleLocationsStateCreateInfoEXT structure, then that structure controls the sample locations used when rasterizing primitives with the pipeline.

The VkPipelineSampleLocationsStateCreateInfoEXT structure is defined as:

typedef struct VkPipelineSampleLocationsStateCreateInfoEXT {
    VkStructureType             sType;
    const void*                 pNext;
    VkBool32                    sampleLocationsEnable;
    VkSampleLocationsInfoEXT    sampleLocationsInfo;
} VkPipelineSampleLocationsStateCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • sampleLocationsEnable controls whether custom sample locations are used. If sampleLocationsEnable is VK_FALSE, the default sample locations are used and the values specified in sampleLocationsInfo are ignored.

  • sampleLocationsInfo is the sample locations to use during rasterization if sampleLocationsEnable is VK_TRUE and the graphics pipeline isn’t created with VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_SAMPLE_LOCATIONS_STATE_CREATE_INFO_EXT

  • sampleLocationsInfo must be a valid VkSampleLocationsInfoEXT structure

The VkSampleLocationsInfoEXT structure is defined as:

typedef struct VkSampleLocationsInfoEXT {
    VkStructureType               sType;
    const void*                   pNext;
    VkSampleCountFlagBits         sampleLocationsPerPixel;
    VkExtent2D                    sampleLocationGridSize;
    uint32_t                      sampleLocationsCount;
    const VkSampleLocationEXT*    pSampleLocations;
} VkSampleLocationsInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • sampleLocationsPerPixel is a VkSampleCountFlagBits specifying the number of sample locations per pixel.

  • sampleLocationGridSize is the size of the sample location grid to select custom sample locations for.

  • sampleLocationsCount is the number of sample locations in pSampleLocations.

  • pSampleLocations is an array of sampleLocationsCount VkSampleLocationEXT structures.

This structure can be used either to specify the sample locations to be used for rendering or to specify the set of sample locations an image subresource has been last rendered with for the purposes of layout transitions of depth/stencil images created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT.

The sample locations in pSampleLocations specify sampleLocationsPerPixel number of sample locations for each pixel in the grid of the size specified in sampleLocationGridSize. The sample location for sample i at the pixel grid location (x,y) is taken from pSampleLocations[(x + y * sampleLocationGridSize.width) * sampleLocationsPerPixel + i].

Valid Usage
  • sampleLocationsPerPixel must be a bit value that is set in VkPhysicalDeviceSampleLocationsPropertiesEXT::sampleLocationSampleCounts

  • sampleLocationsCount must equal sampleLocationsPerPixel × sampleLocationGridSize.width × sampleLocationGridSize.height

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SAMPLE_LOCATIONS_INFO_EXT

  • sampleLocationsPerPixel must be a valid VkSampleCountFlagBits value

  • pSampleLocations must be a valid pointer to an array of sampleLocationsCount VkSampleLocationEXT structures

  • sampleLocationsCount must be greater than 0

The VkSampleLocationEXT structure is defined as:

typedef struct VkSampleLocationEXT {
    float    x;
    float    y;
} VkSampleLocationEXT;
  • x is the horizontal coordinate of the sample’s location.

  • y is the vertical coordinate of the sample’s location.

The domain space of the sample location coordinates has an upper-left origin within the pixel in framebuffer space.

The values specified in a VkSampleLocationEXT structure are always clamped to the implementation-dependent sample location coordinate range [sampleLocationCoordinateRange[0],sampleLocationCoordinateRange[1]] that can be queried by chaining the VkPhysicalDeviceSampleLocationsPropertiesEXT structure to the pNext chain of VkPhysicalDeviceProperties2.

The custom sample locations used for rasterization when VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsEnable is VK_TRUE are specified by the VkPipelineSampleLocationsStateCreateInfoEXT::sampleLocationsInfo property of the bound graphics pipeline, if the pipeline was not created with VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT enabled.

Otherwise, the sample locations used for rasterization are set by calling vkCmdSetSampleLocationsEXT:

void vkCmdSetSampleLocationsEXT(
    VkCommandBuffer                             commandBuffer,
    const VkSampleLocationsInfoEXT*             pSampleLocationsInfo);
  • commandBuffer is the command buffer into which the command will be recorded.

  • pSampleLocationsInfo is the sample locations state to set.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT dynamic state enabled

  • The sampleLocationsPerPixel member of pSampleLocationsInfo must equal the rasterizationSamples member of the VkPipelineMultisampleStateCreateInfo structure the bound graphics pipeline has been created with

  • If VkPhysicalDeviceSampleLocationsPropertiesEXT::variableSampleLocations is VK_FALSE then the current render pass must have been begun by specifying a VkRenderPassSampleLocationsBeginInfoEXT structure whose pPostSubpassSampleLocations member contains an element with a subpassIndex matching the current subpass index and the sampleLocationsInfo member of that element must match the sample locations state pointed to by pSampleLocationsInfo

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pSampleLocationsInfo must be a valid pointer to a valid VkSampleLocationsInfoEXT structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

24.5. Sample Shading

Sample shading can be used to specify a minimum number of unique samples to process for each fragment. If sample shading is enabled an implementation must provide a minimum of max(⌈ minSampleShadingFactor × totalSamples ⌉, 1) unique associated data for each fragment, where minSampleShadingFactor is the minimum fraction of sample shading and totalSamples is the number of samples of the color attachments used in the subpass or, if the subpass does not use any color attachments, the value of VkPipelineMultisampleStateCreateInfo::pname.rasterizationSamples specified at pipeline creation time. These are associated with the samples in an implementation-dependent manner. When minSampleShadingFactor is 1.0, a separate set of associated data are evaluated for each sample, and each set of values is evaluated at the sample location.

Sample shading is enabled for a graphics pipeline:

  • If the interface of the fragment shader entry point of the graphics pipeline includes an input variable decorated with SampleId or SamplePosition. In this case minSampleShadingFactor takes the value 1.0.

  • Else if the sampleShadingEnable member of the VkPipelineMultisampleStateCreateInfo structure specified when creating the graphics pipeline is set to VK_TRUE. In this case minSampleShadingFactor takes the value of VkPipelineMultisampleStateCreateInfo::minSampleShading.

Otherwise, sample shading is considered disabled.

24.6. Points

A point is drawn by generating a set of fragments in the shape of a square centered around the vertex of the point. Each vertex has an associated point size that controls the width/height of that square. The point size is taken from the (potentially clipped) shader built-in PointSize written by:

  • the geometry shader, if active;

  • the tessellation evaluation shader, if active and no geometry shader is active;

  • the vertex shader, otherwise

and clamped to the implementation-dependent point size range [pointSizeRange[0],pointSizeRange[1]]. If the value written to PointSize is less than or equal to zero, or if no value was written to PointSize, results are undefined.

Not all point sizes need be supported, but the size 1.0 must be supported. The range of supported sizes and the size of evenly-spaced gradations within that range are implementation-dependent. The range and gradations are obtained from the pointSizeRange and pointSizeGranularity members of VkPhysicalDeviceLimits. If, for instance, the size range is from 0.1 to 2.0 and the gradation size is 0.1, then the size 0.1, 0.2, …​, 1.9, 2.0 are supported. Additional point sizes may also be supported. There is no requirement that these sizes be equally spaced. If an unsupported size is requested, the nearest supported size is used instead.

24.6.1. Basic Point Rasterization

Point rasterization produces a fragment for each framebuffer pixel with one or more sample points that intersect a region centered at the point’s (xf,yf). This region is a square with side equal to the current point size. Coverage bits that correspond to sample points that intersect the region are 1, other coverage bits are 0.

All fragments produced in rasterizing a point are assigned the same associated data, which are those of the vertex corresponding to the point. However, the fragment shader built-in PointCoord contains point sprite texture coordinates. The s and t point sprite texture coordinates vary from zero to one across the point horizontally left-to-right and top-to-bottom, respectively. The following formulas are used to evaluate s and t:

\[s = {1 \over 2} + { \left( x_p - x_f \right) \over \text{size} }\]
\[t = {1 \over 2} + { \left( y_p - y_f \right) \over \text{size} }\]

where size is the point’s size, (xp,yp) is the location at which the point sprite coordinates are evaluated - this may be the framebuffer coordinates of the pixel center (i.e. at the half-integer) or the location of a sample, and (xf,yf) is the exact, unrounded framebuffer coordinate of the vertex for the point. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the pixel center must be used.

24.7. Line Segments

A line is drawn by generating a set of fragments overlapping a rectangle centered on the line segment. Each line segment has an associated width that controls the width of that rectangle.

The line width is specified by the VkPipelineRasterizationStateCreateInfo::lineWidth property of the currently active pipeline, if the pipeline was not created with VK_DYNAMIC_STATE_LINE_WIDTH enabled.

Otherwise, the line width is set by calling vkCmdSetLineWidth:

void vkCmdSetLineWidth(
    VkCommandBuffer                             commandBuffer,
    float                                       lineWidth);
  • commandBuffer is the command buffer into which the command will be recorded.

  • lineWidth is the width of rasterized line segments.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_LINE_WIDTH dynamic state enabled

  • If the wide lines feature is not enabled, lineWidth must be 1.0

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

Not all line widths need be supported for line segment rasterization, but width 1.0 antialiased segments must be provided. The range and gradations are obtained from the lineWidthRange and lineWidthGranularity members of VkPhysicalDeviceLimits. If, for instance, the size range is from 0.1 to 2.0 and the gradation size is 0.1, then the size 0.1, 0.2, …​, 1.9, 2.0 are supported. Additional line widths may also be supported. There is no requirement that these widths be equally spaced. If an unsupported width is requested, the nearest supported width is used instead.

24.7.1. Basic Line Segment Rasterization

Rasterized line segments produce fragments which intersect a rectangle centered on the line segment. Two of the edges are parallel to the specified line segment; each is at a distance of one-half the current width from that segment in directions perpendicular to the direction of the line. The other two edges pass through the line endpoints and are perpendicular to the direction of the specified line segment. Coverage bits that correspond to sample points that intersect the rectangle are 1, other coverage bits are 0.

Next we specify how the data associated with each rasterized fragment are obtained. Let pr = (xd, yd) be the framebuffer coordinates at which associated data are evaluated. This may be the pixel center of a fragment or the location of a sample within the fragment. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the pixel center must be used. Let pa = (xa, ya) and pb = (xb,yb) be initial and final endpoints of the line segment, respectively. Set

\[t = {{( \mathbf{p}_r - \mathbf{p}_a ) \cdot ( \mathbf{p}_b - \mathbf{p}_a )} \over {\| \mathbf{p}_b - \mathbf{p}_a \|^2 }}\]

(Note that t = 0 at p_a and t = 1 at pb. Also note that this calculation projects the vector from pa to pr onto the line, and thus computes the normalized distance of the fragment along the line.)

The value of an associated datum f for the fragment, whether it be a shader output or the clip w coordinate, must be determined using perspective interpolation:

\[f = {{ (1-t) {f_a / w_a} + t { f_b / w_b} } \over {(1-t) / w_a + t / w_b }}\]

where fa and fb are the data associated with the starting and ending endpoints of the segment, respectively; wa and wb are the clip w coordinates of the starting and ending endpoints of the segments, respectively.

Depth values for lines must be determined using linear interpolation:

z = (1 - t) za + t zb

where za and zb are the depth values of the starting and ending endpoints of the segment, respectively.

The NoPerspective and Flat interpolation decorations can be used with fragment shader inputs to declare how they are interpolated. When neither decoration is applied, perspective interpolation is performed as described above. When the NoPerspective decoration is used, linear interpolation is performed in the same fashion as for depth values, as described above. When the Flat decoration is used, no interpolation is performed, and outputs are taken from the corresponding input value of the provoking vertex corresponding to that primitive.

The above description documents the preferred method of line rasterization, and must be used when the implementation advertises the strictLines limit in VkPhysicalDeviceLimits as VK_TRUE.

When strictLines is VK_FALSE, the edges of the lines are generated as a parallelogram surrounding the original line. The major axis is chosen by noting the axis in which there is the greatest distance between the line start and end points. If the difference is equal in both directions then the X axis is chosen as the major axis. Edges 2 and 3 are aligned to the minor axis and are centered on the endpoints of the line as in Non strict lines, and each is lineWidth long. Edges 0 and 1 are parallel to the line and connect the endpoints of edges 2 and 3. Coverage bits that correspond to sample points that intersect the parallelogram are 1, other coverage bits are 0.

Samples that fall exactly on the edge of the parallelogram follow the polygon rasterization rules.

Interpolation occurs as if the parallelogram was decomposed into two triangles where each pair of vertices at each end of the line has identical attributes.

non strict lines
Figure 27. Non strict lines

24.8. Polygons

A polygon results from the decomposition of a triangle strip, triangle fan or a series of independent triangles. Like points and line segments, polygon rasterization is controlled by several variables in the VkPipelineRasterizationStateCreateInfo structure.

24.8.1. Basic Polygon Rasterization

The first step of polygon rasterization is to determine whether the triangle is back-facing or front-facing. This determination is made based on the sign of the (clipped or unclipped) polygon’s area computed in framebuffer coordinates. One way to compute this area is:

\[a = -{1 \over 2}\sum_{i=0}^{n-1} x_f^i y_f^{i \oplus 1} - x_f^{i \oplus 1} y_f^i\]

where \(x_f^i\) and \(y_f^i\) are the x and y framebuffer coordinates of the ith vertex of the n-vertex polygon (vertices are numbered starting at zero for the purposes of this computation) and i ⊕ 1 is (i + 1) mod n.

The interpretation of the sign of a is determined by the VkPipelineRasterizationStateCreateInfo::frontFace property of the currently active pipeline. Possible values are:

typedef enum VkFrontFace {
    VK_FRONT_FACE_COUNTER_CLOCKWISE = 0,
    VK_FRONT_FACE_CLOCKWISE = 1,
} VkFrontFace;
  • VK_FRONT_FACE_COUNTER_CLOCKWISE specifies that a triangle with positive area is considered front-facing.

  • VK_FRONT_FACE_CLOCKWISE specifies that a triangle with negative area is considered front-facing.

Any triangle which is not front-facing is back-facing, including zero-area triangles.

Once the orientation of triangles is determined, they are culled according to the VkPipelineRasterizationStateCreateInfo::cullMode property of the currently active pipeline. Possible values are:

typedef enum VkCullModeFlagBits {
    VK_CULL_MODE_NONE = 0,
    VK_CULL_MODE_FRONT_BIT = 0x00000001,
    VK_CULL_MODE_BACK_BIT = 0x00000002,
    VK_CULL_MODE_FRONT_AND_BACK = 0x00000003,
} VkCullModeFlagBits;
  • VK_CULL_MODE_NONE specifies that no triangles are discarded

  • VK_CULL_MODE_FRONT_BIT specifies that front-facing triangles are discarded

  • VK_CULL_MODE_BACK_BIT specifies that back-facing triangles are discarded

  • VK_CULL_MODE_FRONT_AND_BACK specifies that all triangles are discarded.

Following culling, fragments are produced for any triangles which have not been discarded.

typedef VkFlags VkCullModeFlags;

VkCullModeFlags is a bitmask type for setting a mask of zero or more VkCullModeFlagBits.

The rule for determining which fragments are produced by polygon rasterization is called point sampling. The two-dimensional projection obtained by taking the x and y framebuffer coordinates of the polygon’s vertices is formed. Fragments are produced for any pixels for which any sample points lie inside of this polygon. Coverage bits that correspond to sample points that satisfy the point sampling criteria are 1, other coverage bits are 0. Special treatment is given to a sample whose sample location lies on a polygon edge. In such a case, if two polygons lie on either side of a common edge (with identical endpoints) on which a sample point lies, then exactly one of the polygons must result in a covered sample for that fragment during rasterization. As for the data associated with each fragment produced by rasterizing a polygon, we begin by specifying how these values are produced for fragments in a triangle. Define barycentric coordinates for a triangle. Barycentric coordinates are a set of three numbers, a, b, and c, each in the range [0,1], with a + b + c = 1. These coordinates uniquely specify any point p within the triangle or on the triangle’s boundary as

p = a pa + b pb + c pc

where pa, pb, and pc are the vertices of the triangle. a, b, and c are determined by:

\[a = {{\mathrm{A}(p p_b p_c)} \over {\mathrm{A}(p_a p_b p_c)}}, \quad b = {{\mathrm{A}(p p_a p_c)} \over {\mathrm{A}(p_a p_b p_c)}}, \quad c = {{\mathrm{A}(p p_a p_b)} \over {\mathrm{A}(p_a p_b p_c)}},\]

where A(lmn) denotes the area in framebuffer coordinates of the triangle with vertices l, m, and n.

Denote an associated datum at pa, pb, or pc as fa, fb, or fc, respectively.

The value of an associated datum f for a fragment produced by rasterizing a triangle, whether it be a shader output or the clip w coordinate, must be determined using perspective interpolation:

\[f = { a {f_a / w_a} + b {f_b / w_b} + c {f_c / w_c} } \over { {a / w_a} + {b / w_b} + {c / w_c} }\]

where wa, wb, and wc are the clip w coordinates of pa, pb, and pc, respectively. a, b, and c are the barycentric coordinates of the location at which the data are produced - this must be a pixel center or the location of a sample. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the pixel center must be used.

Depth values for triangles must be determined using linear interpolation:

z = a za + b zb + c zc

where za, zb, and zc are the depth values of pa, pb, and pc, respectively.

The NoPerspective and Flat interpolation decorations can be used with fragment shader inputs to declare how they are interpolated. When neither decoration is applied, perspective interpolation is performed as described above. When the NoPerspective decoration is used, linear interpolation is performed in the same fashion as for depth values, as described above. When the Flat decoration is used, no interpolation is performed, and outputs are taken from the corresponding input value of the provoking vertex corresponding to that primitive.

When the VK_AMD_shader_explicit_vertex_parameter device extension is enabled the CustomInterpAMD interpolation decoration can also be used with fragment shader inputs which indicate that the decorated inputs can only be accessed by the extended instruction InterpolateAtVertexAMD and allows accessing the value of the inputs for individual vertices of the primitive.

For a polygon with more than three edges, such as are produced by clipping a triangle, a convex combination of the values of the datum at the polygon’s vertices must be used to obtain the value assigned to each fragment produced by the rasterization algorithm. That is, it must be the case that at every fragment

\[f = \sum_{i=1}^{n} a_i f_i\]

where n is the number of vertices in the polygon and fi is the value of f at vertex i. For each i, 0 ≤ ai ≤ 1 and \(\sum_{i=1}^{n}a_i = 1\). The values of ai may differ from fragment to fragment, but at vertex i, ai = 1 and aj = 0 for j ≠ i.

Note

One algorithm that achieves the required behavior is to triangulate a polygon (without adding any vertices) and then treat each triangle individually as already discussed. A scan-line rasterizer that linearly interpolates data along each edge and then linearly interpolates data across each horizontal span from edge to edge also satisfies the restrictions (in this case, the numerator and denominator of equation [triangle_perspective_interpolation] are iterated independently and a division performed for each fragment).

24.8.2. Polygon Mode

Possible values of the VkPipelineRasterizationStateCreateInfo::polygonMode property of the currently active pipeline, specifying the method of rasterization for polygons, are:

typedef enum VkPolygonMode {
    VK_POLYGON_MODE_FILL = 0,
    VK_POLYGON_MODE_LINE = 1,
    VK_POLYGON_MODE_POINT = 2,
    VK_POLYGON_MODE_FILL_RECTANGLE_NV = 1000153000,
} VkPolygonMode;
  • VK_POLYGON_MODE_POINT specifies that polygon vertices are drawn as points.

  • VK_POLYGON_MODE_LINE specifies that polygon edges are drawn as line segments.

  • VK_POLYGON_MODE_FILL specifies that polygons are rendered using the polygon rasterization rules in this section.

  • VK_POLYGON_MODE_FILL_RECTANGLE_NV specifies that polygons are rendered using polygon rasterization rules, modified to consider a sample within the primitive if the sample location is inside the axis-aligned bounding box of the triangle after projection. Note that the barycentric weights used in attribute interpolation can extend outside the range [0,1] when these primitives are shaded. Special treatment is given to a sample position on the boundary edge of the bounding box. In such a case, if two rectangles lie on either side of a common edge (with identical endpoints) on which a sample position lies, then exactly one of the triangles must produce a fragment that covers that sample during rasterization.

    Polygons rendered in VK_POLYGON_MODE_FILL_RECTANGLE_NV mode may be clipped by the frustum or by user clip planes. If clipping is applied, the triangle is culled rather than clipped.

    Area calculation and facingness are determined for VK_POLYGON_MODE_FILL_RECTANGLE_NV mode using the triangle’s vertices.

These modes affect only the final rasterization of polygons: in particular, a polygon’s vertices are shaded and the polygon is clipped and possibly culled before these modes are applied.

24.8.3. Depth Bias

The depth values of all fragments generated by the rasterization of a polygon can be offset by a single value that is computed for that polygon. This behavior is controlled by the depthBiasEnable, depthBiasConstantFactor, depthBiasClamp, and depthBiasSlopeFactor members of VkPipelineRasterizationStateCreateInfo, or by the corresponding parameters to the vkCmdSetDepthBias command if depth bias state is dynamic.

void vkCmdSetDepthBias(
    VkCommandBuffer                             commandBuffer,
    float                                       depthBiasConstantFactor,
    float                                       depthBiasClamp,
    float                                       depthBiasSlopeFactor);
  • commandBuffer is the command buffer into which the command will be recorded.

  • depthBiasConstantFactor is a scalar factor controlling the constant depth value added to each fragment.

  • depthBiasClamp is the maximum (or minimum) depth bias of a fragment.

  • depthBiasSlopeFactor is a scalar factor applied to a fragment’s slope in depth bias calculations.

If depthBiasEnable is VK_FALSE, no depth bias is applied and the fragment’s depth values are unchanged.

depthBiasSlopeFactor scales the maximum depth slope of the polygon, and depthBiasConstantFactor scales an implementation-dependent constant that relates to the usable resolution of the depth buffer. The resulting values are summed to produce the depth bias value which is then clamped to a minimum or maximum value specified by depthBiasClamp. depthBiasSlopeFactor, depthBiasConstantFactor, and depthBiasClamp can each be positive, negative, or zero.

The maximum depth slope m of a triangle is

\[m = \sqrt{ \left({{\partial z_f} \over {\partial x_f}}\right)^2 + \left({{\partial z_f} \over {\partial y_f}}\right)^2}\]

where (xf, yf, zf) is a point on the triangle. m may be approximated as

\[m = \max\left( \left| { {\partial z_f} \over {\partial x_f} } \right|, \left| { {\partial z_f} \over {\partial y_f} } \right| \right).\]

The minimum resolvable difference r is an implementation-dependent parameter that depends on the depth buffer representation. It is the smallest difference in framebuffer coordinate z values that is guaranteed to remain distinct throughout polygon rasterization and in the depth buffer. All pairs of fragments generated by the rasterization of two polygons with otherwise identical vertices, but zf values that differ by r, will have distinct depth values.

For fixed-point depth buffer representations, r is constant throughout the range of the entire depth buffer. For floating-point depth buffers, there is no single minimum resolvable difference. In this case, the minimum resolvable difference for a given polygon is dependent on the maximum exponent, e, in the range of z values spanned by the primitive. If n is the number of bits in the floating-point mantissa, the minimum resolvable difference, r, for the given primitive is defined as

r = 2e-n

If a triangle is rasterized using the VK_POLYGON_MODE_FILL_RECTANGLE_NV polygon mode, then this minimum resolvable difference may not be resolvable for samples outside of the triangle, where the depth is extrapolated.

If no depth buffer is present, r is undefined.

The bias value o for a polygon is

\[o = \begin{cases} m \times depthBiasSlopeFactor + r \times depthBiasConstantFactor & depthBiasClamp = 0\ or\ NaN \\ \min(m \times depthBiasSlopeFactor + r \times depthBiasConstantFactor, depthBiasClamp) & depthBiasClamp > 0 \\ \max(m \times depthBiasSlopeFactor + r \times depthBiasConstantFactor, depthBiasClamp) & depthBiasClamp < 0 \\ \end{cases}\]

m is computed as described above. If the depth buffer uses a fixed-point representation, m is a function of depth values in the range [0,1], and o is applied to depth values in the same range.

For fixed-point depth buffers, fragment depth values are always limited to the range [0,1] by clamping after depth bias addition is performed. Unless the VK_EXT_depth_range_unrestricted extension is enabled, fragment depth values are clamped even when the depth buffer uses a floating-point representation.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_DEPTH_BIAS dynamic state enabled

  • If the depth bias clamping feature is not enabled, depthBiasClamp must be 0.0

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

24.8.4. Conservative Rasterization

Polygon rasterization can be made conservative by setting conservativeRasterizationMode to VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT or VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT in VkPipelineRasterizationConservativeStateCreateInfoEXT. The VkPipelineRasterizationConservativeStateCreateInfoEXT state is set by adding an instance of this structure to the pNext chain of an instance of the VkPipelineRasterizationStateCreateInfo structure when creating the graphics pipeline. Enabling these modes also affects line and point rasterization if the implementation sets VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativePointAndLineRasterization to VK_TRUE.

VkPipelineRasterizationConservativeStateCreateInfoEXT is defined as:

typedef struct VkPipelineRasterizationConservativeStateCreateInfoEXT {
    VkStructureType                                           sType;
    const void*                                               pNext;
    VkPipelineRasterizationConservativeStateCreateFlagsEXT    flags;
    VkConservativeRasterizationModeEXT                        conservativeRasterizationMode;
    float                                                     extraPrimitiveOverestimationSize;
} VkPipelineRasterizationConservativeStateCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • conservativeRasterizationMode is the conservative rasterization mode to use.

  • extraPrimitiveOverestimationSize is the extra size in pixels to increase the generating primitive during conservative rasterization at each of its edges in X and Y equally in screen space beyond the base overestimation specified in VkPhysicalDeviceConservativeRasterizationPropertiesEXT::primitiveOverestimationSize.

Valid Usage
  • extraPrimitiveOverestimationSize must be in the range of 0.0 to VkPhysicalDeviceConservativeRasterizationPropertiesEXT::maxExtraPrimitiveOverestimationSize inclusive

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_CONSERVATIVE_STATE_CREATE_INFO_EXT

  • flags must be 0

  • conservativeRasterizationMode must be a valid VkConservativeRasterizationModeEXT value

typedef VkFlags VkPipelineRasterizationConservativeStateCreateFlagsEXT;

VkPipelineRasterizationConservativeStateCreateFlagsEXT is a bitmask type for setting a mask, but is currently reserved for future use.

Possible values of VkPipelineRasterizationConservativeStateCreateInfoEXT::conservativeRasterizationMode, specifying the conservative rasterization mode are:

typedef enum VkConservativeRasterizationModeEXT {
    VK_CONSERVATIVE_RASTERIZATION_MODE_DISABLED_EXT = 0,
    VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT = 1,
    VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT = 2,
} VkConservativeRasterizationModeEXT;
  • VK_CONSERVATIVE_RASTERIZATION_MODE_DISABLED_EXT specifies that conservative rasterization is disabled and rasterization proceeds as normal.

  • VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT specifies that conservative rasterization is enabled in overestimation mode.

  • VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT specifies that conservative rasterization is enabled in underestimation mode.

When overestimate conservative rasterization is enabled, rather than evaluating coverage at individual sample locations, a determination is made of whether any portion of the pixel (including its edges and corners) is covered by the primitive. If any portion of the pixel is covered, then all bits of the coverage sample mask for the fragment are enabled. If the implementation supports VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativeRasterizationPostDepthCoverage and the PostDepthCoverage execution mode is specified the SampleMask built-in input variable will reflect the coverage after the early per-fragment depth and stencil tests are applied.

For the purposes of evaluating which pixels are covered by the primitive, implementations can increase the size of the primitive by up to VkPhysicalDeviceConservativeRasterizationPropertiesEXT::primitiveOverestimationSize pixels at each of the primitive edges. This may increase the number of fragments generated by this primitive and represents an overestimation of the pixel coverage.

This overestimation size can be increased further by setting the extraPrimitiveOverestimationSize value above 0.0 in steps of VkPhysicalDeviceConservativeRasterizationPropertiesEXT::extraPrimitiveOverestimationSizeGranularity up to and including VkPhysicalDeviceConservativeRasterizationPropertiesEXT::extraPrimitiveOverestimationSize. This will: further increase the number of fragments generated by this primitive.

The actual precision of the overestimation size used for conservative rasterization may vary between implementations and produce results that only approximate the primitiveOverestimationSize and extraPrimitiveOverestimationSizeGranularity properties.

For triangles if VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT is enabled, fragments will be generated if the primitive area covers any portion of the pixel, including its edges or corners. The tie-breaking rule described in Basic Polygon Rasterization does not apply during conservative rasterization and coverage is set for all fragments generated from shared edges of polygons. Degenerate triangles that evaluate to zero area after rasterization, even for pixels that contain a vertex or edge of the zero-area polygon, will be culled if VkPhysicalDeviceConservativeRasterizationPropertiesEXT::degenerateTrianglesRasterized is VK_FALSE or will generate fragments if degenerateTrianglesRasterized is VK_TRUE. The fragment input values for these degenerate triangles take their attribute and depth values from the provoking vertex. Degenerate triangles are considered backfacing and the application can enable backface culling if desired. Triangles that are zero area before rasterization may be culled regardless.

For lines if VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT is enabled, and the implementation sets VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativePointAndLineRasterization to VK_TRUE, fragments will be generated if the line covers any portion of the pixel, including its edges or corners. Degenerate lines that evaluate to zero length after rasterization will be culled if VkPhysicalDeviceConservativeRasterizationPropertiesEXT::degenerateLinesRasterized is VK_FALSE or will generate fragments if degenerateLinesRasterized is VK_TRUE. The fragments input values for these degenerate lines take their attribute and depth values from the provoking vertex. Lines that are zero length before rasterization may be culled regardless.

For points if VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT is enabled, and the implementation sets VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativePointAndLineRasterization to VK_TRUE, fragments will be generated if the point square covers any portion of the pixel square, including its edges or corners.

When underestimate conservative rasterization is enabled, rather than evaluating coverage at individual sample locations, a determination is made of whether all of the pixel (including its edges and corners) is covered by the primitive. If the entire pixel is covered, then a fragment is generated with all bits of its coverage sample mask enabled, otherwise the fragment is discarded even if some portion of the pixel is covered. If the implementation supports VkPhysicalDeviceConservativeRasterizationPropertiesEXT::conservativeRasterizationPostDepthCoverage and the PostDepthCoverage execution mode is specified the SampleMask built-in input variable will reflect the coverage after the early per-fragment depth and stencil tests are applied.

For triangles, if VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT is enabled, fragments will only be generated if they are fully covered by the generating primitive, including its edges and corners.

For lines, if VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT is enabled, fragments will be generated if the entire pixel, including its edges and corners is covered by the line.

For points, if VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT is enabled, fragments will only be generated for pixel squares that are fully covered by the point square.

For both overestimate and underestimate conservative rasterization modes a pixel square that is fully covered by the generating primitive must set FullyCoveredEXT to VK_TRUE if the implementation enables the VkPhysicalDeviceConservativeRasterizationPropertiesEXT::fullyCoveredFragmentShaderInputVariable feature.

25. Fragment Operations

Fragment operations execute on a per-fragment or per-sample basis, affecting whether or how a fragment or sample is written to the framebuffer. Some operations execute before fragment shading, and others after. Fragment operations always adhere to rasterization order.

25.1. Early Per-Fragment Tests

Once fragments are produced by rasterization, a number of per-fragment operations are performed prior to fragment shader execution. If a fragment is discarded during any of these operations, it will not be processed by any subsequent stage, including fragment shader execution.

The scissor test and sample mask generation are both always performed during early fragment tests.

Fragment operations are performed in the following order:

If early per-fragment operations are enabled by the fragment shader, these operations are also performed:

If post-depth coverage operation is enabled by the fragment shader, the SampleMask coverage is determined after the early stencil and depth tests.

25.2. Discard Rectangles Test

The discard rectangles test determines if fragment’s framebuffer coordinates (xf,yf) are inclusive or exclusive to a set of discard-space rectangles. The discard rectangles are set with the VkPipelineDiscardRectangleStateCreateInfoEXT pipeline state, which is defined as:

typedef struct VkPipelineDiscardRectangleStateCreateInfoEXT {
    VkStructureType                                  sType;
    const void*                                      pNext;
    VkPipelineDiscardRectangleStateCreateFlagsEXT    flags;
    VkDiscardRectangleModeEXT                        discardRectangleMode;
    uint32_t                                         discardRectangleCount;
    const VkRect2D*                                  pDiscardRectangles;
} VkPipelineDiscardRectangleStateCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • discardRectangleMode is the mode used to determine whether fragments that lie within the discard rectangle are discarded or not.

  • discardRectangleCount is the number of discard rectangles used by the pipeline.

  • pDiscardRectangles is a pointer to an array of VkRect2D structures, defining the discard rectangles. If the discard rectangle state is dynamic, this member is ignored.

Valid Usage
  • discardRectangleCount must be between 0 and VkPhysicalDeviceDiscardRectanglePropertiesEXT::maxDiscardRectangles, inclusive

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_DISCARD_RECTANGLE_STATE_CREATE_INFO_EXT

  • flags must be 0

  • discardRectangleMode must be a valid VkDiscardRectangleModeEXT value

typedef VkFlags VkPipelineDiscardRectangleStateCreateFlagsEXT;

VkPipelineDiscardRectangleStateCreateFlagsEXT is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPipelineDiscardRectangleStateCreateInfoEXT state is set by adding an instance of this structure to the pNext chain of an instance of the VkGraphicsPipelineCreateInfo structure and setting the graphics pipeline state with vkCreateGraphicsPipelines.

If the bound pipeline state object was not created with the VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT dynamic state enabled, discard rectangles are specified using the pDiscardRectangles member of VkPipelineDiscardRectangleStateCreateInfoEXT linked to the pipeline state object.

If the pipeline state object was created with the VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT dynamic state enabled, the discard rectangles are dynamically set and changed with the command:

void vkCmdSetDiscardRectangleEXT(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    firstDiscardRectangle,
    uint32_t                                    discardRectangleCount,
    const VkRect2D*                             pDiscardRectangles);
  • commandBuffer is the command buffer into which the command will be recorded.

  • firstDiscardRectangle is the index of the first discard rectangle whose state is updated by the command.

  • discardRectangleCount is the number of discard rectangles whose state are updated by the command.

  • pDiscardRectangles is a pointer to an array of VkRect2D structures specifying discard rectangles.

The discard rectangle taken from element i of pDiscardRectangles replace the current state for the discard rectangle index firstDiscardRectangle + i, for i in [0, discardRectangleCount).

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT dynamic state enabled

  • The sum of firstDiscardRectangle and discardRectangleCount must be less than or equal to VkPhysicalDeviceDiscardRectanglePropertiesEXT::maxDiscardRectangles

  • The x and y member of offset in each VkRect2D element of pDiscardRectangles must be greater than or equal to 0

  • Evaluation of (offset.x + extent.width) in each VkRect2D element of pDiscardRectangles must not cause a signed integer addition overflow

  • Evaluation of (offset.y + extent.height) in each VkRect2D element of pDiscardRectangles must not cause a signed integer addition overflow

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pDiscardRectangles must be a valid pointer to an array of discardRectangleCount VkRect2D structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • discardRectangleCount must be greater than 0

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

The VkOffset2D::x and VkOffset2D::y values of the discard rectangle VkRect2D specify the upper-left origin of the discard rectangle box. The lower-right corner of the discard rectangle box is specified as the VkExtent2D::width and VkExtent2D::height from the upper-left origin.

If offset.x ≤ xf < offset.x + extent.width and offset.y ≤ yf < offset.y + extent.height for the selected discard rectangle, then the fragment is within the discard rectangle box. When the discard rectangle mode is VK_DISCARD_RECTANGLE_MODE_INCLUSIVE_EXT a fragment within at least one of the active discard rectangle boxes passes the discard rectangle test; otherwise the fragment fails the discard rectangle test and is discarded. When the discard rectangle mode is VK_DISCARD_RECTANGLE_MODE_EXCLUSIVE_EXT a fragment within at least one of the active discard rectangle boxes fails the discard rectangle test, and the fragment is discarded; otherwise the fragment passes the discard rectangles test. The discard rectangles test only applies to drawing commands, not to other commands like clears or copies.

Possible values of VkPipelineDiscardRectangleStateCreateInfoEXT::discardRectangleMode, specifying the behavior of the discard rectangle test, are:

typedef enum VkDiscardRectangleModeEXT {
    VK_DISCARD_RECTANGLE_MODE_INCLUSIVE_EXT = 0,
    VK_DISCARD_RECTANGLE_MODE_EXCLUSIVE_EXT = 1,
} VkDiscardRectangleModeEXT;
  • VK_DISCARD_RECTANGLE_MODE_INCLUSIVE_EXT specifies that a fragment within any discard rectangle satisfies the test.

  • VK_DISCARD_RECTANGLE_MODE_EXCLUSIVE_EXT specifies that a fragment not within any of the discard rectangles satisfies the test.

25.3. Scissor Test

The scissor test determines if a fragment’s framebuffer coordinates (xf,yf) lie within the scissor rectangle corresponding to the viewport index (see Controlling the Viewport) used by the primitive that generated the fragment. If the pipeline state object is created without VK_DYNAMIC_STATE_SCISSOR enabled then the scissor rectangles are set by the VkPipelineViewportStateCreateInfo state of the pipeline state object. Otherwise, to dynamically set the scissor rectangles call:

void vkCmdSetScissor(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    firstScissor,
    uint32_t                                    scissorCount,
    const VkRect2D*                             pScissors);
  • commandBuffer is the command buffer into which the command will be recorded.

  • firstScissor is the index of the first scissor whose state is updated by the command.

  • scissorCount is the number of scissors whose rectangles are updated by the command.

  • pScissors is a pointer to an array of VkRect2D structures defining scissor rectangles.

The scissor rectangles taken from element i of pScissors replace the current state for the scissor index firstScissor + i, for i in [0, scissorCount).

Each scissor rectangle is described by a VkRect2D structure, with the offset.x and offset.y values determining the upper left corner of the scissor rectangle, and the extent.width and extent.height values determining the size in pixels.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_SCISSOR dynamic state enabled

  • firstScissor must be less than VkPhysicalDeviceLimits::maxViewports

  • The sum of firstScissor and scissorCount must be between 1 and VkPhysicalDeviceLimits::maxViewports, inclusive

  • If the multiple viewports feature is not enabled, firstScissor must be 0

  • If the multiple viewports feature is not enabled, scissorCount must be 1

  • The x and y members of offset must be greater than or equal to 0

  • Evaluation of (offset.x + extent.width) must not cause a signed integer addition overflow

  • Evaluation of (offset.y + extent.height) must not cause a signed integer addition overflow

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pScissors must be a valid pointer to an array of scissorCount VkRect2D structures

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

  • scissorCount must be greater than 0

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

If offset.x ≤ xf < offset.x + extent.width and offset.y ≤ yf < offset.y + extent.height for the selected scissor rectangle, then the scissor test passes. Otherwise, the test fails and the fragment is discarded. For points, lines, and polygons, the scissor rectangle for a primitive is selected in the same manner as the viewport (see Controlling the Viewport). The scissor rectangles test only applies to drawing commands, not to other commands like clears or copies.

It is legal for offset.x + extent.width or offset.y + extent.height to exceed the dimensions of the framebuffer - the scissor test still applies as defined above. Rasterization does not produce fragments outside of the framebuffer, so such fragments never have the scissor test performed on them.

The scissor test is always performed. Applications can effectively disable the scissor test by specifying a scissor rectangle that encompasses the entire framebuffer.

25.4. Sample Mask

This step modifies fragment coverage values based on the values in the pSampleMask array member of VkPipelineMultisampleStateCreateInfo, as described previously in section Graphics Pipelines.

pSampleMask contains an array of static coverage information that is ANDed with the coverage information generated during rasterization. Bits that are zero disable coverage for the corresponding sample. Bit B of mask word M corresponds to sample 32 × M + B. The array is sized to a length of rasterizationSamples / 32 ⌉ words. If pSampleMask is NULL, it is treated as if the mask has all bits enabled, i.e. no coverage is removed from fragments.

The elements of the sample mask array are of type VkSampleMask, each representing 32 bits of coverage information:

typedef uint32_t VkSampleMask;

25.5. Early Fragment Test Mode

The depth bounds test, stencil test, depth test, and occlusion query sample counting are performed before fragment shading if and only if early fragment tests are enabled by the fragment shader (see Early Fragment Tests). When early per-fragment operations are enabled, these operations are performed prior to fragment shader execution, and the stencil buffer, depth buffer, and occlusion query sample counts will be updated accordingly; these operations will not be performed again after fragment shader execution.

If a pipeline’s fragment shader has early fragment tests disabled, these operations are performed only after fragment program execution, in the order described below. If a pipeline does not contain a fragment shader, these operations are performed only once.

If early fragment tests are enabled, any depth value computed by the fragment shader has no effect. Additionally, the depth test (including depth writes), stencil test (including stencil writes) and sample counting operations are performed even for fragments or samples that would be discarded after fragment shader execution due to per-fragment operations such as alpha-to-coverage tests, or due to the fragment being discarded by the shader itself.

25.6. Late Per-Fragment Tests

After programmable fragment processing, per-fragment operations are performed before blending and color output to the framebuffer.

A fragment is produced by rasterization with framebuffer coordinates of (xf,yf) and depth z, as described in Rasterization. The fragment is then modified by programmable fragment processing, which adds associated data as described in Shaders. The fragment is then further modified, and possibly discarded by the late per-fragment operations described in this chapter. Finally, if the fragment was not discarded, it is used to update the framebuffer at the fragment’s framebuffer coordinates for any samples that remain covered.

The depth bounds test, stencil test, and depth test are performed for each pixel sample, rather than just once for each fragment. Stencil and depth operations are performed for a pixel sample only if that sample’s fragment coverage bit is a value of 1 when the fragment executes the corresponding stage of the graphics pipeline. If the corresponding coverage bit is 0, no operations are performed for that sample. Failure of the depth bounds, stencil, or depth test results in termination of the processing of that sample by means of disabling coverage for that sample, rather than discarding of the fragment. If, at any point, a fragment’s coverage becomes zero for all samples, then the fragment is discarded. All operations are performed on the depth and stencil values stored in the depth/stencil attachment of the framebuffer. The contents of the color attachments are not modified at this point.

The depth bounds test, stencil test, depth test, and occlusion query operations described in Depth Bounds Test, Stencil Test, Depth Test, Sample Counting are instead performed prior to fragment processing, as described in Early Fragment Test Mode, if requested by the fragment shader.

25.7. Mixed attachment samples

Special rules apply to per-fragment operations when the number of samples of the color attachments differs from the number of samples of the depth/stencil attachment used in a subpass.

Let C be the number of color attachment samples and D be the number of depth/stencil attachment samples used by a given subpass.

If C < D then only the first C number of samples are guaranteed to have a corresponding fragment shader invocation and thus a corresponding color output value, unless the fragment shaders produce inputs to the late per-fragment tests (e.g. by outputting to a variable decorated with the FragDepth built-in decoration). Implementations are allowed to produce fragment shader invocations for samples with indices greater than or equal to C but (other than potential side effects) the color outputs of fragment shader invocations corresponding to such samples are discarded.

25.8. Multisample Coverage

If a fragment shader is active and its entry point’s interface includes a built-in output variable decorated with SampleMask and also decorated with OverrideCoverageNV the fragment coverage is replaced with the sample mask bits set in the shader. Otherwise if the built-in output variable decorated with SampleMask is not also decorated with OverrideCoverageNV then the fragment coverage is ANDed with the bits of the sample mask to generate a new fragment coverage value. If such a fragment shader did not assign a value to SampleMask due to flow of control, the value ANDed with the fragment coverage is undefined. If no fragment shader is active, or if the active fragment shader does not include SampleMask in its interface, the fragment coverage is not modified.

Next, the fragment alpha and coverage values are modified based on the alphaToCoverageEnable and alphaToOneEnable members of the VkPipelineMultisampleStateCreateInfo structure.

All alpha values in this section refer only to the alpha component of the fragment shader output that has a Location and Index decoration of zero (see the Fragment Output Interface section). If that shader output has an integer or unsigned integer type, then these operations are skipped.

If alphaToCoverageEnable is enabled, a temporary coverage value with rasterizationSamples bits is generated where each bit is determined by the fragment’s alpha value. The temporary coverage value is then ANDed with the fragment coverage value to generate a new fragment coverage value.

No specific algorithm is specified for converting the alpha value to a temporary coverage mask. It is intended that the number of 1’s in this value be proportional to the alpha value (clamped to [0,1]), with all 1’s corresponding to a value of 1.0 and all 0’s corresponding to 0.0. The algorithm may be different at different pixel locations.

Note

Using different algorithms at different pixel location may help to avoid artifacts caused by regular coverage sample locations.

Next, if alphaToOneEnable is enabled, each alpha value is replaced by the maximum representable alpha value for fixed-point color buffers, or by 1.0 for floating-point buffers. Otherwise, the alpha values are not changed.

25.9. Depth and Stencil Operations

Pipeline state controlling the depth bounds tests, stencil test, and depth test is specified through the members of the VkPipelineDepthStencilStateCreateInfo structure.

The VkPipelineDepthStencilStateCreateInfo structure is defined as:

typedef struct VkPipelineDepthStencilStateCreateInfo {
    VkStructureType                           sType;
    const void*                               pNext;
    VkPipelineDepthStencilStateCreateFlags    flags;
    VkBool32                                  depthTestEnable;
    VkBool32                                  depthWriteEnable;
    VkCompareOp                               depthCompareOp;
    VkBool32                                  depthBoundsTestEnable;
    VkBool32                                  stencilTestEnable;
    VkStencilOpState                          front;
    VkStencilOpState                          back;
    float                                     minDepthBounds;
    float                                     maxDepthBounds;
} VkPipelineDepthStencilStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • depthTestEnable controls whether depth testing is enabled.

  • depthWriteEnable controls whether depth writes are enabled when depthTestEnable is VK_TRUE. Depth writes are always disabled when depthTestEnable is VK_FALSE.

  • depthCompareOp is the comparison operator used in the depth test.

  • depthBoundsTestEnable controls whether depth bounds testing is enabled.

  • stencilTestEnable controls whether stencil testing is enabled.

  • front and back control the parameters of the stencil test.

  • minDepthBounds and maxDepthBounds define the range of values used in the depth bounds test.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_DEPTH_STENCIL_STATE_CREATE_INFO

  • pNext must be NULL

  • flags must be 0

  • depthCompareOp must be a valid VkCompareOp value

  • front must be a valid VkStencilOpState structure

  • back must be a valid VkStencilOpState structure

typedef VkFlags VkPipelineDepthStencilStateCreateFlags;

VkPipelineDepthStencilStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

25.10. Depth Bounds Test

The depth bounds test conditionally disables coverage of a sample based on the outcome of a comparison between the value za in the depth attachment at location (xf,yf) (for the appropriate sample) and a range of values. The test is enabled or disabled by the depthBoundsTestEnable member of VkPipelineDepthStencilStateCreateInfo: If the pipeline state object is created without the VK_DYNAMIC_STATE_DEPTH_BOUNDS dynamic state enabled then the range of values used in the depth bounds test are defined by the minDepthBounds and maxDepthBounds members of the VkPipelineDepthStencilStateCreateInfo structure. Otherwise, to dynamically set the depth bounds range values call:

void vkCmdSetDepthBounds(
    VkCommandBuffer                             commandBuffer,
    float                                       minDepthBounds,
    float                                       maxDepthBounds);
  • commandBuffer is the command buffer into which the command will be recorded.

  • minDepthBounds is the lower bound of the range of depth values used in the depth bounds test.

  • maxDepthBounds is the upper bound of the range.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_DEPTH_BOUNDS dynamic state enabled

  • Unless the VK_EXT_depth_range_unrestricted extension is enabled minDepthBounds must be between 0.0 and 1.0, inclusive

  • Unless the VK_EXT_depth_range_unrestricted extension is enabled maxDepthBounds must be between 0.0 and 1.0, inclusive

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

If minDepthBounds ≤ zamaxDepthBounds}, then the depth bounds test passes. Otherwise, the test fails and the sample’s coverage bit is cleared in the fragment. If there is no depth framebuffer attachment or if the depth bounds test is disabled, it is as if the depth bounds test always passes.

25.11. Stencil Test

The stencil test conditionally disables coverage of a sample based on the outcome of a comparison between the stencil value in the depth/stencil attachment at location (xf,yf) (for the appropriate sample) and a reference value. The stencil test also updates the value in the stencil attachment, depending on the test state, the stencil value and the stencil write masks. The test is enabled or disabled by the stencilTestEnable member of VkPipelineDepthStencilStateCreateInfo.

When disabled, the stencil test and associated modifications are not made, and the sample’s coverage is not modified.

The stencil test is controlled with the front and back members of VkPipelineDepthStencilStateCreateInfo which are of type VkStencilOpState.

The VkStencilOpState structure is defined as:

typedef struct VkStencilOpState {
    VkStencilOp    failOp;
    VkStencilOp    passOp;
    VkStencilOp    depthFailOp;
    VkCompareOp    compareOp;
    uint32_t       compareMask;
    uint32_t       writeMask;
    uint32_t       reference;
} VkStencilOpState;
  • failOp is a VkStencilOp value specifying the action performed on samples that fail the stencil test.

  • passOp is a VkStencilOp value specifying the action performed on samples that pass both the depth and stencil tests.

  • depthFailOp is a VkStencilOp value specifying the action performed on samples that pass the stencil test and fail the depth test.

  • compareOp is a VkCompareOp value specifying the comparison operator used in the stencil test.

  • compareMask selects the bits of the unsigned integer stencil values participating in the stencil test.

  • writeMask selects the bits of the unsigned integer stencil values updated by the stencil test in the stencil framebuffer attachment.

  • reference is an integer reference value that is used in the unsigned stencil comparison.

Valid Usage (Implicit)

There are two sets of stencil-related state, the front stencil state set and the back stencil state set. Stencil tests and writes use the front set of stencil state when processing front-facing fragments and use the back set of stencil state when processing back-facing fragments. Fragments rasterized from non-polygon primitives (points and lines) are always considered front-facing. Fragments rasterized from polygon primitives inherit their facingness from the polygon, even if the polygon is rasterized as points or lines due to the current VkPolygonMode. Whether a polygon is front- or back-facing is determined in the same manner used for face culling (see Basic Polygon Rasterization).

The operation of the stencil test is also affected by the compareMask, writeMask, and reference members of VkStencilOpState set in the pipeline state object if the pipeline state object is created without the VK_DYNAMIC_STATE_STENCIL_COMPARE_MASK, VK_DYNAMIC_STATE_STENCIL_WRITE_MASK, and VK_DYNAMIC_STATE_STENCIL_REFERENCE dynamic states enabled, respectively.

If the pipeline state object is created with the VK_DYNAMIC_STATE_STENCIL_COMPARE_MASK dynamic state enabled, then to dynamically set the stencil compare mask call:

void vkCmdSetStencilCompareMask(
    VkCommandBuffer                             commandBuffer,
    VkStencilFaceFlags                          faceMask,
    uint32_t                                    compareMask);
  • commandBuffer is the command buffer into which the command will be recorded.

  • faceMask is a bitmask of VkStencilFaceFlagBits specifying the set of stencil state for which to update the compare mask.

  • compareMask is the new value to use as the stencil compare mask.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_STENCIL_COMPARE_MASK dynamic state enabled

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • faceMask must be a valid combination of VkStencilFaceFlagBits values

  • faceMask must not be 0

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

Bits which can be set in the vkCmdSetStencilCompareMask::faceMask parameter, and similar parameters of other commands specifying which stencil state to update stencil masks for, are:

typedef enum VkStencilFaceFlagBits {
    VK_STENCIL_FACE_FRONT_BIT = 0x00000001,
    VK_STENCIL_FACE_BACK_BIT = 0x00000002,
    VK_STENCIL_FRONT_AND_BACK = 0x00000003,
} VkStencilFaceFlagBits;
  • VK_STENCIL_FACE_FRONT_BIT specifies that only the front set of stencil state is updated.

  • VK_STENCIL_FACE_BACK_BIT specifies that only the back set of stencil state is updated.

  • VK_STENCIL_FRONT_AND_BACK is the combination of VK_STENCIL_FACE_FRONT_BIT and VK_STENCIL_FACE_BACK_BIT, and specifies that both sets of stencil state are updated.

typedef VkFlags VkStencilFaceFlags;

VkStencilFaceFlags is a bitmask type for setting a mask of zero or more VkStencilFaceFlagBits.

If the pipeline state object is created with the VK_DYNAMIC_STATE_STENCIL_WRITE_MASK dynamic state enabled, then to dynamically set the stencil write mask call:

void vkCmdSetStencilWriteMask(
    VkCommandBuffer                             commandBuffer,
    VkStencilFaceFlags                          faceMask,
    uint32_t                                    writeMask);
  • commandBuffer is the command buffer into which the command will be recorded.

  • faceMask is a bitmask of VkStencilFaceFlagBits specifying the set of stencil state for which to update the write mask, as described above for vkCmdSetStencilCompareMask.

  • writeMask is the new value to use as the stencil write mask.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_STENCIL_WRITE_MASK dynamic state enabled

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • faceMask must be a valid combination of VkStencilFaceFlagBits values

  • faceMask must not be 0

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

If the pipeline state object is created with the VK_DYNAMIC_STATE_STENCIL_REFERENCE dynamic state enabled, then to dynamically set the stencil reference value call:

void vkCmdSetStencilReference(
    VkCommandBuffer                             commandBuffer,
    VkStencilFaceFlags                          faceMask,
    uint32_t                                    reference);
  • commandBuffer is the command buffer into which the command will be recorded.

  • faceMask is a bitmask of VkStencilFaceFlagBits specifying the set of stencil state for which to update the reference value, as described above for vkCmdSetStencilCompareMask.

  • reference is the new value to use as the stencil reference value.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_STENCIL_REFERENCE dynamic state enabled

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • faceMask must be a valid combination of VkStencilFaceFlagBits values

  • faceMask must not be 0

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

reference is an integer reference value that is used in the unsigned stencil comparison. The reference value used by stencil comparison must be within the range [0,2s-1] , where s is the number of bits in the stencil framebuffer attachment, otherwise the reference value is considered undefined. The s least significant bits of compareMask are bitwise ANDed with both the reference and the stored stencil value, and the resulting masked values are those that participate in the comparison controlled by compareOp. Let R be the masked reference value and S be the masked stored stencil value.

Possible values of VkStencilOpState::compareOp, specifying the stencil comparison function, are:

typedef enum VkCompareOp {
    VK_COMPARE_OP_NEVER = 0,
    VK_COMPARE_OP_LESS = 1,
    VK_COMPARE_OP_EQUAL = 2,
    VK_COMPARE_OP_LESS_OR_EQUAL = 3,
    VK_COMPARE_OP_GREATER = 4,
    VK_COMPARE_OP_NOT_EQUAL = 5,
    VK_COMPARE_OP_GREATER_OR_EQUAL = 6,
    VK_COMPARE_OP_ALWAYS = 7,
} VkCompareOp;
  • VK_COMPARE_OP_NEVER specifies that the test never passes.

  • VK_COMPARE_OP_LESS specifies that the test passes when R < S.

  • VK_COMPARE_OP_EQUAL specifies that the test passes when R = S.

  • VK_COMPARE_OP_LESS_OR_EQUAL specifies that the test passes when R ≤ S.

  • VK_COMPARE_OP_GREATER specifies that the test passes when R > S.

  • VK_COMPARE_OP_NOT_EQUAL specifies that the test passes when R ≠ S.

  • VK_COMPARE_OP_GREATER_OR_EQUAL specifies that the test passes when R ≥ S.

  • VK_COMPARE_OP_ALWAYS specifies that the test always passes.

Possible values of the failOp, passOp, and depthFailOp members of VkStencilOpState, specifying what happens to the stored stencil value if this or certain subsequent tests fail or pass, are:

typedef enum VkStencilOp {
    VK_STENCIL_OP_KEEP = 0,
    VK_STENCIL_OP_ZERO = 1,
    VK_STENCIL_OP_REPLACE = 2,
    VK_STENCIL_OP_INCREMENT_AND_CLAMP = 3,
    VK_STENCIL_OP_DECREMENT_AND_CLAMP = 4,
    VK_STENCIL_OP_INVERT = 5,
    VK_STENCIL_OP_INCREMENT_AND_WRAP = 6,
    VK_STENCIL_OP_DECREMENT_AND_WRAP = 7,
} VkStencilOp;
  • VK_STENCIL_OP_KEEP keeps the current value.

  • VK_STENCIL_OP_ZERO sets the value to 0.

  • VK_STENCIL_OP_REPLACE sets the value to reference.

  • VK_STENCIL_OP_INCREMENT_AND_CLAMP increments the current value and clamps to the maximum representable unsigned value.

  • VK_STENCIL_OP_DECREMENT_AND_CLAMP decrements the current value and clamps to 0.

  • VK_STENCIL_OP_INVERT bitwise-inverts the current value.

  • VK_STENCIL_OP_INCREMENT_AND_WRAP increments the current value and wraps to 0 when the maximum value would have been exceeded.

  • VK_STENCIL_OP_DECREMENT_AND_WRAP decrements the current value and wraps to the maximum possible value when the value would go below 0.

For purposes of increment and decrement, the stencil bits are considered as an unsigned integer.

If the stencil test fails, the sample’s coverage bit is cleared in the fragment. If there is no stencil framebuffer attachment, stencil modification cannot occur, and it is as if the stencil tests always pass.

If the stencil test passes, the writeMask member of the VkStencilOpState structures controls how the updated stencil value is written to the stencil framebuffer attachment.

The least significant s bits of writeMask, where s is the number of bits in the stencil framebuffer attachment, specify an integer mask. Where a 1 appears in this mask, the corresponding bit in the stencil value in the depth/stencil attachment is written; where a 0 appears, the bit is not written. The writeMask value uses either the front-facing or back-facing state based on the facingness of the fragment. Fragments generated by front-facing primitives use the front mask and fragments generated by back-facing primitives use the back mask.

25.12. Depth Test

The depth test conditionally disables coverage of a sample based on the outcome of a comparison between the fragment’s depth value at the sample location and the sample’s depth value in the depth/stencil attachment at location (xf,yf). The comparison is enabled or disabled with the depthTestEnable member of the VkPipelineDepthStencilStateCreateInfo structure. When disabled, the depth comparison and subsequent possible updates to the value of the depth component of the depth/stencil attachment are bypassed and the fragment is passed to the next operation. The stencil value, however, can be modified as indicated above as if the depth test passed. If enabled, the comparison takes place and the depth/stencil attachment value can subsequently be modified.

The comparison is specified with the depthCompareOp member of VkPipelineDepthStencilStateCreateInfo. Let zf be the incoming fragment’s depth value for a sample, and let za be the depth/stencil attachment value in memory for that sample. The depth test passes under the following conditions:

  • VK_COMPARE_OP_NEVER: the test never passes.

  • VK_COMPARE_OP_LESS: the test passes when zf < za.

  • VK_COMPARE_OP_EQUAL: the test passes when zf = za.

  • VK_COMPARE_OP_LESS_OR_EQUAL: the test passes when zf ≤ za.

  • VK_COMPARE_OP_GREATER: the test passes when zf > za.

  • VK_COMPARE_OP_NOT_EQUAL: the test passes when zf ≠ za.

  • VK_COMPARE_OP_GREATER_OR_EQUAL: the test passes when zf ≥ za.

  • VK_COMPARE_OP_ALWAYS: the test always passes.

If VkPipelineRasterizationStateCreateInfo::depthClampEnable is enabled, before the incoming fragment’s zf is compared to za, zf is clamped to [min(n,f),max(n,f)], where n and f are the minDepth and maxDepth depth range values of the viewport used by this fragment, respectively.

If the depth test fails, the sample’s coverage bit is cleared in the fragment. The stencil value at the sample’s location is updated according to the function currently in effect for depth test failure.

If the depth test passes, the sample’s (possibly clamped) zf value is conditionally written to the depth framebuffer attachment based on the depthWriteEnable member of VkPipelineDepthStencilStateCreateInfo. If depthWriteEnable is VK_TRUE the value is written, and if it is VK_FALSE the value is not written. If the depth framebuffer attachment is a fixed-point format and the depth value is outside of the 0.0 to 1.0 range the depth value is clamped between 0.0 and 1.0 inclusive before writing. The stencil value at the sample’s location is updated according to the function currently in effect for depth test success.

If there is no depth framebuffer attachment, it is as if the depth test always passes.

25.13. Sample Counting

Occlusion queries use query pool entries to track the number of samples that pass all the per-fragment tests. The mechanism of collecting an occlusion query value is described in Occlusion Queries.

The occlusion query sample counter increments by one for each sample with a coverage value of 1 in each fragment that survives all the per-fragment tests, including scissor, sample mask, alpha to coverage, stencil, and depth tests.

25.14. Fragment Coverage To Color

If the pNext chain of VkPipelineMultisampleStateCreateInfo includes a VkPipelineCoverageToColorStateCreateInfoNV structure, then that structure controls whether the fragment coverage is substituted for a fragment color output and, if so, which output is replaced.

The VkPipelineCoverageToColorStateCreateInfoNV structure is defined as:

typedef struct VkPipelineCoverageToColorStateCreateInfoNV {
    VkStructureType                                sType;
    const void*                                    pNext;
    VkPipelineCoverageToColorStateCreateFlagsNV    flags;
    VkBool32                                       coverageToColorEnable;
    uint32_t                                       coverageToColorLocation;
} VkPipelineCoverageToColorStateCreateInfoNV;
  • sType is the type of this structure

  • pNext is NULL or a pointer to an extension-specific structure

  • flags is reserved for future use.

  • coverageToColorEnable controls whether the fragment coverage value replaces a fragment color output.

  • coverageToColorLocation controls which fragment shader color output value is replaced.

If coverageToColorEnable is VK_TRUE, the fragment coverage information is treated as a bitmask with one bit for each sample (as in the Sample Mask section), and this bitmask replaces the first component of the color value corresponding to the fragment shader output location with Location equal to coverageToColorLocation and Index equal to zero. If the color attachment format has fewer bits than the sample coverage, the low bits of the sample coverage bitmask are taken without any clamping. If the color attachment format has more bits than the sample coverage, the high bits of the sample coverage bitmask are filled with zeros.

If Sample Shading is in use, the coverage bitmask only has bits set for samples that correspond to the fragment shader invocation that shades those samples.

This pipeline stage occurs after sample counting and before blending, and is always performed after fragment shading regardless of the setting of EarlyFragmentTests.

If coverageToColorEnable is VK_FALSE, these operations are skipped. If this structure is not present, it is as if coverageToColorEnable is VK_FALSE.

Valid Usage
  • If coverageToColorEnable is VK_TRUE, then the render pass subpass indicated by VkGraphicsPipelineCreateInfo::renderPass and VkGraphicsPipelineCreateInfo::subpass must have a color attachment at the location selected by coverageToColorLocation, with a VkFormat of VK_FORMAT_R8_UINT, VK_FORMAT_R8_SINT, VK_FORMAT_R16_UINT, VK_FORMAT_R16_SINT, VK_FORMAT_R32_UINT, or VK_FORMAT_R32_SINT

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_COVERAGE_TO_COLOR_STATE_CREATE_INFO_NV

  • flags must be 0

typedef VkFlags VkPipelineCoverageToColorStateCreateFlagsNV;

VkPipelineCoverageToColorStateCreateFlagsNV is a bitmask type for setting a mask, but is currently reserved for future use.

25.15. Coverage Reduction

Coverage reduction generates a color sample mask from the coverage mask, with one bit for each sample in the color attachment(s) for the subpass. If a bit in the color sample mask is 0, then blending and writing to the framebuffer are not performed for that sample.

If the pipeline’s VkPipelineMultisampleStateCreateInfo::rasterizationSamples is greater than one and the VkAttachmentDescription::samples of the color attachments is one, then the fragment’s coverage is reduced from rasterizationSamples bits to a single bit, where the color sample mask is 1 if any bit in the fragment’s coverage is on, and 0 otherwise.

If the pipeline’s VkPipelineMultisampleStateCreateInfo::rasterizationSamples is greater than the VkAttachmentDescription::samples of the color attachments in the subpass, then the fragment’s coverage is reduced from rasterizationSamples bits to a color sample mask with VkAttachmentDescription::samples bits. There is an implementation-dependent association of raster samples to color samples. The reduced color sample mask is computed such that the bit for each color sample is 1 if any of the associated bits in the fragment’s coverage is on, and 0 otherwise.

25.15.1. Coverage Modulation

As part of coverage reduction, fragment color values can also be modulated (multiplied) by a value that is a function of fraction of covered rasterization samples associated with that color sample.

Pipeline state controlling coverage reduction is specified through the members of the VkPipelineCoverageModulationStateCreateInfoNV structure.

The VkPipelineCoverageModulationStateCreateInfoNV structure is defined as:

typedef struct VkPipelineCoverageModulationStateCreateInfoNV {
    VkStructureType                                   sType;
    const void*                                       pNext;
    VkPipelineCoverageModulationStateCreateFlagsNV    flags;
    VkCoverageModulationModeNV                        coverageModulationMode;
    VkBool32                                          coverageModulationTableEnable;
    uint32_t                                          coverageModulationTableCount;
    const float*                                      pCoverageModulationTable;
} VkPipelineCoverageModulationStateCreateInfoNV;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • coverageModulationMode controls which color components are modulated and is of type VkCoverageModulationModeNV.

  • coverageModulationTableEnable controls whether the modulation factor is looked up from a table in pCoverageModulationTable.

  • coverageModulationTableCount is the number of elements in pCoverageModulationTable.

  • pCoverageModulationTable is a table of modulation factors containing a value for each number of covered samples.

If coverageModulationTableEnable is VK_FALSE, then for each color sample the associated bits of the fragment’s coverage are counted and divided by the number of associated bits to produce a modulation factor R in the range (0,1] (a value of zero would have been killed due to a color coverage of 0). Specifically:

  • N = value of rasterizationSamples

  • M = value of VkAttachmentDescription::samples for any color attachments

  • R = popcount(associated coverage bits) / (N / M)

If coverageModulationTableEnable is VK_TRUE, the value R is computed using a programmable lookup table. The lookup table has N / M elements, and the element of the table is selected by:

  • R = pCoverageModulationTable[popcount(associated coverage bits)-1]

Note that the table does not have an entry for popcount(associated coverage bits) = 0, because such samples would have been killed.

The values of pCoverageModulationTable may be rounded to an implementation-dependent precision, which is at least as fine as 1 / N, and clamped to [0,1].

For each color attachment with a floating point or normalized color format, each fragment output color value is replicated to M values which can each be modulated (multiplied) by that color sample’s associated value of R. Which components are modulated is controlled by coverageModulationMode.

If this structure is not present, it is as if coverageModulationMode is VK_COVERAGE_MODULATION_MODE_NONE_NV.

Valid Usage
  • If coverageModulationTableEnable is VK_TRUE, coverageModulationTableCount must be equal to the number of rasterization samples divided by the number of color samples in the subpass.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_COVERAGE_MODULATION_STATE_CREATE_INFO_NV

  • flags must be 0

  • coverageModulationMode must be a valid VkCoverageModulationModeNV value

  • coverageModulationTableCount must be greater than 0

typedef VkFlags VkPipelineCoverageModulationStateCreateFlagsNV;

VkPipelineCoverageModulationStateCreateFlagsNV is a bitmask type for setting a mask, but is currently reserved for future use.

Possible values of VkPipelineCoverageModulationStateCreateInfoNV::coverageModulationMode, specifying which color components are modulated, are:

typedef enum VkCoverageModulationModeNV {
    VK_COVERAGE_MODULATION_MODE_NONE_NV = 0,
    VK_COVERAGE_MODULATION_MODE_RGB_NV = 1,
    VK_COVERAGE_MODULATION_MODE_ALPHA_NV = 2,
    VK_COVERAGE_MODULATION_MODE_RGBA_NV = 3,
} VkCoverageModulationModeNV;
  • VK_COVERAGE_MODULATION_MODE_NONE_NV specifies that no components are multiplied by the modulation factor.

  • VK_COVERAGE_MODULATION_MODE_RGB_NV specifies that the red, green, and blue components are multiplied by the modulation factor.

  • VK_COVERAGE_MODULATION_MODE_ALPHA_NV specifies that the alpha component is multiplied by the modulation factor.

  • VK_COVERAGE_MODULATION_MODE_RGBA_NV specifies that all components are multiplied by the modulation factor.

26. The Framebuffer

26.1. Blending

Blending combines the incoming source fragment’s R, G, B, and A values with the destination R, G, B, and A values of each sample stored in the framebuffer at the fragment’s (xf,yf) location. Blending is performed for each pixel sample, rather than just once for each fragment.

Source and destination values are combined according to the blend operation, quadruplets of source and destination weighting factors determined by the blend factors, and a blend constant, to obtain a new set of R, G, B, and A values, as described below.

Blending is computed and applied separately to each color attachment used by the subpass, with separate controls for each attachment.

Prior to performing the blend operation, signed and unsigned normalized fixed-point color components undergo an implied conversion to floating-point as specified by Conversion from Normalized Fixed-Point to Floating-Point. Blending computations are treated as if carried out in floating-point, and basic blend operations are performed with a precision and dynamic range no lower than that used to represent destination components. Advanced blending operations are performed with a precision and dynamic range no lower than the smaller of that used to represent destination components or that used to represent 16-bit floating-point values.

Blending applies only to fixed-point and floating-point color attachments. If the color attachment has an integer format, blending is not applied.

The pipeline blend state is included in the VkPipelineColorBlendStateCreateInfo structure during graphics pipeline creation:

The VkPipelineColorBlendStateCreateInfo structure is defined as:

typedef struct VkPipelineColorBlendStateCreateInfo {
    VkStructureType                               sType;
    const void*                                   pNext;
    VkPipelineColorBlendStateCreateFlags          flags;
    VkBool32                                      logicOpEnable;
    VkLogicOp                                     logicOp;
    uint32_t                                      attachmentCount;
    const VkPipelineColorBlendAttachmentState*    pAttachments;
    float                                         blendConstants[4];
} VkPipelineColorBlendStateCreateInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • logicOpEnable controls whether to apply Logical Operations.

  • logicOp selects which logical operation to apply.

  • attachmentCount is the number of VkPipelineColorBlendAttachmentState elements in pAttachments. This value must equal the colorAttachmentCount for the subpass in which this pipeline is used.

  • pAttachments: is a pointer to array of per target attachment states.

  • blendConstants is an array of four values used as the R, G, B, and A components of the blend constant that are used in blending, depending on the blend factor.

Each element of the pAttachments array is a VkPipelineColorBlendAttachmentState structure specifying per-target blending state for each individual color attachment. If the independent blending feature is not enabled on the device, all VkPipelineColorBlendAttachmentState elements in the pAttachments array must be identical.

Valid Usage
  • If the independent blending feature is not enabled, all elements of pAttachments must be identical

  • If the logic operations feature is not enabled, logicOpEnable must be VK_FALSE

  • If logicOpEnable is VK_TRUE, logicOp must be a valid VkLogicOp value

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_STATE_CREATE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkPipelineColorBlendAdvancedStateCreateInfoEXT

  • flags must be 0

  • If attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkPipelineColorBlendAttachmentState structures

typedef VkFlags VkPipelineColorBlendStateCreateFlags;

VkPipelineColorBlendStateCreateFlags is a bitmask type for setting a mask, but is currently reserved for future use.

The VkPipelineColorBlendAttachmentState structure is defined as:

typedef struct VkPipelineColorBlendAttachmentState {
    VkBool32                 blendEnable;
    VkBlendFactor            srcColorBlendFactor;
    VkBlendFactor            dstColorBlendFactor;
    VkBlendOp                colorBlendOp;
    VkBlendFactor            srcAlphaBlendFactor;
    VkBlendFactor            dstAlphaBlendFactor;
    VkBlendOp                alphaBlendOp;
    VkColorComponentFlags    colorWriteMask;
} VkPipelineColorBlendAttachmentState;
  • blendEnable controls whether blending is enabled for the corresponding color attachment. If blending is not enabled, the source fragment’s color for that attachment is passed through unmodified.

  • srcColorBlendFactor selects which blend factor is used to determine the source factors (Sr,Sg,Sb).

  • dstColorBlendFactor selects which blend factor is used to determine the destination factors (Dr,Dg,Db).

  • colorBlendOp selects which blend operation is used to calculate the RGB values to write to the color attachment.

  • srcAlphaBlendFactor selects which blend factor is used to determine the source factor Sa.

  • dstAlphaBlendFactor selects which blend factor is used to determine the destination factor Da.

  • alphaBlendOp selects which blend operation is use to calculate the alpha values to write to the color attachment.

  • colorWriteMask is a bitmask of VkColorComponentFlagBits specifying which of the R, G, B, and/or A components are enabled for writing, as described for the Color Write Mask.

Valid Usage
  • If the dual source blending feature is not enabled, srcColorBlendFactor must not be VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, or VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA

  • If the dual source blending feature is not enabled, dstColorBlendFactor must not be VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, or VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA

  • If the dual source blending feature is not enabled, srcAlphaBlendFactor must not be VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, or VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA

  • If the dual source blending feature is not enabled, dstAlphaBlendFactor must not be VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, or VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA

  • If either of colorBlendOp or alphaBlendOp is an advanced blend operation, then colorBlendOp must equal alphaBlendOp

  • If VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT::advancedBlendIndependentBlend is VK_FALSE and colorBlendOp is an advanced blend operation, then colorBlendOp must be the same for all attachments.

  • If VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT::advancedBlendIndependentBlend is VK_FALSE and alphaBlendOp is an advanced blend operation, then alphaBlendOp must be the same for all attachments.

  • If VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT::advancedBlendAllOperations is VK_FALSE, then colorBlendOp must not be VK_BLEND_OP_ZERO_EXT, VK_BLEND_OP_SRC_EXT, VK_BLEND_OP_DST_EXT, VK_BLEND_OP_SRC_OVER_EXT, VK_BLEND_OP_DST_OVER_EXT, VK_BLEND_OP_SRC_IN_EXT, VK_BLEND_OP_DST_IN_EXT, VK_BLEND_OP_SRC_OUT_EXT, VK_BLEND_OP_DST_OUT_EXT, VK_BLEND_OP_SRC_ATOP_EXT, VK_BLEND_OP_DST_ATOP_EXT, VK_BLEND_OP_XOR_EXT, VK_BLEND_OP_INVERT_EXT, VK_BLEND_OP_INVERT_RGB_EXT, VK_BLEND_OP_LINEARDODGE_EXT, VK_BLEND_OP_LINEARBURN_EXT, VK_BLEND_OP_VIVIDLIGHT_EXT, VK_BLEND_OP_LINEARLIGHT_EXT, VK_BLEND_OP_PINLIGHT_EXT, VK_BLEND_OP_HARDMIX_EXT, VK_BLEND_OP_PLUS_EXT, VK_BLEND_OP_PLUS_CLAMPED_EXT, VK_BLEND_OP_PLUS_CLAMPED_ALPHA_EXT, VK_BLEND_OP_PLUS_DARKER_EXT, VK_BLEND_OP_MINUS_EXT, VK_BLEND_OP_MINUS_CLAMPED_EXT, VK_BLEND_OP_CONTRAST_EXT, VK_BLEND_OP_INVERT_OVG_EXT, VK_BLEND_OP_RED_EXT, VK_BLEND_OP_GREEN_EXT, or VK_BLEND_OP_BLUE_EXT

  • If colorBlendOp or alphaBlendOp is an advanced blend operation, then VkSubpassDescription::colorAttachmentCount of the subpass this pipeline is compiled against must be less than or equal to VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT::advancedBlendMaxColorAttachments

Valid Usage (Implicit)

26.1.1. Blend Factors

The source and destination color and alpha blending factors are selected from the enum:

typedef enum VkBlendFactor {
    VK_BLEND_FACTOR_ZERO = 0,
    VK_BLEND_FACTOR_ONE = 1,
    VK_BLEND_FACTOR_SRC_COLOR = 2,
    VK_BLEND_FACTOR_ONE_MINUS_SRC_COLOR = 3,
    VK_BLEND_FACTOR_DST_COLOR = 4,
    VK_BLEND_FACTOR_ONE_MINUS_DST_COLOR = 5,
    VK_BLEND_FACTOR_SRC_ALPHA = 6,
    VK_BLEND_FACTOR_ONE_MINUS_SRC_ALPHA = 7,
    VK_BLEND_FACTOR_DST_ALPHA = 8,
    VK_BLEND_FACTOR_ONE_MINUS_DST_ALPHA = 9,
    VK_BLEND_FACTOR_CONSTANT_COLOR = 10,
    VK_BLEND_FACTOR_ONE_MINUS_CONSTANT_COLOR = 11,
    VK_BLEND_FACTOR_CONSTANT_ALPHA = 12,
    VK_BLEND_FACTOR_ONE_MINUS_CONSTANT_ALPHA = 13,
    VK_BLEND_FACTOR_SRC_ALPHA_SATURATE = 14,
    VK_BLEND_FACTOR_SRC1_COLOR = 15,
    VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR = 16,
    VK_BLEND_FACTOR_SRC1_ALPHA = 17,
    VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA = 18,
} VkBlendFactor;

The semantics of each enum value is described in the table below:

Table 33. Blend Factors
VkBlendFactor RGB Blend Factors (Sr,Sg,Sb) or (Dr,Dg,Db) Alpha Blend Factor (Sa or Da)

VK_BLEND_FACTOR_ZERO

(0,0,0)

0

VK_BLEND_FACTOR_ONE

(1,1,1)

1

VK_BLEND_FACTOR_SRC_COLOR

(Rs0,Gs0,Bs0)

As0

VK_BLEND_FACTOR_ONE_MINUS_SRC_COLOR

(1-Rs0,1-Gs0,1-Bs0)

1-As0

VK_BLEND_FACTOR_DST_COLOR

(Rd,Gd,Bd)

Ad

VK_BLEND_FACTOR_ONE_MINUS_DST_COLOR

(1-Rd,1-Gd,1-Bd)

1-Ad

VK_BLEND_FACTOR_SRC_ALPHA

(As0,As0,As0)

As0

VK_BLEND_FACTOR_ONE_MINUS_SRC_ALPHA

(1-As0,1-As0,1-As0)

1-As0

VK_BLEND_FACTOR_DST_ALPHA

(Ad,Ad,Ad)

Ad

VK_BLEND_FACTOR_ONE_MINUS_DST_ALPHA

(1-Ad,1-Ad,1-Ad)

1-Ad

VK_BLEND_FACTOR_CONSTANT_COLOR

(Rc,Gc,Bc)

Ac

VK_BLEND_FACTOR_ONE_MINUS_CONSTANT_COLOR

(1-Rc,1-Gc,1-Bc)

1-Ac

VK_BLEND_FACTOR_CONSTANT_ALPHA

(Ac,Ac,Ac)

Ac

VK_BLEND_FACTOR_ONE_MINUS_CONSTANT_ALPHA

(1-Ac,1-Ac,1-Ac)

1-Ac

VK_BLEND_FACTOR_SRC_ALPHA_SATURATE

(f,f,f); f = min(As0,1-Ad)

1

VK_BLEND_FACTOR_SRC1_COLOR

(Rs1,Gs1,Bs1)

As1

VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR

(1-Rs1,1-Gs1,1-Bs1)

1-As1

VK_BLEND_FACTOR_SRC1_ALPHA

(As1,As1,As1)

As1

VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA

(1-As1,1-As1,1-As1)

1-As1

In this table, the following conventions are used:

  • Rs0,Gs0,Bs0 and As0 represent the first source color R, G, B, and A components, respectively, for the fragment output location corresponding to the color attachment being blended.

  • Rs1,Gs1,Bs1 and As1 represent the second source color R, G, B, and A components, respectively, used in dual source blending modes, for the fragment output location corresponding to the color attachment being blended.

  • Rd,Gd,Bd and Ad represent the R, G, B, and A components of the destination color. That is, the color currently in the corresponding color attachment for this fragment/sample.

  • Rc,Gc,Bc and Ac represent the blend constant R, G, B, and A components, respectively.

If the pipeline state object is created without the VK_DYNAMIC_STATE_BLEND_CONSTANTS dynamic state enabled then the blend constant (Rc,Gc,Bc,Ac) is specified via the blendConstants member of VkPipelineColorBlendStateCreateInfo.

Otherwise, to dynamically set and change the blend constant, call:

void vkCmdSetBlendConstants(
    VkCommandBuffer                             commandBuffer,
    const float                                 blendConstants[4]);
  • commandBuffer is the command buffer into which the command will be recorded.

  • blendConstants is an array of four values specifying the R, G, B, and A components of the blend constant color used in blending, depending on the blend factor.

Valid Usage
  • The bound graphics pipeline must have been created with the VK_DYNAMIC_STATE_BLEND_CONSTANTS dynamic state enabled

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics operations

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics

26.1.2. Dual-Source Blending

Blend factors that use the secondary color input (Rs1,Gs1,Bs1,As1) (VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, and VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA) may consume implementation resources that could otherwise be used for rendering to multiple color attachments. Therefore, the number of color attachments that can be used in a framebuffer may be lower when using dual-source blending.

Dual-source blending is only supported if the dualSrcBlend feature is enabled.

The maximum number of color attachments that can be used in a subpass when using dual-source blending functions is implementation-dependent and is reported as the maxFragmentDualSrcAttachments member of VkPhysicalDeviceLimits.

When using a fragment shader with dual-source blending functions, the color outputs are bound to the first and second inputs of the blender using the Index decoration, as described in Fragment Output Interface. If the second color input to the blender is not written in the shader, or if no output is bound to the second input of a blender, the result of the blending operation is not defined.

26.1.3. Blend Operations

Once the source and destination blend factors have been selected, they along with the source and destination components are passed to the blending operations. RGB and alpha components can use different operations. Possible values of VkBlendOp, specifying the operations, are:

typedef enum VkBlendOp {
    VK_BLEND_OP_ADD = 0,
    VK_BLEND_OP_SUBTRACT = 1,
    VK_BLEND_OP_REVERSE_SUBTRACT = 2,
    VK_BLEND_OP_MIN = 3,
    VK_BLEND_OP_MAX = 4,
    VK_BLEND_OP_ZERO_EXT = 1000148000,
    VK_BLEND_OP_SRC_EXT = 1000148001,
    VK_BLEND_OP_DST_EXT = 1000148002,
    VK_BLEND_OP_SRC_OVER_EXT = 1000148003,
    VK_BLEND_OP_DST_OVER_EXT = 1000148004,
    VK_BLEND_OP_SRC_IN_EXT = 1000148005,
    VK_BLEND_OP_DST_IN_EXT = 1000148006,
    VK_BLEND_OP_SRC_OUT_EXT = 1000148007,
    VK_BLEND_OP_DST_OUT_EXT = 1000148008,
    VK_BLEND_OP_SRC_ATOP_EXT = 1000148009,
    VK_BLEND_OP_DST_ATOP_EXT = 1000148010,
    VK_BLEND_OP_XOR_EXT = 1000148011,
    VK_BLEND_OP_MULTIPLY_EXT = 1000148012,
    VK_BLEND_OP_SCREEN_EXT = 1000148013,
    VK_BLEND_OP_OVERLAY_EXT = 1000148014,
    VK_BLEND_OP_DARKEN_EXT = 1000148015,
    VK_BLEND_OP_LIGHTEN_EXT = 1000148016,
    VK_BLEND_OP_COLORDODGE_EXT = 1000148017,
    VK_BLEND_OP_COLORBURN_EXT = 1000148018,
    VK_BLEND_OP_HARDLIGHT_EXT = 1000148019,
    VK_BLEND_OP_SOFTLIGHT_EXT = 1000148020,
    VK_BLEND_OP_DIFFERENCE_EXT = 1000148021,
    VK_BLEND_OP_EXCLUSION_EXT = 1000148022,
    VK_BLEND_OP_INVERT_EXT = 1000148023,
    VK_BLEND_OP_INVERT_RGB_EXT = 1000148024,
    VK_BLEND_OP_LINEARDODGE_EXT = 1000148025,
    VK_BLEND_OP_LINEARBURN_EXT = 1000148026,
    VK_BLEND_OP_VIVIDLIGHT_EXT = 1000148027,
    VK_BLEND_OP_LINEARLIGHT_EXT = 1000148028,
    VK_BLEND_OP_PINLIGHT_EXT = 1000148029,
    VK_BLEND_OP_HARDMIX_EXT = 1000148030,
    VK_BLEND_OP_HSL_HUE_EXT = 1000148031,
    VK_BLEND_OP_HSL_SATURATION_EXT = 1000148032,
    VK_BLEND_OP_HSL_COLOR_EXT = 1000148033,
    VK_BLEND_OP_HSL_LUMINOSITY_EXT = 1000148034,
    VK_BLEND_OP_PLUS_EXT = 1000148035,
    VK_BLEND_OP_PLUS_CLAMPED_EXT = 1000148036,
    VK_BLEND_OP_PLUS_CLAMPED_ALPHA_EXT = 1000148037,
    VK_BLEND_OP_PLUS_DARKER_EXT = 1000148038,
    VK_BLEND_OP_MINUS_EXT = 1000148039,
    VK_BLEND_OP_MINUS_CLAMPED_EXT = 1000148040,
    VK_BLEND_OP_CONTRAST_EXT = 1000148041,
    VK_BLEND_OP_INVERT_OVG_EXT = 1000148042,
    VK_BLEND_OP_RED_EXT = 1000148043,
    VK_BLEND_OP_GREEN_EXT = 1000148044,
    VK_BLEND_OP_BLUE_EXT = 1000148045,
} VkBlendOp;

The semantics of each basic blend operations is described in the table below:

Table 34. Basic Blend Operations
VkBlendOp RGB Components Alpha Component

VK_BLEND_OP_ADD

R = Rs0 × Sr + Rd × Dr
G = Gs0 × Sg + Gd × Dg
B = Bs0 × Sb + Bd × Db

A = As0 × Sa + Ad × Da

VK_BLEND_OP_SUBTRACT

R = Rs0 × Sr - Rd × Dr
G = Gs0 × Sg - Gd × Dg
B = Bs0 × Sb - Bd × Db

A = As0 × Sa - Ad × Da

VK_BLEND_OP_REVERSE_SUBTRACT

R = Rd × Dr - Rs0 × Sr
G = Gd × Dg - Gs0 × Sg
B = Bd × Db - Bs0 × Sb

A = Ad × Da - As0 × Sa

VK_BLEND_OP_MIN

R = min(Rs0,Rd)
G = min(Gs0,Gd)
B = min(Bs0,Bd)

A = min(As0,Ad)

VK_BLEND_OP_MAX

R = max(Rs0,Rd)
G = max(Gs0,Gd)
B = max(Bs0,Bd)

A = max(As0,Ad)

In this table, the following conventions are used:

  • Rs0, Gs0, Bs0 and As0 represent the first source color R, G, B, and A components, respectively.

  • Rd, Gd, Bd and Ad represent the R, G, B, and A components of the destination color. That is, the color currently in the corresponding color attachment for this fragment/sample.

  • Sr, Sg, Sb and Sa represent the source blend factor R, G, B, and A components, respectively.

  • Dr, Dg, Db and Da represent the destination blend factor R, G, B, and A components, respectively.

The blending operation produces a new set of values R, G, B and A, which are written to the framebuffer attachment. If blending is not enabled for this attachment, then R, G, B and A are assigned Rs0, Gs0, Bs0 and As0, respectively.

If the color attachment is fixed-point, the components of the source and destination values and blend factors are each clamped to [0,1] or [-1,1] respectively for an unsigned normalized or signed normalized color attachment prior to evaluating the blend operations. If the color attachment is floating-point, no clamping occurs.

If the numeric format of a framebuffer attachment uses sRGB encoding, the R, G, and B destination color values (after conversion from fixed-point to floating-point) are considered to be encoded for the sRGB color space and hence are linearized prior to their use in blending. Each R, G, and B component is converted from nonlinear to linear as described in the “sRGB EOTF” section of the Khronos Data Format Specification. If the format is not sRGB, no linearization is performed.

If the numeric format of a framebuffer attachment uses sRGB encoding, then the final R, G and B values are converted into the nonlinear sRGB representation before being written to the framebuffer attachment as described in the “sRGB EOTF -1” section of the Khronos Data Format Specification.

If the framebuffer color attachment numeric format is not sRGB encoded then the resulting cs values for R, G and B are unmodified. The value of A is never sRGB encoded. That is, the alpha component is always stored in memory as linear.

If the framebuffer color attachment is VK_ATTACHMENT_UNUSED, no writes are performed through that attachment. Framebuffer color attachments greater than or equal to VkSubpassDescription::colorAttachmentCount perform no writes.

26.1.4. Advanced Blend Operations

If the pNext chain of VkPipelineColorBlendStateCreateInfo includes a VkPipelineColorBlendAdvancedStateCreateInfoEXT structure, then that structure includes parameters that affect advanced blend operations.

The VkPipelineColorBlendAdvancedStateCreateInfoEXT structure is defined as:

typedef struct VkPipelineColorBlendAdvancedStateCreateInfoEXT {
    VkStructureType      sType;
    const void*          pNext;
    VkBool32             srcPremultiplied;
    VkBool32             dstPremultiplied;
    VkBlendOverlapEXT    blendOverlap;
} VkPipelineColorBlendAdvancedStateCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • srcPremultiplied specifies whether the source color of the blend operation is treated as premultiplied.

  • dstPremultiplied specifies whether the destination color of the blend operation is treated as premultiplied.

  • blendOverlap is a VkBlendOverlapEXT value specifying how the source and destination sample’s coverage is correlated.

If this structure is not present, srcPremultiplied and dstPremultiplied are both considered to be VK_TRUE, and blendOverlap is considered to be VK_BLEND_OVERLAP_UNCORRELATED_EXT.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_ADVANCED_STATE_CREATE_INFO_EXT

  • blendOverlap must be a valid VkBlendOverlapEXT value

When using one of the operations in table f/X/Y/Z Advanced Blend Operations or Hue-Saturation-Luminosity Advanced Blend Operations, blending is performed according to the following equations:

\[\begin{aligned} R & = & f(R_s',R_d')*p_0(A_s,A_d) & + & Y*R_s'*p_1(A_s,A_d) & + & Z*R_d'*p_2(A_s,A_d) \\ G & = & f(G_s',G_d')*p_0(A_s,A_d) & + & Y*G_s'*p_1(A_s,A_d) & + & Z*G_d'*p_2(A_s,A_d) \\ B & = & f(B_s',B_d')*p_0(A_s,A_d) & + & Y*B_s'*p_1(A_s,A_d) & + & Z*B_d'*p_2(A_s,A_d) \\ A & = & X*p_0(A_s,A_d) & + & Y*p_1(A_s,A_d) & + & Z*p_2(A_s,A_d) \end{aligned}\]

where the function f and terms X, Y, and Z are specified in the table. The R, G, and B components of the source color used for blending are derived according to srcPremultiplied. If srcPremultiplied is set to VK_TRUE, the fragment color components are considered to have been premultiplied by the A component prior to blending. The base source color (Rs',Gs',Bs') is obtained by dividing through by the A component:

\[\begin{aligned} (R_s', G_s', B_s') & = \begin{cases} (0, 0, 0) & A_s = 0 \\ (\frac{R_s}{A_s}, \frac{G_s}{A_s}, \frac{B_s}{A_s}) & \text{otherwise} \end{cases} \end{aligned}\]

If srcPremultiplied is VK_FALSE, the fragment color components are used as the base color:

\[\begin{aligned} (R_s', G_s', B_s') & = (R_s, G_s, B_s) \end{aligned}\]

The R, G, and B components of the destination color used for blending are derived according to dstPremultiplied. If dstPremultiplied is set to VK_TRUE, the destination components are considered to have been premultiplied by the A component prior to blending. The base destination color (Rd',Gd',Bd') is obtained by dividing through by the A component:

\[\begin{aligned} (R_d', G_d', B_d') & = \begin{cases} (0, 0, 0) & A_d = 0 \\ (\frac{R_d}{A_d}, \frac{G_d}{A_d}, \frac{B_d}{A_d}) & \text{otherwise} \end{cases} \end{aligned}\]

If dstPremultiplied is VK_FALSE, the destination color components are used as the base color:

\[\begin{aligned} (R_d', G_d', B_d') & = (R_d, G_d, B_d) \end{aligned}\]

When blending using advanced blend operations, we expect that the R, G, and B components of premultiplied source and destination color inputs be stored as the product of non-premultiplied R, G, and B component values and the A component of the color. If any R, G, or B component of a premultiplied input color is non-zero and the A component is zero, the color is considered ill-formed, and the corresponding component of the blend result is undefined.

The weighting functions p0, p1, and p2 are defined in table Advanced Blend Overlap Modes. In these functions, the A components of the source and destination colors are taken to indicate the portion of the pixel covered by the fragment (source) and the fragments previously accumulated in the pixel (destination). The functions p0, p1, and p2 approximate the relative portion of the pixel covered by the intersection of the source and destination, covered only by the source, and covered only by the destination, respectively.

Possible values of VkPipelineColorBlendAdvancedStateCreateInfoEXT::blendOverlap, specifying the blend overlap functions, are:

typedef enum VkBlendOverlapEXT {
    VK_BLEND_OVERLAP_UNCORRELATED_EXT = 0,
    VK_BLEND_OVERLAP_DISJOINT_EXT = 1,
    VK_BLEND_OVERLAP_CONJOINT_EXT = 2,
} VkBlendOverlapEXT;
  • VK_BLEND_OVERLAP_UNCORRELATED_EXT specifies that there is no correlation between the source and destination coverage.

  • VK_BLEND_OVERLAP_CONJOINT_EXT specifies that the source and destination coverage are considered to have maximal overlap.

  • VK_BLEND_OVERLAP_DISJOINT_EXT specifies that the source and destination coverage are considered to have minimal overlap.

Table 35. Advanced Blend Overlap Modes
Overlap Mode Weighting Equations

VK_BLEND_OVERLAP_UNCORRELATED_EXT

\[ \begin{aligned} p_0(A_s,A_d) & = A_sA_d \\ p_1(A_s,A_d) & = A_s(1-A_d) \\ p_2(A_s,A_d) & = A_d(1-A_s) \\ \end{aligned}\]

VK_BLEND_OVERLAP_CONJOINT_EXT

\[ \begin{aligned} p_0(A_s,A_d) & = min(A_s,A_d) \\ p_1(A_s,A_d) & = max(A_s-A_d,0) \\ p_2(A_s,A_d) & = max(A_d-A_s,0) \\ \end{aligned}\]

VK_BLEND_OVERLAP_DISJOINT_EXT

\[ \begin{aligned} p_0(A_s,A_d) & = max(A_s+A_d-1,0) \\ p_1(A_s,A_d) & = min(A_s,1-A_d) \\ p_2(A_s,A_d) & = min(A_d,1-A_s) \\ \end{aligned}\]
Table 36. f/X/Y/Z Advanced Blend Operations
Mode Blend Coefficients

VK_BLEND_OP_ZERO_EXT

\[ \begin{aligned} (X,Y,Z) & = (0,0,0) \\ f(C_s,C_d) & = 0 \end{aligned}\]

VK_BLEND_OP_SRC_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,0) \\ f(C_s,C_d) & = C_s \end{aligned}\]

VK_BLEND_OP_DST_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,1) \\ f(C_s,C_d) & = C_d \end{aligned}\]

VK_BLEND_OP_SRC_OVER_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = C_s \end{aligned}\]

VK_BLEND_OP_DST_OVER_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = C_d \end{aligned}\]

VK_BLEND_OP_SRC_IN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,0) \\ f(C_s,C_d) & = C_s \end{aligned}\]

VK_BLEND_OP_DST_IN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,0) \\ f(C_s,C_d) & = C_d \end{aligned}\]

VK_BLEND_OP_SRC_OUT_EXT

\[ \begin{aligned} (X,Y,Z) & = (0,1,0) \\ f(C_s,C_d) & = 0 \end{aligned}\]

VK_BLEND_OP_DST_OUT_EXT

\[ \begin{aligned} (X,Y,Z) & = (0,0,1) \\ f(C_s,C_d) & = 0 \end{aligned}\]

VK_BLEND_OP_SRC_ATOP_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,1) \\ f(C_s,C_d) & = C_s \end{aligned}\]

VK_BLEND_OP_DST_ATOP_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,0) \\ f(C_s,C_d) & = C_d \end{aligned}\]

VK_BLEND_OP_XOR_EXT

\[ \begin{aligned} (X,Y,Z) & = (0,1,1) \\ f(C_s,C_d) & = 0 \end{aligned}\]

VK_BLEND_OP_MULTIPLY_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = C_sC_d \end{aligned}\]

VK_BLEND_OP_SCREEN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = C_s+C_d-C_sC_d \end{aligned}\]

VK_BLEND_OP_OVERLAY_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 2 C_sC_d & C_d \leq 0.5 \\ 1-2 (1-C_s)(1-C_d) & \text{otherwise} \end{cases} \end{aligned}\]

VK_BLEND_OP_DARKEN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = min(C_s,C_d) \end{aligned}\]

VK_BLEND_OP_LIGHTEN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = max(C_s,C_d) \end{aligned}\]

VK_BLEND_OP_COLORDODGE_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 0 & C_d \leq 0 \\ min(1,\frac{C_d}{1-C_s}) & C_d \gt 0 \text{ and } C_s \lt 1 \\ 1 & C_d \gt 0 \text{ and } C_s \geq 1 \end{cases} \end{aligned}\]

VK_BLEND_OP_COLORBURN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 1 & C_d \geq 1 \\ 1 - min(1,\frac{1-C_d}{C_s}) & C_d \lt 1 \text{ and } C_s \gt 0 \\ 0 & C_d \lt 1 \text{ and } C_s \leq 0 \end{cases} \end{aligned}\]

VK_BLEND_OP_HARDLIGHT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 2 C_sC_d & C_s \leq 0.5 \\ 1-2 (1-C_s)(1-C_d) & \text{otherwise} \end{cases} \end{aligned}\]

VK_BLEND_OP_SOFTLIGHT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} C_d-(1-2 C_s)C_d(1-C_d) & C_s \leq 0.5 \\ C_d+(2 C_s-1)C_d((16 C_d-12)C_d+3) & C_s \gt 0.5 \text{ and } C_d \leq 0.25 \\ C_d+(2 C_s-1)(\sqrt{C_d}-C_d) & C_s \gt 0.5 \text{ and } C_d \gt 0.25 \end{cases} \end{aligned}\]

VK_BLEND_OP_DIFFERENCE_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \lvert C_d-C_s \rvert \end{aligned}\]

VK_BLEND_OP_EXCLUSION_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = C_s+C_d-2C_sC_d \end{aligned}\]

VK_BLEND_OP_INVERT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,1) \\ f(C_s,C_d) & = 1-C_d \end{aligned}\]

VK_BLEND_OP_INVERT_RGB_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,0,1) \\ f(C_s,C_d) & = C_s(1-C_d) \end{aligned}\]

VK_BLEND_OP_LINEARDODGE_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} C_s+C_d & C_s+C_d \leq 1 \\ 1 & \text{otherwise} \end{cases} \end{aligned}\]

VK_BLEND_OP_LINEARBURN_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} C_s+C_d-1 & C_s+C_d \gt 1 \\ 0 & \text{otherwise} \end{cases} \end{aligned}\]

VK_BLEND_OP_VIVIDLIGHT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 1-min(1,\frac{1-C_d}{2C_s}) & 0 \lt C_s \lt 0.5 \\ 0 & C_s \leq 0 \\ min(1,\frac{C_d}{2(1-C_s)}) & 0.5 \leq C_s \lt 1 \\ 1 & C_s \geq 1 \end{cases} \end{aligned}\]

VK_BLEND_OP_LINEARLIGHT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 1 & 2C_s+C_d \gt 2 \\ 2C_s+C_d-1 & 1 \lt 2C_s+C_d \leq 2 \\ 0 & 2C_s+C_d \leq 1 \end{cases} \end{aligned}\]

VK_BLEND_OP_PINLIGHT_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 0 & 2C_s-1 \gt C_d \text{ and } C_s \lt 0.5 \\ 2C_s-1 & 2C_s-1 \gt C_d \text{ and } C_s \geq 0.5 \\ 2C_s & 2C_s-1 \leq C_d \text{ and } C_s \lt 0.5C_d \\ C_d & 2C_s-1 \leq C_d \text{ and } C_s \geq 0.5C_d \end{cases} \end{aligned}\]

VK_BLEND_OP_HARDMIX_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = \begin{cases} 0 & C_s+C_d \lt 1 \\ 1 & \text{otherwise} \end{cases} \end{aligned}\]

When using one of the HSL blend operations in table Hue-Saturation-Luminosity Advanced Blend Operations as the blend operation, the RGB color components produced by the function f are effectively obtained by converting both the non-premultiplied source and destination colors to the HSL (hue, saturation, luminosity) color space, generating a new HSL color by selecting H, S, and L components from the source or destination according to the blend operation, and then converting the result back to RGB. In the equations below, a blended RGB color is produced according to the following pseudocode:

  float minv3(vec3 c) {
    return min(min(c.r, c.g), c.b);
  }
  float maxv3(vec3 c) {
    return max(max(c.r, c.g), c.b);
  }
  float lumv3(vec3 c) {
    return dot(c, vec3(0.30, 0.59, 0.11));
  }
  float satv3(vec3 c) {
    return maxv3(c) - minv3(c);
  }

  // If any color components are outside [0,1], adjust the color to
  // get the components in range.
  vec3 ClipColor(vec3 color) {
    float lum = lumv3(color);
    float mincol = minv3(color);
    float maxcol = maxv3(color);
    if (mincol < 0.0) {
      color = lum + ((color-lum)*lum) / (lum-mincol);
    }
    if (maxcol > 1.0) {
      color = lum + ((color-lum)*lum) / (maxcol-lum);
    }
    return color;
  }

  // Take the base RGB color <cbase> and override its luminosity
  // with that of the RGB color <clum>.
  vec3 SetLum(vec3 cbase, vec3 clum) {
    float lbase = lumv3(cbase);
    float llum = lumv3(clum);
    float ldiff = llum - lbase;
    vec3 color = cbase + vec3(ldiff);
    return ClipColor(color);
  }

  // Take the base RGB color <cbase> and override its saturation with
  // that of the RGB color <csat>.  The override the luminosity of the
  // result with that of the RGB color <clum>.
  vec3 SetLumSat(vec3 cbase, vec3 csat, vec3 clum)
  {
    float minbase = minv3(cbase);
    float sbase = satv3(cbase);
    float ssat = satv3(csat);
    vec3 color;
    if (sbase > 0) {
      // Equivalent (modulo rounding errors) to setting the
      // smallest (R,G,B) component to 0, the largest to <ssat>,
      // and interpolating the "middle" component based on its
      // original value relative to the smallest/largest.
      color = (cbase - minbase) * ssat / sbase;
    } else {
      color = vec3(0.0);
    }
    return SetLum(color, clum);
  }
Table 37. Hue-Saturation-Luminosity Advanced Blend Operations
Mode Result

VK_BLEND_OP_HSL_HUE_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = SetLumSat(C_s,C_d,C_d) \end{aligned}\]

VK_BLEND_OP_HSL_SATURATION_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = SetLumSat(C_d,C_s,C_d) \end{aligned}\]

VK_BLEND_OP_HSL_COLOR_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = SetLum(C_s,C_d) \end{aligned}\]

VK_BLEND_OP_HSL_LUMINOSITY_EXT

\[ \begin{aligned} (X,Y,Z) & = (1,1,1) \\ f(C_s,C_d) & = SetLum(C_d,C_s) \end{aligned}\]

When using one of the operations in table Additional RGB Blend Operations as the blend operation, the source and destination colors used by these blending operations are interpreted according to srcPremultiplied and dstPremultiplied. The blending operations below are evaluated where the RGB source and destination color components are both considered to have been premultiplied by the corresponding A component.

\[\begin{aligned} (R_s', G_s', B_s') & = \begin{cases} (R_s, G_s, B_s) & \text{if srcPremultiplied is VK\_TRUE} \\ (R_sA_s, G_sA_s, B_sA_s) & \text{if srcPremultiplied is VK\_FALSE} \end{cases} \\ (R_d', G_d', B_d') & = \begin{cases} (R_d, G_d, B_d) & \text{if dstPremultiplied is VK\_TRUE} \\ (R_dA_d, G_dA_d, B_dA_d) & \text{if dstPremultiplied is VK\_FALSE} \end{cases} \end{aligned}\]
Table 38. Additional RGB Blend Operations
Mode Result

VK_BLEND_OP_PLUS_EXT

\[ \begin{aligned} (R,G,B,A) = ( & R_s'+R_d', \\ & G_s'+G_d', \\ & B_s'+B_d', \\ & A_s+A_d) \end{aligned}\]

VK_BLEND_OP_PLUS_CLAMPED_EXT

\[ \begin{aligned} (R,G,B,A) = ( & min(1,R_s'+R_d'), \\ & min(1,G_s'+G_d'), \\ & min(1,B_s'+B_d'), \\ & min(1,A_s+A_d)) \end{aligned}\]

VK_BLEND_OP_PLUS_CLAMPED_ALPHA_EXT

\[ \begin{aligned} (R,G,B,A) = ( & min(min(1,A_s+A_d),R_s'+R_d'), \\ & min(min(1,A_s+A_d),G_s'+G_d'), \\ & min(min(1,A_s+A_d),B_s'+B_d'), \\ & min(1,A_s+A_d)) \end{aligned}\]

VK_BLEND_OP_PLUS_DARKER_EXT

\[ \begin{aligned} (R,G,B,A) = ( & max(0,min(1,A_s+A_d)-((A_s-R_s')+(A_d-R_d'))), \\ & max(0,min(1,A_s+A_d)-((A_s-G_s')+(A_d-G_d'))), \\ & max(0,min(1,A_s+A_d)-((A_s-B_s')+(A_d-B_d'))), \\ & min(1,A_s+A_d)) \end{aligned}\]

VK_BLEND_OP_MINUS_EXT

\[ \begin{aligned} (R,G,B,A) = ( & R_d'-R_s', \\ & G_d'-G_s', \\ & B_d'-B_s', \\ & A_d-A_s) \end{aligned}\]

VK_BLEND_OP_MINUS_CLAMPED_EXT

\[ \begin{aligned} (R,G,B,A) = ( & max(0,R_d'-R_s'), \\ & max(0,G_d'-G_s'), \\ & max(0,B_d'-B_s'), \\ & max(0,A_d-A_s)) \end{aligned}\]

VK_BLEND_OP_CONTRAST_EXT

\[ \begin{aligned} (R,G,B,A) = ( & \frac{A_d}{2} + 2(R_d'-\frac{A_d}{2})(R_s'-\frac{A_s}{2}), \\ & \frac{A_d}{2} + 2(G_d'-\frac{A_d}{2})(G_s'-\frac{A_s}{2}), \\ & \frac{A_d}{2} + 2(B_d'-\frac{A_d}{2})(B_s'-\frac{A_s}{2}), \\ & A_d) \end{aligned}\]

VK_BLEND_OP_INVERT_OVG_EXT

\[ \begin{aligned} (R,G,B,A) = ( & A_s(1-R_d') + (1-A_s)R_d', \\ & A_s(1-G_d') + (1-A_s)G_d', \\ & A_s(1-B_d') + (1-A_s)B_d', \\ & A_s+A_d-A_sA_d) \end{aligned}\]

VK_BLEND_OP_RED_EXT

\[ \begin{aligned} (R,G,B,A) & = (R_s', G_d', B_d', A_d) \end{aligned}\]

VK_BLEND_OP_GREEN_EXT

\[ \begin{aligned} (R,G,B,A) & = (R_d', G_s', B_d', A_d) \end{aligned}\]

VK_BLEND_OP_BLUE_EXT

\[ \begin{aligned} (R,G,B,A) & = (R_d', G_d', B_s', A_d) \end{aligned}\]

26.2. Logical Operations

The application can enable a logical operation between the fragment’s color values and the existing value in the framebuffer attachment. This logical operation is applied prior to updating the framebuffer attachment. Logical operations are applied only for signed and unsigned integer and normalized integer framebuffers. Logical operations are not applied to floating-point or sRGB format color attachments.

Logical operations are controlled by the logicOpEnable and logicOp members of VkPipelineColorBlendStateCreateInfo. If logicOpEnable is VK_TRUE, then a logical operation selected by logicOp is applied between each color attachment and the fragment’s corresponding output value, and blending of all attachments is treated as if it were disabled. Any attachments using color formats for which logical operations are not supported simply pass through the color values unmodified. The logical operation is applied independently for each of the red, green, blue, and alpha components. The logicOp is selected from the following operations:

typedef enum VkLogicOp {
    VK_LOGIC_OP_CLEAR = 0,
    VK_LOGIC_OP_AND = 1,
    VK_LOGIC_OP_AND_REVERSE = 2,
    VK_LOGIC_OP_COPY = 3,
    VK_LOGIC_OP_AND_INVERTED = 4,
    VK_LOGIC_OP_NO_OP = 5,
    VK_LOGIC_OP_XOR = 6,
    VK_LOGIC_OP_OR = 7,
    VK_LOGIC_OP_NOR = 8,
    VK_LOGIC_OP_EQUIVALENT = 9,
    VK_LOGIC_OP_INVERT = 10,
    VK_LOGIC_OP_OR_REVERSE = 11,
    VK_LOGIC_OP_COPY_INVERTED = 12,
    VK_LOGIC_OP_OR_INVERTED = 13,
    VK_LOGIC_OP_NAND = 14,
    VK_LOGIC_OP_SET = 15,
} VkLogicOp;

The logical operations supported by Vulkan are summarized in the following table in which

  • ¬ is bitwise invert,

  • is bitwise and,

  • is bitwise or,

  • is bitwise exclusive or,

  • s is the fragment’s Rs0, Gs0, Bs0 or As0 component value for the fragment output corresponding to the color attachment being updated, and

  • d is the color attachment’s R, G, B or A component value:

Table 39. Logical Operations
Mode Operation

VK_LOGIC_OP_CLEAR

0

VK_LOGIC_OP_AND

s ∧ d

VK_LOGIC_OP_AND_REVERSE

s ∧ ¬ d

VK_LOGIC_OP_COPY

s

VK_LOGIC_OP_AND_INVERTED

¬ s ∧ d

VK_LOGIC_OP_NO_OP

d

VK_LOGIC_OP_XOR

s ⊕ d

VK_LOGIC_OP_OR

s ∨ d

VK_LOGIC_OP_NOR

¬ (s ∨ d)

VK_LOGIC_OP_EQUIVALENT

¬ (s ⊕ d)

VK_LOGIC_OP_INVERT

¬ d

VK_LOGIC_OP_OR_REVERSE

s ∨ ¬ d

VK_LOGIC_OP_COPY_INVERTED

¬ s

VK_LOGIC_OP_OR_INVERTED

¬ s ∨ d

VK_LOGIC_OP_NAND

¬ (s ∧ d)

VK_LOGIC_OP_SET

all 1s

The result of the logical operation is then written to the color attachment as controlled by the component write mask, described in Blend Operations.

26.3. Color Write Mask

Bits which can be set in VkPipelineColorBlendAttachmentState::colorWriteMask to determine whether the final color values R, G, B and A are written to the framebuffer attachment are:

typedef enum VkColorComponentFlagBits {
    VK_COLOR_COMPONENT_R_BIT = 0x00000001,
    VK_COLOR_COMPONENT_G_BIT = 0x00000002,
    VK_COLOR_COMPONENT_B_BIT = 0x00000004,
    VK_COLOR_COMPONENT_A_BIT = 0x00000008,
} VkColorComponentFlagBits;
  • VK_COLOR_COMPONENT_R_BIT specifies that the R value is written to the color attachment for the appropriate sample. Otherwise, the value in memory is unmodified.

  • VK_COLOR_COMPONENT_G_BIT specifies that the G value is written to the color attachment for the appropriate sample. Otherwise, the value in memory is unmodified.

  • VK_COLOR_COMPONENT_B_BIT specifies that the B value is written to the color attachment for the appropriate sample. Otherwise, the value in memory is unmodified.

  • VK_COLOR_COMPONENT_A_BIT specifies that the A value is written to the color attachment for the appropriate sample. Otherwise, the value in memory is unmodified.

The color write mask operation is applied regardless of whether blending is enabled.

typedef VkFlags VkColorComponentFlags;

VkColorComponentFlags is a bitmask type for setting a mask of zero or more VkColorComponentFlagBits.

27. Dispatching Commands

Dispatching commands (commands with Dispatch in the name) provoke work in a compute pipeline. Dispatching commands are recorded into a command buffer and when executed by a queue, will produce work which executes according to the bound compute pipeline. A compute pipeline must be bound to a command buffer before any dispatch commands are recorded in that command buffer.

To record a dispatch, call:

void vkCmdDispatch(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    groupCountX,
    uint32_t                                    groupCountY,
    uint32_t                                    groupCountZ);
  • commandBuffer is the command buffer into which the command will be recorded.

  • groupCountX is the number of local workgroups to dispatch in the X dimension.

  • groupCountY is the number of local workgroups to dispatch in the Y dimension.

  • groupCountZ is the number of local workgroups to dispatch in the Z dimension.

When the command is executed, a global workgroup consisting of groupCountX × groupCountY × groupCountZ local workgroups is assembled.

Valid Usage
  • groupCountX must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[0]

  • groupCountY must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[1]

  • groupCountZ must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[2]

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_COMPUTE, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • A valid compute pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_COMPUTE

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE, a push constant value must have been set for VK_PIPELINE_BIND_POINT_COMPUTE, with a VkPipelineLayout that is compatible for push constants with the one used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_POINT_COMPUTE writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the compute pipeline stage in the VkPipeline object bound to VK_PIPELINE_POINT_COMPUTE reads from any image or buffer, the image or buffer must not be a protected image or protected buffer.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support compute operations

  • This command must only be called outside of a render pass instance

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Compute

Compute

To record an indirect command dispatch, call:

void vkCmdDispatchIndirect(
    VkCommandBuffer                             commandBuffer,
    VkBuffer                                    buffer,
    VkDeviceSize                                offset);
  • commandBuffer is the command buffer into which the command will be recorded.

  • buffer is the buffer containing dispatch parameters.

  • offset is the byte offset into buffer where parameters begin.

vkCmdDispatchIndirect behaves similarly to vkCmdDispatch except that the parameters are read by the device from a buffer during execution. The parameters of the dispatch are encoded in a VkDispatchIndirectCommand structure taken from buffer starting at offset.

Valid Usage
  • If buffer is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object

  • For each set n that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE, a descriptor set must have been bound to n at VK_PIPELINE_BIND_POINT_COMPUTE, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid if they are statically used by the bound VkPipeline object, specified via vkCmdBindPipeline

  • A valid compute pipeline must be bound to the current command buffer with VK_PIPELINE_BIND_POINT_COMPUTE

  • buffer must have been created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set

  • offset must be a multiple of 4

  • The sum of offset and the size of VkDispatchIndirectCommand must be less than or equal to the size of buffer

  • For each push constant that is statically used by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE, a push constant value must have been set for VK_PIPELINE_BIND_POINT_COMPUTE, with a VkPipelineLayout that is compatible for push constants with the one used to create the current VkPipeline, as described in Pipeline Layout Compatibility

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • If any VkSampler object that is accessed from a shader by the VkPipeline bound to VK_PIPELINE_BIND_POINT_COMPUTE uses unnormalized coordinates, it must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE accesses a uniform buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • If the robust buffer access feature is not enabled, and any shader stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE accesses a storage buffer, it must not access values outside of the range of that buffer specified in the bound descriptor set

  • Any VkImageView being sampled with VK_FILTER_LINEAR as a result of this command must be of a format which supports linear filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must be of a format which supports cubic filtering, as specified by the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG flag in VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures returned by vkGetAndroidHardwareBufferPropertiesANDROID for external format images, or by VkFormatProperties::linearTilingFeatures or VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties for non-external format linearly or optimally tiled images, respectively

  • Any VkImageView being sampled with VK_FILTER_CUBIC_IMG as a result of this command must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • If commandBuffer is an unprotected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_BIND_POINT_COMPUTE reads from or writes to any image or buffer, that image or buffer must not be a protected image or protected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage in the VkPipeline object bound to VK_PIPELINE_POINT_COMPUTE writes to any image or buffer, that image or buffer must not be an unprotected image or unprotected buffer.

  • If commandBuffer is a protected command buffer, and any pipeline stage other than the compute pipeline stage in the VkPipeline object bound to VK_PIPELINE_POINT_COMPUTE reads from any image or buffer, the image or buffer must not be a protected image or protected buffer.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • buffer must be a valid VkBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support compute operations

  • This command must only be called outside of a render pass instance

  • Both of buffer, and commandBuffer must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Compute

Compute

The VkDispatchIndirectCommand structure is defined as:

typedef struct VkDispatchIndirectCommand {
    uint32_t    x;
    uint32_t    y;
    uint32_t    z;
} VkDispatchIndirectCommand;
  • x is the number of local workgroups to dispatch in the X dimension.

  • y is the number of local workgroups to dispatch in the Y dimension.

  • z is the number of local workgroups to dispatch in the Z dimension.

The members of VkDispatchIndirectCommand have the same meaning as the corresponding parameters of vkCmdDispatch.

Valid Usage
  • x must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[0]

  • y must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[1]

  • z must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[2]

To record a dispatch using non-zero base values for the components of WorkgroupId, call:

void vkCmdDispatchBase(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    baseGroupX,
    uint32_t                                    baseGroupY,
    uint32_t                                    baseGroupZ,
    uint32_t                                    groupCountX,
    uint32_t                                    groupCountY,
    uint32_t                                    groupCountZ);

or the equivalent command

void vkCmdDispatchBaseKHR(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    baseGroupX,
    uint32_t                                    baseGroupY,
    uint32_t                                    baseGroupZ,
    uint32_t                                    groupCountX,
    uint32_t                                    groupCountY,
    uint32_t                                    groupCountZ);
  • commandBuffer is the command buffer into which the command will be recorded.

  • baseGroupX is the start value for the X component of WorkgroupId.

  • baseGroupY is the start value for the Y component of WorkgroupId.

  • baseGroupZ is the start value for the Z component of WorkgroupId.

  • groupCountX is the number of local workgroups to dispatch in the X dimension.

  • groupCountY is the number of local workgroups to dispatch in the Y dimension.

  • groupCountZ is the number of local workgroups to dispatch in the Z dimension.

When the command is executed, a global workgroup consisting of groupCountX × groupCountY × groupCountZ local workgroups is assembled, with WorkgroupId values ranging from [baseGroup, baseGroup + groupCount) in each component. vkCmdDispatch is equivalent to vkCmdDispatchBase(0,0,0,groupCountX,groupCountY,groupCountZ).

Valid Usage
  • All valid usage rules from vkCmdDispatch apply

  • baseGroupX must be less than VkPhysicalDeviceLimits::maxComputeWorkGroupCount[0]

  • baseGroupX must be less than VkPhysicalDeviceLimits::maxComputeWorkGroupCount[1]

  • baseGroupZ must be less than VkPhysicalDeviceLimits::maxComputeWorkGroupCount[2]

  • groupCountX must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[0] minus baseGroupX

  • groupCountY must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[1] minus baseGroupY

  • groupCountZ must be less than or equal to VkPhysicalDeviceLimits::maxComputeWorkGroupCount[2] minus baseGroupZ

  • If any of baseGroupX, baseGroupY, or baseGroupZ are not zero, then the bound compute pipeline must have been created with the VK_PIPELINE_CREATE_DISPATCH_BASE flag.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support compute operations

  • This command must only be called outside of a render pass instance

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Outside

Compute

28. Device-Generated Commands

This chapter discusses the generation of command buffer content on the device. These principle steps are to be taken to generate commands on the device:

  • Make resource bindings accessible for the device via registering in an VkObjectTableNVX.

  • Define via VkIndirectCommandsLayoutNVX the sequence of commands which should be generated.

  • Fill one or more VkBuffer with the appropriate content that gets interpreted by VkIndirectCommandsLayoutNVX.

  • Reserve command space via vkCmdReserveSpaceForCommandsNVX in a secondary VkCommandBuffer where the generated commands should be recorded.

  • Generate the actual commands via vkCmdProcessCommandsNVX passing all required data.

Execution of such generated commands can either be triggered directly with the generation process, or by executing the secondary VkCommandBuffer that was chosen as optional target. The latter allows re-using generated commands as well. Similar to VkDescriptorSet, special care should be taken for the lifetime of resources referenced in VkObjectTableNVX, which may be accessed at either generation or execution time.

vkCmdProcessCommandsNVX executes in a separate logical pipeline from either graphics or compute. When generating commands into a secondary command buffer, the command generation must be explicitly synchronized against the secondary command buffer’s execution. When not using a secondary command buffer, the command generation is automatically synchronized against the command execution.

28.1. Features and Limitations

To query the support of related features and limitations, call:

void vkGetPhysicalDeviceGeneratedCommandsPropertiesNVX(
    VkPhysicalDevice                            physicalDevice,
    VkDeviceGeneratedCommandsFeaturesNVX*       pFeatures,
    VkDeviceGeneratedCommandsLimitsNVX*         pLimits);
  • physicalDevice is the handle to the physical device whose properties will be queried.

  • pFeatures points to an instance of the VkDeviceGeneratedCommandsFeaturesNVX structure, that will be filled with returned information.

  • pLimits points to an instance of the VkDeviceGeneratedCommandsLimitsNVX structure, that will be filled with returned information.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pFeatures must be a valid pointer to a VkDeviceGeneratedCommandsFeaturesNVX structure

  • pLimits must be a valid pointer to a VkDeviceGeneratedCommandsLimitsNVX structure

The VkDeviceGeneratedCommandsFeaturesNVX structure is defined as:

typedef struct VkDeviceGeneratedCommandsFeaturesNVX {
    VkStructureType    sType;
    const void*        pNext;
    VkBool32           computeBindingPointSupport;
} VkDeviceGeneratedCommandsFeaturesNVX;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • computeBindingPointSupport specifies whether the VkObjectTableNVX supports entries with VK_OBJECT_ENTRY_USAGE_GRAPHICS_BIT_NVX bit set and VkIndirectCommandsLayoutNVX supports VK_PIPELINE_BIND_POINT_COMPUTE.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_FEATURES_NVX

  • pNext must be NULL

The VkDeviceGeneratedCommandsLimitsNVX structure is defined as:

typedef struct VkDeviceGeneratedCommandsLimitsNVX {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           maxIndirectCommandsLayoutTokenCount;
    uint32_t           maxObjectEntryCounts;
    uint32_t           minSequenceCountBufferOffsetAlignment;
    uint32_t           minSequenceIndexBufferOffsetAlignment;
    uint32_t           minCommandsTokenBufferOffsetAlignment;
} VkDeviceGeneratedCommandsLimitsNVX;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • maxIndirectCommandsLayoutTokenCount the maximum number of tokens in VkIndirectCommandsLayoutNVX.

  • maxObjectEntryCounts the maximum number of entries per resource type in VkObjectTableNVX.

  • minSequenceCountBufferOffsetAlignment the minimum alignment for memory addresses optionally used in vkCmdProcessCommandsNVX.

  • minSequenceIndexBufferOffsetAlignment the minimum alignment for memory addresses optionally used in vkCmdProcessCommandsNVX.

  • minCommandsTokenBufferOffsetAlignment the minimum alignment for memory addresses optionally used in vkCmdProcessCommandsNVX.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_LIMITS_NVX

  • pNext must be NULL

28.2. Binding Object Table

The device-side bindings are registered inside a table:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkObjectTableNVX)

This is required as the CPU-side object pointers, for example when binding a VkPipeline or VkDescriptorSet, cannot be used by the device. The combination of VkObjectTableNVX and uint32_t table indices stored inside a VkBuffer serve that purpose during device command generation.

At creation time the table is defined with a fixed amount of registration slots for the individual resource types. A detailed resource binding can then later be registered via vkRegisterObjectsNVX at any uint32_t index below the allocated maximum.

28.2.1. Table Creation

To create object tables, call:

VkResult vkCreateObjectTableNVX(
    VkDevice                                    device,
    const VkObjectTableCreateInfoNVX*           pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkObjectTableNVX*                           pObjectTable);
  • device is the logical device that creates the object table.

  • pCreateInfo is a pointer to an instance of the VkObjectTableCreateInfoNVX structure containing parameters affecting creation of the table.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pObjectTable points to a VkObjectTableNVX handle in which the resulting object table is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkObjectTableCreateInfoNVX structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pObjectTable must be a valid pointer to a VkObjectTableNVX handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkObjectTableCreateInfoNVX structure is defined as:

typedef struct VkObjectTableCreateInfoNVX {
    VkStructureType                      sType;
    const void*                          pNext;
    uint32_t                             objectCount;
    const VkObjectEntryTypeNVX*          pObjectEntryTypes;
    const uint32_t*                      pObjectEntryCounts;
    const VkObjectEntryUsageFlagsNVX*    pObjectEntryUsageFlags;
    uint32_t                             maxUniformBuffersPerDescriptor;
    uint32_t                             maxStorageBuffersPerDescriptor;
    uint32_t                             maxStorageImagesPerDescriptor;
    uint32_t                             maxSampledImagesPerDescriptor;
    uint32_t                             maxPipelineLayouts;
} VkObjectTableCreateInfoNVX;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectCount is the number of entry configurations that the object table supports.

  • pObjectEntryTypes is an array of VkObjectEntryTypeNVX values providing the entry type of a given configuration.

  • pObjectEntryCounts is an array of counts of how many objects can be registered in the table.

  • pObjectEntryUsageFlags is an array of bitmasks of VkObjectEntryUsageFlagBitsNVX specifying the binding usage of the entry.

  • maxUniformBuffersPerDescriptor is the maximum number of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC used by any single registered VkDescriptorSet in this table.

  • maxStorageBuffersPerDescriptor is the maximum number of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC used by any single registered VkDescriptorSet in this table.

  • maxStorageImagesPerDescriptor is the maximum number of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER used by any single registered VkDescriptorSet in this table.

  • maxSampledImagesPerDescriptor is the maximum number of VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT used by any single registered VkDescriptorSet in this table.

  • maxPipelineLayouts is the maximum number of unique VkPipelineLayout used by any registered VkDescriptorSet or VkPipeline in this table.

Valid Usage
  • If the VkDeviceGeneratedCommandsFeaturesNVX::computeBindingPointSupport feature is not enabled, pObjectEntryUsageFlags must not contain VK_OBJECT_ENTRY_USAGE_COMPUTE_BIT_NVX

  • Any value within pObjectEntryCounts must not exceed VkDeviceGeneratedCommandsLimitsNVX::maxObjectEntryCounts

  • maxUniformBuffersPerDescriptor must be within the limits supported by the device.

  • maxStorageBuffersPerDescriptor must be within the limits supported by the device.

  • maxStorageImagesPerDescriptor must be within the limits supported by the device.

  • maxSampledImagesPerDescriptor must be within the limits supported by the device.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_OBJECT_TABLE_CREATE_INFO_NVX

  • pNext must be NULL

  • pObjectEntryTypes must be a valid pointer to an array of objectCount valid VkObjectEntryTypeNVX values

  • pObjectEntryCounts must be a valid pointer to an array of objectCount uint32_t values

  • pObjectEntryUsageFlags must be a valid pointer to an array of objectCount valid combinations of VkObjectEntryUsageFlagBitsNVX values

  • Each element of pObjectEntryUsageFlags must not be 0

  • objectCount must be greater than 0

Possible values of elements of the VkObjectTableCreateInfoNVX::pObjectEntryTypes array, specifying the entry type of a configuration, are:

typedef enum VkObjectEntryTypeNVX {
    VK_OBJECT_ENTRY_TYPE_DESCRIPTOR_SET_NVX = 0,
    VK_OBJECT_ENTRY_TYPE_PIPELINE_NVX = 1,
    VK_OBJECT_ENTRY_TYPE_INDEX_BUFFER_NVX = 2,
    VK_OBJECT_ENTRY_TYPE_VERTEX_BUFFER_NVX = 3,
    VK_OBJECT_ENTRY_TYPE_PUSH_CONSTANT_NVX = 4,
} VkObjectEntryTypeNVX;
  • VK_OBJECT_ENTRY_TYPE_DESCRIPTOR_SET_NVX specifies a VkDescriptorSet resource entry that is registered via VkObjectTableDescriptorSetEntryNVX.

  • VK_OBJECT_ENTRY_TYPE_PIPELINE_NVX specifies a VkPipeline resource entry that is registered via VkObjectTablePipelineEntryNVX.

  • VK_OBJECT_ENTRY_TYPE_INDEX_BUFFER_NVX specifies a VkBuffer resource entry that is registered via VkObjectTableIndexBufferEntryNVX.

  • VK_OBJECT_ENTRY_TYPE_VERTEX_BUFFER_NVX specifies a VkBuffer resource entry that is registered via VkObjectTableVertexBufferEntryNVX.

  • VK_OBJECT_ENTRY_TYPE_PUSH_CONSTANT_NVX specifies the resource entry is registered via VkObjectTablePushConstantEntryNVX.

Bits which can be set in elements of the VkObjectTableCreateInfoNVX::pObjectEntryUsageFlags array, specifying binding usage of an entry, are:

typedef enum VkObjectEntryUsageFlagBitsNVX {
    VK_OBJECT_ENTRY_USAGE_GRAPHICS_BIT_NVX = 0x00000001,
    VK_OBJECT_ENTRY_USAGE_COMPUTE_BIT_NVX = 0x00000002,
} VkObjectEntryUsageFlagBitsNVX;
  • VK_OBJECT_ENTRY_USAGE_GRAPHICS_BIT_NVX specifies that the resource is bound to VK_PIPELINE_BIND_POINT_GRAPHICS

  • VK_OBJECT_ENTRY_USAGE_COMPUTE_BIT_NVX specifies that the resource is bound to VK_PIPELINE_BIND_POINT_COMPUTE

typedef VkFlags VkObjectEntryUsageFlagsNVX;

VkObjectEntryUsageFlagsNVX is a bitmask type for setting a mask of zero or more VkObjectEntryUsageFlagBitsNVX.

To destroy an object table, call:

void vkDestroyObjectTableNVX(
    VkDevice                                    device,
    VkObjectTableNVX                            objectTable,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the table.

  • objectTable is the table to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to objectTable must have completed execution.

  • If VkAllocationCallbacks were provided when objectTable was created, a compatible set of callbacks must be provided here.

  • If no VkAllocationCallbacks were provided when objectTable was created, pAllocator must be NULL.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • objectTable must be a valid VkObjectTableNVX handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • objectTable must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to objectTable must be externally synchronized

28.2.2. Registering Objects

Resource bindings of Vulkan objects are registered at an arbitrary uint32_t index within an object table. As long as the object table references such objects, they must not be deleted.

VkResult vkRegisterObjectsNVX(
    VkDevice                                    device,
    VkObjectTableNVX                            objectTable,
    uint32_t                                    objectCount,
    const VkObjectTableEntryNVX* const*         ppObjectTableEntries,
    const uint32_t*                             pObjectIndices);
  • device is the logical device that creates the object table.

  • objectTable is the table for which the resources are registered.

  • objectCount is the number of resources to register.

  • ppObjectTableEntries provides an array for detailed binding informations, each array element is a pointer to a struct of type VkObjectTablePipelineEntryNVX, VkObjectTableDescriptorSetEntryNVX, VkObjectTableVertexBufferEntryNVX, VkObjectTableIndexBufferEntryNVX or VkObjectTablePushConstantEntryNVX (see below for details).

  • pObjectIndices are the indices at which each resource is registered.

Valid Usage
  • The contents of pObjectTableEntry must yield plausible bindings supported by the device.

  • At any pObjectIndices there must not be a registered resource already.

  • Any value inside pObjectIndices must be below the appropriate VkObjectTableCreateInfoNVX::pObjectEntryCounts limits provided at objectTable creation time.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • objectTable must be a valid VkObjectTableNVX handle

  • ppObjectTableEntries must be a valid pointer to an array of objectCount valid VkObjectTableEntryNVX structures

  • pObjectIndices must be a valid pointer to an array of objectCount uint32_t values

  • objectCount must be greater than 0

  • objectTable must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to objectTable must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Common to all resource entries are:

typedef struct VkObjectTableEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
} VkObjectTableEntryNVX;
  • type defines the entry type

  • flags defines which VkPipelineBindPoint the resource can be used with. Some entry types allow only a single flag to be set.

Valid Usage
  • If the VkDeviceGeneratedCommandsFeaturesNVX::computeBindingPointSupport feature is not enabled, flags must not contain VK_OBJECT_ENTRY_USAGE_COMPUTE_BIT_NVX

Valid Usage (Implicit)
typedef struct VkObjectTablePipelineEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
    VkPipeline                    pipeline;
} VkObjectTablePipelineEntryNVX;
  • pipeline specifies the VkPipeline that this resource entry references.

Valid Usage
  • type must be VK_OBJECT_ENTRY_TYPE_PIPELINE_NVX

Valid Usage (Implicit)
typedef struct VkObjectTableDescriptorSetEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
    VkPipelineLayout              pipelineLayout;
    VkDescriptorSet               descriptorSet;
} VkObjectTableDescriptorSetEntryNVX;
  • pipelineLayout specifies the VkPipelineLayout that the descriptorSet is used with.

  • descriptorSet specifies the VkDescriptorSet that can be bound with this entry.

Valid Usage
  • type must be VK_OBJECT_ENTRY_TYPE_DESCRIPTOR_SET_NVX

Valid Usage (Implicit)
  • type must be a valid VkObjectEntryTypeNVX value

  • flags must be a valid combination of VkObjectEntryUsageFlagBitsNVX values

  • flags must not be 0

  • pipelineLayout must be a valid VkPipelineLayout handle

  • descriptorSet must be a valid VkDescriptorSet handle

  • Both of descriptorSet, and pipelineLayout must have been created, allocated, or retrieved from the same VkDevice

typedef struct VkObjectTableVertexBufferEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
    VkBuffer                      buffer;
} VkObjectTableVertexBufferEntryNVX;
  • buffer specifies the VkBuffer that can be bound as vertex bufer

Valid Usage
  • type must be VK_OBJECT_ENTRY_TYPE_VERTEX_BUFFER_NVX

Valid Usage (Implicit)
typedef struct VkObjectTableIndexBufferEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
    VkBuffer                      buffer;
    VkIndexType                   indexType;
} VkObjectTableIndexBufferEntryNVX;
  • buffer specifies the VkBuffer that can be bound as index buffer

  • indexType specifies the VkIndexType used with this index buffer

Valid Usage
  • type must be VK_OBJECT_ENTRY_TYPE_INDEX_BUFFER_NVX

Valid Usage (Implicit)
typedef struct VkObjectTablePushConstantEntryNVX {
    VkObjectEntryTypeNVX          type;
    VkObjectEntryUsageFlagsNVX    flags;
    VkPipelineLayout              pipelineLayout;
    VkShaderStageFlags            stageFlags;
} VkObjectTablePushConstantEntryNVX;
  • pipelineLayout specifies the VkPipelineLayout that the pushconstants are used with

  • stageFlags specifies the VkShaderStageFlags that the pushconstants are used with

Valid Usage
  • type must be VK_OBJECT_ENTRY_TYPE_PUSH_CONSTANT_NVX

Valid Usage (Implicit)

Use the following command to unregister resources from an object table:

VkResult vkUnregisterObjectsNVX(
    VkDevice                                    device,
    VkObjectTableNVX                            objectTable,
    uint32_t                                    objectCount,
    const VkObjectEntryTypeNVX*                 pObjectEntryTypes,
    const uint32_t*                             pObjectIndices);
  • device is the logical device that creates the object table.

  • objectTable is the table from which the resources are unregistered.

  • objectCount is the number of resources being removed from the object table.

  • pObjectEntryType provides an array of VkObjectEntryTypeNVX for the resources being removed.

  • pObjectIndices provides the array of object indices to be removed.

Valid Usage
  • At any pObjectIndices there must be a registered resource already.

  • The pObjectEntryTypes of the resource at pObjectIndices must match.

  • All operations on the device using the registered resource must have been completed.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • objectTable must be a valid VkObjectTableNVX handle

  • pObjectEntryTypes must be a valid pointer to an array of objectCount valid VkObjectEntryTypeNVX values

  • pObjectIndices must be a valid pointer to an array of objectCount uint32_t values

  • objectCount must be greater than 0

  • objectTable must have been created, allocated, or retrieved from device

Host Synchronization
  • Host access to objectTable must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

28.3. Indirect Commands Layout

The device-side command generation happens through an iterative processing of an atomic sequence comprised of command tokens, which are represented by:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkIndirectCommandsLayoutNVX)

28.3.1. Tokenized Command Processing

The processing is in principle illustrated below:

void cmdProcessSequence(cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, s)
{
  for (c = 0; c < indirectCommandsLayout.tokenCount; c++)
  {
    indirectCommandsLayout.pTokens[c].command (cmd, objectTable, pIndirectCommandsTokens[c], s);
  }
}

void cmdProcessAllSequences(cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, sequencesCount)
{
  for (s = 0; s < sequencesCount; s++)
  {
    cmdProcessSequence(cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, s);
  }
}

The processing of each sequence is considered stateless, therefore all state changes must occur prior work provoking commands within the sequence. A single sequence is either strictly targeting VK_PIPELINE_BIND_POINT_GRAPHICS or VK_PIPELINE_BIND_POINT_COMPUTE.

The primary input data for each token is provided through VkBuffer content at command generation time using vkCmdProcessCommandsNVX, however some functional arguments, for example binding sets, are specified at layout creation time. The input size is different for each token.

Possible values of those elements of the VkIndirectCommandsLayoutCreateInfoNVX::pTokens array which specify command tokens (other elements of the array specify command parameters) are:

typedef enum VkIndirectCommandsTokenTypeNVX {
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_PIPELINE_NVX = 0,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_DESCRIPTOR_SET_NVX = 1,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_INDEX_BUFFER_NVX = 2,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_VERTEX_BUFFER_NVX = 3,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_PUSH_CONSTANT_NVX = 4,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_INDEXED_NVX = 5,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_NVX = 6,
    VK_INDIRECT_COMMANDS_TOKEN_TYPE_DISPATCH_NVX = 7,
} VkIndirectCommandsTokenTypeNVX;
Table 40. Supported indirect command tokens
Token type Equivalent command

VK_INDIRECT_COMMANDS_TOKEN_TYPE_PIPELINE_NVX

vkCmdBindPipeline

VK_INDIRECT_COMMANDS_TOKEN_TYPE_DESCRIPTOR_SET_NVX

vkCmdBindDescriptorSets

VK_INDIRECT_COMMANDS_TOKEN_TYPE_INDEX_BUFFER_NVX

vkCmdBindIndexBuffer

VK_INDIRECT_COMMANDS_TOKEN_TYPE_VERTEX_BUFFER_NVX

vkCmdBindVertexBuffers

VK_INDIRECT_COMMANDS_TOKEN_TYPE_PUSH_CONSTANT_NVX

vkCmdPushConstants

VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_INDEXED_NVX

vkCmdDrawIndexedIndirect

VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_NVX

vkCmdDrawIndirect

VK_INDIRECT_COMMANDS_TOKEN_TYPE_DISPATCH_NVX

vkCmdDispatchIndirect

The VkIndirectCommandsLayoutTokenNVX structure specifies details to the function arguments that need to be known at layout creation time:

typedef struct VkIndirectCommandsLayoutTokenNVX {
    VkIndirectCommandsTokenTypeNVX    tokenType;
    uint32_t                          bindingUnit;
    uint32_t                          dynamicCount;
    uint32_t                          divisor;
} VkIndirectCommandsLayoutTokenNVX;
  • type specifies the token command type.

  • bindingUnit has a different meaning depending on the type, please refer pseudo code further down for details.

  • dynamicCount has a different meaning depending on the type, please refer pseudo code further down for details.

  • divisor defines the rate at which the input data buffers are accessed.

Valid Usage
  • bindingUnit must stay within device supported limits for the appropriate commands.

  • dynamicCount must stay within device supported limits for the appropriate commands.

  • divisor must be greater than 0 and a power of two.

Valid Usage (Implicit)

The VkIndirectCommandsTokenNVX structure specifies the input data for a token at processing time.

typedef struct VkIndirectCommandsTokenNVX {
    VkIndirectCommandsTokenTypeNVX    tokenType;
    VkBuffer                          buffer;
    VkDeviceSize                      offset;
} VkIndirectCommandsTokenNVX;
  • tokenType specifies the token command type.

  • buffer specifies the VkBuffer storing the functional arguments for each squence. These argumetns can be written by the device.

  • offset specified an offset into buffer where the arguments start.

Valid Usage
  • The buffer’s usage flag must have the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set.

  • The offset must be aligned to VkDeviceGeneratedCommandsLimitsNVX::minCommandsTokenBufferOffsetAlignment.

Valid Usage (Implicit)

The following code provides detailed information on how an individual sequence is processed:

void cmdProcessSequence(cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, s)
{
  for (uint32_t c = 0; c < indirectCommandsLayout.tokenCount; c++){
    input   = pIndirectCommandsTokens[c];
    i       = s / indirectCommandsLayout.pTokens[c].divisor;

    switch(input.type){
      VK_INDIRECT_COMMANDS_TOKEN_TYPE_PIPELINE_NVX:
        size_t    stride  = sizeof(uint32_t);
        uint32_t* data    = input.buffer.pointer( input.offset + stride * i );
        uint32_t  object  = data[0];

        vkCmdBindPipeline(cmd, indirectCommandsLayout.pipelineBindPoint,
          objectTable.pipelines[ object ].pipeline);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_DESCRIPTOR_SET_NVX:
        size_t    stride  = sizeof(uint32_t) + sizeof(uint32_t) * indirectCommandsLayout.pTokens[c].dynamicCount;
        uint32_t* data    = input.buffer.pointer( input.offset + stride * i);
        uint32_t  object  = data[0];

        vkCmdBindDescriptorSets(cmd, indirectCommandsLayout.pipelineBindPoint,
          objectTable.descriptorsets[ object ].layout,
          indirectCommandsLayout.pTokens[ c ].bindingUnit,
          1, &objectTable.descriptorsets[ object ].descriptorSet,
          indirectCommandsLayout.pTokens[ c ].dynamicCount, data + 1);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_PUSH_CONSTANT_NVX:
        size_t    stride  = sizeof(uint32_t) + indirectCommandsLayout.pTokens[c].dynamicCount;
        uint32_t* data    = input.buffer.pointer( input.offset + stride * i );
        uint32_t  object  = data[0];

        vkCmdPushConstants(cmd,
          objectTable.pushconstants[ object ].layout,
          objectTable.pushconstants[ object ].stageFlags,
          indirectCommandsLayout.pTokens[ c ].bindingUnit, indirectCommandsLayout.pTokens[c].dynamicCount, data + 1);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_INDEX_BUFFER_NVX:
        size_t   s tride  = sizeof(uint32_t) + sizeof(uint32_t) * indirectCommandsLayout.pTokens[c].dynamicCount;
        uint32_t* data    = input.buffer.pointer( input.offset + stride * i );
        uint32_t  object  = data[0];

        vkCmdBindIndexBuffer(cmd,
          objectTable.vertexbuffers[ object ].buffer,
          indirectCommandsLayout.pTokens[ c ].dynamicCount ? data[1] : 0,
          objectTable.vertexbuffers[ object ].indexType);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_VERTEX_BUFFER_NVX:
        size_t    stride  = sizeof(uint32_t) + sizeof(uint32_t) * indirectCommandsLayout.pTokens[c].dynamicCount;
        uint32_t* data    = input.buffer.pointer( input.offset + stride * i );
        uint32_t  object  = data[0];

        vkCmdBindVertexBuffers(cmd,
          indirectCommandsLayout.pTokens[ c ].bindingUnit, 1,
          &objectTable.vertexbuffers[ object ].buffer,
          indirectCommandsLayout.pTokens[ c ].dynamicCount ? data + 1 : {0}); // device size handled as uint32_t
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_INDEXED_NVX:
        vkCmdDrawIndexedIndirect(cmd,
          input.buffer,
          sizeof(VkDrawIndexedIndirectCommand) * i + input.offset, 1, 0);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_NVX:
        vkCmdDrawIndirect(cmd,
          input.buffer,
          sizeof(VkDrawIndirectCommand) * i  + input.offset, 1, 0);
      break;

      VK_INDIRECT_COMMANDS_TOKEN_TYPE_DISPATCH_NVX:
        vkCmdDispatchIndirect(cmd,
          input.buffer,
          sizeof(VkDispatchIndirectCommand) * i  + input.offset);
      break;
    }
  }
}

28.3.2. Creation and Deletion

Indirect command layouts are created by:

VkResult vkCreateIndirectCommandsLayoutNVX(
    VkDevice                                    device,
    const VkIndirectCommandsLayoutCreateInfoNVX* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkIndirectCommandsLayoutNVX*                pIndirectCommandsLayout);
  • device is the logical device that creates the indirect command layout.

  • pCreateInfo is a pointer to an instance of the VkIndirectCommandsLayoutCreateInfoNVX structure containing parameters affecting creation of the indirect command layout.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pIndirectCommandsLayout points to a VkIndirectCommandsLayoutNVX handle in which the resulting indirect command layout is returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkIndirectCommandsLayoutCreateInfoNVX structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pIndirectCommandsLayout must be a valid pointer to a VkIndirectCommandsLayoutNVX handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkIndirectCommandsLayoutCreateInfoNVX structure is defined as:

typedef struct VkIndirectCommandsLayoutCreateInfoNVX {
    VkStructureType                            sType;
    const void*                                pNext;
    VkPipelineBindPoint                        pipelineBindPoint;
    VkIndirectCommandsLayoutUsageFlagsNVX      flags;
    uint32_t                                   tokenCount;
    const VkIndirectCommandsLayoutTokenNVX*    pTokens;
} VkIndirectCommandsLayoutCreateInfoNVX;

The following code illustrates some of the key flags:

void cmdProcessAllSequences(cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, sequencesCount, indexbuffer, indexbufferoffset)
{
  for (s = 0; s < sequencesCount; s++)
  {
    sequence = s;

    if (indirectCommandsLayout.flags & VK_INDIRECT_COMMANDS_LAYOUT_USAGE_UNORDERED_SEQUENCES_BIT_NVX) {
      sequence = incoherent_implementation_dependent_permutation[ sequence ];
    }
    if (indirectCommandsLayout.flags & VK_INDIRECT_COMMANDS_LAYOUT_USAGE_INDEXED_SEQUENCES_BIT_NVX) {
      sequence = indexbuffer.load_uint32( sequence * sizeof(uint32_t) + indexbufferoffset);
    }

    cmdProcessSequence( cmd, objectTable, indirectCommandsLayout, pIndirectCommandsTokens, sequence );
  }
}
Valid Usage
  • tokenCount must be greater than 0 and below VkDeviceGeneratedCommandsLimitsNVX::maxIndirectCommandsLayoutTokenCount

  • If the VkDeviceGeneratedCommandsFeaturesNVX::computeBindingPointSupport feature is not enabled, then pipelineBindPoint must not be VK_PIPELINE_BIND_POINT_COMPUTE

  • If pTokens contains an entry of VK_INDIRECT_COMMANDS_TOKEN_TYPE_PIPELINE_NVX it must be the first element of the array and there must be only a single element of such token type.

  • All state binding tokens in pTokens must occur prior work provoking tokens (VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_NVX, VK_INDIRECT_COMMANDS_TOKEN_TYPE_DRAW_INDEXED_NVX, VK_INDIRECT_COMMANDS_TOKEN_TYPE_DISPATCH_NVX).

  • The content of pTokens must include one single work provoking token that is compatible with the pipelineBindPoint.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_INDIRECT_COMMANDS_LAYOUT_CREATE_INFO_NVX

  • pNext must be NULL

  • pipelineBindPoint must be a valid VkPipelineBindPoint value

  • flags must be a valid combination of VkIndirectCommandsLayoutUsageFlagBitsNVX values

  • flags must not be 0

  • pTokens must be a valid pointer to an array of tokenCount valid VkIndirectCommandsLayoutTokenNVX structures

  • tokenCount must be greater than 0

Bits which can be set in VkIndirectCommandsLayoutCreateInfoNVX::flags, specifying usage hints of an indirect command layout, are:

typedef enum VkIndirectCommandsLayoutUsageFlagBitsNVX {
    VK_INDIRECT_COMMANDS_LAYOUT_USAGE_UNORDERED_SEQUENCES_BIT_NVX = 0x00000001,
    VK_INDIRECT_COMMANDS_LAYOUT_USAGE_SPARSE_SEQUENCES_BIT_NVX = 0x00000002,
    VK_INDIRECT_COMMANDS_LAYOUT_USAGE_EMPTY_EXECUTIONS_BIT_NVX = 0x00000004,
    VK_INDIRECT_COMMANDS_LAYOUT_USAGE_INDEXED_SEQUENCES_BIT_NVX = 0x00000008,
} VkIndirectCommandsLayoutUsageFlagBitsNVX;
  • VK_INDIRECT_COMMANDS_LAYOUT_USAGE_UNORDERED_SEQUENCES_BIT_NVX specifies that the processing of sequences can happen at an implementation-dependent order, which is not guaranteed to be coherent across multiple invocations.

  • VK_INDIRECT_COMMANDS_LAYOUT_USAGE_SPARSE_SEQUENCES_BIT_NVX specifies that there is likely a high difference between allocated number of sequences and actually used.

  • VK_INDIRECT_COMMANDS_LAYOUT_USAGE_EMPTY_EXECUTIONS_BIT_NVX specifies that there are likely many draw or dispatch calls that are zero-sized (zero grid dimension, no primitives to render).

  • VK_INDIRECT_COMMANDS_LAYOUT_USAGE_INDEXED_SEQUENCES_BIT_NVX specifies that the input data for the sequences is not implicitly indexed from 0..sequencesUsed but a user provided VkBuffer encoding the index is provided.

typedef VkFlags VkIndirectCommandsLayoutUsageFlagsNVX;

VkIndirectCommandsLayoutUsageFlagsNVX is a bitmask type for setting a mask of zero or more VkIndirectCommandsLayoutUsageFlagBitsNVX.

Indirect command layouts are destroyed by:

void vkDestroyIndirectCommandsLayoutNVX(
    VkDevice                                    device,
    VkIndirectCommandsLayoutNVX                 indirectCommandsLayout,
    const VkAllocationCallbacks*                pAllocator);
  • device is the logical device that destroys the layout.

  • indirectCommandsLayout is the table to destroy.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • All submitted commands that refer to indirectCommandsLayout must have completed execution

  • If VkAllocationCallbacks were provided when objectTable was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when objectTable was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • indirectCommandsLayout must be a valid VkIndirectCommandsLayoutNVX handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • indirectCommandsLayout must have been created, allocated, or retrieved from device

28.4. Indirect Commands Generation

Command space for generated commands recorded into a secondary command buffer must be reserved by calling:

void vkCmdReserveSpaceForCommandsNVX(
    VkCommandBuffer                             commandBuffer,
    const VkCmdReserveSpaceForCommandsInfoNVX*  pReserveSpaceInfo);
  • commandBuffer is the secondary command buffer in which the space for device-generated commands is reserved.

  • pProcessCommandsInfo is a pointer to an instance of the vkCmdReserveSpaceForCommandsNVX structure containing parameters affecting the reservation of command buffer space.

Valid Usage
  • The provided commandBuffer must not have had a prior space reservation since its creation or the last reset.

  • The state of the commandBuffer must be legal to execute all commands within the sequence provided by the indirectCommandsLayout member of pProcessCommandsInfo.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pReserveSpaceInfo must be a valid pointer to a valid VkCmdReserveSpaceForCommandsInfoNVX structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called inside of a render pass instance

  • commandBuffer must be a secondary VkCommandBuffer

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Secondary

Inside

Graphics
Compute

typedef struct VkCmdReserveSpaceForCommandsInfoNVX {
    VkStructureType                sType;
    const void*                    pNext;
    VkObjectTableNVX               objectTable;
    VkIndirectCommandsLayoutNVX    indirectCommandsLayout;
    uint32_t                       maxSequencesCount;
} VkCmdReserveSpaceForCommandsInfoNVX;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectTable is the VkObjectTableNVX to be used for the generation process. Only registered objects at the time vkCmdReserveSpaceForCommandsNVX is called, will be taken into account for the reservation.

  • indirectCommandsLayout is the VkIndirectCommandsLayoutNVX that must also be used at generation time.

  • maxSequencesCount is the maximum number of sequences for which command buffer space will be reserved.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_CMD_RESERVE_SPACE_FOR_COMMANDS_INFO_NVX

  • pNext must be NULL

  • objectTable must be a valid VkObjectTableNVX handle

  • indirectCommandsLayout must be a valid VkIndirectCommandsLayoutNVX handle

  • Both of indirectCommandsLayout, and objectTable must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to objectTable must be externally synchronized

The generated commands will behave as if they were recorded within the call to vkCmdReserveSpaceForCommandsNVX, that means they can inherit state defined in the command buffer prior this call. However, given the stateless nature of the generated sequences, they will not affect commands after the reserved space. Treat the state that can be affected by the provided VkIndirectCommandsLayoutNVX as undefined.

The actual generation on the device is handled with:

void vkCmdProcessCommandsNVX(
    VkCommandBuffer                             commandBuffer,
    const VkCmdProcessCommandsInfoNVX*          pProcessCommandsInfo);
  • commandBuffer is the primary command buffer in which the generation process takes space.

  • pProcessCommandsInfo is a pointer to an instance of the VkCmdProcessCommandsInfoNVX structure containing parameters affecting the processing of commands.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pProcessCommandsInfo must be a valid pointer to a valid VkCmdProcessCommandsInfoNVX structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • This command must only be called inside of a render pass instance

Host Synchronization
  • Host access to commandBuffer must be externally synchronized

  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Inside

Graphics
Compute

typedef struct VkCmdProcessCommandsInfoNVX {
    VkStructureType                      sType;
    const void*                          pNext;
    VkObjectTableNVX                     objectTable;
    VkIndirectCommandsLayoutNVX          indirectCommandsLayout;
    uint32_t                             indirectCommandsTokenCount;
    const VkIndirectCommandsTokenNVX*    pIndirectCommandsTokens;
    uint32_t                             maxSequencesCount;
    VkCommandBuffer                      targetCommandBuffer;
    VkBuffer                             sequencesCountBuffer;
    VkDeviceSize                         sequencesCountOffset;
    VkBuffer                             sequencesIndexBuffer;
    VkDeviceSize                         sequencesIndexOffset;
} VkCmdProcessCommandsInfoNVX;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectTable is the VkObjectTableNVX to be used for the generation process. Only registered objects at the time vkCmdReserveSpaceForCommandsNVX is called, will be taken into account for the reservation.

  • indirectCommandsLayout is the VkIndirectCommandsLayoutNVX that provides the command sequence to generate.

  • indirectCommandsTokenCount defines the number of input tokens used.

  • pIndirectCommandsTokens provides an array of VkIndirectCommandsTokenNVX that reference the input data for each token command.

  • maxSequencesCount is the maximum number of sequences for which command buffer space will be reserved. If sequencesCountBuffer is VK_NULL_HANDLE, this is also the actual number of sequences generated.

  • targetCommandBuffer can be the secondary VkCommandBuffer in which the commands should be recorded. If targetCommandBuffer is NULL an implicit reservation as well as execution takes place on the processing VkCommandBuffer.

  • sequencesCountBuffer can be VkBuffer from which the actual amount of sequences is sourced from as uint32_t value.

  • sequencesCountOffset is the byte offset into sequencesCountBuffer where the count value is stored.

  • sequencesIndexBuffer must be set if indirectCommandsLayout’s VK_INDIRECT_COMMANDS_LAYOUT_USAGE_INDEXED_SEQUENCES_BIT is set and provides the used sequence indices as uint32_t array. Otherwise it must be VK_NULL_HANDLE.

  • sequencesIndexOffset is the byte offset into sequencesIndexBuffer where the index values start.

Valid Usage
  • The provided objectTable must include all objects referenced by the generation process.

  • indirectCommandsTokenCount must match the indirectCommandsLayout’s tokenCount.

  • The tokenType member of each entry in the pIndirectCommandsTokens array must match the values used at creation time of indirectCommandsLayout

  • If targetCommandBuffer is provided, it must have reserved command space.

  • If targetCommandBuffer is provided, the objectTable must match the reservation’s objectTable and must have had all referenced objects registered at reservation time.

  • If targetCommandBuffer is provided, the indirectCommandsLayout must match the reservation’s indirectCommandsLayout.

  • If targetCommandBuffer is provided, the maxSequencesCount must not exceed the reservation’s maxSequencesCount.

  • If sequencesCountBuffer is used, its usage flag must have VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set.

  • If sequencesCountBuffer is used, sequencesCountOffset must be aligned to VkDeviceGeneratedCommandsLimitsNVX::minSequenceCountBufferOffsetAlignment.

  • If sequencesIndexBuffer is used, its usage flag must have VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT bit set.

  • If sequencesIndexBuffer is used, sequencesIndexOffset must be aligned to VkDeviceGeneratedCommandsLimitsNVX::minSequenceIndexBufferOffsetAlignment.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_CMD_PROCESS_COMMANDS_INFO_NVX

  • pNext must be NULL

  • objectTable must be a valid VkObjectTableNVX handle

  • indirectCommandsLayout must be a valid VkIndirectCommandsLayoutNVX handle

  • pIndirectCommandsTokens must be a valid pointer to an array of indirectCommandsTokenCount valid VkIndirectCommandsTokenNVX structures

  • If targetCommandBuffer is not NULL, targetCommandBuffer must be a valid VkCommandBuffer handle

  • If sequencesCountBuffer is not VK_NULL_HANDLE, sequencesCountBuffer must be a valid VkBuffer handle

  • If sequencesIndexBuffer is not VK_NULL_HANDLE, sequencesIndexBuffer must be a valid VkBuffer handle

  • indirectCommandsTokenCount must be greater than 0

  • Each of indirectCommandsLayout, objectTable, sequencesCountBuffer, sequencesIndexBuffer, and targetCommandBuffer that are valid handles must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to objectTable must be externally synchronized

  • Host access to targetCommandBuffer must be externally synchronized

Referencing the functions defined in Indirect Commands Layout, vkCmdProcessCommandsNVX behaves as:

// For targetCommandBuffers the existing reservedSpace is reset & overwritten.

VkCommandBuffer cmd = targetCommandBuffer ?
      targetCommandBuffer.reservedSpace :
      commandBuffer;

uint32_t sequencesCount = sequencesCountBuffer ?
      min(maxSequencesCount, sequencesCountBuffer.load_uint32(sequencesCountOffset) :
      maxSequencesCount;


cmdProcessAllSequences(cmd, objectTable,
                       indirectCommandsLayout, pIndirectCommandsTokens,
                       sequencesCount,
                       sequencesIndexBuffer, sequencesIndexOffset);

// The stateful commands within indirectCommandsLayout will not
// affect the state of subsequent commands in the target
// command buffer (cmd)
Note

It is important to note that the state that may be affected through generated commands must be considered undefined for the commands following them. It is not possible to setup generated state and provoking work that uses this state outside of the generated sequence.

29. Sparse Resources

As documented in Resource Memory Association, VkBuffer and VkImage resources in Vulkan must be bound completely and contiguously to a single VkDeviceMemory object. This binding must be done before the resource is used, and the binding is immutable for the lifetime of the resource.

Sparse resources relax these restrictions and provide these additional features:

  • Sparse resources can be bound non-contiguously to one or more VkDeviceMemory allocations.

  • Sparse resources can be re-bound to different memory allocations over the lifetime of the resource.

  • Sparse resources can have descriptors generated and used orthogonally with memory binding commands.

29.1. Sparse Resource Features

Sparse resources have several features that must be enabled explicitly at resource creation time. The features are enabled by including bits in the flags parameter of VkImageCreateInfo or VkBufferCreateInfo. Each feature also has one or more corresponding feature enables specified in VkPhysicalDeviceFeatures.

  • Sparse binding is the base feature, and provides the following capabilities:

    • Resources can be bound at some defined (sparse block) granularity.

    • The entire resource must be bound to memory before use regardless of regions actually accessed.

    • No specific mapping of image region to memory offset is defined, i.e. the location that each texel corresponds to in memory is implementation-dependent.

    • Sparse buffers have a well-defined mapping of buffer range to memory range, where an offset into a range of the buffer that is bound to a single contiguous range of memory corresponds to an identical offset within that range of memory.

    • Requested via the VK_IMAGE_CREATE_SPARSE_BINDING_BIT and VK_BUFFER_CREATE_SPARSE_BINDING_BIT bits.

    • A sparse image created using VK_IMAGE_CREATE_SPARSE_BINDING_BIT (but not VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT) supports all formats that non-sparse usage supports, and supports both VK_IMAGE_TILING_OPTIMAL and VK_IMAGE_TILING_LINEAR tiling.

  • Sparse Residency builds on (and requires) the sparseBinding feature. It includes the following capabilities:

    • Resources do not have to be completely bound to memory before use on the device.

    • Images have a prescribed sparse image block layout, allowing specific rectangular regions of the image to be bound to specific offsets in memory allocations.

    • Consistency of access to unbound regions of the resource is defined by the absence or presence of VkPhysicalDeviceSparseProperties::residencyNonResidentStrict. If this property is present, accesses to unbound regions of the resource are well defined and behave as if the data bound is populated with all zeros; writes are discarded. When this property is absent, accesses are considered safe, but reads will return undefined values.

    • Requested via the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT and VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT bits.

    • Sparse residency support is advertised on a finer grain via the following features:

      • sparseResidencyBuffer: Support for creating VkBuffer objects with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidencyImage2D: Support for creating 2D single-sampled VkImage objects with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidencyImage3D: Support for creating 3D VkImage objects with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidency2Samples: Support for creating 2D VkImage objects with 2 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidency4Samples: Support for creating 2D VkImage objects with 4 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidency8Samples: Support for creating 2D VkImage objects with 8 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      • sparseResidency16Samples: Support for creating 2D VkImage objects with 16 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

      Implementations supporting sparseResidencyImage2D are only required to support sparse 2D, single-sampled images. Support for sparse 3D and MSAA images is optional and can be enabled via sparseResidencyImage3D, sparseResidency2Samples, sparseResidency4Samples, sparseResidency8Samples, and sparseResidency16Samples.

    • A sparse image created using VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT supports all non-compressed color formats with power-of-two element size that non-sparse usage supports. Additional formats may also be supported and can be queried via vkGetPhysicalDeviceSparseImageFormatProperties. VK_IMAGE_TILING_LINEAR tiling is not supported.

  • Sparse aliasing provides the following capability that can be enabled per resource:

    Allows physical memory ranges to be shared between multiple locations in the same sparse resource or between multiple sparse resources, with each binding of a memory location observing a consistent interpretation of the memory contents.

    See Sparse Memory Aliasing for more information.

29.2. Sparse Buffers and Fully-Resident Images

Both VkBuffer and VkImage objects created with the VK_IMAGE_CREATE_SPARSE_BINDING_BIT or VK_BUFFER_CREATE_SPARSE_BINDING_BIT bits can be thought of as a linear region of address space. In the VkImage case if VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT is not used, this linear region is entirely opaque, meaning that there is no application-visible mapping between texel location and memory offset.

Unless VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT or VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT are also used, the entire resource must be bound to one or more VkDeviceMemory objects before use.

29.2.1. Sparse Buffer and Fully-Resident Image Block Size

The sparse block size in bytes for sparse buffers and fully-resident images is reported as VkMemoryRequirements::alignment. alignment represents both the memory alignment requirement and the binding granularity (in bytes) for sparse resources.

29.3. Sparse Partially-Resident Buffers

VkBuffer objects created with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT bit allow the buffer to be made only partially resident. Partially resident VkBuffer objects are allocated and bound identically to VkBuffer objects using only the VK_BUFFER_CREATE_SPARSE_BINDING_BIT feature. The only difference is the ability for some regions of the buffer to be unbound during device use.

29.4. Sparse Partially-Resident Images

VkImage objects created with the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT bit allow specific rectangular regions of the image called sparse image blocks to be bound to specific ranges of memory. This allows the application to manage residency at either image subresource or sparse image block granularity. Each image subresource (outside of the mip tail) starts on a sparse block boundary and has dimensions that are integer multiples of the corresponding dimensions of the sparse image block.

Note

Applications can use these types of images to control LOD based on total memory consumption. If memory pressure becomes an issue the application can unbind and disable specific mipmap levels of images without having to recreate resources or modify texel data of unaffected levels.

The application can also use this functionality to access subregions of the image in a “megatexture” fashion. The application can create a large image and only populate the region of the image that is currently being used in the scene.

29.4.1. Accessing Unbound Regions

The following member of VkPhysicalDeviceSparseProperties affects how data in unbound regions of sparse resources are handled by the implementation:

  • residencyNonResidentStrict

If this property is not present, reads of unbound regions of the image will return undefined values. Both reads and writes are still considered safe and will not affect other resources or populated regions of the image.

If this property is present, all reads of unbound regions of the image will behave as if the region was bound to memory populated with all zeros; writes will be discarded.

Formatted accesses to unbound memory may still alter some component values in the natural way for those accesses, e.g. substituting a value of one for alpha in formats that do not have an alpha component.

Example: Reading the alpha component of an unbacked VK_FORMAT_R8_UNORM image will return a value of 1.0f.

See Physical Device Enumeration for instructions for retrieving physical device properties.

Implementor’s Note

For implementations that cannot natively handle access to unbound regions of a resource, the implementation may allocate and bind memory to the unbound regions. Reads and writes to unbound regions will access the implementation-managed memory instead.

Given that reads of unbound regions are undefined in this scenario, implementations may use the same physical memory for all unbound regions of multiple resources within the same process.

29.4.2. Mip Tail Regions

Sparse images created using VK_IMAGE_CREATE_SPARSE_BINDING_BIT (without also using VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT) have no specific mapping of image region or image subresource to memory offset defined, so the entire image can be thought of as a linear opaque address region. However, images created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT do have a prescribed sparse image block layout, and hence each image subresource must start on a sparse block boundary. Within each array layer, the set of mip levels that have a smaller size than the sparse block size in bytes are grouped together into a mip tail region.

If the VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT flag is present in the flags member of VkSparseImageFormatProperties, for the image’s format, then any mip level which has dimensions that are not integer multiples of the corresponding dimensions of the sparse image block, and all subsequent mip levels, are also included in the mip tail region.

The following member of VkPhysicalDeviceSparseProperties may affect how the implementation places mip levels in the mip tail region:

  • residencyAlignedMipSize

Each mip tail region is bound to memory as an opaque region (i.e. must be bound using a VkSparseImageOpaqueMemoryBindInfo structure) and may be of a size greater than or equal to the sparse block size in bytes. This size is guaranteed to be an integer multiple of the sparse block size in bytes.

An implementation may choose to allow each array-layer’s mip tail region to be bound to memory independently or require that all array-layer’s mip tail regions be treated as one. This is dictated by VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT in VkSparseImageMemoryRequirements::flags.

The following diagrams depict how VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT and VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT alter memory usage and requirements.

sparseimage
Figure 28. Sparse Image

In the absence of VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT and VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT, each array layer contains a mip tail region containing texel data for all mip levels smaller than the sparse image block in any dimension.

Mip levels that are as large or larger than a sparse image block in all dimensions can be bound individually. Right-edges and bottom-edges of each level are allowed to have partially used sparse blocks. Any bound partially-used-sparse-blocks must still have their full sparse block size in bytes allocated in memory.

sparseimage singlemiptail
Figure 29. Sparse Image with Single Mip Tail

When VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT is present all array layers will share a single mip tail region.

sparseimage alignedmipsize
Figure 30. Sparse Image with Aligned Mip Size
Note

The mip tail regions are presented here in 2D arrays simply for figure size reasons. Each mip tail is logically a single array of sparse blocks with an implementation-dependent mapping of texels or compressed texel blocks to sparse blocks.

When VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT is present the first mip level that would contain partially used sparse blocks begins the mip tail region. This level and all subsequent levels are placed in the mip tail. Only the first N mip levels whose dimensions are an exact multiple of the sparse image block dimensions can be bound and unbound on a sparse block basis.

sparseimage alignedmipsize singlemiptail
Figure 31. Sparse Image with Aligned Mip Size and Single Mip Tail
Note

The mip tail region is presented here in a 2D array simply for figure size reasons. It is logically a single array of sparse blocks with an implementation-dependent mapping of texels or compressed texel blocks to sparse blocks.

When both VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT and VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT are present the constraints from each of these flags are in effect.

29.4.3. Standard Sparse Image Block Shapes

Standard sparse image block shapes define a standard set of dimensions for sparse image blocks that depend on the format of the image. Layout of texels or compressed texel blocks within a sparse image block is implementation dependent. All currently defined standard sparse image block shapes are 64 KB in size.

For block-compressed formats (e.g. VK_FORMAT_BC5_UNORM_BLOCK), the texel size is the size of the compressed texel block (e.g. 128-bit for BC5) thus the dimensions of the standard sparse image block shapes apply in terms of compressed texel blocks.

Note

For block-compressed formats, the dimensions of a sparse image block in terms of texels can be calculated by multiplying the sparse image block dimensions by the compressed texel block dimensions.

Table 41. Standard Sparse Image Block Shapes (Single Sample)
TEXEL SIZE (bits) Block Shape (2D) Block Shape (3D)

8-Bit

256 × 256 × 1

64 × 32 × 32

16-Bit

256 × 128 × 1

32 × 32 × 32

32-Bit

128 × 128 × 1

32 × 32 × 16

64-Bit

128 × 64 × 1

32 × 16 × 16

128-Bit

64 × 64 × 1

16 × 16 × 16

Table 42. Standard Sparse Image Block Shapes (MSAA)
TEXEL SIZE (bits) Block Shape (2X) Block Shape (4X) Block Shape (8X) Block Shape (16X)

8-Bit

128 × 256 × 1

128 × 128 × 1

64 × 128 × 1

64 × 64 × 1

16-Bit

128 × 128 × 1

128 × 64 × 1

64 × 64 × 1

64 × 32 × 1

32-Bit

64 × 128 × 1

64 × 64 × 1

32 × 64 × 1

32 × 32 × 1

64-Bit

64 × 64 × 1

64 × 32 × 1

32 × 32 × 1

32 × 16 × 1

128-Bit

32 × 64 × 1

32 × 32 × 1

16 × 32 × 1

16 × 16 × 1

Implementations that support the standard sparse image block shape for all formats listed in the Standard Sparse Image Block Shapes (Single Sample) and Standard Sparse Image Block Shapes (MSAA) tables may advertise the following VkPhysicalDeviceSparseProperties:

  • residencyStandard2DBlockShape

  • residencyStandard2DMultisampleBlockShape

  • residencyStandard3DBlockShape

Reporting each of these features does not imply that all possible image types are supported as sparse. Instead, this indicates that no supported sparse image of the corresponding type will use custom sparse image block dimensions for any formats that have a corresponding standard sparse image block shape.

29.4.4. Custom Sparse Image Block Shapes

An implementation that does not support a standard image block shape for a particular sparse partially-resident image may choose to support a custom sparse image block shape for it instead. The dimensions of such a custom sparse image block shape are reported in VkSparseImageFormatProperties::imageGranularity. As with standard sparse image block shapes, the size in bytes of the custom sparse image block shape will be reported in VkMemoryRequirements::alignment.

Custom sparse image block dimensions are reported through vkGetPhysicalDeviceSparseImageFormatProperties and vkGetImageSparseMemoryRequirements.

An implementation must not support both the standard sparse image block shape and a custom sparse image block shape for the same image. The standard sparse image block shape must be used if it is supported.

29.4.5. Multiple Aspects

Partially resident images are allowed to report separate sparse properties for different aspects of the image. One example is for depth/stencil images where the implementation separates the depth and stencil data into separate planes. Another reason for multiple aspects is to allow the application to manage memory allocation for implementation-private metadata associated with the image. See the figure below:

sparseimage multiaspect
Figure 32. Multiple Aspect Sparse Image
Note

The mip tail regions are presented here in 2D arrays simply for figure size reasons. Each mip tail is logically a single array of sparse blocks with an implementation-dependent mapping of texels or compressed texel blocks to sparse blocks.

In the figure above the depth, stencil, and metadata aspects all have unique sparse properties. The per-texel stencil data is ¼ the size of the depth data, hence the stencil sparse blocks include 4 × the number of texels. The sparse block size in bytes for all of the aspects is identical and defined by VkMemoryRequirements::alignment.

Metadata

The metadata aspect of an image has the following constraints:

  • All metadata is reported in the mip tail region of the metadata aspect.

  • All metadata must be bound prior to device use of the sparse image.

29.5. Sparse Memory Aliasing

By default sparse resources have the same aliasing rules as non-sparse resources. See Memory Aliasing for more information.

VkDevice objects that have the sparseResidencyAliased feature enabled are able to use the VK_BUFFER_CREATE_SPARSE_ALIASED_BIT and VK_IMAGE_CREATE_SPARSE_ALIASED_BIT flags for resource creation. These flags allow resources to access physical memory bound into multiple locations within one or more sparse resources in a data consistent fashion. This means that reading physical memory from multiple aliased locations will return the same value.

Care must be taken when performing a write operation to aliased physical memory. Memory dependencies must be used to separate writes to one alias from reads or writes to another alias. Writes to aliased memory that are not properly guarded against accesses to different aliases will have undefined results for all accesses to the aliased memory.

Applications that wish to make use of data consistent sparse memory aliasing must abide by the following guidelines:

  • All sparse resources that are bound to aliased physical memory must be created with the VK_BUFFER_CREATE_SPARSE_ALIASED_BIT / VK_IMAGE_CREATE_SPARSE_ALIASED_BIT flag.

  • All resources that access aliased physical memory must interpret the memory in the same way. This implies the following:

    • Buffers and images cannot alias the same physical memory in a data consistent fashion. The physical memory ranges must be used exclusively by buffers or used exclusively by images for data consistency to be guaranteed.

    • Memory in sparse image mip tail regions cannot access aliased memory in a data consistent fashion.

    • Sparse images that alias the same physical memory must have compatible formats and be using the same sparse image block shape in order to access aliased memory in a data consistent fashion.

Failure to follow any of the above guidelines will require the application to abide by the normal, non-sparse resource aliasing rules. In this case memory cannot be accessed in a data consistent fashion.

Note

Enabling sparse resource memory aliasing can be a way to lower physical memory use, but it may reduce performance on some implementations. An application developer can test on their target HW and balance the memory / performance trade-offs measured.

29.6. Sparse Resource Implementation Guidelines

This section is Informative. It is included to aid in implementors’ understanding of sparse resources.

Device Virtual Address

The basic sparseBinding feature allows the resource to reserve its own device virtual address range at resource creation time rather than relying on a bind operation to set this. Without any other creation flags, no other constraints are relaxed compared to normal resources. All pages must be bound to physical memory before the device accesses the resource.

The sparse residency features allow sparse resources to be used even when not all pages are bound to memory. Implementations that support access to unbound pages without causing a fault may support residencyNonResidentStrict.

Not faulting on access to unbound pages is not enough to support residencyNonResidentStrict. An implementation must also guarantee that reads after writes to unbound regions of the resource always return data for the read as if the memory contains zeros. Depending on any caching hierarchy of the implementation this may not always be possible.

Any implementation that does not fault, but does not guarantee correct read values must not support residencyNonResidentStrict.

Any implementation that cannot access unbound pages without causing a fault will require the implementation to bind the entire device virtual address range to physical memory. Any pages that the application does not bind to memory may be bound to one (or more) “dummy” physical page(s) allocated by the implementation. Given the following properties:

  • A process must not access memory from another process

  • Reads return undefined values

It is sufficient for each host process to allocate these dummy pages and use them for all resources in that process. Implementations may allocate more often (per instance, per device, or per resource).

Binding Memory

The byte size reported in VkMemoryRequirements::size must be greater than or equal to the amount of physical memory required to fully populate the resource. Some implementations require “holes” in the device virtual address range that are never accessed. These holes may be included in the size reported for the resource.

Including or not including the device virtual address holes in the resource size will alter how the implementation provides support for VkSparseImageOpaqueMemoryBindInfo. This operation must be supported for all sparse images, even ones created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

editing-note

@ntrevett suggested expanding the NOTE tag below to encompass everything from “The cost is…​” in the first bullet point through the current note. TBD.

  • If the holes are included in the size, this bind function becomes very easy. In most cases the resourceOffset is simply a device virtual address offset and the implementation can easily determine what device virtual address to bind. The cost is that the application may allocate more physical memory for the resource than it needs.

  • If the holes are not included in the size, the application can allocate less physical memory than otherwise for the resource. However, in this case the implementation must account for the holes when mapping resourceOffset to the actual device virtual address intended to be mapped.

Note

If the application always uses VkSparseImageMemoryBindInfo to bind memory for the non-tail mip levels, any holes that are present in the resource size may never be bound.

Since VkSparseImageMemoryBindInfo uses texel locations to determine which device virtual addresses to bind, it is impossible to bind device virtual address holes with this operation.

Binding Metadata Memory

All metadata for sparse images have their own sparse properties and are embedded in the mip tail region for said properties. See the Multiaspect section for details.

Given that metadata is in a mip tail region, and the mip tail region must be reported as contiguous (either globally or per-array-layer), some implementations will have to resort to complicated offset → device virtual address mapping for handling VkSparseImageOpaqueMemoryBindInfo.

To make this easier on the implementation, the VK_SPARSE_MEMORY_BIND_METADATA_BIT explicitly specifies when metadata is bound with VkSparseImageOpaqueMemoryBindInfo. When this flag is not present, the resourceOffset may be treated as a strict device virtual address offset.

When VK_SPARSE_MEMORY_BIND_METADATA_BIT is present, the resourceOffset must have been derived explicitly from the imageMipTailOffset in the sparse resource properties returned for the metadata aspect. By manipulating the value returned for imageMipTailOffset, the resourceOffset does not have to correlate directly to a device virtual address offset, and may instead be whatever values makes it easiest for the implementation to derive the correct device virtual address.

29.7. Sparse Resource API

The APIs related to sparse resources are grouped into the following categories:

29.7.1. Physical Device Features

Some sparse-resource related features are reported and enabled in VkPhysicalDeviceFeatures. These features must be supported and enabled on the VkDevice object before applications can use them. See Physical Device Features for information on how to get and set enabled device features, and for more detailed explanations of these features.

Sparse Physical Device Features
  • sparseBinding: Support for creating VkBuffer and VkImage objects with the VK_BUFFER_CREATE_SPARSE_BINDING_BIT and VK_IMAGE_CREATE_SPARSE_BINDING_BIT flags, respectively.

  • sparseResidencyBuffer: Support for creating VkBuffer objects with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT flag.

  • sparseResidencyImage2D: Support for creating 2D single-sampled VkImage objects with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidencyImage3D: Support for creating 3D VkImage objects with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidency2Samples: Support for creating 2D VkImage objects with 2 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidency4Samples: Support for creating 2D VkImage objects with 4 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidency8Samples: Support for creating 2D VkImage objects with 8 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidency16Samples: Support for creating 2D VkImage objects with 16 samples and VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.

  • sparseResidencyAliased: Support for creating VkBuffer and VkImage objects with the VK_BUFFER_CREATE_SPARSE_ALIASED_BIT and VK_IMAGE_CREATE_SPARSE_ALIASED_BIT flags, respectively.

29.7.2. Physical Device Sparse Properties

Some features of the implementation are not possible to disable, and are reported to allow applications to alter their sparse resource usage accordingly. These read-only capabilities are reported in the VkPhysicalDeviceProperties::sparseProperties member, which is a structure of type VkPhysicalDeviceSparseProperties.

The VkPhysicalDeviceSparseProperties structure is defined as:

typedef struct VkPhysicalDeviceSparseProperties {
    VkBool32    residencyStandard2DBlockShape;
    VkBool32    residencyStandard2DMultisampleBlockShape;
    VkBool32    residencyStandard3DBlockShape;
    VkBool32    residencyAlignedMipSize;
    VkBool32    residencyNonResidentStrict;
} VkPhysicalDeviceSparseProperties;
  • residencyStandard2DBlockShape is VK_TRUE if the physical device will access all single-sample 2D sparse resources using the standard sparse image block shapes (based on image format), as described in the Standard Sparse Image Block Shapes (Single Sample) table. If this property is not supported the value returned in the imageGranularity member of the VkSparseImageFormatProperties structure for single-sample 2D images is not required to match the standard sparse image block dimensions listed in the table.

  • residencyStandard2DMultisampleBlockShape is VK_TRUE if the physical device will access all multisample 2D sparse resources using the standard sparse image block shapes (based on image format), as described in the Standard Sparse Image Block Shapes (MSAA) table. If this property is not supported, the value returned in the imageGranularity member of the VkSparseImageFormatProperties structure for multisample 2D images is not required to match the standard sparse image block dimensions listed in the table.

  • residencyStandard3DBlockShape is VK_TRUE if the physical device will access all 3D sparse resources using the standard sparse image block shapes (based on image format), as described in the Standard Sparse Image Block Shapes (Single Sample) table. If this property is not supported, the value returned in the imageGranularity member of the VkSparseImageFormatProperties structure for 3D images is not required to match the standard sparse image block dimensions listed in the table.

  • residencyAlignedMipSize is VK_TRUE if images with mip level dimensions that are not integer multiples of the corresponding dimensions of the sparse image block may be placed in the mip tail. If this property is not reported, only mip levels with dimensions smaller than the imageGranularity member of the VkSparseImageFormatProperties structure will be placed in the mip tail. If this property is reported the implementation is allowed to return VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT in the flags member of VkSparseImageFormatProperties, indicating that mip level dimensions that are not integer multiples of the corresponding dimensions of the sparse image block will be placed in the mip tail.

  • residencyNonResidentStrict specifies whether the physical device can consistently access non-resident regions of a resource. If this property is VK_TRUE, access to non-resident regions of resources will be guaranteed to return values as if the resource were populated with 0; writes to non-resident regions will be discarded.

29.7.3. Sparse Image Format Properties

Given that certain aspects of sparse image support, including the sparse image block dimensions, may be implementation-dependent, vkGetPhysicalDeviceSparseImageFormatProperties can be used to query for sparse image format properties prior to resource creation. This command is used to check whether a given set of sparse image parameters is supported and what the sparse image block shape will be.

Sparse Image Format Properties API

The VkSparseImageFormatProperties structure is defined as:

typedef struct VkSparseImageFormatProperties {
    VkImageAspectFlags          aspectMask;
    VkExtent3D                  imageGranularity;
    VkSparseImageFormatFlags    flags;
} VkSparseImageFormatProperties;
  • aspectMask is a bitmask VkImageAspectFlagBits specifying which aspects of the image the properties apply to.

  • imageGranularity is the width, height, and depth of the sparse image block in texels or compressed texel blocks.

  • flags is a bitmask of VkSparseImageFormatFlagBits specifying additional information about the sparse resource.

Bits which can be set in VkSparseImageFormatProperties::flags, specifying additional information about the sparse resource, are:

typedef enum VkSparseImageFormatFlagBits {
    VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT = 0x00000001,
    VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT = 0x00000002,
    VK_SPARSE_IMAGE_FORMAT_NONSTANDARD_BLOCK_SIZE_BIT = 0x00000004,
} VkSparseImageFormatFlagBits;
  • VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT specifies that the image uses a single mip tail region for all array layers.

  • VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT specifies that the first mip level whose dimensions are not integer multiples of the corresponding dimensions of the sparse image block begins the mip tail region.

  • VK_SPARSE_IMAGE_FORMAT_NONSTANDARD_BLOCK_SIZE_BIT specifies that the image uses non-standard sparse image block dimensions, and the imageGranularity values do not match the standard sparse image block dimensions for the given format.

typedef VkFlags VkSparseImageFormatFlags;

VkSparseImageFormatFlags is a bitmask type for setting a mask of zero or more VkSparseImageFormatFlagBits.

vkGetPhysicalDeviceSparseImageFormatProperties returns an array of VkSparseImageFormatProperties. Each element will describe properties for one set of image aspects that are bound simultaneously in the image. This is usually one element for each aspect in the image, but for interleaved depth/stencil images there is only one element describing the combined aspects.

void vkGetPhysicalDeviceSparseImageFormatProperties(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkImageType                                 type,
    VkSampleCountFlagBits                       samples,
    VkImageUsageFlags                           usage,
    VkImageTiling                               tiling,
    uint32_t*                                   pPropertyCount,
    VkSparseImageFormatProperties*              pProperties);
  • physicalDevice is the physical device from which to query the sparse image capabilities.

  • format is the image format.

  • type is the dimensionality of image.

  • samples is the number of samples per texel as defined in VkSampleCountFlagBits.

  • usage is a bitmask describing the intended usage of the image.

  • tiling is the tiling arrangement of the data elements in memory.

  • pPropertyCount is a pointer to an integer related to the number of sparse format properties available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkSparseImageFormatProperties structures.

If pProperties is NULL, then the number of sparse format properties available is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If pPropertyCount is less than the number of sparse format properties available, at most pPropertyCount structures will be written.

If VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT is not supported for the given arguments, pPropertyCount will be set to zero upon return, and no data will be written to pProperties.

Multiple aspects are returned for depth/stencil images that are implemented as separate planes by the implementation. The depth and stencil data planes each have unique VkSparseImageFormatProperties data.

Depth/stencil images with depth and stencil data interleaved into a single plane will return a single VkSparseImageFormatProperties structure with the aspectMask set to VK_IMAGE_ASPECT_DEPTH_BIT | VK_IMAGE_ASPECT_STENCIL_BIT.

Valid Usage
  • samples must be a bit value that is set in VkImageFormatProperties::sampleCounts returned by vkGetPhysicalDeviceImageFormatProperties with format, type, tiling, and usage equal to those in this command and flags equal to the value that is set in VkImageCreateInfo::flags when the image is created

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • format must be a valid VkFormat value

  • type must be a valid VkImageType value

  • samples must be a valid VkSampleCountFlagBits value

  • usage must be a valid combination of VkImageUsageFlagBits values

  • usage must not be 0

  • tiling must be a valid VkImageTiling value

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkSparseImageFormatProperties structures

vkGetPhysicalDeviceSparseImageFormatProperties2 returns an array of VkSparseImageFormatProperties2. Each element will describe properties for one set of image aspects that are bound simultaneously in the image. This is usually one element for each aspect in the image, but for interleaved depth/stencil images there is only one element describing the combined aspects.

void vkGetPhysicalDeviceSparseImageFormatProperties2(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceSparseImageFormatInfo2* pFormatInfo,
    uint32_t*                                   pPropertyCount,
    VkSparseImageFormatProperties2*             pProperties);

or the equivalent command

void vkGetPhysicalDeviceSparseImageFormatProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceSparseImageFormatInfo2* pFormatInfo,
    uint32_t*                                   pPropertyCount,
    VkSparseImageFormatProperties2*             pProperties);
  • physicalDevice is the physical device from which to query the sparse image capabilities.

  • pFormatInfo is a pointer to a structure of type VkPhysicalDeviceSparseImageFormatInfo2 containing input parameters to the command.

  • pPropertyCount is a pointer to an integer related to the number of sparse format properties available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkSparseImageFormatProperties2 structures.

vkGetPhysicalDeviceSparseImageFormatProperties2 behaves identically to vkGetPhysicalDeviceSparseImageFormatProperties, with the ability to return extended information by adding extension structures to the pNext chain of its pProperties parameter.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pFormatInfo must be a valid pointer to a valid VkPhysicalDeviceSparseImageFormatInfo2 structure

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkSparseImageFormatProperties2 structures

The VkPhysicalDeviceSparseImageFormatInfo2 structure is defined as:

typedef struct VkPhysicalDeviceSparseImageFormatInfo2 {
    VkStructureType          sType;
    const void*              pNext;
    VkFormat                 format;
    VkImageType              type;
    VkSampleCountFlagBits    samples;
    VkImageUsageFlags        usage;
    VkImageTiling            tiling;
} VkPhysicalDeviceSparseImageFormatInfo2;

or the equivalent

typedef VkPhysicalDeviceSparseImageFormatInfo2 VkPhysicalDeviceSparseImageFormatInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • format is the image format.

  • type is the dimensionality of image.

  • samples is the number of samples per texel as defined in VkSampleCountFlagBits.

  • usage is a bitmask describing the intended usage of the image.

  • tiling is the tiling arrangement of the data elements in memory.

Valid Usage
  • samples must be a bit value that is set in VkImageFormatProperties::sampleCounts returned by vkGetPhysicalDeviceImageFormatProperties with format, type, tiling, and usage equal to those in this command and flags equal to the value that is set in VkImageCreateInfo::flags when the image is created

Valid Usage (Implicit)

The VkSparseImageFormatProperties2 structure is defined as:

typedef struct VkSparseImageFormatProperties2 {
    VkStructureType                  sType;
    void*                            pNext;
    VkSparseImageFormatProperties    properties;
} VkSparseImageFormatProperties2;

or the equivalent

typedef VkSparseImageFormatProperties2 VkSparseImageFormatProperties2KHR;
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2

  • pNext must be NULL

29.7.4. Sparse Resource Creation

Sparse resources require that one or more sparse feature flags be specified (as part of the VkPhysicalDeviceFeatures structure described previously in the Physical Device Features section) at CreateDevice time. When the appropriate device features are enabled, the VK_BUFFER_CREATE_SPARSE_* and VK_IMAGE_CREATE_SPARSE_* flags can be used. See vkCreateBuffer and vkCreateImage for details of the resource creation APIs.

Note

Specifying VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT or VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT requires specifying VK_BUFFER_CREATE_SPARSE_BINDING_BIT or VK_IMAGE_CREATE_SPARSE_BINDING_BIT, respectively, as well. This means that resources must be created with the appropriate *_SPARSE_BINDING_BIT to be used with the sparse binding command (vkQueueBindSparse).

29.7.5. Sparse Resource Memory Requirements

Sparse resources have specific memory requirements related to binding sparse memory. These memory requirements are reported differently for VkBuffer objects and VkImage objects.

Buffer and Fully-Resident Images

Buffers (both fully and partially resident) and fully-resident images can be bound to memory using only the data from VkMemoryRequirements. For all sparse resources the VkMemoryRequirements::alignment member specifies both the bindable sparse block size in bytes and required alignment of VkDeviceMemory.

Partially Resident Images

Partially resident images have a different method for binding memory. As with buffers and fully resident images, the VkMemoryRequirements::alignment field specifies the bindable sparse block size in bytes for the image.

Requesting sparse memory requirements for VkImage objects using vkGetImageSparseMemoryRequirements will return an array of one or more VkSparseImageMemoryRequirements structures. Each structure describes the sparse memory requirements for a group of aspects of the image.

The sparse image must have been created using the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT flag to retrieve valid sparse image memory requirements.

Sparse Image Memory Requirements

The VkSparseImageMemoryRequirements structure is defined as:

typedef struct VkSparseImageMemoryRequirements {
    VkSparseImageFormatProperties    formatProperties;
    uint32_t                         imageMipTailFirstLod;
    VkDeviceSize                     imageMipTailSize;
    VkDeviceSize                     imageMipTailOffset;
    VkDeviceSize                     imageMipTailStride;
} VkSparseImageMemoryRequirements;
  • formatProperties.aspectMask is the set of aspects of the image that this sparse memory requirement applies to. This will usually have a single aspect specified. However, depth/stencil images may have depth and stencil data interleaved in the same sparse block, in which case both VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT would be present.

  • formatProperties.imageGranularity describes the dimensions of a single bindable sparse image block in texel units. For aspect VK_IMAGE_ASPECT_METADATA_BIT, all dimensions will be zero. All metadata is located in the mip tail region.

  • formatProperties.flags is a bitmask of VkSparseImageFormatFlagBits:

    • If VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT is set the image uses a single mip tail region for all array layers.

    • If VK_SPARSE_IMAGE_FORMAT_ALIGNED_MIP_SIZE_BIT is set the dimensions of mip levels must be integer multiples of the corresponding dimensions of the sparse image block for levels not located in the mip tail.

    • If VK_SPARSE_IMAGE_FORMAT_NONSTANDARD_BLOCK_SIZE_BIT is set the image uses non-standard sparse image block dimensions. The formatProperties.imageGranularity values do not match the standard sparse image block dimension corresponding to the image’s format.

  • imageMipTailFirstLod is the first mip level at which image subresources are included in the mip tail region.

  • imageMipTailSize is the memory size (in bytes) of the mip tail region. If formatProperties.flags contains VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT, this is the size of the whole mip tail, otherwise this is the size of the mip tail of a single array layer. This value is guaranteed to be a multiple of the sparse block size in bytes.

  • imageMipTailOffset is the opaque memory offset used with VkSparseImageOpaqueMemoryBindInfo to bind the mip tail region(s).

  • imageMipTailStride is the offset stride between each array-layer’s mip tail, if formatProperties.flags does not contain VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT (otherwise the value is undefined).

To query sparse memory requirements for an image, call:

void vkGetImageSparseMemoryRequirements(
    VkDevice                                    device,
    VkImage                                     image,
    uint32_t*                                   pSparseMemoryRequirementCount,
    VkSparseImageMemoryRequirements*            pSparseMemoryRequirements);
  • device is the logical device that owns the image.

  • image is the VkImage object to get the memory requirements for.

  • pSparseMemoryRequirementCount is a pointer to an integer related to the number of sparse memory requirements available or queried, as described below.

  • pSparseMemoryRequirements is either NULL or a pointer to an array of VkSparseImageMemoryRequirements structures.

If pSparseMemoryRequirements is NULL, then the number of sparse memory requirements available is returned in pSparseMemoryRequirementCount. Otherwise, pSparseMemoryRequirementCount must point to a variable set by the user to the number of elements in the pSparseMemoryRequirements array, and on return the variable is overwritten with the number of structures actually written to pSparseMemoryRequirements. If pSparseMemoryRequirementCount is less than the number of sparse memory requirements available, at most pSparseMemoryRequirementCount structures will be written.

If the image was not created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT then pSparseMemoryRequirementCount will be set to zero and pSparseMemoryRequirements will not be written to.

Note

It is legal for an implementation to report a larger value in VkMemoryRequirements::size than would be obtained by adding together memory sizes for all VkSparseImageMemoryRequirements returned by vkGetImageSparseMemoryRequirements. This may occur when the implementation requires unused padding in the address range describing the resource.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • image must be a valid VkImage handle

  • pSparseMemoryRequirementCount must be a valid pointer to a uint32_t value

  • If the value referenced by pSparseMemoryRequirementCount is not 0, and pSparseMemoryRequirements is not NULL, pSparseMemoryRequirements must be a valid pointer to an array of pSparseMemoryRequirementCount VkSparseImageMemoryRequirements structures

  • image must have been created, allocated, or retrieved from device

To query sparse memory requirements for an image, call:

void vkGetImageSparseMemoryRequirements2(
    VkDevice                                    device,
    const VkImageSparseMemoryRequirementsInfo2* pInfo,
    uint32_t*                                   pSparseMemoryRequirementCount,
    VkSparseImageMemoryRequirements2*           pSparseMemoryRequirements);

or the equivalent command

void vkGetImageSparseMemoryRequirements2KHR(
    VkDevice                                    device,
    const VkImageSparseMemoryRequirementsInfo2* pInfo,
    uint32_t*                                   pSparseMemoryRequirementCount,
    VkSparseImageMemoryRequirements2*           pSparseMemoryRequirements);
  • device is the logical device that owns the image.

  • pInfo is a pointer to an instance of the VkImageSparseMemoryRequirementsInfo2 structure containing parameters required for the memory requirements query.

  • pSparseMemoryRequirementCount is a pointer to an integer related to the number of sparse memory requirements available or queried, as described below.

  • pSparseMemoryRequirements is either NULL or a pointer to an array of VkSparseImageMemoryRequirements2 structures.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pInfo must be a valid pointer to a valid VkImageSparseMemoryRequirementsInfo2 structure

  • pSparseMemoryRequirementCount must be a valid pointer to a uint32_t value

  • If the value referenced by pSparseMemoryRequirementCount is not 0, and pSparseMemoryRequirements is not NULL, pSparseMemoryRequirements must be a valid pointer to an array of pSparseMemoryRequirementCount VkSparseImageMemoryRequirements2 structures

The VkImageSparseMemoryRequirementsInfo2 structure is defined as:

typedef struct VkImageSparseMemoryRequirementsInfo2 {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
} VkImageSparseMemoryRequirementsInfo2;

or the equivalent

typedef VkImageSparseMemoryRequirementsInfo2 VkImageSparseMemoryRequirementsInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • image is the image to query.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2

  • pNext must be NULL

  • image must be a valid VkImage handle

The VkSparseImageMemoryRequirements2 structure is defined as:

typedef struct VkSparseImageMemoryRequirements2 {
    VkStructureType                    sType;
    void*                              pNext;
    VkSparseImageMemoryRequirements    memoryRequirements;
} VkSparseImageMemoryRequirements2;

or the equivalent

typedef VkSparseImageMemoryRequirements2 VkSparseImageMemoryRequirements2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • memoryRequirements is a structure of type VkSparseImageMemoryRequirements describing the memory requirements of the sparse image.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2

  • pNext must be NULL

29.7.6. Binding Resource Memory

Non-sparse resources are backed by a single physical allocation prior to device use (via vkBindImageMemory or vkBindBufferMemory), and their backing must not be changed. On the other hand, sparse resources can be bound to memory non-contiguously and these bindings can be altered during the lifetime of the resource.

Note

It is important to note that freeing a VkDeviceMemory object with vkFreeMemory will not cause resources (or resource regions) bound to the memory object to become unbound. Access to resources that are bound to memory objects that have been freed will result in undefined behavior, potentially including application termination.

Implementations must ensure that no access to physical memory owned by the system or another process will occur in this scenario. In other words, accessing resources bound to freed memory may result in application termination, but must not result in system termination or in reading non-process-accessible memory.

Sparse memory bindings execute on a queue that includes the VK_QUEUE_SPARSE_BINDING_BIT bit. Applications must use synchronization primitives to guarantee that other queues do not access ranges of memory concurrently with a binding change. Accessing memory in a range while it is being rebound results in undefined behavior. It is valid to access other ranges of the same resource while a bind operation is executing.

Note

Implementations must provide a guarantee that simultaneously binding sparse blocks while another queue accesses those same sparse blocks via a sparse resource must not access memory owned by another process or otherwise corrupt the system.

While some implementations may include VK_QUEUE_SPARSE_BINDING_BIT support in queue families that also include graphics and compute support, other implementations may only expose a VK_QUEUE_SPARSE_BINDING_BIT-only queue family. In either case, applications must use synchronization primitives to explicitly request any ordering dependencies between sparse memory binding operations and other graphics/compute/transfer operations, as sparse binding operations are not automatically ordered against command buffer execution, even within a single queue.

When binding memory explicitly for the VK_IMAGE_ASPECT_METADATA_BIT the application must use the VK_SPARSE_MEMORY_BIND_METADATA_BIT in the VkSparseMemoryBind::flags field when binding memory. Binding memory for metadata is done the same way as binding memory for the mip tail, with the addition of the VK_SPARSE_MEMORY_BIND_METADATA_BIT flag.

Binding the mip tail for any aspect must only be performed using VkSparseImageOpaqueMemoryBindInfo. If formatProperties.flags contains VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT, then it can be bound with a single VkSparseMemoryBind structure, with resourceOffset = imageMipTailOffset and size = imageMipTailSize.

If formatProperties.flags does not contain VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT then the offset for the mip tail in each array layer is given as:

arrayMipTailOffset = imageMipTailOffset + arrayLayer * imageMipTailStride;

and the mip tail can be bound with layerCount VkSparseMemoryBind structures, each using size = imageMipTailSize and resourceOffset = arrayMipTailOffset as defined above.

Sparse memory binding is handled by the following APIs and related data structures.

Sparse Memory Binding Functions

The VkSparseMemoryBind structure is defined as:

typedef struct VkSparseMemoryBind {
    VkDeviceSize               resourceOffset;
    VkDeviceSize               size;
    VkDeviceMemory             memory;
    VkDeviceSize               memoryOffset;
    VkSparseMemoryBindFlags    flags;
} VkSparseMemoryBind;
  • resourceOffset is the offset into the resource.

  • size is the size of the memory region to be bound.

  • memory is the VkDeviceMemory object that the range of the resource is bound to. If memory is VK_NULL_HANDLE, the range is unbound.

  • memoryOffset is the offset into the VkDeviceMemory object to bind the resource range to. If memory is VK_NULL_HANDLE, this value is ignored.

  • flags is a bitmask of VkSparseMemoryBindFlagBits specifying usage of the binding operation.

The binding range [resourceOffset, resourceOffset + size) has different constraints based on flags. If flags contains VK_SPARSE_MEMORY_BIND_METADATA_BIT, the binding range must be within the mip tail region of the metadata aspect. This metadata region is defined by:

metadataRegion = [base, base + imageMipTailSize)

base = imageMipTailOffset + imageMipTailStride × n

and imageMipTailOffset, imageMipTailSize, and imageMipTailStride values are from the VkSparseImageMemoryRequirements corresponding to the metadata aspect of the image, and n is a valid array layer index for the image,

imageMipTailStride is considered to be zero for aspects where VkSparseImageMemoryRequirements::formatProperties.flags contains VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT.

If flags does not contain VK_SPARSE_MEMORY_BIND_METADATA_BIT, the binding range must be within the range [0,VkMemoryRequirements::size).

Valid Usage
  • If memory is not VK_NULL_HANDLE, memory and memoryOffset must match the memory requirements of the resource, as described in section Resource Memory Association

  • If memory is not VK_NULL_HANDLE, memory must not have been created with a memory type that reports VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT bit set

  • size must be greater than 0

  • resourceOffset must be less than the size of the resource

  • size must be less than or equal to the size of the resource minus resourceOffset

  • memoryOffset must be less than the size of memory

  • size must be less than or equal to the size of memory minus memoryOffset

Valid Usage (Implicit)

Bits which can be set in VkSparseMemoryBind::flags, specifying usage of a sparse memory binding operation, are:

typedef enum VkSparseMemoryBindFlagBits {
    VK_SPARSE_MEMORY_BIND_METADATA_BIT = 0x00000001,
} VkSparseMemoryBindFlagBits;
  • VK_SPARSE_MEMORY_BIND_METADATA_BIT specifies that the memory being bound is only for the metadata aspect.

typedef VkFlags VkSparseMemoryBindFlags;

VkSparseMemoryBindFlags is a bitmask type for setting a mask of zero or more VkSparseMemoryBindFlagBits.

Memory is bound to VkBuffer objects created with the VK_BUFFER_CREATE_SPARSE_BINDING_BIT flag using the following structure:

typedef struct VkSparseBufferMemoryBindInfo {
    VkBuffer                     buffer;
    uint32_t                     bindCount;
    const VkSparseMemoryBind*    pBinds;
} VkSparseBufferMemoryBindInfo;
  • buffer is the VkBuffer object to be bound.

  • bindCount is the number of VkSparseMemoryBind structures in the pBinds array.

  • pBinds is a pointer to array of VkSparseMemoryBind structures.

Valid Usage (Implicit)
  • buffer must be a valid VkBuffer handle

  • pBinds must be a valid pointer to an array of bindCount valid VkSparseMemoryBind structures

  • bindCount must be greater than 0

Memory is bound to opaque regions of VkImage objects created with the VK_IMAGE_CREATE_SPARSE_BINDING_BIT flag using the following structure:

typedef struct VkSparseImageOpaqueMemoryBindInfo {
    VkImage                      image;
    uint32_t                     bindCount;
    const VkSparseMemoryBind*    pBinds;
} VkSparseImageOpaqueMemoryBindInfo;
  • image is the VkImage object to be bound.

  • bindCount is the number of VkSparseMemoryBind structures in the pBinds array.

  • pBinds is a pointer to array of VkSparseMemoryBind structures.

Valid Usage
  • If the flags member of any element of pBinds contains VK_SPARSE_MEMORY_BIND_METADATA_BIT, the binding range defined must be within the mip tail region of the metadata aspect of image

Valid Usage (Implicit)
  • image must be a valid VkImage handle

  • pBinds must be a valid pointer to an array of bindCount valid VkSparseMemoryBind structures

  • bindCount must be greater than 0

Note

This operation is normally used to bind memory to fully-resident sparse images or for mip tail regions of partially resident images. However, it can also be used to bind memory for the entire binding range of partially resident images.

In case flags does not contain VK_SPARSE_MEMORY_BIND_METADATA_BIT, the resourceOffset is in the range [0, VkMemoryRequirements::size), This range includes data from all aspects of the image, including metadata. For most implementations this will probably mean that the resourceOffset is a simple device address offset within the resource. It is possible for an application to bind a range of memory that includes both resource data and metadata. However, the application would not know what part of the image the memory is used for, or if any range is being used for metadata.

When flags contains VK_SPARSE_MEMORY_BIND_METADATA_BIT, the binding range specified must be within the mip tail region of the metadata aspect. In this case the resourceOffset is not required to be a simple device address offset within the resource. However, it is defined to be within [imageMipTailOffset, imageMipTailOffset + imageMipTailSize) for the metadata aspect. See VkSparseMemoryBind for the full constraints on binding region with this flag present.

editing-note

(Jon) The preceding NOTE refers to flags, which is presumably a reference to VkSparseMemoryBind above, even though that is not contextually clear.

Memory can be bound to sparse image blocks of VkImage objects created with the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT flag using the following structure:

typedef struct VkSparseImageMemoryBindInfo {
    VkImage                           image;
    uint32_t                          bindCount;
    const VkSparseImageMemoryBind*    pBinds;
} VkSparseImageMemoryBindInfo;
  • image is the VkImage object to be bound

  • bindCount is the number of VkSparseImageMemoryBind structures in pBinds array

  • pBinds is a pointer to array of VkSparseImageMemoryBind structures

Valid Usage
  • The subresource.mipLevel member of each element of pBinds must be less than the mipLevels specified in VkImageCreateInfo when image was created

  • The subresource.arrayLayer member of each element of pBinds must be less than the arrayLayers specified in VkImageCreateInfo when image was created

Valid Usage (Implicit)
  • image must be a valid VkImage handle

  • pBinds must be a valid pointer to an array of bindCount valid VkSparseImageMemoryBind structures

  • bindCount must be greater than 0

The VkSparseImageMemoryBind structure is defined as:

typedef struct VkSparseImageMemoryBind {
    VkImageSubresource         subresource;
    VkOffset3D                 offset;
    VkExtent3D                 extent;
    VkDeviceMemory             memory;
    VkDeviceSize               memoryOffset;
    VkSparseMemoryBindFlags    flags;
} VkSparseImageMemoryBind;
  • subresource is the aspectMask and region of interest in the image.

  • offset are the coordinates of the first texel within the image subresource to bind.

  • extent is the size in texels of the region within the image subresource to bind. The extent must be a multiple of the sparse image block dimensions, except when binding sparse image blocks along the edge of an image subresource it can instead be such that any coordinate of offset + extent equals the corresponding dimensions of the image subresource.

  • memory is the VkDeviceMemory object that the sparse image blocks of the image are bound to. If memory is VK_NULL_HANDLE, the sparse image blocks are unbound.

  • memoryOffset is an offset into VkDeviceMemory object. If memory is VK_NULL_HANDLE, this value is ignored.

  • flags are sparse memory binding flags.

Valid Usage
  • If the sparse aliased residency feature is not enabled, and if any other resources are bound to ranges of memory, the range of memory being bound must not overlap with those bound ranges

  • memory and memoryOffset must match the memory requirements of the calling command’s image, as described in section Resource Memory Association

  • subresource must be a valid image subresource for image (see Image Views)

  • offset.x must be a multiple of the sparse image block width (VkSparseImageFormatProperties::imageGranularity.width) of the image

  • extent.width must either be a multiple of the sparse image block width of the image, or else (extent.width + offset.x) must equal the width of the image subresource

  • offset.y must be a multiple of the sparse image block height (VkSparseImageFormatProperties::imageGranularity.height) of the image

  • extent.height must either be a multiple of the sparse image block height of the image, or else (extent.height + offset.y) must equal the height of the image subresource

  • offset.z must be a multiple of the sparse image block depth (VkSparseImageFormatProperties::imageGranularity.depth) of the image

  • extent.depth must either be a multiple of the sparse image block depth of the image, or else (extent.depth + offset.z) must equal the depth of the image subresource

Valid Usage (Implicit)
  • subresource must be a valid VkImageSubresource structure

  • If memory is not VK_NULL_HANDLE, memory must be a valid VkDeviceMemory handle

  • flags must be a valid combination of VkSparseMemoryBindFlagBits values

To submit sparse binding operations to a queue, call:

VkResult vkQueueBindSparse(
    VkQueue                                     queue,
    uint32_t                                    bindInfoCount,
    const VkBindSparseInfo*                     pBindInfo,
    VkFence                                     fence);
  • queue is the queue that the sparse binding operations will be submitted to.

  • bindInfoCount is the number of elements in the pBindInfo array.

  • pBindInfo is an array of VkBindSparseInfo structures, each specifying a sparse binding submission batch.

  • fence is an optional handle to a fence to be signaled. If fence is not VK_NULL_HANDLE, it defines a fence signal operation.

vkQueueBindSparse is a queue submission command, with each batch defined by an element of pBindInfo as an instance of the VkBindSparseInfo structure. Batches begin execution in the order they appear in pBindInfo, but may complete out of order.

Within a batch, a given range of a resource must not be bound more than once. Across batches, if a range is to be bound to one allocation and offset and then to another allocation and offset, then the application must guarantee (usually using semaphores) that the binding operations are executed in the correct order, as well as to order binding operations against the execution of command buffer submissions.

As no operation to vkQueueBindSparse causes any pipeline stage to access memory, synchronization primitives used in this command effectively only define execution dependencies.

Additional information about fence and semaphore operation is described in the synchronization chapter.

Valid Usage
  • If fence is not VK_NULL_HANDLE, fence must be unsignaled

  • If fence is not VK_NULL_HANDLE, fence must not be associated with any other queue command that has not yet completed execution on that queue

  • Each element of the pSignalSemaphores member of each element of pBindInfo must be unsignaled when the semaphore signal operation it defines is executed on the device

  • When a semaphore unsignal operation defined by any element of the pWaitSemaphores member of any element of pBindInfo executes on queue, no other queue must be waiting on the same semaphore.

  • All elements of the pWaitSemaphores member of all elements of pBindInfo must be semaphores that are signaled, or have semaphore signal operations previously submitted for execution.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

  • If bindInfoCount is not 0, pBindInfo must be a valid pointer to an array of bindInfoCount valid VkBindSparseInfo structures

  • If fence is not VK_NULL_HANDLE, fence must be a valid VkFence handle

  • The queue must support sparse binding operations

  • Both of fence, and queue that are valid handles must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization
  • Host access to queue must be externally synchronized

  • Host access to pBindInfo[].pWaitSemaphores[] must be externally synchronized

  • Host access to pBindInfo[].pSignalSemaphores[] must be externally synchronized

  • Host access to pBindInfo[].pBufferBinds[].buffer must be externally synchronized

  • Host access to pBindInfo[].pImageOpaqueBinds[].image must be externally synchronized

  • Host access to pBindInfo[].pImageBinds[].image must be externally synchronized

  • Host access to fence must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

The VkBindSparseInfo structure is defined as:

typedef struct VkBindSparseInfo {
    VkStructureType                             sType;
    const void*                                 pNext;
    uint32_t                                    waitSemaphoreCount;
    const VkSemaphore*                          pWaitSemaphores;
    uint32_t                                    bufferBindCount;
    const VkSparseBufferMemoryBindInfo*         pBufferBinds;
    uint32_t                                    imageOpaqueBindCount;
    const VkSparseImageOpaqueMemoryBindInfo*    pImageOpaqueBinds;
    uint32_t                                    imageBindCount;
    const VkSparseImageMemoryBindInfo*          pImageBinds;
    uint32_t                                    signalSemaphoreCount;
    const VkSemaphore*                          pSignalSemaphores;
} VkBindSparseInfo;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • waitSemaphoreCount is the number of semaphores upon which to wait before executing the sparse binding operations for the batch.

  • pWaitSemaphores is a pointer to an array of semaphores upon which to wait on before the sparse binding operations for this batch begin execution. If semaphores to wait on are provided, they define a semaphore wait operation.

  • bufferBindCount is the number of sparse buffer bindings to perform in the batch.

  • pBufferBinds is a pointer to an array of VkSparseBufferMemoryBindInfo structures.

  • imageOpaqueBindCount is the number of opaque sparse image bindings to perform.

  • pImageOpaqueBinds is a pointer to an array of VkSparseImageOpaqueMemoryBindInfo structures, indicating opaque sparse image bindings to perform.

  • imageBindCount is the number of sparse image bindings to perform.

  • pImageBinds is a pointer to an array of VkSparseImageMemoryBindInfo structures, indicating sparse image bindings to perform.

  • signalSemaphoreCount is the number of semaphores to be signaled once the sparse binding operations specified by the structure have completed execution.

  • pSignalSemaphores is a pointer to an array of semaphores which will be signaled when the sparse binding operations for this batch have completed execution. If semaphores to be signaled are provided, they define a semaphore signal operation.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_BIND_SPARSE_INFO

  • pNext must be NULL or a pointer to a valid instance of VkDeviceGroupBindSparseInfo

  • If waitSemaphoreCount is not 0, pWaitSemaphores must be a valid pointer to an array of waitSemaphoreCount valid VkSemaphore handles

  • If bufferBindCount is not 0, pBufferBinds must be a valid pointer to an array of bufferBindCount valid VkSparseBufferMemoryBindInfo structures

  • If imageOpaqueBindCount is not 0, pImageOpaqueBinds must be a valid pointer to an array of imageOpaqueBindCount valid VkSparseImageOpaqueMemoryBindInfo structures

  • If imageBindCount is not 0, pImageBinds must be a valid pointer to an array of imageBindCount valid VkSparseImageMemoryBindInfo structures

  • If signalSemaphoreCount is not 0, pSignalSemaphores must be a valid pointer to an array of signalSemaphoreCount valid VkSemaphore handles

  • Both of the elements of pSignalSemaphores, and the elements of pWaitSemaphores that are valid handles must have been created, allocated, or retrieved from the same VkDevice

If the pNext chain of VkBindSparseInfo includes a VkDeviceGroupBindSparseInfo structure, then that structure includes device indices specifying which instance of the resources and memory are bound.

The VkDeviceGroupBindSparseInfo structure is defined as:

typedef struct VkDeviceGroupBindSparseInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           resourceDeviceIndex;
    uint32_t           memoryDeviceIndex;
} VkDeviceGroupBindSparseInfo;

or the equivalent

typedef VkDeviceGroupBindSparseInfo VkDeviceGroupBindSparseInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • resourceDeviceIndex is a device index indicating which instance of the resource is bound.

  • memoryDeviceIndex is a device index indicating which instance of the memory the resource instance is bound to.

These device indices apply to all buffer and image memory binds included in the batch that points to this structure. The semaphore waits and signals for the batch are executed only by the physical device specified by the resourceDeviceIndex.

If this structure is not present, resourceDeviceIndex and memoryDeviceIndex are assumed to be zero.

Valid Usage
  • resourceDeviceIndex and memoryDeviceIndex must both be valid device indices.

  • Each memory allocation bound in this batch must have allocated an instance for memoryDeviceIndex.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO

29.8. Examples

The following examples illustrate basic creation of sparse images and binding them to physical memory.

29.8.1. Basic Sparse Resources

This basic example creates a normal VkImage object but uses fine-grained memory allocation to back the resource with multiple memory ranges.

VkDevice                device;
VkQueue                 queue;
VkImage                 sparseImage;
VkAllocationCallbacks*  pAllocator = NULL;
VkMemoryRequirements    memoryRequirements = {};
VkDeviceSize            offset = 0;
VkSparseMemoryBind      binds[MAX_CHUNKS] = {}; // MAX_CHUNKS is NOT part of Vulkan
uint32_t                bindCount = 0;

// ...

// Allocate image object
const VkImageCreateInfo sparseImageInfo =
{
    VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO,        // sType
    NULL,                                       // pNext
    VK_IMAGE_CREATE_SPARSE_BINDING_BIT | ...,   // flags
    ...
};
vkCreateImage(device, &sparseImageInfo, pAllocator, &sparseImage);

// Get memory requirements
vkGetImageMemoryRequirements(
    device,
    sparseImage,
    &memoryRequirements);

// Bind memory in fine-grained fashion, find available memory ranges
// from potentially multiple VkDeviceMemory pools.
// (Illustration purposes only, can be optimized for perf)
while (memoryRequirements.size && bindCount < MAX_CHUNKS)
{
    VkSparseMemoryBind* pBind = &binds[bindCount];
    pBind->resourceOffset = offset;

    AllocateOrGetMemoryRange(
        device,
        &memoryRequirements,
        &pBind->memory,
        &pBind->memoryOffset,
        &pBind->size);

    // memory ranges must be sized as multiples of the alignment
    assert(IsMultiple(pBind->size, memoryRequirements.alignment));
    assert(IsMultiple(pBind->memoryOffset, memoryRequirements.alignment));

    memoryRequirements.size -= pBind->size;
    offset                  += pBind->size;
    bindCount++;
}

// Ensure all image has backing
if (memoryRequirements.size)
{
    // Error condition - too many chunks
}

const VkSparseImageOpaqueMemoryBindInfo opaqueBindInfo =
{
    sparseImage,                                // image
    bindCount,                                  // bindCount
    binds                                       // pBinds
};

const VkBindSparseInfo bindSparseInfo =
{
    VK_STRUCTURE_TYPE_BIND_SPARSE_INFO,         // sType
    NULL,                                       // pNext
    ...
    1,                                          // imageOpaqueBindCount
    &opaqueBindInfo,                            // pImageOpaqueBinds
    ...
};

// vkQueueBindSparse is externally synchronized per queue object.
AcquireQueueOwnership(queue);

// Actually bind memory
vkQueueBindSparse(queue, 1, &bindSparseInfo, VK_NULL_HANDLE);

ReleaseQueueOwnership(queue);

29.8.2. Advanced Sparse Resources

This more advanced example creates an arrayed color attachment / texture image and binds only LOD zero and the required metadata to physical memory.

VkDevice                            device;
VkQueue                             queue;
VkImage                             sparseImage;
VkAllocationCallbacks*              pAllocator = NULL;
VkMemoryRequirements                memoryRequirements = {};
uint32_t                            sparseRequirementsCount = 0;
VkSparseImageMemoryRequirements*    pSparseReqs = NULL;
VkSparseMemoryBind                  binds[MY_IMAGE_ARRAY_SIZE] = {};
VkSparseImageMemoryBind             imageBinds[MY_IMAGE_ARRAY_SIZE] = {};
uint32_t                            bindCount = 0;

// Allocate image object (both renderable and sampleable)
const VkImageCreateInfo sparseImageInfo =
{
    VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO,        // sType
    NULL,                                       // pNext
    VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT | ..., // flags
    ...
    VK_FORMAT_R8G8B8A8_UNORM,                   // format
    ...
    MY_IMAGE_ARRAY_SIZE,                        // arrayLayers
    ...
    VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT |
    VK_IMAGE_USAGE_SAMPLED_BIT,                 // usage
    ...
};
vkCreateImage(device, &sparseImageInfo, pAllocator, &sparseImage);

// Get memory requirements
vkGetImageMemoryRequirements(
    device,
    sparseImage,
    &memoryRequirements);

// Get sparse image aspect properties
vkGetImageSparseMemoryRequirements(
    device,
    sparseImage,
    &sparseRequirementsCount,
    NULL);

pSparseReqs = (VkSparseImageMemoryRequirements*)
    malloc(sparseRequirementsCount * sizeof(VkSparseImageMemoryRequirements));

vkGetImageSparseMemoryRequirements(
    device,
    sparseImage,
    &sparseRequirementsCount,
    pSparseReqs);

// Bind LOD level 0 and any required metadata to memory
for (uint32_t i = 0; i < sparseRequirementsCount; ++i)
{
    if (pSparseReqs[i].formatProperties.aspectMask &
        VK_IMAGE_ASPECT_METADATA_BIT)
    {
        // Metadata must not be combined with other aspects
        assert(pSparseReqs[i].formatProperties.aspectMask ==
               VK_IMAGE_ASPECT_METADATA_BIT);

        if (pSparseReqs[i].formatProperties.flags &
            VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT)
        {
            VkSparseMemoryBind* pBind = &binds[bindCount];
            pBind->memorySize = pSparseReqs[i].imageMipTailSize;
            bindCount++;

            // ... Allocate memory range

            pBind->resourceOffset = pSparseReqs[i].imageMipTailOffset;
            pBind->memoryOffset = /* allocated memoryOffset */;
            pBind->memory = /* allocated memory */;
            pBind->flags = VK_SPARSE_MEMORY_BIND_METADATA_BIT;

        }
        else
        {
            // Need a mip tail region per array layer.
            for (uint32_t a = 0; a < sparseImageInfo.arrayLayers; ++a)
            {
                VkSparseMemoryBind* pBind = &binds[bindCount];
                pBind->memorySize = pSparseReqs[i].imageMipTailSize;
                bindCount++;

                // ... Allocate memory range

                pBind->resourceOffset = pSparseReqs[i].imageMipTailOffset +
                                        (a * pSparseReqs[i].imageMipTailStride);

                pBind->memoryOffset = /* allocated memoryOffset */;
                pBind->memory = /* allocated memory */
                pBind->flags = VK_SPARSE_MEMORY_BIND_METADATA_BIT;
            }
        }
    }
    else
    {
        // resource data
        VkExtent3D lod0BlockSize =
        {
            AlignedDivide(
                sparseImageInfo.extent.width,
                pSparseReqs[i].formatProperties.imageGranularity.width);
            AlignedDivide(
                sparseImageInfo.extent.height,
                pSparseReqs[i].formatProperties.imageGranularity.height);
            AlignedDivide(
                sparseImageInfo.extent.depth,
                pSparseReqs[i].formatProperties.imageGranularity.depth);
        }
        size_t totalBlocks =
            lod0BlockSize.width *
            lod0BlockSize.height *
            lod0BlockSize.depth;

        // Each block is the same size as the alignment requirement,
        // calculate total memory size for level 0
        VkDeviceSize lod0MemSize = totalBlocks * memoryRequirements.alignment;

        // Allocate memory for each array layer
        for (uint32_t a = 0; a < sparseImageInfo.arrayLayers; ++a)
        {
            // ... Allocate memory range

            VkSparseImageMemoryBind* pBind = &imageBinds[a];
            pBind->subresource.aspectMask = pSparseReqs[i].formatProperties.aspectMask;
            pBind->subresource.mipLevel = 0;
            pBind->subresource.arrayLayer = a;

            pBind->offset = (VkOffset3D){0, 0, 0};
            pBind->extent = sparseImageInfo.extent;
            pBind->memoryOffset = /* allocated memoryOffset */;
            pBind->memory = /* allocated memory */;
            pBind->flags = 0;
        }
    }

    free(pSparseReqs);
}

const VkSparseImageOpaqueMemoryBindInfo opaqueBindInfo =
{
    sparseImage,                                // image
    bindCount,                                  // bindCount
    binds                                       // pBinds
};

const VkSparseImageMemoryBindInfo imageBindInfo =
{
    sparseImage,                                // image
    sparseImageInfo.arrayLayers,                // bindCount
    imageBinds                                  // pBinds
};

const VkBindSparseInfo bindSparseInfo =
{
    VK_STRUCTURE_TYPE_BIND_SPARSE_INFO,         // sType
    NULL,                                       // pNext
    ...
    1,                                          // imageOpaqueBindCount
    &opaqueBindInfo,                            // pImageOpaqueBinds
    1,                                          // imageBindCount
    &imageBindInfo,                             // pImageBinds
    ...
};

// vkQueueBindSparse is externally synchronized per queue object.
AcquireQueueOwnership(queue);

// Actually bind memory
vkQueueBindSparse(queue, 1, &bindSparseInfo, VK_NULL_HANDLE);

ReleaseQueueOwnership(queue);

30. Window System Integration (WSI)

This chapter discusses the window system integration (WSI) between the Vulkan API and the various forms of displaying the results of rendering to a user. Since the Vulkan API can be used without displaying results, WSI is provided through the use of optional Vulkan extensions. This chapter provides an overview of WSI. See the appendix for additional details of each WSI extension, including which extensions must be enabled in order to use each of the functions described in this chapter.

30.1. WSI Platform

A platform is an abstraction for a window system, OS, etc. Some examples include MS Windows, Android, and Wayland. The Vulkan API may be integrated in a unique manner for each platform.

The Vulkan API does not define any type of platform object. Platform-specific WSI extensions are defined, which contain platform-specific functions for using WSI. Use of these extensions is guarded by preprocessor symbols as defined in the Window System-Specific Header Control appendix.

In order for an application to be compiled to use WSI with a given platform, it must either:

  • #define the appropriate preprocessor symbol prior to including the vulkan.h header file, or

  • include vulkan_core.h and any native platform headers, followed by the appropriate platform-specific header.

The preprocessor symbols and platform-specific headers are defined in the Window System Extensions and Headers table.

Each platform-specific extension is an instance extension. The application must enable instance extensions with vkCreateInstance before using them.

30.2. WSI Surface

Native platform surface or window objects are abstracted by surface objects, which are represented by VkSurfaceKHR handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSurfaceKHR)

The VK_KHR_surface extension declares the VkSurfaceKHR object, and provides a function for destroying VkSurfaceKHR objects. Separate platform-specific extensions each provide a function for creating a VkSurfaceKHR object for the respective platform. From the application’s perspective this is an opaque handle, just like the handles of other Vulkan objects.

Note

On certain platforms, the Vulkan loader and ICDs may have conventions that treat the handle as a pointer to a struct that contains the platform-specific information about the surface. This will be described in the documentation for the loader-ICD interface, and in the vk_icd.h header file of the LoaderAndTools source-code repository. This does not affect the loader-layer interface; layers may wrap VkSurfaceKHR objects.

editing-note

TODO: Consider replacing the above note editing note with a pointer to the loader spec when it exists. However, the information is not relevant to users of the API nor does it affect conformance of a Vulkan implementation to this spec.

30.2.1. Android Platform

To create a VkSurfaceKHR object for an Android native window, call:

VkResult vkCreateAndroidSurfaceKHR(
    VkInstance                                  instance,
    const VkAndroidSurfaceCreateInfoKHR*        pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkAndroidSurfaceCreateInfoKHR structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

During the lifetime of a surface created using a particular ANativeWindow handle any attempts to create another surface for the same ANativeWindow and any attempts to connect to the same ANativeWindow through other platform mechanisms will fail.

Note

In particular, only one VkSurfaceKHR can exist at a time for a given window. Similarly, a native window cannot be used by both a VkSurfaceKHR and EGLSurface simultaneously.

If successful, vkCreateAndroidSurfaceKHR increments the ANativeWindow’s reference count, and vkDestroySurfaceKHR will decrement it.

On Android, when a swapchain’s imageExtent does not match the surface’s currentExtent, the presentable images will be scaled to the surface’s dimensions during presentation. minImageExtent is (1,1), and maxImageExtent is the maximum image size supported by the consumer. For the system compositor, currentExtent is the window size (i.e. the consumer’s preferred size).

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkAndroidSurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

The VkAndroidSurfaceCreateInfoKHR structure is defined as:

typedef struct VkAndroidSurfaceCreateInfoKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkAndroidSurfaceCreateFlagsKHR    flags;
    struct ANativeWindow*             window;
} VkAndroidSurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • window is a pointer to the ANativeWindow to associate the surface with.

Valid Usage
  • window must point to a valid Android ANativeWindow.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_ANDROID_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

30.2.2. Mir Platform

To create a VkSurfaceKHR object for a Mir window, call:

VkResult vkCreateMirSurfaceKHR(
    VkInstance                                  instance,
    const VkMirSurfaceCreateInfoKHR*            pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkMirSurfaceCreateInfoKHR structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkMirSurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkMirSurfaceCreateInfoKHR structure is defined as:

typedef struct VkMirSurfaceCreateInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    VkMirSurfaceCreateFlagsKHR    flags;
    MirConnection*                connection;
    MirSurface*                   mirSurface;
} VkMirSurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • connection and surface are pointers to the MirConnection and MirSurface for the window to associate the surface with.

Valid Usage
  • connection must point to a valid MirConnection.

  • surface must point to a valid MirSurface.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MIR_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

On Mir, when a swapchain’s imageExtent does not match the surface’s currentExtent, the presentable images will be scaled to the surface’s dimensions during presentation. minImageExtent is (1,1), and maxImageExtent is the maximum supported surface size.

30.2.3. Wayland Platform

To create a VkSurfaceKHR object for a Wayland surface, call:

VkResult vkCreateWaylandSurfaceKHR(
    VkInstance                                  instance,
    const VkWaylandSurfaceCreateInfoKHR*        pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkWaylandSurfaceCreateInfoKHR structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkWaylandSurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkWaylandSurfaceCreateInfoKHR structure is defined as:

typedef struct VkWaylandSurfaceCreateInfoKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkWaylandSurfaceCreateFlagsKHR    flags;
    struct wl_display*                display;
    struct wl_surface*                surface;
} VkWaylandSurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • display and surface are pointers to the Wayland wl_display and wl_surface to associate the surface with.

Valid Usage
  • display must point to a valid Wayland wl_display.

  • surface must point to a valid Wayland wl_surface.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_WAYLAND_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

On Wayland, currentExtent is undefined (0xFFFFFFFF, 0xFFFFFFFF). Whatever the application sets a swapchain’s imageExtent to will be the size of the window, after the first image is presented. minImageExtent is (1,1), and maxImageExtent is the maximum supported surface size. Any calls to vkGetPhysicalDeviceSurfacePresentModesKHR on a surface created with vkCreateWaylandSurfaceKHR are required to return VK_PRESENT_MODE_MAILBOX_KHR as one of the valid present modes.

Some Vulkan functions may send protocol over the specified wl_display connection when using a swapchain or presentable images created from a VkSurfaceKHR referring to a wl_surface. Applications must therefore ensure that both the wl_display and the wl_surface remain valid for the lifetime of any VkSwapchainKHR objects created from a particular wl_display and wl_surface. Also, calling vkQueuePresentKHR will result in Vulkan sending wl_surface.commit requests to the underlying wl_surface of each VkSwapchainKHR objects referenced by pPresentInfo. If the swapchain is created with a present mode of VK_PRESENT_MODE_MAILBOX_KHR or VK_PRESENT_MODE_IMMEDIATE_KHR, then the corresponding wl_surface.attach, wl_surface.damage, and wl_surface.commit request must be issued by the implementation during the call to vkQueuePresentKHR and must not be issued by the implementation outside of vkQueuePresentKHR. This ensures that any Wayland requests sent by the client after the call to vkQueuePresentKHR returns will be received by the compositor after the wl_surface.commit. Regardless of the mode of swapchain creation, a new wl_event_queue must be created for each successful vkCreateWaylandSurfaceKHR call, and every Wayland object created by the implementation must be assigned to this event queue. If the platform provides Wayland 1.11 or greater, this must be implemented by the use of Wayland proxy object wrappers, to avoid race conditions.

If the application wishes to synchronize any window changes with a particular frame, such requests must be sent to the Wayland display server prior to calling vkQueuePresentKHR. For full control over interactions between Vulkan rendering and other Wayland protocol requests and events, a present mode of VK_PRESENT_MODE_MAILBOX_KHR should be used.

30.2.4. Win32 Platform

To create a VkSurfaceKHR object for a Win32 window, call:

VkResult vkCreateWin32SurfaceKHR(
    VkInstance                                  instance,
    const VkWin32SurfaceCreateInfoKHR*          pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkWin32SurfaceCreateInfoKHR structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkWin32SurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkWin32SurfaceCreateInfoKHR structure is defined as:

typedef struct VkWin32SurfaceCreateInfoKHR {
    VkStructureType                 sType;
    const void*                     pNext;
    VkWin32SurfaceCreateFlagsKHR    flags;
    HINSTANCE                       hinstance;
    HWND                            hwnd;
} VkWin32SurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • hinstance and hwnd are the Win32 HINSTANCE and HWND for the window to associate the surface with.

Valid Usage
  • hinstance must be a valid Win32 HINSTANCE.

  • hwnd must be a valid Win32 HWND.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_WIN32_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

With Win32, minImageExtent, maxImageExtent, and currentExtent must always equal the window size.

The currentExtent of a Win32 surface must have both width and height greater than 0, or both of them 0.

Note

Due to above restrictions, it is only possible to create a new swapchain on this platform with imageExtent being equal to the current size of the window.

The window size may become (0, 0) on this platform (e.g. when the window is minimized), and so a swapchain cannot be created until the size changes.

30.2.5. XCB Platform

To create a VkSurfaceKHR object for an X11 window, using the XCB client-side library, call:

VkResult vkCreateXcbSurfaceKHR(
    VkInstance                                  instance,
    const VkXcbSurfaceCreateInfoKHR*            pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkXcbSurfaceCreateInfoKHR structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkXcbSurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkXcbSurfaceCreateInfoKHR structure is defined as:

typedef struct VkXcbSurfaceCreateInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    VkXcbSurfaceCreateFlagsKHR    flags;
    xcb_connection_t*             connection;
    xcb_window_t                  window;
} VkXcbSurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • connection is a pointer to an xcb_connection_t to the X server.

  • window is the xcb_window_t for the X11 window to associate the surface with.

Valid Usage
  • connection must point to a valid X11 xcb_connection_t.

  • window must be a valid X11 xcb_window_t.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_XCB_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

With Xcb, minImageExtent, maxImageExtent, and currentExtent must always equal the window size.

The currentExtent of an Xcb surface must have both width and height greater than 0, or both of them 0.

Note

Due to above restrictions, it is only possible to create a new swapchain on this platform with imageExtent being equal to the current size of the window.

The window size may become (0, 0) on this platform (e.g. when the window is minimized), and so a swapchain cannot be created until the size changes.

Some Vulkan functions may send protocol over the specified xcb connection when using a swapchain or presentable images created from a VkSurfaceKHR referring to an xcb window. Applications must therefore ensure the xcb connection is available to Vulkan for the duration of any functions that manipulate such swapchains or their presentable images, and any functions that build or queue command buffers that operate on such presentable images. Specifically, applications using Vulkan with xcb-based swapchains must

  • Avoid holding a server grab on an xcb connection while waiting for Vulkan operations to complete using a swapchain derived from a different xcb connection referring to the same X server instance. Failing to do so may result in deadlock.

30.2.6. Xlib Platform

To create a VkSurfaceKHR object for an X11 window, using the Xlib client-side library, call:

VkResult vkCreateXlibSurfaceKHR(
    VkInstance                                  instance,
    const VkXlibSurfaceCreateInfoKHR*           pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance to associate the surface with.

  • pCreateInfo is a pointer to an instance of the VkXlibSurfaceCreateInfoKHR structure containing the parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkXlibSurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkXlibSurfaceCreateInfoKHR structure is defined as:

typedef struct VkXlibSurfaceCreateInfoKHR {
    VkStructureType                sType;
    const void*                    pNext;
    VkXlibSurfaceCreateFlagsKHR    flags;
    Display*                       dpy;
    Window                         window;
} VkXlibSurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • dpy is a pointer to an Xlib Display connection to the X server.

  • window is an Xlib Window to associate the surface with.

Valid Usage
  • dpy must point to a valid Xlib Display.

  • window must be a valid Xlib Window.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_XLIB_SURFACE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

With Xlib, minImageExtent, maxImageExtent, and currentExtent must always equal the window size.

The currentExtent of an Xlib surface must have both width and height greater than 0, or both of them 0.

Note

Due to above restrictions, it is only possible to create a new swapchain on this platform with imageExtent being equal to the current size of the window.

The window size may become (0, 0) on this platform (e.g. when the window is minimized), and so a swapchain cannot be created until the size changes.

Some Vulkan functions may send protocol over the specified Xlib Display connection when using a swapchain or presentable images created from a VkSurfaceKHR referring to an Xlib window. Applications must therefore ensure the display connection is available to Vulkan for the duration of any functions that manipulate such swapchains or their presentable images, and any functions that build or queue command buffers that operate on such presentable images. Specifically, applications using Vulkan with Xlib-based swapchains must

  • Avoid holding a server grab on a display connection while waiting for Vulkan operations to complete using a swapchain derived from a different display connection referring to the same X server instance. Failing to do so may result in deadlock.

Some implementations may require threads to implement some presentation modes so applications must call XInitThreads() before calling any other Xlib functions.

30.2.7. iOS Platform

To create a VkSurfaceKHR object for an iOS UIView, call:

VkResult vkCreateIOSSurfaceMVK(
    VkInstance                                  instance,
    const VkIOSSurfaceCreateInfoMVK*            pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance with which to associate the surface.

  • pCreateInfo is a pointer to an instance of the VkIOSSurfaceCreateInfoMVK structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkIOSSurfaceCreateInfoMVK structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

The VkIOSSurfaceCreateInfoMVK structure is defined as:

typedef struct VkIOSSurfaceCreateInfoMVK {
    VkStructureType               sType;
    const void*                   pNext;
    VkIOSSurfaceCreateFlagsMVK    flags;
    const void*                   pView;
} VkIOSSurfaceCreateInfoMVK;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • pView is a reference to a UIView object which will display this surface. This UIView must be backed by a CALayer instance of type CAMetalLayer.

Valid Usage
  • pView must be a valid UIView and must be backed by a CALayer instance of type CAMetalLayer.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_IOS_SURFACE_CREATE_INFO_MVK

  • pNext must be NULL

  • flags must be 0

30.2.8. macOS Platform

To create a VkSurfaceKHR object for a macOS NSView, call:

VkResult vkCreateMacOSSurfaceMVK(
    VkInstance                                  instance,
    const VkMacOSSurfaceCreateInfoMVK*          pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance with which to associate the surface.

  • pCreateInfo is a pointer to an instance of the VkMacOSSurfaceCreateInfoMVK structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkMacOSSurfaceCreateInfoMVK structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

The VkMacOSSurfaceCreateInfoMVK structure is defined as:

typedef struct VkMacOSSurfaceCreateInfoMVK {
    VkStructureType                 sType;
    const void*                     pNext;
    VkMacOSSurfaceCreateFlagsMVK    flags;
    const void*                     pView;
} VkMacOSSurfaceCreateInfoMVK;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • pView is a reference to a NSView object which will display this surface. This NSView must be backed by a CALayer instance of type CAMetalLayer.

Valid Usage
  • pView must be a valid NSView and must be backed by a CALayer instance of type CAMetalLayer.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK

  • pNext must be NULL

  • flags must be 0

30.2.9. VI Platform

To create a VkSurfaceKHR object for an nn::vi::Layer, query the layer’s native handle using nn::vi::GetNativeWindow, and then call:

VkResult vkCreateViSurfaceNN(
    VkInstance                                  instance,
    const VkViSurfaceCreateInfoNN*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance with which to associate the surface.

  • pCreateInfo is a pointer to an instance of the VkViSurfaceCreateInfoNN structure containing parameters affecting the creation of the surface object.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface object is returned.

During the lifetime of a surface created using a particular nn::vi::NativeWindowHandle any attempts to create another surface for the same nn::vi::Layer and any attempts to connect to the same nn::vi::Layer through other platform mechanisms will have undefined results.

The currentExtent of a VI surface is always undefined. Applications are expected to choose an appropriate size for the swapchain’s imageExtent (e.g., by matching the the result of a call to nn::vi::GetDisplayResolution).

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkViSurfaceCreateInfoNN structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

The VkViSurfaceCreateInfoNN structure is defined as:

typedef struct VkViSurfaceCreateInfoNN {
    VkStructureType             sType;
    const void*                 pNext;
    VkViSurfaceCreateFlagsNN    flags;
    void*                       window;
} VkViSurfaceCreateInfoNN;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use.

  • window is the nn::vi::NativeWindowHandle for the nn::vi::Layer with which to associate the surface.

Valid Usage
  • window must be a valid nn::vi::NativeWindowHandle

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_VI_SURFACE_CREATE_INFO_NN

  • pNext must be NULL

  • flags must be 0

30.2.10. Platform-Independent Information

Once created, VkSurfaceKHR objects can be used in this and other extensions, in particular the VK_KHR_swapchain extension.

Several WSI functions return VK_ERROR_SURFACE_LOST_KHR if the surface becomes no longer available. After such an error, the surface (and any child swapchain, if one exists) should be destroyed, as there is no way to restore them to a not-lost state. Applications may attempt to create a new VkSurfaceKHR using the same native platform window object, but whether such re-creation will succeed is platform-dependent and may depend on the reason the surface became unavailable. A lost surface does not otherwise cause devices to be lost.

To destroy a VkSurfaceKHR object, call:

void vkDestroySurfaceKHR(
    VkInstance                                  instance,
    VkSurfaceKHR                                surface,
    const VkAllocationCallbacks*                pAllocator);
  • instance is the instance used to create the surface.

  • surface is the surface to destroy.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

Destroying a VkSurfaceKHR merely severs the connection between Vulkan and the native surface, and does not imply destroying the native surface, closing a window, or similar behavior.

Valid Usage
  • All VkSwapchainKHR objects created for surface must have been destroyed prior to destroying surface

  • If VkAllocationCallbacks were provided when surface was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when surface was created, pAllocator must be NULL

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • If surface is not VK_NULL_HANDLE, surface must be a valid VkSurfaceKHR handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • If surface is a valid handle, it must have been created, allocated, or retrieved from instance

Host Synchronization
  • Host access to surface must be externally synchronized

30.3. Presenting Directly to Display Devices

In some environments applications can also present Vulkan rendering directly to display devices without using an intermediate windowing system. This can be useful for embedded applications, or implementing the rendering/presentation backend of a windowing system using Vulkan. The VK_KHR_display extension provides the functionality necessary to enumerate display devices and create VkSurfaceKHR objects that target displays.

30.3.1. Display Enumeration

Displays are represented by VkDisplayKHR handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDisplayKHR)

Various functions are provided for enumerating the available display devices present on a Vulkan physical device. To query information about the available displays, call:

VkResult vkGetPhysicalDeviceDisplayPropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pPropertyCount,
    VkDisplayPropertiesKHR*                     pProperties);
  • physicalDevice is a physical device.

  • pPropertyCount is a pointer to an integer related to the number of display devices available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkDisplayPropertiesKHR structures.

If pProperties is NULL, then the number of display devices available for physicalDevice is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If the value of pPropertyCount is less than the number of display devices for physicalDevice, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of display devices available for physicalDevice, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkDisplayPropertiesKHR structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDisplayPropertiesKHR structure is defined as:

typedef struct VkDisplayPropertiesKHR {
    VkDisplayKHR                  display;
    const char*                   displayName;
    VkExtent2D                    physicalDimensions;
    VkExtent2D                    physicalResolution;
    VkSurfaceTransformFlagsKHR    supportedTransforms;
    VkBool32                      planeReorderPossible;
    VkBool32                      persistentContent;
} VkDisplayPropertiesKHR;
  • display is a handle that is used to refer to the display described here. This handle will be valid for the lifetime of the Vulkan instance.

  • displayName is a pointer to a NULL-terminated string containing the name of the display. Generally, this will be the name provided by the display’s EDID. It can be NULL if no suitable name is available. If not NULL, the memory it points to must remain accessible as long as display is valid.

  • physicalDimensions describes the physical width and height of the visible portion of the display, in millimeters.

  • physicalResolution describes the physical, native, or preferred resolution of the display.

Note

For devices which have no natural value to return here, implementations should return the maximum resolution supported.

  • supportedTransforms tells which transforms are supported by this display. This will contain one or more of the bits from VkSurfaceTransformFlagsKHR.

  • planeReorderPossible tells whether the planes on this display can have their z order changed. If this is VK_TRUE, the application can re-arrange the planes on this display in any order relative to each other.

  • persistentContent tells whether the display supports self-refresh/internal buffering. If this is true, the application can submit persistent present operations on swapchains created against this display.

Note

Persistent presents may have higher latency, and may use less power when the screen content is updated infrequently, or when only a portion of the screen needs to be updated in most frames.

Acquiring and Releasing Displays

On some platforms, access to displays is limited to a single process or native driver instance. On such platforms, some or all of the displays may not be available to Vulkan if they are already in use by a native windowing system or other application.

To acquire permission to directly access a display in Vulkan from an X11 server, call:

VkResult vkAcquireXlibDisplayEXT(
    VkPhysicalDevice                            physicalDevice,
    Display*                                    dpy,
    VkDisplayKHR                                display);
  • physicalDevice The physical device the display is on.

  • dpy A connection to the X11 server that currently owns display.

  • display The display the caller wishes to control in Vulkan.

All permissions necessary to control the display are granted to the Vulkan instance associated with physicalDevice until the display is released or the X11 connection specified by dpy is terminated. Permission to access the display may be temporarily revoked during periods when the X11 server from which control was acquired itself looses access to display. During such periods, operations which require access to the display must fail with an approriate error code. If the X11 server associated with dpy does not own display, or if permission to access it has already been acquired by another entity, the call must return the error code VK_ERROR_INITIALIZATION_FAILED.

Note

One example of when an X11 server loses access to a display is when it loses ownership of its virtual terminal.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • dpy must be a valid pointer to a Display value

  • display must be a valid VkDisplayKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_INITIALIZATION_FAILED

When acquiring displays from an X11 server, an application may also wish to enumerate and identify them using a native handle rather than a VkDisplayKHR handle. To determine the VkDisplayKHR handle corresponding to an X11 RandR Output, call:

VkResult vkGetRandROutputDisplayEXT(
    VkPhysicalDevice                            physicalDevice,
    Display*                                    dpy,
    RROutput                                    rrOutput,
    VkDisplayKHR*                               pDisplay);
  • physicalDevice The physical device to query the display handle on.

  • dpy A connection to the X11 server from which rrOutput was queried.

  • rrOutput An X11 RandR output ID.

  • pDisplay The corresponding VkDisplayKHR handle will be returned here.

If there is no VkDisplayKHR corresponding to rrOutput on physicalDevice, VK_NULL_HANDLE must be returned in pDisplay.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • dpy must be a valid pointer to a Display value

  • pDisplay must be a valid pointer to a VkDisplayKHR handle

Return Codes
Success
  • VK_SUCCESS

To release a previously acquired display, call:

VkResult vkReleaseDisplayEXT(
    VkPhysicalDevice                            physicalDevice,
    VkDisplayKHR                                display);
  • physicalDevice The physical device the display is on.

  • display The display to release control of.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • display must be a valid VkDisplayKHR handle

Return Codes
Success
  • VK_SUCCESS

Display Planes

Images are presented to individual planes on a display. Devices must support at least one plane on each display. Planes can be stacked and blended to composite multiple images on one display. Devices may support only a fixed stacking order and fixed mapping between planes and displays, or they may allow arbitrary application specified stacking orders and mappings between planes and displays. To query the properties of device display planes, call:

VkResult vkGetPhysicalDeviceDisplayPlanePropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pPropertyCount,
    VkDisplayPlanePropertiesKHR*                pProperties);
  • physicalDevice is a physical device.

  • pPropertyCount is a pointer to an integer related to the number of display planes available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkDisplayPlanePropertiesKHR structures.

If pProperties is NULL, then the number of display planes available for physicalDevice is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If the value of pPropertyCount is less than the number of display planes for physicalDevice, at most pPropertyCount structures will be written.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkDisplayPlanePropertiesKHR structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDisplayPlanePropertiesKHR structure is defined as:

typedef struct VkDisplayPlanePropertiesKHR {
    VkDisplayKHR    currentDisplay;
    uint32_t        currentStackIndex;
} VkDisplayPlanePropertiesKHR;
  • currentDisplay is the handle of the display the plane is currently associated with. If the plane is not currently attached to any displays, this will be VK_NULL_HANDLE.

  • currentStackIndex is the current z-order of the plane. This will be between 0 and the value returned by vkGetPhysicalDeviceDisplayPlanePropertiesKHR in pPropertyCount.

To determine which displays a plane is usable with, call

VkResult vkGetDisplayPlaneSupportedDisplaysKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    planeIndex,
    uint32_t*                                   pDisplayCount,
    VkDisplayKHR*                               pDisplays);
  • physicalDevice is a physical device.

  • planeIndex is the plane which the application wishes to use, and must be in the range [0, physical device plane count - 1].

  • pDisplayCount is a pointer to an integer related to the number of displays available or queried, as described below.

  • pDisplays is either NULL or a pointer to an array of VkDisplayKHR handles.

If pDisplays is NULL, then the number of displays usable with the specified planeIndex for physicalDevice is returned in pDisplayCount. Otherwise, pDisplayCount must point to a variable set by the user to the number of elements in the pDisplays array, and on return the variable is overwritten with the number of handles actually written to pDisplays. If the value of pDisplayCount is less than the number of display planes for physicalDevice, at most pDisplayCount handles will be written. If pDisplayCount is smaller than the number of displays usable with the specified planeIndex for physicalDevice, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage
  • planeIndex must be less than the number of display planes supported by the device as determined by calling vkGetPhysicalDeviceDisplayPlanePropertiesKHR

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pDisplayCount must be a valid pointer to a uint32_t value

  • If the value referenced by pDisplayCount is not 0, and pDisplays is not NULL, pDisplays must be a valid pointer to an array of pDisplayCount VkDisplayKHR handles

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Additional properties of displays are queried using specialized query functions.

Display Modes

Display modes are represented by VkDisplayModeKHR handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDisplayModeKHR)

Each display has one or more supported modes associated with it by default. These built-in modes are queried by calling:

VkResult vkGetDisplayModePropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    VkDisplayKHR                                display,
    uint32_t*                                   pPropertyCount,
    VkDisplayModePropertiesKHR*                 pProperties);
  • physicalDevice is the physical device associated with display.

  • display is the display to query.

  • pPropertyCount is a pointer to an integer related to the number of display modes available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkDisplayModePropertiesKHR structures.

If pProperties is NULL, then the number of display modes available on the specified display for physicalDevice is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If the value of pPropertyCount is less than the number of display modes for physicalDevice, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of display modes available on the specified display for physicalDevice, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • display must be a valid VkDisplayKHR handle

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkDisplayModePropertiesKHR structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDisplayModePropertiesKHR structure is defined as:

typedef struct VkDisplayModePropertiesKHR {
    VkDisplayModeKHR              displayMode;
    VkDisplayModeParametersKHR    parameters;
} VkDisplayModePropertiesKHR;
  • displayMode is a handle to the display mode described in this structure. This handle will be valid for the lifetime of the Vulkan instance.

  • parameters is a VkDisplayModeParametersKHR structure describing the display parameters associated with displayMode.

The VkDisplayModeParametersKHR structure is defined as:

typedef struct VkDisplayModeParametersKHR {
    VkExtent2D    visibleRegion;
    uint32_t      refreshRate;
} VkDisplayModeParametersKHR;
  • visibleRegion is the 2D extents of the visible region.

  • refreshRate is a uint32_t that is the number of times the display is refreshed each second multiplied by 1000.

Note

For example, a 60Hz display mode would report a refreshRate of 60,000.

Additional modes may also be created by calling:

VkResult vkCreateDisplayModeKHR(
    VkPhysicalDevice                            physicalDevice,
    VkDisplayKHR                                display,
    const VkDisplayModeCreateInfoKHR*           pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDisplayModeKHR*                           pMode);
  • physicalDevice is the physical device associated with display.

  • display is the display to create an additional mode for.

  • pCreateInfo is a VkDisplayModeCreateInfoKHR structure describing the new mode to create.

  • pAllocator is the allocator used for host memory allocated for the display mode object when there is no more specific allocator available (see Memory Allocation).

  • pMode returns the handle of the mode created.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • display must be a valid VkDisplayKHR handle

  • pCreateInfo must be a valid pointer to a valid VkDisplayModeCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pMode must be a valid pointer to a VkDisplayModeKHR handle

Host Synchronization
  • Host access to display must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INITIALIZATION_FAILED

The VkDisplayModeCreateInfoKHR structure is defined as:

typedef struct VkDisplayModeCreateInfoKHR {
    VkStructureType                sType;
    const void*                    pNext;
    VkDisplayModeCreateFlagsKHR    flags;
    VkDisplayModeParametersKHR     parameters;
} VkDisplayModeCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use, and must be zero.

  • parameters is a VkDisplayModeParametersKHR structure describing the display parameters to use in creating the new mode. If the parameters are not compatible with the specified display, the implementation must return VK_ERROR_INITIALIZATION_FAILED.

Valid Usage
  • The width and height members of the visibleRegion member of parameters must be greater than 0

  • The refreshRate member of parameters must be greater than 0

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DISPLAY_MODE_CREATE_INFO_KHR

  • pNext must be NULL

  • flags must be 0

Applications that wish to present directly to a display must select which layer, or “plane” of the display they wish to target, and a mode to use with the display. Each display supports at least one plane. The capabilities of a given mode and plane combination are determined by calling:

VkResult vkGetDisplayPlaneCapabilitiesKHR(
    VkPhysicalDevice                            physicalDevice,
    VkDisplayModeKHR                            mode,
    uint32_t                                    planeIndex,
    VkDisplayPlaneCapabilitiesKHR*              pCapabilities);
  • physicalDevice is the physical device associated with display

  • mode is the display mode the application intends to program when using the specified plane. Note this parameter also implicitly specifies a display.

  • planeIndex is the plane which the application intends to use with the display, and is less than the number of display planes supported by the device.

  • pCapabilities is a pointer to a VkDisplayPlaneCapabilitiesKHR structure in which the capabilities are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • mode must be a valid VkDisplayModeKHR handle

  • pCapabilities must be a valid pointer to a VkDisplayPlaneCapabilitiesKHR structure

Host Synchronization
  • Host access to mode must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDisplayPlaneCapabilitiesKHR structure is defined as:

typedef struct VkDisplayPlaneCapabilitiesKHR {
    VkDisplayPlaneAlphaFlagsKHR    supportedAlpha;
    VkOffset2D                     minSrcPosition;
    VkOffset2D                     maxSrcPosition;
    VkExtent2D                     minSrcExtent;
    VkExtent2D                     maxSrcExtent;
    VkOffset2D                     minDstPosition;
    VkOffset2D                     maxDstPosition;
    VkExtent2D                     minDstExtent;
    VkExtent2D                     maxDstExtent;
} VkDisplayPlaneCapabilitiesKHR;
  • supportedAlpha is a bitmask of VkDisplayPlaneAlphaFlagBitsKHR describing the supported alpha blending modes.

  • minSrcPosition is the minimum source rectangle offset supported by this plane using the specified mode.

  • maxSrcPosition is the maximum source rectangle offset supported by this plane using the specified mode. The x and y components of maxSrcPosition must each be greater than or equal to the x and y components of minSrcPosition, respectively.

  • minSrcExtent is the minimum source rectangle size supported by this plane using the specified mode.

  • maxSrcExtent is the maximum source rectangle size supported by this plane using the specified mode.

  • minDstPosition, maxDstPosition, minDstExtent, maxDstExtent all have similar semantics to their corresponding *Src* equivalents, but apply to the output region within the mode rather than the input region within the source image. Unlike the *Src* offsets, minDstPosition and maxDstPosition may contain negative values.

The minimum and maximum position and extent fields describe the implementation limits, if any, as they apply to the specified display mode and plane. Vendors may support displaying a subset of a swapchain’s presentable images on the specified display plane. This is expressed by returning minSrcPosition, maxSrcPosition, minSrcExtent, and maxSrcExtent values that indicate a range of possible positions and sizes may be used to specify the region within the presentable images that source pixels will be read from when creating a swapchain on the specified display mode and plane.

Vendors may also support mapping the presentable images’ content to a subset or superset of the visible region in the specified display mode. This is expressed by returning minDstPosition, maxDstPosition, minDstExtent and maxDstExtent values that indicate a range of possible positions and sizes may be used to describe the region within the display mode that the source pixels will be mapped to.

Other vendors may support only a 1-1 mapping between pixels in the presentable images and the display mode. This may be indicated by returning (0,0) for minSrcPosition, maxSrcPosition, minDstPosition, and maxDstPosition, and (display mode width, display mode height) for minSrcExtent, maxSrcExtent, minDstExtent, and maxDstExtent.

These values indicate the limits of the implementation’s individual fields. Not all combinations of values within the offset and extent ranges returned in VkDisplayPlaneCapabilitiesKHR are guaranteed to be supported. Vendors may still fail presentation requests that specify unsupported combinations.

30.3.2. Display Control

To set the power state of a display, call:

VkResult vkDisplayPowerControlEXT(
    VkDevice                                    device,
    VkDisplayKHR                                display,
    const VkDisplayPowerInfoEXT*                pDisplayPowerInfo);
  • device is a logical device associated with display.

  • display is the display whose power state is modified.

  • pDisplayPowerInfo is an instance of VkDisplayPowerInfoEXT specifying the new power state of display.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • display must be a valid VkDisplayKHR handle

  • pDisplayPowerInfo must be a valid pointer to a valid VkDisplayPowerInfoEXT structure

Return Codes
Success
  • VK_SUCCESS

The VkDisplayPowerInfoEXT structure is defined as:

typedef struct VkDisplayPowerInfoEXT {
    VkStructureType           sType;
    const void*               pNext;
    VkDisplayPowerStateEXT    powerState;
} VkDisplayPowerInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • powerState is a VkDisplayPowerStateEXT value specifying the new power state of the display.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DISPLAY_POWER_INFO_EXT

  • pNext must be NULL

  • powerState must be a valid VkDisplayPowerStateEXT value

Possible values of VkDisplayPowerInfoEXT::powerState, specifying the new power state of a display, are:

typedef enum VkDisplayPowerStateEXT {
    VK_DISPLAY_POWER_STATE_OFF_EXT = 0,
    VK_DISPLAY_POWER_STATE_SUSPEND_EXT = 1,
    VK_DISPLAY_POWER_STATE_ON_EXT = 2,
} VkDisplayPowerStateEXT;
  • VK_DISPLAY_POWER_STATE_OFF_EXT specifies that the display is powered down.

  • VK_DISPLAY_POWER_STATE_SUSPEND_EXT specifies that the display is put into a low power mode, from which it may be able to transition back to VK_DISPLAY_POWER_STATE_ON_EXT more quickly than if it were in VK_DISPLAY_POWER_STATE_OFF_EXT. This state may be the same as VK_DISPLAY_POWER_STATE_OFF_EXT.

  • VK_DISPLAY_POWER_STATE_ON_EXT specifies that the display is powered on.

30.3.3. Display Surfaces

A complete display configuration includes a mode, one or more display planes and any parameters describing their behavior, and parameters describing some aspects of the images associated with those planes. Display surfaces describe the configuration of a single plane within a complete display configuration. To create a VkSurfaceKHR structure for a display surface, call:

VkResult vkCreateDisplayPlaneSurfaceKHR(
    VkInstance                                  instance,
    const VkDisplaySurfaceCreateInfoKHR*        pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSurfaceKHR*                               pSurface);
  • instance is the instance corresponding to the physical device the targeted display is on.

  • pCreateInfo is a pointer to an instance of the VkDisplaySurfaceCreateInfoKHR structure specifying which mode, plane, and other parameters to use, as described below.

  • pAllocator is the allocator used for host memory allocated for the surface object when there is no more specific allocator available (see Memory Allocation).

  • pSurface points to a VkSurfaceKHR handle in which the created surface is returned.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkDisplaySurfaceCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSurface must be a valid pointer to a VkSurfaceKHR handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDisplaySurfaceCreateInfoKHR structure is defined as:

typedef struct VkDisplaySurfaceCreateInfoKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkDisplaySurfaceCreateFlagsKHR    flags;
    VkDisplayModeKHR                  displayMode;
    uint32_t                          planeIndex;
    uint32_t                          planeStackIndex;
    VkSurfaceTransformFlagBitsKHR     transform;
    float                             globalAlpha;
    VkDisplayPlaneAlphaFlagBitsKHR    alphaMode;
    VkExtent2D                        imageExtent;
} VkDisplaySurfaceCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is reserved for future use, and must be zero.

  • displayMode is a VkDisplayModeKHR handle specifying the mode to use when displaying this surface.

  • planeIndex is the plane on which this surface appears.

  • planeStackIndex is the z-order of the plane.

  • transform is a VkSurfaceTransformFlagBitsKHR value specifying the transformation to apply to images as part of the scanout operation.

  • globalAlpha is the global alpha value. This value is ignored if alphaMode is not VK_DISPLAY_PLANE_ALPHA_GLOBAL_BIT_KHR.

  • alphaMode is a VkDisplayPlaneAlphaFlagBitsKHR value specifying the type of alpha blending to use.

  • imageExtent The size of the presentable images to use with the surface.

Note

Creating a display surface must not modify the state of the displays, planes, or other resources it names. For example, it must not apply the specified mode to be set on the associated display. Application of display configuration occurs as a side effect of presenting to a display surface.

Valid Usage
  • planeIndex must be less than the number of display planes supported by the device as determined by calling vkGetPhysicalDeviceDisplayPlanePropertiesKHR

  • If the planeReorderPossible member of the VkDisplayPropertiesKHR structure returned by vkGetPhysicalDeviceDisplayPropertiesKHR for the display corresponding to displayMode is VK_TRUE then planeStackIndex must be less than the number of display planes supported by the device as determined by calling vkGetPhysicalDeviceDisplayPlanePropertiesKHR; otherwise planeStackIndex must equal the currentStackIndex member of VkDisplayPlanePropertiesKHR returned by vkGetPhysicalDeviceDisplayPlanePropertiesKHR for the display plane corresponding to displayMode

  • If alphaMode is VK_DISPLAY_PLANE_ALPHA_GLOBAL_BIT_KHR then globalAlpha must be between 0 and 1, inclusive

  • alphaMode must be 0 or one of the bits present in the supportedAlpha member of VkDisplayPlaneCapabilitiesKHR returned by vkGetDisplayPlaneCapabilitiesKHR for the display plane corresponding to displayMode

  • The width and height members of imageExtent must be less than the maxImageDimensions2D member of VkPhysicalDeviceLimits

Valid Usage (Implicit)

Possible values of VkDisplaySurfaceCreateInfoKHR::alphaMode, specifying the type of alpha blending to use on a display, are:

typedef enum VkDisplayPlaneAlphaFlagBitsKHR {
    VK_DISPLAY_PLANE_ALPHA_OPAQUE_BIT_KHR = 0x00000001,
    VK_DISPLAY_PLANE_ALPHA_GLOBAL_BIT_KHR = 0x00000002,
    VK_DISPLAY_PLANE_ALPHA_PER_PIXEL_BIT_KHR = 0x00000004,
    VK_DISPLAY_PLANE_ALPHA_PER_PIXEL_PREMULTIPLIED_BIT_KHR = 0x00000008,
} VkDisplayPlaneAlphaFlagBitsKHR;
  • VK_DISPLAY_PLANE_ALPHA_OPAQUE_BIT_KHR specifies that the source image will be treated as opaque.

  • VK_DISPLAY_PLANE_ALPHA_GLOBAL_BIT_KHR specifies that a global alpha value must be specified that will be applied to all pixels in the source image.

  • VK_DISPLAY_PLANE_ALPHA_PER_PIXEL_BIT_KHR specifies that the alpha value will be determined by the alpha channel of the source image’s pixels. If the source format contains no alpha values, no blending will be applied. The source alpha values are not premultiplied into the source image’s other color channels.

  • VK_DISPLAY_PLANE_ALPHA_PER_PIXEL_PREMULTIPLIED_BIT_KHR is equivalent to VK_DISPLAY_PLANE_ALPHA_PER_PIXEL_BIT_KHR, except the source alpha values are assumed to be premultiplied into the source image’s other color channels.

typedef VkFlags VkDisplayPlaneAlphaFlagsKHR;

VkDisplayPlaneAlphaFlagsKHR is a bitmask type for setting a mask of zero or more VkDisplayPlaneAlphaFlagBitsKHR.

30.4. Querying for WSI Support

Not all physical devices will include WSI support. Within a physical device, not all queue families will support presentation. WSI support and compatibility can be determined in a platform-neutral manner (which determines support for presentation to a particular surface object) and additionally may be determined in platform-specific manners (which determine support for presentation on the specified physical device but do not guarantee support for presentation to a particular surface object).

To determine whether a queue family of a physical device supports presentation to a given surface, call:

VkResult vkGetPhysicalDeviceSurfaceSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    VkSurfaceKHR                                surface,
    VkBool32*                                   pSupported);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family.

  • surface is the surface.

  • pSupported is a pointer to a VkBool32, which is set to VK_TRUE to indicate support, and VK_FALSE otherwise.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pSupported must be a valid pointer to a VkBool32 value

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

30.4.1. Android Platform

On Android, all physical devices and queue families must be capable of presentation with any native window. As a result there is no Android-specific query for these capabilities.

30.4.2. Mir Platform

To determine whether a queue family of a physical device supports presentation to the Mir compositor, call:

VkBool32 vkGetPhysicalDeviceMirPresentationSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    MirConnection*                              connection);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family index.

  • connection is a pointer to the MirConnection, and identifies the desired Mir compositor.

This platform-specific function can be called prior to creating a surface.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • connection must be a valid pointer to a MirConnection value

30.4.3. Wayland Platform

To determine whether a queue family of a physical device supports presentation to a Wayland compositor, call:

VkBool32 vkGetPhysicalDeviceWaylandPresentationSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    struct wl_display*                          display);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family index.

  • display is a pointer to the wl_display associated with a Wayland compositor.

This platform-specific function can be called prior to creating a surface.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • display must be a valid pointer to a wl_display value

30.4.4. Win32 Platform

To determine whether a queue family of a physical device supports presentation to the Microsoft Windows desktop, call:

VkBool32 vkGetPhysicalDeviceWin32PresentationSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family index.

This platform-specific function can be called prior to creating a surface.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

30.4.5. XCB Platform

To determine whether a queue family of a physical device supports presentation to an X11 server, using the XCB client-side library, call:

VkBool32 vkGetPhysicalDeviceXcbPresentationSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    xcb_connection_t*                           connection,
    xcb_visualid_t                              visual_id);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family index.

  • connection is a pointer to an xcb_connection_t to the X server. visual_id is an X11 visual (xcb_visualid_t).

This platform-specific function can be called prior to creating a surface.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • connection must be a valid pointer to a xcb_connection_t value

30.4.6. Xlib Platform

To determine whether a queue family of a physical device supports presentation to an X11 server, using the Xlib client-side library, call:

VkBool32 vkGetPhysicalDeviceXlibPresentationSupportKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    Display*                                    dpy,
    VisualID                                    visualID);
  • physicalDevice is the physical device.

  • queueFamilyIndex is the queue family index.

  • dpy is a pointer to an Xlib Display connection to the server.

  • visualId is an X11 visual (VisualID).

This platform-specific function can be called prior to creating a surface.

Valid Usage
  • queueFamilyIndex must be less than pQueueFamilyPropertyCount returned by vkGetPhysicalDeviceQueueFamilyProperties for the given physicalDevice

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • dpy must be a valid pointer to a Display value

30.4.7. iOS Platform

On iOS, all physical devices and queue families must be capable of presentation with any layer. As a result there is no iOS-specific query for these capabilities.

30.4.8. macOS Platform

On macOS, all physical devices and queue families must be capable of presentation with any layer. As a result there is no macOS-specific query for these capabilities.

30.4.9. VI Platform

On VI, all physical devices and queue families must be capable of presentation with any layer. As a result there is no VI-specific query for these capabilities.

30.5. Surface Queries

The capabilities of a swapchain targetting a surface are the intersection of the capabilities of the WSI platform, the native window or display, and the physical device. The resulting capabilities can be obtained with the queries listed below in this section. Capabilities that correspond to image creation parameters are not independent of each other: combinations of parameters that are not supported as reported by vkGetPhysicalDeviceImageFormatProperties are not supported by the surface on that physical device, even if the capabilities taken individually are supported as part of some other parameter combinations.

To query the basic capabilities of a surface, needed in order to create a swapchain, call:

VkResult vkGetPhysicalDeviceSurfaceCapabilitiesKHR(
    VkPhysicalDevice                            physicalDevice,
    VkSurfaceKHR                                surface,
    VkSurfaceCapabilitiesKHR*                   pSurfaceCapabilities);
  • physicalDevice is the physical device that will be associated with the swapchain to be created, as described for vkCreateSwapchainKHR.

  • surface is the surface that will be associated with the swapchain.

  • pSurfaceCapabilities is a pointer to an instance of the VkSurfaceCapabilitiesKHR structure in which the capabilities are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pSurfaceCapabilities must be a valid pointer to a VkSurfaceCapabilitiesKHR structure

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

The VkSurfaceCapabilitiesKHR structure is defined as:

typedef struct VkSurfaceCapabilitiesKHR {
    uint32_t                         minImageCount;
    uint32_t                         maxImageCount;
    VkExtent2D                       currentExtent;
    VkExtent2D                       minImageExtent;
    VkExtent2D                       maxImageExtent;
    uint32_t                         maxImageArrayLayers;
    VkSurfaceTransformFlagsKHR       supportedTransforms;
    VkSurfaceTransformFlagBitsKHR    currentTransform;
    VkCompositeAlphaFlagsKHR         supportedCompositeAlpha;
    VkImageUsageFlags                supportedUsageFlags;
} VkSurfaceCapabilitiesKHR;
  • minImageCount is the minimum number of images the specified device supports for a swapchain created for the surface, and will be at least one.

  • maxImageCount is the maximum number of images the specified device supports for a swapchain created for the surface, and will be either 0, or greater than or equal to minImageCount. A value of 0 means that there is no limit on the number of images, though there may be limits related to the total amount of memory used by presentable images.

  • currentExtent is the current width and height of the surface, or the special value (0xFFFFFFFF, 0xFFFFFFFF) indicating that the surface size will be determined by the extent of a swapchain targeting the surface.

  • minImageExtent contains the smallest valid swapchain extent for the surface on the specified device. The width and height of the extent will each be less than or equal to the corresponding width and height of currentExtent, unless currentExtent has the special value described above.

  • maxImageExtent contains the largest valid swapchain extent for the surface on the specified device. The width and height of the extent will each be greater than or equal to the corresponding width and height of minImageExtent. The width and height of the extent will each be greater than or equal to the corresponding width and height of currentExtent, unless currentExtent has the special value described above.

  • maxImageArrayLayers is the maximum number of layers presentable images can have for a swapchain created for this device and surface, and will be at least one.

  • supportedTransforms is a bitmask of VkSurfaceTransformFlagBitsKHR indicating the presentation transforms supported for the surface on the specified device. At least one bit will be set.

  • currentTransform is VkSurfaceTransformFlagBitsKHR value indicating the surface’s current transform relative to the presentation engine’s natural orientation.

  • supportedCompositeAlpha is a bitmask of VkCompositeAlphaFlagBitsKHR, representing the alpha compositing modes supported by the presentation engine for the surface on the specified device, and at least one bit will be set. Opaque composition can be achieved in any alpha compositing mode by either using an image format that has no alpha component, or by ensuring that all pixels in the presentable images have an alpha value of 1.0.

  • supportedUsageFlags is a bitmask of VkImageUsageFlagBits representing the ways the application can use the presentable images of a swapchain created with VkPresentModeKHR set to VK_PRESENT_MODE_IMMEDIATE_KHR, VK_PRESENT_MODE_MAILBOX_KHR, VK_PRESENT_MODE_FIFO_KHR or VK_PRESENT_MODE_FIFO_RELAXED_KHR for the surface on the specified device. VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT must be included in the set but implementations may support additional usages.

Note

Supported usage flags of a presentable image when using VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR presentation mode are provided by VkSharedPresentSurfaceCapabilitiesKHR::sharedPresentSupportedUsageFlags.

Note

Formulas such as min(N, maxImageCount) are not correct, since maxImageCount may be zero.

To query the basic capabilities of a surface defined by the core or extensions, call:

VkResult vkGetPhysicalDeviceSurfaceCapabilities2KHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceSurfaceInfo2KHR*      pSurfaceInfo,
    VkSurfaceCapabilities2KHR*                  pSurfaceCapabilities);

vkGetPhysicalDeviceSurfaceCapabilities2KHR behaves similarly to vkGetPhysicalDeviceSurfaceCapabilitiesKHR, with the ability to specify extended inputs via chained input structures, and to return extended information via chained output structures.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pSurfaceInfo must be a valid pointer to a valid VkPhysicalDeviceSurfaceInfo2KHR structure

  • pSurfaceCapabilities must be a valid pointer to a VkSurfaceCapabilities2KHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

The VkPhysicalDeviceSurfaceInfo2KHR structure is defined as:

typedef struct VkPhysicalDeviceSurfaceInfo2KHR {
    VkStructureType    sType;
    const void*        pNext;
    VkSurfaceKHR       surface;
} VkPhysicalDeviceSurfaceInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • surface is the surface that will be associated with the swapchain.

The members of VkPhysicalDeviceSurfaceInfo2KHR correspond to the arguments to vkGetPhysicalDeviceSurfaceCapabilitiesKHR, with sType and pNext added for extensibility.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SURFACE_INFO_2_KHR

  • pNext must be NULL

  • surface must be a valid VkSurfaceKHR handle

The VkSurfaceCapabilities2KHR structure is defined as:

typedef struct VkSurfaceCapabilities2KHR {
    VkStructureType             sType;
    void*                       pNext;
    VkSurfaceCapabilitiesKHR    surfaceCapabilities;
} VkSurfaceCapabilities2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • surfaceCapabilities is a structure of type VkSurfaceCapabilitiesKHR describing the capabilities of the specified surface.

Valid Usage (Implicit)

The VkSharedPresentSurfaceCapabilitiesKHR structure is defined as:

typedef struct VkSharedPresentSurfaceCapabilitiesKHR {
    VkStructureType      sType;
    void*                pNext;
    VkImageUsageFlags    sharedPresentSupportedUsageFlags;
} VkSharedPresentSurfaceCapabilitiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • sharedPresentSupportedUsageFlags is a bitmask of VkImageUsageFlagBits representing the ways the application can use the shared presentable image from a swapchain created with VkPresentModeKHR set to VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR for the surface on the specified device. VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT must be included in the set but implementations may support additional usages.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SHARED_PRESENT_SURFACE_CAPABILITIES_KHR

To query the basic capabilities of a surface, needed in order to create a swapchain, call:

VkResult vkGetPhysicalDeviceSurfaceCapabilities2EXT(
    VkPhysicalDevice                            physicalDevice,
    VkSurfaceKHR                                surface,
    VkSurfaceCapabilities2EXT*                  pSurfaceCapabilities);
  • physicalDevice is the physical device that will be associated with the swapchain to be created, as described for vkCreateSwapchainKHR.

  • surface is the surface that will be associated with the swapchain.

  • pSurfaceCapabilities is a pointer to an instance of the VkSurfaceCapabilities2EXT structure in which the capabilities are returned.

vkGetPhysicalDeviceSurfaceCapabilities2EXT behaves similarly to vkGetPhysicalDeviceSurfaceCapabilitiesKHR, with the ability to return extended information by adding extension structures to the pNext chain of its pSurfaceCapabilities parameter.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pSurfaceCapabilities must be a valid pointer to a VkSurfaceCapabilities2EXT structure

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

The VkSurfaceCapabilities2EXT structure is defined as:

typedef struct VkSurfaceCapabilities2EXT {
    VkStructureType                  sType;
    void*                            pNext;
    uint32_t                         minImageCount;
    uint32_t                         maxImageCount;
    VkExtent2D                       currentExtent;
    VkExtent2D                       minImageExtent;
    VkExtent2D                       maxImageExtent;
    uint32_t                         maxImageArrayLayers;
    VkSurfaceTransformFlagsKHR       supportedTransforms;
    VkSurfaceTransformFlagBitsKHR    currentTransform;
    VkCompositeAlphaFlagsKHR         supportedCompositeAlpha;
    VkImageUsageFlags                supportedUsageFlags;
    VkSurfaceCounterFlagsEXT         supportedSurfaceCounters;
} VkSurfaceCapabilities2EXT;

All members of VkSurfaceCapabilities2EXT are identical to the corresponding members of VkSurfaceCapabilitiesKHR where one exists. The remaining members are:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • supportedSurfaceCounters is a bitmask of VkSurfaceCounterFlagBitsEXT indicating the supported surface counter types.

Valid Usage
  • supportedSurfaceCounters must not include VK_SURFACE_COUNTER_VBLANK_EXT unless the surface queried is a display surface.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SURFACE_CAPABILITIES_2_EXT

  • pNext must be NULL

Bits which can be set in VkSurfaceCapabilities2EXT::supportedSurfaceCounters, indicating supported surface counter types, are:

typedef enum VkSurfaceCounterFlagBitsEXT {
    VK_SURFACE_COUNTER_VBLANK_EXT = 0x00000001,
} VkSurfaceCounterFlagBitsEXT;
  • VK_SURFACE_COUNTER_VBLANK_EXT specifies a counter incrementing once every time a vertical blanking period occurs on the display associated with the surface.

typedef VkFlags VkSurfaceCounterFlagsEXT;

VkSurfaceCounterFlagsEXT is a bitmask type for setting a mask of zero or more VkSurfaceCounterFlagBitsEXT.

Bits which may be set in VkSurfaceCapabilitiesKHR::supportedTransforms indicating the presentation transforms supported for the surface on the specified device, and possible values of VkSurfaceCapabilitiesKHR::currentTransform is indicating the surface’s current transform relative to the presentation engine’s natural orientation, are:

typedef enum VkSurfaceTransformFlagBitsKHR {
    VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR = 0x00000001,
    VK_SURFACE_TRANSFORM_ROTATE_90_BIT_KHR = 0x00000002,
    VK_SURFACE_TRANSFORM_ROTATE_180_BIT_KHR = 0x00000004,
    VK_SURFACE_TRANSFORM_ROTATE_270_BIT_KHR = 0x00000008,
    VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_BIT_KHR = 0x00000010,
    VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_90_BIT_KHR = 0x00000020,
    VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_180_BIT_KHR = 0x00000040,
    VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_270_BIT_KHR = 0x00000080,
    VK_SURFACE_TRANSFORM_INHERIT_BIT_KHR = 0x00000100,
} VkSurfaceTransformFlagBitsKHR;
  • VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR specifies that image content is presented without being transformed.

  • VK_SURFACE_TRANSFORM_ROTATE_90_BIT_KHR specifies that image content is rotated 90 degrees clockwise.

  • VK_SURFACE_TRANSFORM_ROTATE_180_BIT_KHR specifies that image content is rotated 180 degrees clockwise.

  • VK_SURFACE_TRANSFORM_ROTATE_270_BIT_KHR specifies that image content is rotated 270 degrees clockwise.

  • VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_BIT_KHR specifies that image content is mirrored horizontally.

  • VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_90_BIT_KHR specifies that image content is mirrored horizontally, then rotated 90 degrees clockwise.

  • VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_180_BIT_KHR specifies that image content is mirrored horizontally, then rotated 180 degrees clockwise.

  • VK_SURFACE_TRANSFORM_HORIZONTAL_MIRROR_ROTATE_270_BIT_KHR specifies that image content is mirrored horizontally, then rotated 270 degrees clockwise.

  • VK_SURFACE_TRANSFORM_INHERIT_BIT_KHR specifies that the presentation transform is not specified, and is instead determined by platform-specific considerations and mechanisms outside Vulkan.

typedef VkFlags VkSurfaceTransformFlagsKHR;

VkSurfaceTransformFlagsKHR is a bitmask type for setting a mask of zero or more VkSurfaceTransformFlagBitsKHR.

The supportedCompositeAlpha member is of type VkCompositeAlphaFlagBitsKHR, which contains the following values:

typedef enum VkCompositeAlphaFlagBitsKHR {
    VK_COMPOSITE_ALPHA_OPAQUE_BIT_KHR = 0x00000001,
    VK_COMPOSITE_ALPHA_PRE_MULTIPLIED_BIT_KHR = 0x00000002,
    VK_COMPOSITE_ALPHA_POST_MULTIPLIED_BIT_KHR = 0x00000004,
    VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR = 0x00000008,
} VkCompositeAlphaFlagBitsKHR;

These values are described as follows:

  • VK_COMPOSITE_ALPHA_OPAQUE_BIT_KHR: The alpha channel, if it exists, of the images is ignored in the compositing process. Instead, the image is treated as if it has a constant alpha of 1.0.

  • VK_COMPOSITE_ALPHA_PRE_MULTIPLIED_BIT_KHR: The alpha channel, if it exists, of the images is respected in the compositing process. The non-alpha channels of the image are expected to already be multiplied by the alpha channel by the application.

  • VK_COMPOSITE_ALPHA_POST_MULTIPLIED_BIT_KHR: The alpha channel, if it exists, of the images is respected in the compositing process. The non-alpha channels of the image are not expected to already be multiplied by the alpha channel by the application; instead, the compositor will multiply the non-alpha channels of the image by the alpha channel during compositing.

  • VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR: The way in which the presentation engine treats the alpha channel in the images is unknown to the Vulkan API. Instead, the application is responsible for setting the composite alpha blending mode using native window system commands. If the application does not set the blending mode using native window system commands, then a platform-specific default will be used.

typedef VkFlags VkCompositeAlphaFlagsKHR;

VkCompositeAlphaFlagsKHR is a bitmask type for setting a mask of zero or more VkCompositeAlphaFlagBitsKHR.

To query the supported swapchain format-color space pairs for a surface, call:

VkResult vkGetPhysicalDeviceSurfaceFormatsKHR(
    VkPhysicalDevice                            physicalDevice,
    VkSurfaceKHR                                surface,
    uint32_t*                                   pSurfaceFormatCount,
    VkSurfaceFormatKHR*                         pSurfaceFormats);
  • physicalDevice is the physical device that will be associated with the swapchain to be created, as described for vkCreateSwapchainKHR.

  • surface is the surface that will be associated with the swapchain.

  • pSurfaceFormatCount is a pointer to an integer related to the number of format pairs available or queried, as described below.

  • pSurfaceFormats is either NULL or a pointer to an array of VkSurfaceFormatKHR structures.

If pSurfaceFormats is NULL, then the number of format pairs supported for the given surface is returned in pSurfaceFormatCount. The number of format pairs supported will be greater than or equal to 1. Otherwise, pSurfaceFormatCount must point to a variable set by the user to the number of elements in the pSurfaceFormats array, and on return the variable is overwritten with the number of structures actually written to pSurfaceFormats. If the value of pSurfaceFormatCount is less than the number of format pairs supported, at most pSurfaceFormatCount structures will be written. If pSurfaceFormatCount is smaller than the number of format pairs supported for the given surface, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pSurfaceFormatCount must be a valid pointer to a uint32_t value

  • If the value referenced by pSurfaceFormatCount is not 0, and pSurfaceFormats is not NULL, pSurfaceFormats must be a valid pointer to an array of pSurfaceFormatCount VkSurfaceFormatKHR structures

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

The VkSurfaceFormatKHR structure is defined as:

typedef struct VkSurfaceFormatKHR {
    VkFormat           format;
    VkColorSpaceKHR    colorSpace;
} VkSurfaceFormatKHR;
  • format is a VkFormat that is compatible with the specified surface.

  • colorSpace is a presentation VkColorSpaceKHR that is compatible with the surface.

To query the supported swapchain format tuples for a surface, call:

VkResult vkGetPhysicalDeviceSurfaceFormats2KHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceSurfaceInfo2KHR*      pSurfaceInfo,
    uint32_t*                                   pSurfaceFormatCount,
    VkSurfaceFormat2KHR*                        pSurfaceFormats);
  • physicalDevice is the physical device that will be associated with the swapchain to be created, as described for vkCreateSwapchainKHR.

  • pSurfaceInfo points to an instance of the VkPhysicalDeviceSurfaceInfo2KHR structure, describing the surface and other fixed parameters that would be consumed by vkCreateSwapchainKHR.

  • pSurfaceFormatCount is a pointer to an integer related to the number of format tuples available or queried, as described below.

  • pSurfaceFormats is either NULL or a pointer to an array of VkSurfaceFormat2KHR structures.

If pSurfaceFormats is NULL, then the number of format tuples supported for the given surface is returned in pSurfaceFormatCount. The number of format tuples supported will be greater than or equal to 1. Otherwise, pSurfaceFormatCount must point to a variable set by the user to the number of elements in the pSurfaceFormats array, and on return the variable is overwritten with the number of structures actually written to pSurfaceFormats. If the value of pSurfaceFormatCount is less than the number of format tuples supported, at most pSurfaceFormatCount structures will be written. If pSurfaceFormatCount is smaller than the number of format tuples supported for the surface parameters described in pSurfaceInfo, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pSurfaceInfo must be a valid pointer to a valid VkPhysicalDeviceSurfaceInfo2KHR structure

  • pSurfaceFormatCount must be a valid pointer to a uint32_t value

  • If the value referenced by pSurfaceFormatCount is not 0, and pSurfaceFormats is not NULL, pSurfaceFormats must be a valid pointer to an array of pSurfaceFormatCount VkSurfaceFormat2KHR structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

The VkSurfaceFormat2KHR structure is defined as:

typedef struct VkSurfaceFormat2KHR {
    VkStructureType       sType;
    void*                 pNext;
    VkSurfaceFormatKHR    surfaceFormat;
} VkSurfaceFormat2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • surfaceFormat is an instance of VkSurfaceFormatKHR describing a format-color space pair that is compatible with the specified surface.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SURFACE_FORMAT_2_KHR

  • pNext must be NULL

While the format of a presentable image refers to the encoding of each pixel, the colorSpace determines how the presentation engine interprets the pixel values. A color space in this document refers to a specific color space (defined by the chromaticities of its primaries and a white point in CIE Lab), and a transfer function that is applied before storing or transmitting color data in the given color space.

Possible values of VkSurfaceFormatKHR::colorSpace, specifying supported color spaces of a presentation engine, are:

typedef enum VkColorSpaceKHR {
    VK_COLOR_SPACE_SRGB_NONLINEAR_KHR = 0,
    VK_COLOR_SPACE_DISPLAY_P3_NONLINEAR_EXT = 1000104001,
    VK_COLOR_SPACE_EXTENDED_SRGB_LINEAR_EXT = 1000104002,
    VK_COLOR_SPACE_DCI_P3_LINEAR_EXT = 1000104003,
    VK_COLOR_SPACE_DCI_P3_NONLINEAR_EXT = 1000104004,
    VK_COLOR_SPACE_BT709_LINEAR_EXT = 1000104005,
    VK_COLOR_SPACE_BT709_NONLINEAR_EXT = 1000104006,
    VK_COLOR_SPACE_BT2020_LINEAR_EXT = 1000104007,
    VK_COLOR_SPACE_HDR10_ST2084_EXT = 1000104008,
    VK_COLOR_SPACE_DOLBYVISION_EXT = 1000104009,
    VK_COLOR_SPACE_HDR10_HLG_EXT = 1000104010,
    VK_COLOR_SPACE_ADOBERGB_LINEAR_EXT = 1000104011,
    VK_COLOR_SPACE_ADOBERGB_NONLINEAR_EXT = 1000104012,
    VK_COLOR_SPACE_PASS_THROUGH_EXT = 1000104013,
    VK_COLOR_SPACE_EXTENDED_SRGB_NONLINEAR_EXT = 1000104014,
} VkColorSpaceKHR;
  • VK_COLOR_SPACE_SRGB_NONLINEAR_KHR specifies support for the sRGB color space.

  • VK_COLOR_SPACE_DISPLAY_P3_NONLINEAR_EXT specifies support for the Display-P3 color space and applies an sRGB-like transfer function (defined below).

  • VK_COLOR_SPACE_EXTENDED_SRGB_LINEAR_EXT specifies support for the extended sRGB color space and applies a linear transfer function.

  • VK_COLOR_SPACE_EXTENDED_SRGB_NONLINEAR_EXT specifies support for the extended sRGB color space and applies an sRGB transfer function.

  • VK_COLOR_SPACE_DCI_P3_LINEAR_EXT specifies support for the DCI-P3 color space and applies a linear OETF.

  • VK_COLOR_SPACE_DCI_P3_NONLINEAR_EXT specifies support for the DCI-P3 color space and applies the Gamma 2.6 OETF.

  • VK_COLOR_SPACE_BT709_LINEAR_EXT specifies support for the BT709 color space and applies a linear OETF.

  • VK_COLOR_SPACE_BT709_NONLINEAR_EXT specifies support for the BT709 color space and applies the SMPTE 170M OETF.

  • VK_COLOR_SPACE_BT2020_LINEAR_EXT specifies support for the BT2020 color space and applies a linear OETF.

  • VK_COLOR_SPACE_HDR10_ST2084_EXT specifies support for the HDR10 (BT2020 color) space and applies the SMPTE ST2084 Perceptual Quantizer (PQ) OETF.

  • VK_COLOR_SPACE_DOLBYVISION_EXT specifies support for the Dolby Vision (BT2020 color space), proprietary encoding, and applies the SMPTE ST2084 OETF.

  • VK_COLOR_SPACE_HDR10_HLG_EXT specifies support for the HDR10 (BT2020 color space) and applies the Hybrid Log Gamma (HLG) OETF.

  • VK_COLOR_SPACE_ADOBERGB_LINEAR_EXT specifies support for the AdobeRGB color space and applies a linear OETF.

  • VK_COLOR_SPACE_ADOBERGB_NONLINEAR_EXT specifies support for the AdobeRGB color space and applies the Gamma 2.2 OETF.

  • VK_COLOR_SPACE_PASS_THROUGH_EXT specifies that color components are used “as is”. This is intended to allow applications to supply data for color spaces not described here.

The color components of Non-linear color space swap chain images have had the appropriate transfer function applied. Vulkan requires that all implementations support the sRGB transfer function when using an SRGB pixel format. Other transfer functions, such as SMPTE 170M or SMPTE2084, must not be performed by the implementation, but can be performed by the application shader. This extension defines enums for VkColorSpaceKHR that correspond to the following color spaces:

Table 43. Color Spaces and Attributes
Name Red Primary Green Primary Blue Primary White-point Transfer function

DCI-P3

0.680, 0.320

0.265, 0.690

0.150, 0.060

0.3127, 0.3290 (D65)

Gamma 2.6

Display-P3

0.680, 0.320

0.265, 0.690

0.150, 0.060

0.3127, 0.3290 (D65)

Display-P3

BT709

0.640, 0.330

0.300, 0.600

0.150, 0.060

0.3127, 0.3290 (D65)

SMPTE 170M

sRGB

0.640, 0.330

0.300, 0.600

0.150, 0.060

0.3127, 0.3290 (D65)

sRGB

extended sRGB

0.640, 0.330

0.300, 0.600

0.150, 0.060

0.3127, 0.3290 (D65)

extended sRGB

HDR10_ST2084

0.708, 0.292

0.170, 0.797

0.131, 0.046

0.3127, 0.3290 (D65)

ST2084

DOLBYVISION

0.708, 0.292

0.170, 0.797

0.131, 0.046

0.3127, 0.3290 (D65)

ST2084

HDR10_HLG

0.708, 0.292

0.170, 0.797

0.131, 0.046

0.3127, 0.3290 (D65)

HLG

AdobeRGB

0.640, 0.330

0.210, 0.710

0.150, 0.060

0.3127, 0.3290 (D65)

AdobeRGB

For Opto-Electrical Transfer Function (OETF), unless otherwise specified, the values of L and E are defined as:

L - linear luminance of image \(0 \leq L \leq 1\) for conventional colorimetry

E - corresponding electrical signal (value stored in memory)

30.5.1. sRGB transfer function

\[\begin{aligned} E & = \begin{cases} 1.055 \times L^{1 \over 2.4} - 0.055 & \text{for}\ 0.0031308 \leq L \leq 1 \\ 12.92 \times L & \text{for}\ 0 \leq L < 0.0031308 \end{cases} \end{aligned}\]

30.5.2. Display-P3 EOTF

\[\begin{aligned} E & = \begin{cases} (a \times L + b)^{2.4} & \text{for}\ 0.039 \leq L \leq 1 \\ b \times L & \text{for}\ 0 \leq L < 0.039 \end{cases} \end{aligned}\]

\(a = 0.948\)
\(b = 0.052\)
\(c = 0.077\)

30.5.3. Display-P3 OETF

\[\begin{aligned} E & = \begin{cases} 1.055 \times L^{1 \over 2.4} - 0.055 & \text{for}\ 0.0030186 \leq L \leq 1 \\ 12.92 \times L & \text{for}\ 0 \leq L < 0.0030186 \end{cases} \end{aligned}\]
Note

For most uses, the sRGB OETF is equivalent.

30.5.4. Extended sRGB OETF

\[\begin{aligned} E & = \begin{cases} 1.055 \times L^{1 \over 2.4} - 0.055 & \text{for}\ 0.0031308 \leq L \leq 7.5913 \\ 12.92 \times L & \text{for}\ 0 \leq L < 0.0031308 \\ -f(-L) & \text{for}\ L < 0 \end{cases} \end{aligned}\]

L - luminance of image is within [-0.6038, 7.5913].

E can be negative and/or > 1. That is how extended sRGB specifies colors outside the standard sRGB gamut. This means extended sRGB needs a floating point pixel format to cover the intended color range.

30.5.5. SMPTE 170M OETF

\[\begin{aligned} E & = \begin{cases} \alpha \times L^{0.45} - (1 - \alpha) & \text{for}\ \beta \leq L \leq 1 \\ 4.5 \times L & \text{for}\ 0 \leq L < \beta \end{cases} \end{aligned}\]

\(\alpha = 1.099 \text{ and } \beta = 0.018 \text{ for 10-bits and less per sample system (the values given in Rec. 709)}\)
\(\alpha = 1.0993 \text{ and } \beta = 0.0181 \text{ for 12-bits per sample system}\)

30.5.6. SMPTE ST2084 OETF (Inverse-EOTF)

\[ E = (\frac{c_1 + c_2 \times L^{m_1}}{1 + c_3 \times L^{m_1}})^{m_2} \]

where:

\(m_1 = 2610 / 4096 \times \frac{1}{4} = 0.1593017578125\)
\(m_2 = 2523 / 4096 \times 128 = 78.84375\)
\(c_1 = 3424 / 4096 = 0.8359375 = c3 - c2 + 1\)
\(c_2 = 2413 / 4096 \times 32 = 18.8515625\)
\(c_3 = 2392 / 4096 \times 32 = 18.6875\)

30.5.7. Hybrid Log Gamma (HLG)

\[\begin{aligned} E & = \begin{cases} r \sqrt{L} & \text{for}\ 0 \leq L \leq 1 \\ a \times \ln(L - b) + c & \text{for}\ 1 < L \end{cases} \end{aligned}\]

L — is the signal normalized by the reference white level
r — is the reference white level and has a signal value of 0.5
a = 0.17883277 and b = 0.28466892 and c = 0.55991073

30.5.8. Adobe RGB (1998) OETF

\(E = L^\frac{1}{2.19921875}\)

30.5.9. Gamma 2.6 OETF

\(E = L^\frac{1}{2.6}\)

An implementation supporting this extension indicates support for these color spaces via VkSurfaceFormatKHR structures returned from vkGetPhysicalDeviceSurfaceFormatsKHR.

Specifying the supported surface color space when calling vkCreateSwapchainKHR will create a swapchain using that color space.

Vulkan requires that all implementations support the sRGB Opto-Electrical Transfer Function (OETF) and Electro-optical transfer function (EOTF) when using an SRGB pixel format. Other transfer functions, such as SMPTE 170M, must not be performed by the implementation, but can be performed by the application shader.

If pSurfaceFormats includes an entry whose value for colorSpace is VK_COLOR_SPACE_SRGB_NONLINEAR_KHR and whose value for format is a UNORM (or SRGB) format and the corresponding SRGB (or UNORM) format is a color renderable format for VK_IMAGE_TILING_OPTIMAL, then pSurfaceFormats must also contain an entry with the same value for colorSpace and format equal to the corresponding SRGB (or UNORM) format.

Note

If pSurfaceFormats includes just one entry, whose value for format is VK_FORMAT_UNDEFINED, surface has no preferred format. In this case, the application can use any valid VkFormat value.

Note

In the initial release of the VK_KHR_surface and VK_KHR_swapchain extensions, the token VK_COLORSPACE_SRGB_NONLINEAR_KHR was used. Starting in the 2016-05-13 updates to the extension branches, matching release 1.0.13 of the core API specification, VK_COLOR_SPACE_SRGB_NONLINEAR_KHR is used instead for consistency with Vulkan naming rules. The older enum is still available for backwards compatibility.

To query the supported presentation modes for a surface, call:

VkResult vkGetPhysicalDeviceSurfacePresentModesKHR(
    VkPhysicalDevice                            physicalDevice,
    VkSurfaceKHR                                surface,
    uint32_t*                                   pPresentModeCount,
    VkPresentModeKHR*                           pPresentModes);
  • physicalDevice is the physical device that will be associated with the swapchain to be created, as described for vkCreateSwapchainKHR.

  • surface is the surface that will be associated with the swapchain.

  • pPresentModeCount is a pointer to an integer related to the number of presentation modes available or queried, as described below.

  • pPresentModes is either NULL or a pointer to an array of VkPresentModeKHR values, indicating the supported presentation modes.

If pPresentModes is NULL, then the number of presentation modes supported for the given surface is returned in pPresentModeCount. Otherwise, pPresentModeCount must point to a variable set by the user to the number of elements in the pPresentModes array, and on return the variable is overwritten with the number of values actually written to pPresentModes. If the value of pPresentModeCount is less than the number of presentation modes supported, at most pPresentModeCount values will be written. If pPresentModeCount is smaller than the number of presentation modes supported for the given surface, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pPresentModeCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPresentModeCount is not 0, and pPresentModes is not NULL, pPresentModes must be a valid pointer to an array of pPresentModeCount VkPresentModeKHR values

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

Possible values of elements of the vkGetPhysicalDeviceSurfacePresentModesKHR::pPresentModes array, indicating the supported presentation modes for a surface, are:

typedef enum VkPresentModeKHR {
    VK_PRESENT_MODE_IMMEDIATE_KHR = 0,
    VK_PRESENT_MODE_MAILBOX_KHR = 1,
    VK_PRESENT_MODE_FIFO_KHR = 2,
    VK_PRESENT_MODE_FIFO_RELAXED_KHR = 3,
    VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR = 1000111000,
    VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR = 1000111001,
} VkPresentModeKHR;
  • VK_PRESENT_MODE_IMMEDIATE_KHR specifies that the presentation engine does not wait for a vertical blanking period to update the current image, meaning this mode may result in visible tearing. No internal queuing of presentation requests is needed, as the requests are applied immediately.

  • VK_PRESENT_MODE_MAILBOX_KHR specifies that the presentation engine waits for the next vertical blanking period to update the current image. Tearing cannot be observed. An internal single-entry queue is used to hold pending presentation requests. If the queue is full when a new presentation request is received, the new request replaces the existing entry, and any images associated with the prior entry become available for re-use by the application. One request is removed from the queue and processed during each vertical blanking period in which the queue is non-empty.

  • VK_PRESENT_MODE_FIFO_KHR specifies that the presentation engine waits for the next vertical blanking period to update the current image. Tearing cannot be observed. An internal queue is used to hold pending presentation requests. New requests are appended to the end of the queue, and one request is removed from the beginning of the queue and processed during each vertical blanking period in which the queue is non-empty. This is the only value of presentMode that is required to be supported.

  • VK_PRESENT_MODE_FIFO_RELAXED_KHR specifies that the presentation engine generally waits for the next vertical blanking period to update the current image. If a vertical blanking period has already passed since the last update of the current image then the presentation engine does not wait for another vertical blanking period for the update, meaning this mode may result in visible tearing in this case. This mode is useful for reducing visual stutter with an application that will mostly present a new image before the next vertical blanking period, but may occasionally be late, and present a new image just after the next vertical blanking period. An internal queue is used to hold pending presentation requests. New requests are appended to the end of the queue, and one request is removed from the beginning of the queue and processed during or after each vertical blanking period in which the queue is non-empty.

  • VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR specifies that the presentation engine and application have concurrent access to a single image, which is referred to as a shared presentable image. The presentation engine is only required to update the current image after a new presentation request is received. Therefore the application must make a presentation request whenever an update is required. However, the presentation engine may update the current image at any point, meaning this mode may result in visible tearing.

  • VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR specifies that the presentation engine and application have concurrent access to a single image, which is referred to as a shared presentable image. The presentation engine periodically updates the current image on its regular refresh cycle. The application is only required to make one initial presentation request, after which the presentation engine must update the current image without any need for further presentation requests. The application can indicate the image contents have been updated by making a presentation request, but this does not guarantee the timing of when it will be updated. This mode may result in visible tearing if rendering to the image is not timed correctly.

The supported VkImageUsageFlagBits of the presentable images of a swapchain created for a surface may differ depending on the presentation mode, and can be determined as per the table below:

Table 44. Presentable image usage queries
Presentation mode Image usage flags

VK_PRESENT_MODE_IMMEDIATE_KHR

VkSurfaceCapabilitiesKHR::supportedUsageFlags

VK_PRESENT_MODE_MAILBOX_KHR

VkSurfaceCapabilitiesKHR::supportedUsageFlags

VK_PRESENT_MODE_FIFO_KHR

VkSurfaceCapabilitiesKHR::supportedUsageFlags

VK_PRESENT_MODE_FIFO_RELAXED_KHR

VkSurfaceCapabilitiesKHR::supportedUsageFlags

VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR

VkSharedPresentSurfaceCapabilitiesKHR::sharedPresentSupportedUsageFlags

VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR

VkSharedPresentSurfaceCapabilitiesKHR::sharedPresentSupportedUsageFlags

Note

For reference, the mode indicated by VK_PRESENT_MODE_FIFO_KHR is equivalent to the behavior of {wgl|glX|egl}SwapBuffers with a swap interval of 1, while the mode indicated by VK_PRESENT_MODE_FIFO_RELAXED_KHR is equivalent to the behavior of {wgl|glX}SwapBuffers with a swap interval of -1 (from the {WGL|GLX}_EXT_swap_control_tear extensions).

30.6. Device Group Queries

A logical device that represents multiple physical devices may support presenting from images on more than one physical device, or combining images from multiple physical devices.

To query these capabilities, call:

VkResult vkGetDeviceGroupPresentCapabilitiesKHR(
    VkDevice                                    device,
    VkDeviceGroupPresentCapabilitiesKHR*        pDeviceGroupPresentCapabilities);
  • device is the logical device.

  • pDeviceGroupPresentCapabilities is a pointer to a structure of type VkDeviceGroupPresentCapabilitiesKHR that is filled with the logical device’s capabilities.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pDeviceGroupPresentCapabilities must be a valid pointer to a VkDeviceGroupPresentCapabilitiesKHR structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDeviceGroupPresentCapabilitiesKHR structure is defined as:

typedef struct VkDeviceGroupPresentCapabilitiesKHR {
    VkStructureType                     sType;
    const void*                         pNext;
    uint32_t                            presentMask[VK_MAX_DEVICE_GROUP_SIZE];
    VkDeviceGroupPresentModeFlagsKHR    modes;
} VkDeviceGroupPresentCapabilitiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • presentMask is an array of masks, where the mask at element i is non-zero if physical device i has a presentation engine, and where bit j is set in element i if physical device i can present swapchain images from physical device j. If element i is non-zero, then bit i must be set.

  • modes is a bitmask of VkDeviceGroupPresentModeFlagBitsKHR indicating which device group presentation modes are supported.

modes always has VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR set.

The present mode flags are also used when presenting an image, in VkDeviceGroupPresentInfoKHR::mode.

If a device group only includes a single physical device, then modes must equal VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_CAPABILITIES_KHR

  • pNext must be NULL

Bits which may be set in VkDeviceGroupPresentCapabilitiesKHR::modes to indicate which device group presentation modes are supported are:

typedef enum VkDeviceGroupPresentModeFlagBitsKHR {
    VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR = 0x00000001,
    VK_DEVICE_GROUP_PRESENT_MODE_REMOTE_BIT_KHR = 0x00000002,
    VK_DEVICE_GROUP_PRESENT_MODE_SUM_BIT_KHR = 0x00000004,
    VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_MULTI_DEVICE_BIT_KHR = 0x00000008,
} VkDeviceGroupPresentModeFlagBitsKHR;
  • VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR specifies that any physical device with a presentation engine can present its own swapchain images.

  • VK_DEVICE_GROUP_PRESENT_MODE_REMOTE_BIT_KHR specifies that any physical device with a presentation engine can present swapchain images from any physical device in its presentMask.

  • VK_DEVICE_GROUP_PRESENT_MODE_SUM_BIT_KHR specifies that any physical device with a presentation engine can present the sum of swapchain images from any physical devices in its presentMask.

  • VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_MULTI_DEVICE_BIT_KHR specifies that multiple physical devices with a presentation engine can each present their own swapchain images.

typedef VkFlags VkDeviceGroupPresentModeFlagsKHR;

VkDeviceGroupPresentModeFlagsKHR is a bitmask type for setting a mask of zero or more VkDeviceGroupPresentModeFlagBitsKHR.

Some surfaces may not be capable of using all the device group present modes.

To query the supported device group present modes for a particular surface, call:

VkResult vkGetDeviceGroupSurfacePresentModesKHR(
    VkDevice                                    device,
    VkSurfaceKHR                                surface,
    VkDeviceGroupPresentModeFlagsKHR*           pModes);
  • device is the logical device.

  • surface is the surface.

  • pModes is a pointer to a value of type VkDeviceGroupPresentModeFlagsKHR that is filled with the supported device group present modes for the surface.

The modes returned by this command are not invariant, and may change in response to the surface being moved, resized, or occluded. These modes must be a subset of the modes returned by vkGetDeviceGroupPresentCapabilitiesKHR.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pModes must be a valid pointer to a VkDeviceGroupPresentModeFlagsKHR value

  • Both of device, and surface must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to surface must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_SURFACE_LOST_KHR

When using VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_MULTI_DEVICE_BIT_KHR, the application may need to know which regions of the surface are used when presenting locally on each physical device. Presentation of swapchain images to this surface need only have valid contents in the regions returned by this command.

To query a set of rectangles used in presentation on the physical device, call:

VkResult vkGetPhysicalDevicePresentRectanglesKHR(
    VkPhysicalDevice                            physicalDevice,
    VkSurfaceKHR                                surface,
    uint32_t*                                   pRectCount,
    VkRect2D*                                   pRects);
  • physicalDevice is the physical device.

  • surface is the surface.

  • pRectCount is a pointer to an integer related to the number of rectangles available or queried, as described below.

  • pRects is either NULL or a pointer to an array of VkRect2D structures.

If pRects is NULL, then the number of rectangles used when presenting the given surface is returned in pRectCount. Otherwise, pRectCount must point to a variable set by the user to the number of elements in the pRects array, and on return the variable is overwritten with the number of structures actually written to pRects. If the value of pRectCount is less than the number of rectangles, at most pRectCount structures will be written. If pRectCount is smaller than the number of rectangles used for the given surface, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

The values returned by this command are not invariant, and may change in response to the surface being moved, resized, or occluded.

The rectangles returned by this command must not overlap.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • surface must be a valid VkSurfaceKHR handle

  • pRectCount must be a valid pointer to a uint32_t value

  • If the value referenced by pRectCount is not 0, and pRects is not NULL, pRects must be a valid pointer to an array of pRectCount VkRect2D structures

  • Both of physicalDevice, and surface must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to surface must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

30.7. Display Timing Queries

Traditional game and real-time-animation applications frequently use VK_PRESENT_MODE_FIFO_KHR so that presentable images are updated during the vertical blanking period of a given refresh cycle (RC) of the presentation engine’s display. This avoids the visual anomaly known as tearing.

However, synchronizing the presentation of images with the RC does not prevent all forms of visual anomalies. Stuttering occurs when the geometry for each presentable image isn’t accurately positioned for when that image will be displayed. The geometry may appear to move too little some RCs, and too much for others. Sometimes the animation appears to freeze, when the same image is used for more than one RC.

In order to minimize stuttering, an application needs to correctly position their geometry for when the presentable image will be displayed to the user. To accomplish this, applications need various timing information about the presentation engine’s display. They need to know when presentable images were actually presented, and when they could have been presented. Applications also need to tell the presentation engine to display an image no sooner than a given time. This can allow the application’s animation to look smooth to the user, with no stuttering. The VK_GOOGLE_display_timing extension allows an application to satisfy these needs.

The presentation engine’s display typically refreshes the pixels that are displayed to the user on a periodic basis. The period may be fixed or variable. In many cases, the presentation engine is associated with fixed refresh rate (FRR) display technology, with a fixed refresh rate (RR, e.g. 60Hz). In some cases, the presentation engine is associated with variable refresh rate (VRR) display technology, where each refresh cycle (RC) can vary in length. This extension treats VRR displays as if they are FRR.

To query the duration of a refresh cycle (RC) for the presentation engine’s display, call:

VkResult vkGetRefreshCycleDurationGOOGLE(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    VkRefreshCycleDurationGOOGLE*               pDisplayTimingProperties);
  • device is the device associated with swapchain.

  • swapchain is the swapchain to obtain the refresh duration for.

  • pDisplayTimingProperties is a pointer to an instance of the VkRefreshCycleDurationGOOGLE structure.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • pDisplayTimingProperties must be a valid pointer to a VkRefreshCycleDurationGOOGLE structure

  • Both of device, and swapchain must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_SURFACE_LOST_KHR

The VkRefreshCycleDurationGOOGLE structure is defined as:

typedef struct VkRefreshCycleDurationGOOGLE {
    uint64_t    refreshDuration;
} VkRefreshCycleDurationGOOGLE;
  • refreshDuration is the number of nanoseconds from the start of one refresh cycle to the next.

Note

The rate at which an application renders and presents new images is known as the image present rate (IPR, aka frame rate). The inverse of IPR, or the duration between each image present, is the image present duration (IPD). In order to provide a smooth, stutter-free animation, an application will want its IPD to be a multiple of refreshDuration. For example, if a display has a 60Hz refresh rate, refreshDuration will be a value in nanoseconds that is approximately equal to 16.67ms. In such a case, an application will want an IPD of 16.67ms (1X multiplier of refreshDuration), or 33.33ms (2X multiplier of refreshDuration), or 50.0ms (3X multiplier of refreshDuration), etc.

In order to determine a target IPD for a display (i.e. a multiple of refreshDuration), an application needs to determine when its images are actually displayed. Let’s say that an application has an initial target IPD of 16.67ms (1X multiplier of refreshDuration). It will therefore position the geometry of a new image 16.67ms later than the previous image. Let’s say that this application is running on slower hardware, so that it actually takes 20ms to render each new image. This will create visual anomalies, because the images won’t be displayed to the user every 16.67ms, nor every 20ms. In this case, it is better for the application to adjust its target IPD to 33.33ms (i.e. a 2X multiplier of refreshDuration), and tell the presentation engine to not present images any sooner than every 33.33ms. This will allow the geometry to be correctly positioned for each presentable image.

Adjustments to an application’s IPD may be needed because different views of an application’s geometry can take different amounts of time to render. For example, looking at the sky may take less time to render than looking at multiple, complex items in a room. In general, it is good to not frequently change IPD, as that can cause visual anomalies. Adjustments to a larger IPD because of late images should happen quickly, but adjustments to a smaller IPD should only happen if the actualPresentTime and earliestPresentTime members of the VkPastPresentationTimingGOOGLE structure are consistently different, and if presentMargin is consistently large, over multiple images.

The implementation will maintain a limited amount of history of timing information about previous presents. Because of the asynchronous nature of the presentation engine, the timing information for a given vkQueuePresentKHR command will become available some time later. These time values can be asynchronously queried, and will be returned if available. All time values are in nanoseconds, relative to a monotonically-increasing clock (e.g. CLOCK_MONOTONIC (see clock_gettime(2)) on Android and Linux).

To asynchronously query the presentation engine, for newly-available timing information about one or more previous presents to a given swapchain, call:

VkResult vkGetPastPresentationTimingGOOGLE(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    uint32_t*                                   pPresentationTimingCount,
    VkPastPresentationTimingGOOGLE*             pPresentationTimings);
  • device is the device associated with swapchain.

  • swapchain is the swapchain to obtain presentation timing information duration for.

  • pPresentationTimingCount is a pointer to an integer related to the number of VkPastPresentationTimingGOOGLE structures to query, as described below.

  • pPresentationTimings is either NULL or a pointer to an an array of VkPastPresentationTimingGOOGLE structures.

If pPresentationTimings is NULL, then the number of newly-available timing records for the given swapchain is returned in pPresentationTimingCount. Otherwise, pPresentationTimingCount must point to a variable set by the user to the number of elements in the pPresentationTimings array, and on return the variable is overwritten with the number of structures actually written to pPresentationTimings. If the value of pPresentationTimingCount is less than the number of newly-available timing records, at most pPresentationTimingCount structures will be written. If pPresentationTimingCount is smaller than the number of newly-available timing records for the given swapchain, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • pPresentationTimingCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPresentationTimingCount is not 0, and pPresentationTimings is not NULL, pPresentationTimings must be a valid pointer to an array of pPresentationTimingCount VkPastPresentationTimingGOOGLE structures

  • Both of device, and swapchain must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

  • VK_ERROR_SURFACE_LOST_KHR

The VkPastPresentationTimingGOOGLE structure is defined as:

typedef struct VkPastPresentationTimingGOOGLE {
    uint32_t    presentID;
    uint64_t    desiredPresentTime;
    uint64_t    actualPresentTime;
    uint64_t    earliestPresentTime;
    uint64_t    presentMargin;
} VkPastPresentationTimingGOOGLE;
  • presentID is an application-provided value that was given to a previous vkQueuePresentKHR command via VkPresentTimeGOOGLE::presentID (see below). It can be used to uniquely identify a previous present with the vkQueuePresentKHR command.

  • desiredPresentTime is an application-provided value that was given to a previous vkQueuePresentKHR command via VkPresentTimeGOOGLE::desiredPresentTime. If non-zero, it was used by the application to indicate that an image not be presented any sooner than desiredPresentTime.

  • actualPresentTime is the time when the image of the swapchain was actually displayed.

  • earliestPresentTime is the time when the image of the swapchain could have been displayed. This may differ from actualPresentTime if the application requested that the image be presented no sooner than VkPresentTimeGOOGLE::desiredPresentTime.

  • presentMargin is an indication of how early the vkQueuePresentKHR command was processed compared to how soon it needed to be processed, and still be presented at earliestPresentTime.

The results for a given swapchain and presentID are only returned once from vkGetPastPresentationTimingGOOGLE.

The application can use the VkPastPresentationTimingGOOGLE values to occasionally adjust its timing. For example, if actualPresentTime is later than expected (e.g. one refreshDuration late), the application may increase its target IPD to a higher multiple of refreshDuration (e.g. decrease its frame rate from 60Hz to 30Hz). If actualPresentTime and earliestPresentTime are consistently different, and if presentMargin is consistently large enough, the application may decrease its target IPD to a smaller multiple of refreshDuration (e.g. increase its frame rate from 30Hz to 60Hz). If actualPresentTime and earliestPresentTime are same, and if presentMargin is consistently high, the application may delay the start of its input-render-present loop in order to decrease the latency between user input and the corresponding present (always leaving some margin in case a new image takes longer to render than the previous image). An application that desires its target IPD to always be the same as refreshDuration, can also adjust features until actualPresentTime is never late and presentMargin is satisfactory.

The full VK_GOOGLE_display_timing extension semantics are described for swapchains created with VK_PRESENT_MODE_FIFO_KHR. For example, non-zero values of VkPresentTimeGOOGLE::desiredPresentTime must be honored, and vkGetPastPresentationTimingGOOGLE should return a VkPastPresentationTimingGOOGLE structure with valid values for all images presented with vkQueuePresentKHR. The semantics for other present modes are as follows:

  • VK_PRESENT_MODE_IMMEDIATE_KHR. The presentation engine may ignore non-zero values of VkPresentTimeGOOGLE::desiredPresentTime in favor of presenting immediately. The value of VkPastPresentationTimingGOOGLE::earliestPresentTime must be the same as VkPastPresentationTimingGOOGLE::actualPresentTime, which should be when the presentation engine displayed the image.

  • VK_PRESENT_MODE_MAILBOX_KHR. The intention of using this present mode with this extension is to handle cases where an image is presented late, and the next image is presented soon enough to replace it at the next vertical blanking period. For images that are displayed to the user, the value of VkPastPresentationTimingGOOGLE::actualPresentTime must be when the image was displayed. For images that are not displayed to the user, vkGetPastPresentationTimingGOOGLE may not return a VkPastPresentationTimingGOOGLE structure, or it may return return a VkPastPresentationTimingGOOGLE structure with the value of zero for both VkPastPresentationTimingGOOGLE::actualPresentTime and VkPastPresentationTimingGOOGLE::earliestPresentTime. It is possible that an application can submit images with VkPresentTimeGOOGLE::desiredPresentTime values such that new images may not be displayed. For example, if VkPresentTimeGOOGLE::desiredPresentTime is far enough in the future that an image is not presented before vkQueuePresentKHR is called to present another image, the first image will not be displayed to the user. If the application continues to do that, the presentation may not display new images.

  • VK_PRESENT_MODE_FIFO_RELAXED_KHR. For images that are presented in time to be displayed at the next vertical blanking period, the semantics are identical as for VK_PRESENT_MODE_FIFO_RELAXED_KHR. For images that are presented late, and are displayed after the start of the vertical blanking period (i.e. with tearing), the values of VkPastPresentationTimingGOOGLE may be treated as if the image was displayed at the start of the vertical blanking period, or may be treated the same as for VK_PRESENT_MODE_IMMEDIATE_KHR.

30.8. WSI Swapchain

A swapchain object (a.k.a. swapchain) provides the ability to present rendering results to a surface. Swapchain objects are represented by VkSwapchainKHR handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSwapchainKHR)

A swapchain is an abstraction for an array of presentable images that are associated with a surface. The presentable images are represented by VkImage objects created by the platform. One image (which can be an array image for multiview/stereoscopic-3D surfaces) is displayed at a time, but multiple images can be queued for presentation. An application renders to the image, and then queues the image for presentation to the surface.

A native window cannot be associated with more than one non-retired swapchain at a time. Further, swapchains cannot be created for native windows that have a non-Vulkan graphics API surface associated with them.

Note

The presentation engine is an abstraction for the platform’s compositor or display engine.

The presentation engine may be synchronous or asynchronous with respect to the application and/or logical device.

Some implementations may use the device’s graphics queue or dedicated presentation hardware to perform presentation.

The presentable images of a swapchain are owned by the presentation engine. An application can acquire use of a presentable image from the presentation engine. Use of a presentable image must occur only after the image is returned by vkAcquireNextImageKHR, and before it is presented by vkQueuePresentKHR. This includes transitioning the image layout and rendering commands.

An application can acquire use of a presentable image with vkAcquireNextImageKHR. After acquiring a presentable image and before modifying it, the application must use a synchronization primitive to ensure that the presentation engine has finished reading from the image. The application can then transition the image’s layout, queue rendering commands to it, etc. Finally, the application presents the image with vkQueuePresentKHR, which releases the acquisition of the image.

The presentation engine controls the order in which presentable images are acquired for use by the application.

Note

This allows the platform to handle situations which require out-of-order return of images after presentation. At the same time, it allows the application to generate command buffers referencing all of the images in the swapchain at initialization time, rather than in its main loop.

How this all works is described below.

If a swapchain is created with presentMode set to either VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, a single presentable image can be acquired, referred to as a shared presentable image. A shared presentable image may be concurrently accessed by the application and the presentation engine, without transitioning the image’s layout after it is initially presented.

  • With VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR, the presentation engine is only required to update to the latest contents of a shared presentable image after a present. The application must call vkQueuePresentKHR to guarantee an update. However, the presentation engine may update from it at any time.

  • With VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, the presentation engine will automatically present the latest contents of a shared presentable image during every refresh cycle. The application is only required to make one initial call to vkQueuePresentKHR, after which the presentation engine will update from it without any need for further present calls. The application can indicate the image contents have been updated by calling vkQueuePresentKHR, but this does not guarantee the timing of when updates will occur.

The presentation engine may access a shared presentable image at any time after it is first presented. To avoid tearing, an application should coordinate access with the presentation engine. This requires presentation engine timing information through platform-specific mechanisms and ensuring that color attachment writes are made available during the portion of the presentation engine’s refresh cycle they are intended for.

Note

The VK_KHR_shared_presentable_image extension does not provide functionality for determining the timing of the presentation engine’s refresh cycles.

In order to query a swapchain’s status when rendering to a shared presentable image, call:

VkResult vkGetSwapchainStatusKHR(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain);
  • device is the device associated with swapchain.

  • swapchain is the swapchain to query.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • Both of device, and swapchain must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

  • VK_SUBOPTIMAL_KHR

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

  • VK_ERROR_SURFACE_LOST_KHR

The possible return values for vkGetSwapchainStatusKHR should be interpreted as follows:

  • VK_SUCCESS specifies the presentation engine is presenting the contents of the shared presentable image, as per the swapchain’s VkPresentModeKHR.

  • VK_SUBOPTIMAL_KHR the swapchain no longer matches the surface properties exactly, but the presentation engine is presenting the contents of the shared presentable image, as per the swapchain’s VkPresentModeKHR.

  • VK_ERROR_OUT_OF_DATE_KHR the surface has changed in such a way that it is no longer compatible with the swapchain.

  • VK_ERROR_SURFACE_LOST_KHR the surface is no longer available.

Note

The swapchain state may be cached by implementations, so applications should regularly call vkGetSwapchainStatusKHR when using a swapchain with VkPresentModeKHR set to VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR.

To create a swapchain, call:

VkResult vkCreateSwapchainKHR(
    VkDevice                                    device,
    const VkSwapchainCreateInfoKHR*             pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSwapchainKHR*                             pSwapchain);
  • device is the device to create the swapchain for.

  • pCreateInfo is a pointer to an instance of the VkSwapchainCreateInfoKHR structure specifying the parameters of the created swapchain.

  • pAllocator is the allocator used for host memory allocated for the swapchain object when there is no more specific allocator available (see Memory Allocation).

  • pSwapchain is a pointer to a VkSwapchainKHR handle in which the created swapchain object will be returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfo must be a valid pointer to a valid VkSwapchainCreateInfoKHR structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSwapchain must be a valid pointer to a VkSwapchainKHR handle

Host Synchronization
  • Host access to pCreateInfo.surface must be externally synchronized

  • Host access to pCreateInfo.oldSwapchain must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_SURFACE_LOST_KHR

  • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

The VkSwapchainCreateInfoKHR structure is defined as:

typedef struct VkSwapchainCreateInfoKHR {
    VkStructureType                  sType;
    const void*                      pNext;
    VkSwapchainCreateFlagsKHR        flags;
    VkSurfaceKHR                     surface;
    uint32_t                         minImageCount;
    VkFormat                         imageFormat;
    VkColorSpaceKHR                  imageColorSpace;
    VkExtent2D                       imageExtent;
    uint32_t                         imageArrayLayers;
    VkImageUsageFlags                imageUsage;
    VkSharingMode                    imageSharingMode;
    uint32_t                         queueFamilyIndexCount;
    const uint32_t*                  pQueueFamilyIndices;
    VkSurfaceTransformFlagBitsKHR    preTransform;
    VkCompositeAlphaFlagBitsKHR      compositeAlpha;
    VkPresentModeKHR                 presentMode;
    VkBool32                         clipped;
    VkSwapchainKHR                   oldSwapchain;
} VkSwapchainCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkSwapchainCreateFlagBitsKHR indicating parameters of the swapchain creation.

  • surface is the surface onto which the swapchain will present images. If the creation succeeds, the swapchain becomes associated with surface.

  • minImageCount is the minimum number of presentable images that the application needs. The implementation will either create the swapchain with at least that many images, or it will fail to create the swapchain.

  • imageFormat is a VkFormat value specifying the format the swapchain image(s) will be created with.

  • imageColorSpace is a VkColorSpaceKHR value specifying the way the swapchain interprets image data.

  • imageExtent is the size (in pixels) of the swapchain image(s). The behavior is platform-dependent if the image extent does not match the surface’s currentExtent as returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR.

Note

On some platforms, it is normal that maxImageExtent may become (0, 0), for example when the window is minimized. In such a case, it is not possible to create a swapchain due to the Valid Usage requirements.

  • imageArrayLayers is the number of views in a multiview/stereo surface. For non-stereoscopic-3D applications, this value is 1.

  • imageUsage is a bitmask of VkImageUsageFlagBits describing the intended usage of the (acquired) swapchain images.

  • imageSharingMode is the sharing mode used for the image(s) of the swapchain.

  • queueFamilyIndexCount is the number of queue families having access to the image(s) of the swapchain when imageSharingMode is VK_SHARING_MODE_CONCURRENT.

  • pQueueFamilyIndices is an array of queue family indices having access to the images(s) of the swapchain when imageSharingMode is VK_SHARING_MODE_CONCURRENT.

  • preTransform is a VkSurfaceTransformFlagBitsKHR value describing the transform, relative to the presentation engine’s natural orientation, applied to the image content prior to presentation. If it does not match the currentTransform value returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR, the presentation engine will transform the image content as part of the presentation operation.

  • compositeAlpha is a VkCompositeAlphaFlagBitsKHR value indicating the alpha compositing mode to use when this surface is composited together with other surfaces on certain window systems.

  • presentMode is the presentation mode the swapchain will use. A swapchain’s present mode determines how incoming present requests will be processed and queued internally.

  • clipped specifies whether the Vulkan implementation is allowed to discard rendering operations that affect regions of the surface that are not visible.

    • If set to VK_TRUE, the presentable images associated with the swapchain may not own all of their pixels. Pixels in the presentable images that correspond to regions of the target surface obscured by another window on the desktop, or subject to some other clipping mechanism will have undefined content when read back. Pixel shaders may not execute for these pixels, and thus any side effects they would have had will not occur. VK_TRUE value does not guarantee any clipping will occur, but allows more optimal presentation methods to be used on some platforms.

    • If set to VK_FALSE, presentable images associated with the swapchain will own all of the pixels they contain.

Note

Applications should set this value to VK_TRUE if they do not expect to read back the content of presentable images before presenting them or after reacquiring them, and if their pixel shaders do not have any side effects that require them to run for all pixels in the presentable image.

  • oldSwapchain is VK_NULL_HANDLE, or the existing non-retired swapchain currently associated with surface. Providing a valid oldSwapchain may aid in the resource reuse, and also allows the application to still present any images that are already acquired from it.

Upon calling vkCreateSwapchainKHR with an oldSwapchain that is not VK_NULL_HANDLE, oldSwapchain is retired — even if creation of the new swapchain fails. The new swapchain is created in the non-retired state whether or not oldSwapchain is VK_NULL_HANDLE.

Upon calling vkCreateSwapchainKHR with an oldSwapchain that is not VK_NULL_HANDLE, any images from oldSwapchain that are not acquired by the application may be freed by the implementation, which may occur even if creation of the new swapchain fails. The application can destroy oldSwapchain to free all memory associated with oldSwapchain.

Note

Multiple retired swapchains can be associated with the same VkSurfaceKHR through multiple uses of oldSwapchain that outnumber calls to vkDestroySwapchainKHR.

After oldSwapchain is retired, the application can pass to vkQueuePresentKHR any images it had already acquired from oldSwapchain. E.g., an application may present an image from the old swapchain before an image from the new swapchain is ready to be presented. As usual, vkQueuePresentKHR may fail if oldSwapchain has entered a state that causes VK_ERROR_OUT_OF_DATE_KHR to be returned.

The application can continue to use a shared presentable image obtained from oldSwapchain until a presentable image is acquired from the new swapchain, as long as it has not entered a state that causes it to return VK_ERROR_OUT_OF_DATE_KHR.

Valid Usage
  • surface must be a surface that is supported by the device as determined using vkGetPhysicalDeviceSurfaceSupportKHR

  • minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface

  • minImageCount must be less than or equal to the value returned in the maxImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface if the returned maxImageCount is not zero

  • minImageCount must be 1 if presentMode is either VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR

  • imageFormat and imageColorSpace must match the format and colorSpace members, respectively, of one of the VkSurfaceFormatKHR structures returned by vkGetPhysicalDeviceSurfaceFormatsKHR for the surface

  • imageExtent must be between minImageExtent and maxImageExtent, inclusive, where minImageExtent and maxImageExtent are members of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface

  • imageExtent members width and height must both be non-zero

  • imageArrayLayers must be greater than 0 and less than or equal to the maxImageArrayLayers member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface

  • If presentMode is VK_PRESENT_MODE_IMMEDIATE_KHR, VK_PRESENT_MODE_MAILBOX_KHR, VK_PRESENT_MODE_FIFO_KHR or VK_PRESENT_MODE_FIFO_RELAXED_KHR, imageUsage must be a subset of the supported usage flags present in the supportedUsageFlags member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for surface

  • If presentMode is VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, imageUsage must be a subset of the supported usage flags present in the sharedPresentSupportedUsageFlags member of the VkSharedPresentSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilities2KHR for surface

  • If imageSharingMode is VK_SHARING_MODE_CONCURRENT, pQueueFamilyIndices must be a valid pointer to an array of queueFamilyIndexCount uint32_t values

  • If imageSharingMode is VK_SHARING_MODE_CONCURRENT, queueFamilyIndexCount must be greater than 1

  • If imageSharingMode is VK_SHARING_MODE_CONCURRENT, each element of pQueueFamilyIndices must be unique and must be less than pQueueFamilyPropertyCount returned by either vkGetPhysicalDeviceQueueFamilyProperties or vkGetPhysicalDeviceQueueFamilyProperties2 for the physicalDevice that was used to create device

  • preTransform must be one of the bits present in the supportedTransforms member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface

  • compositeAlpha must be one of the bits present in the supportedCompositeAlpha member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface

  • presentMode must be one of the VkPresentModeKHR values returned by vkGetPhysicalDeviceSurfacePresentModesKHR for the surface

  • If the logical device was created with VkDeviceGroupDeviceCreateInfo::physicalDeviceCount equal to 1, flags must not contain VK_SWAPCHAIN_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR

  • If oldSwapchain is not VK_NULL_HANDLE, oldSwapchain must be a non-retired swapchain associated with native window referred to by surface

  • imageFormat, imageUsage, imageExtent, and imageArrayLayers must be supported for VK_IMAGE_TYPE_2D VK_IMAGE_TILING_OPTIMAL images as reported by vkGetPhysicalDeviceImageFormatProperties.

Valid Usage (Implicit)

Bits which can be set in VkSwapchainCreateInfoKHR::flags, specifying parameters of swapchain creation, are:

typedef enum VkSwapchainCreateFlagBitsKHR {
    VK_SWAPCHAIN_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR = 0x00000001,
    VK_SWAPCHAIN_CREATE_PROTECTED_BIT_KHR = 0x00000002,
} VkSwapchainCreateFlagBitsKHR;
  • VK_SWAPCHAIN_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR specifies that images created from the swapchain (i.e. with the swapchain member of VkImageSwapchainCreateInfoKHR set to this swapchain’s handle) must use VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT.

  • VK_SWAPCHAIN_CREATE_PROTECTED_BIT_KHR specifies that images created from the swapchain are protected images.

typedef VkFlags VkSwapchainCreateFlagsKHR;

VkSwapchainCreateFlagsKHR is a bitmask type for setting a mask of zero or more VkSwapchainCreateFlagBitsKHR.

If the pNext chain of VkSwapchainCreateInfoKHR includes a VkDeviceGroupSwapchainCreateInfoKHR structure, then that structure includes a set of device group present modes that the swapchain can be used with.

The VkDeviceGroupSwapchainCreateInfoKHR structure is defined as:

typedef struct VkDeviceGroupSwapchainCreateInfoKHR {
    VkStructureType                     sType;
    const void*                         pNext;
    VkDeviceGroupPresentModeFlagsKHR    modes;
} VkDeviceGroupSwapchainCreateInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • modes is a bitfield of modes that the swapchain can be used with.

If this structure is not present, modes is considered to be VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR.

Valid Usage (Implicit)

To enable surface counters when creating a swapchain, add VkSwapchainCounterCreateInfoEXT to the pNext chain of VkSwapchainCreateInfoKHR. VkSwapchainCounterCreateInfoEXT is defined as:

typedef struct VkSwapchainCounterCreateInfoEXT {
    VkStructureType             sType;
    const void*                 pNext;
    VkSurfaceCounterFlagsEXT    surfaceCounters;
} VkSwapchainCounterCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • surfaceCounters is a bitmask of VkSurfaceCounterFlagBitsEXT specifying surface counters to enable for the swapchain.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SWAPCHAIN_COUNTER_CREATE_INFO_EXT

  • surfaceCounters must be a valid combination of VkSurfaceCounterFlagBitsEXT values

The requested counters become active when the first presentation command for the associated swapchain is processed by the presentation engine. To query the value of an active counter, use:

VkResult vkGetSwapchainCounterEXT(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    VkSurfaceCounterFlagBitsEXT                 counter,
    uint64_t*                                   pCounterValue);
  • device is the VkDevice associated with swapchain.

  • swapchain is the swapchain from which to query the counter value.

  • counter is the counter to query.

  • pCounterValue will return the current value of the counter.

If a counter is not available because the swapchain is out of date, the implementation may return VK_ERROR_OUT_OF_DATE_KHR.

Valid Usage
  • One or more present commands on swapchain must have been processed by the presentation engine.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • counter must be a valid VkSurfaceCounterFlagBitsEXT value

  • pCounterValue must be a valid pointer to a uint64_t value

  • Both of device, and swapchain must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

As mentioned above, if vkCreateSwapchainKHR succeeds, it will return a handle to a swapchain that contains an array of at least minImageCount presentable images.

While acquired by the application, presentable images can be used in any way that equivalent non-presentable images can be used. A presentable image is equivalent to a non-presentable image created with the following VkImageCreateInfo parameters:

VkImageCreateInfo Field Value

flags

VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT is set if VK_SWAPCHAIN_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR is set

VK_IMAGE_CREATE_PROTECTED_BIT is set if VK_SWAPCHAIN_CREATE_PROTECTED_BIT_KHR is set

all other bits are unset

imageType

VK_IMAGE_TYPE_2D

format

pCreateInfo→imageFormat

extent

{pCreateInfo→imageExtent.width, pCreateInfo→imageExtent.height, 1}

mipLevels

1

arrayLayers

pCreateInfo→imageArrayLayers

samples

VK_SAMPLE_COUNT_1_BIT

tiling

VK_IMAGE_TILING_OPTIMAL

usage

pCreateInfo→imageUsage

sharingMode

pCreateInfo→imageSharingMode

queueFamilyIndexCount

pCreateInfo→queueFamilyIndexCount

pQueueFamilyIndices

pCreateInfo→pQueueFamilyIndices

initialLayout

VK_IMAGE_LAYOUT_UNDEFINED

The surface must not be destroyed until after the swapchain is destroyed.

If oldSwapchain is VK_NULL_HANDLE, and the native window referred to by surface is already associated with a Vulkan swapchain, VK_ERROR_NATIVE_WINDOW_IN_USE_KHR must be returned.

If the native window referred to by surface is already associated with a non-Vulkan graphics API surface, VK_ERROR_NATIVE_WINDOW_IN_USE_KHR must be returned.

The native window referred to by surface must not become associated with a non-Vulkan graphics API surface before all associated Vulkan swapchains have been destroyed.

Like core functions, several WSI functions, including vkCreateSwapchainKHR return VK_ERROR_DEVICE_LOST if the logical device was lost. See Lost Device. As with most core objects, VkSwapchainKHR is a child of the device and is affected by the lost state; it must be destroyed before destroying the VkDevice. However, VkSurfaceKHR is not a child of any VkDevice and is not otherwise affected by the lost device. After successfully recreating a VkDevice, the same VkSurfaceKHR can be used to create a new VkSwapchainKHR, provided the previous one was destroyed.

Note

As mentioned in Lost Device, after a lost device event, the VkPhysicalDevice may also be lost. If other VkPhysicalDevice are available, they can be used together with the same VkSurfaceKHR to create the new VkSwapchainKHR, however the application must query the surface capabilities again, because they may differ on a per-physical device basis.

To destroy a swapchain object call:

void vkDestroySwapchainKHR(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    const VkAllocationCallbacks*                pAllocator);
  • device is the VkDevice associated with swapchain.

  • swapchain is the swapchain to destroy.

  • pAllocator is the allocator used for host memory allocated for the swapchain object when there is no more specific allocator available (see Memory Allocation).

The application must not destroy a swapchain until after completion of all outstanding operations on images that were acquired from the swapchain. swapchain and all associated VkImage handles are destroyed, and must not be acquired or used any more by the application. The memory of each VkImage will only be freed after that image is no longer used by the presentation engine. For example, if one image of the swapchain is being displayed in a window, the memory for that image may not be freed until the window is destroyed, or another swapchain is created for the window. Destroying the swapchain does not invalidate the parent VkSurfaceKHR, and a new swapchain can be created with it.

When a swapchain associated with a display surface is destroyed, if the image most recently presented to the display surface is from the swapchain being destroyed, then either any display resources modified by presenting images from any swapchain associated with the display surface must be reverted by the implementation to their state prior to the first present performed on one of these swapchains, or such resources must be left in their current state.

Valid Usage
  • All uses of presentable images acquired from swapchain must have completed execution

  • If VkAllocationCallbacks were provided when swapchain was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when swapchain was created, pAllocator must be NULL

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • If swapchain is not VK_NULL_HANDLE, swapchain must be a valid VkSwapchainKHR handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • Both of device, and swapchain that are valid handles must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

When the VK_KHR_display_swapchain extension is enabled, multiple swapchains that share presentable images are created by calling:

VkResult vkCreateSharedSwapchainsKHR(
    VkDevice                                    device,
    uint32_t                                    swapchainCount,
    const VkSwapchainCreateInfoKHR*             pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkSwapchainKHR*                             pSwapchains);
  • device is the device to create the swapchains for.

  • swapchainCount is the number of swapchains to create.

  • pCreateInfos is a pointer to an array of VkSwapchainCreateInfoKHR structures specifying the parameters of the created swapchains.

  • pAllocator is the allocator used for host memory allocated for the swapchain objects when there is no more specific allocator available (see Memory Allocation).

  • pSwapchains is a pointer to an array of VkSwapchainKHR handles in which the created swapchain objects will be returned.

vkCreateSharedSwapchains is similar to vkCreateSwapchainKHR, except that it takes an array of VkSwapchainCreateInfoKHR structures, and returns an array of swapchain objects.

The swapchain creation parameters that affect the properties and number of presentable images must match between all the swapchains. If the displays used by any of the swapchains do not use the same presentable image layout or are incompatible in a way that prevents sharing images, swapchain creation will fail with the result code VK_ERROR_INCOMPATIBLE_DISPLAY_KHR. If any error occurs, no swapchains will be created. Images presented to multiple swapchains must be re-acquired from all of them before transitioning away from VK_IMAGE_LAYOUT_PRESENT_SRC_KHR. After destroying one or more of the swapchains, the remaining swapchains and the presentable images can continue to be used.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pCreateInfos must be a valid pointer to an array of swapchainCount valid VkSwapchainCreateInfoKHR structures

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pSwapchains must be a valid pointer to an array of swapchainCount VkSwapchainKHR handles

  • swapchainCount must be greater than 0

Host Synchronization
  • Host access to pCreateInfos[].surface must be externally synchronized

  • Host access to pCreateInfos[].oldSwapchain must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_INCOMPATIBLE_DISPLAY_KHR

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_SURFACE_LOST_KHR

To obtain the array of presentable images associated with a swapchain, call:

VkResult vkGetSwapchainImagesKHR(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    uint32_t*                                   pSwapchainImageCount,
    VkImage*                                    pSwapchainImages);
  • device is the device associated with swapchain.

  • swapchain is the swapchain to query.

  • pSwapchainImageCount is a pointer to an integer related to the number of presentable images available or queried, as described below.

  • pSwapchainImages is either NULL or a pointer to an array of VkImage handles.

If pSwapchainImages is NULL, then the number of presentable images for swapchain is returned in pSwapchainImageCount. Otherwise, pSwapchainImageCount must point to a variable set by the user to the number of elements in the pSwapchainImages array, and on return the variable is overwritten with the number of structures actually written to pSwapchainImages. If the value of pSwapchainImageCount is less than the number of presentable images for swapchain, at most pSwapchainImageCount structures will be written. If pSwapchainImageCount is smaller than the number of presentable images for swapchain, VK_INCOMPLETE will be returned instead of VK_SUCCESS to indicate that not all the available values were returned.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • pSwapchainImageCount must be a valid pointer to a uint32_t value

  • If the value referenced by pSwapchainImageCount is not 0, and pSwapchainImages is not NULL, pSwapchainImages must be a valid pointer to an array of pSwapchainImageCount VkImage handles

  • Both of device, and swapchain must have been created, allocated, or retrieved from the same VkInstance

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

Note

By knowing all presentable images used in the swapchain, the application can create command buffers that reference these images prior to entering its main rendering loop.

The implementation will have already allocated and bound the memory backing the VkImages returned by vkGetSwapchainImagesKHR. The memory for each image will not alias with the memory for other images or with any VkDeviceMemory object. As such, performing any operation affecting the binding of memory to a presentable image results in undefined behavior. All presentable images are initially in the VK_IMAGE_LAYOUT_UNDEFINED layout, thus before using presentable images, the application must transition them to a valid layout for the intended use.

Further, the lifetime of presentable images is controlled by the implementation so destroying a presentable image with vkDestroyImage results in undefined behavior. See vkDestroySwapchainKHR for further details on the lifetime of presentable images.

Images can also be created by using vkCreateImage with VkImageSwapchainCreateInfoKHR and bound to swapchain memory using vkBindImageMemory2KHR with VkBindImageMemorySwapchainInfoKHR. These images can be used anywhere swapchain images are used, and are useful in logical devices with multiple physical devices to create peer memory bindings of swapchain memory. These images and bindings have no effect on what memory is presented. Unlike images retrieved from vkGetSwapchainImagesKHR, these images must be destroyed with vkDestroyImage.

To acquire an available presentable image to use, and retrieve the index of that image, call:

VkResult vkAcquireNextImageKHR(
    VkDevice                                    device,
    VkSwapchainKHR                              swapchain,
    uint64_t                                    timeout,
    VkSemaphore                                 semaphore,
    VkFence                                     fence,
    uint32_t*                                   pImageIndex);
  • device is the device associated with swapchain.

  • swapchain is the non-retired swapchain from which an image is being acquired.

  • timeout specifies how long the function waits, in nanoseconds, if no image is available.

  • semaphore is VK_NULL_HANDLE or a semaphore to signal.

  • fence is VK_NULL_HANDLE or a fence to signal.

  • pImageIndex is a pointer to a uint32_t that is set to the index of the next image to use (i.e. an index into the array of images returned by vkGetSwapchainImagesKHR).

Valid Usage
  • swapchain must not be in the retired state

  • If semaphore is not VK_NULL_HANDLE it must be unsignaled

  • If semaphore is not VK_NULL_HANDLE it must not have any uncompleted signal or wait operations pending

  • If fence is not VK_NULL_HANDLE it must be unsignaled and must not be associated with any other queue command that has not yet completed execution on that queue

  • semaphore and fence must not both be equal to VK_NULL_HANDLE

  • If the number of currently acquired images is greater than the difference between the number of images in swapchain and the value of VkSurfaceCapabilitiesKHR::minImageCount as returned by a call to vkGetPhysicalDeviceSurfaceCapabilities2KHR with the surface used to create swapchain, timeout must not be UINT64_MAX

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • swapchain must be a valid VkSwapchainKHR handle

  • If semaphore is not VK_NULL_HANDLE, semaphore must be a valid VkSemaphore handle

  • If fence is not VK_NULL_HANDLE, fence must be a valid VkFence handle

  • pImageIndex must be a valid pointer to a uint32_t value

  • If semaphore is a valid handle, it must have been created, allocated, or retrieved from device

  • If fence is a valid handle, it must have been created, allocated, or retrieved from device

  • Both of device, and swapchain that are valid handles must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

  • Host access to semaphore must be externally synchronized

  • Host access to fence must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

  • VK_TIMEOUT

  • VK_NOT_READY

  • VK_SUBOPTIMAL_KHR

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

  • VK_ERROR_SURFACE_LOST_KHR

When successful, vkAcquireNextImageKHR acquires a presentable image from swapchain that an application can use, and sets pImageIndex to the index of that image within the swapchain. The presentation engine may not have finished reading from the image at the time it is acquired, so the application must use semaphore and/or fence to ensure that the image layout and contents are not modified until the presentation engine reads have completed. The order in which images are acquired is implementation-dependent, and may be different than the order the images were presented.

If timeout is zero, then vkAcquireNextImageKHR does not wait, and will either successfully acquire an image, or fail and return VK_NOT_READY if no image is available.

If the specified timeout period expires before an image is acquired, vkAcquireNextImageKHR returns VK_TIMEOUT. If timeout is UINT64_MAX, the timeout period is treated as infinite, and vkAcquireNextImageKHR will block until an image is acquired or an error occurs.

An image will eventually be acquired if the number of images that the application has currently acquired (but not yet presented) is less than or equal to the difference between the number of images in swapchain and the value of VkSurfaceCapabilitiesKHR::minImageCount. If the number of currently acquired images is greater than this, vkAcquireNextImage should not be called; if it is, timeout must not be UINT64_MAX.

If an image is acquired successfully, vkAcquireNextImage must either return VK_SUCCESS, or VK_SUBOPTIMAL_KHR if the swapchain no longer matches the surface properties exactly, but can still be used for presentation.

Note

This may happen, for example, if the platform surface has been resized but the platform is able to scale the presented images to the new size to produce valid surface updates. It is up to the application to decide whether it prefers to continue using the current swapchain in this state, or to re-create the swapchain to better match the platform surface properties.

If the swapchain images no longer match native surface properties, either VK_SUBOPTIMAL_KHR or VK_ERROR_OUT_OF_DATE_KHR must be returned. If VK_ERROR_OUT_OF_DATE_KHR is returned, no image is acquired and attempts to present previously acquired images to the swapchain will also fail with VK_ERROR_OUT_OF_DATE_KHR. Applications need to create a new swapchain for the surface to continue presenting if VK_ERROR_OUT_OF_DATE_KHR is returned.

If device loss occurs (see Lost Device) before the timeout has expired, vkAcquireNextImageKHR must return in finite time with either one of the allowed success codes, or VK_ERROR_DEVICE_LOST.

If semaphore is not VK_NULL_HANDLE, the semaphore must be unsignaled, with no signal or wait operations pending. It will become signaled when the application can use the image.

Note

Use of semaphore allows rendering operations to be recorded and submitted before the presentation engine has completed its use of the image.

If fence is not equal to VK_NULL_HANDLE, the fence must be unsignaled, with no signal operations pending. It will become signaled when the application can use the image.

Note

Applications should not rely on vkAcquireNextImageKHR blocking in order to meter their rendering speed. The implementation may return from this function immediately regardless of how many presentation requests are queued, and regardless of when queued presentation requests will complete relative to the call. Instead, applications can use fence to meter their frame generation work to match the presentation rate.

An application must wait until either the semaphore or fence is signaled before accessing the image’s data.

Note

When the presentable image will be accessed by some stage S, the recommended idiom for ensuring correct synchronization is:

  • The VkSubmitInfo used to submit the image layout transition for execution includes vkAcquireNextImageKHR::semaphore in its pWaitSemaphores member, with the corresponding element of pWaitDstStageMask including S.

  • The synchronization command that performs any necessary image layout transition includes S in both the srcStageMask and dstStageMask.

After a successful return, the image indicated by pImageIndex and its data will be unmodified compared to when it was presented.

Note

Exclusive ownership of presentable images corresponding to a swapchain created with VK_SHARING_MODE_EXCLUSIVE as defined in Resource Sharing is not altered by a call to vkAcquireNextImageKHR. That means upon the first acquisition from such a swapchain presentable images are not owned by any queue family, while at subsequent acquisitions the presentable images remain owned by the queue family the image was previously presented on.

The possible return values for vkAcquireNextImageKHR depend on the timeout provided:

  • VK_SUCCESS is returned if an image became available.

  • VK_ERROR_SURFACE_LOST_KHR if the surface becomes no longer available.

  • VK_NOT_READY is returned if timeout is zero and no image was available.

  • VK_TIMEOUT is returned if timeout is greater than zero and less than UINT64_MAX, and no image became available within the time allowed.

  • VK_SUBOPTIMAL_KHR is returned if an image became available, and the swapchain no longer matches the surface properties exactly, but can still be used to present to the surface successfully.

Note

This may happen, for example, if the platform surface has been resized but the platform is able to scale the presented images to the new size to produce valid surface updates. It is up to the application to decide whether it prefers to continue using the current swapchain indefinitely or temporarily in this state, or to re-create the swapchain to better match the platform surface properties.

  • VK_ERROR_OUT_OF_DATE_KHR is returned if the surface has changed in such a way that it is no longer compatible with the swapchain, and further presentation requests using the swapchain will fail. Applications must query the new surface properties and recreate their swapchain if they wish to continue presenting to the surface.

If the native surface and presented image sizes no longer match, presentation may fail. If presentation does succeed, parts of the native surface may be undefined, parts of the presented image may have been clipped before presentation, and/or the image may have been scaled (uniformly or not uniformly) before presentation. It is the application’s responsibility to detect surface size changes and react appropriately. If presentation fails because of a mismatch in the surface and presented image sizes, a VK_ERROR_OUT_OF_DATE_KHR error will be returned.

To acquire an available presentable image to use, and retrieve the index of that image, call:

VkResult vkAcquireNextImage2KHR(
    VkDevice                                    device,
    const VkAcquireNextImageInfoKHR*            pAcquireInfo,
    uint32_t*                                   pImageIndex);
  • device is the device associated with swapchain.

  • pAcquireInfo is a pointer to a structure of type VkAcquireNextImageInfoKHR containing parameters of the acquire.

  • pImageIndex is a pointer to a uint32_t that is set to the index of the next image to use.

Valid Usage
  • If the number of currently acquired images is greater than the difference between the number of images in the swapchain member of pAcquireInfo and the value of VkSurfaceCapabilitiesKHR::minImageCount as returned by a call to vkGetPhysicalDeviceSurfaceCapabilities2KHR with the surface used to create swapchain, the timeout member of pAcquireInfo must not be UINT64_MAX

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pAcquireInfo must be a valid pointer to a valid VkAcquireNextImageInfoKHR structure

  • pImageIndex must be a valid pointer to a uint32_t value

Return Codes
Success
  • VK_SUCCESS

  • VK_TIMEOUT

  • VK_NOT_READY

  • VK_SUBOPTIMAL_KHR

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

  • VK_ERROR_SURFACE_LOST_KHR

The VkAcquireNextImageInfoKHR structure is defined as:

typedef struct VkAcquireNextImageInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkSwapchainKHR     swapchain;
    uint64_t           timeout;
    VkSemaphore        semaphore;
    VkFence            fence;
    uint32_t           deviceMask;
} VkAcquireNextImageInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchain is a non-retired swapchain from which an image is acquired.

  • timeout specifies how long the function waits, in nanoseconds, if no image is available.

  • semaphore is VK_NULL_HANDLE or a semaphore to signal.

  • fence is VK_NULL_HANDLE or a fence to signal.

  • deviceMask is a mask of physical devices for which the swapchain image will be ready to use when the semaphore or fence is signaled.

If vkAcquireNextImageKHR is used, the device mask is considered to include all physical devices in the logical device.

Note

vkAcquireNextImage2KHR signals at most one semaphore, even if the application requests waiting for multiple physical devices to be ready via the deviceMask. However, only a single physical device can wait on that semaphore, since the semaphore becomes unsignaled when the wait succeeds. For other physical devices to wait for the image to be ready, it is necessary for the application to submit semaphore signal operation(s) to that first physical device to signal additional semaphore(s) after the wait succeeds, which the other physical device(s) can wait upon.

Valid Usage
  • swapchain must not be in the retired state

  • If semaphore is not VK_NULL_HANDLE it must be unsignaled

  • If semaphore is not VK_NULL_HANDLE it must not have any uncompleted signal or wait operations pending

  • If fence is not VK_NULL_HANDLE it must be unsignaled and must not be associated with any other queue command that has not yet completed execution on that queue

  • semaphore and fence must not both be equal to VK_NULL_HANDLE

  • deviceMask must be a valid device mask

  • deviceMask must not be zero

  • semaphore and fence must not both be equal to VK_NULL_HANDLE.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_ACQUIRE_NEXT_IMAGE_INFO_KHR

  • pNext must be NULL

  • swapchain must be a valid VkSwapchainKHR handle

  • If semaphore is not VK_NULL_HANDLE, semaphore must be a valid VkSemaphore handle

  • If fence is not VK_NULL_HANDLE, fence must be a valid VkFence handle

  • Each of fence, semaphore, and swapchain that are valid handles must have been created, allocated, or retrieved from the same VkInstance

Host Synchronization
  • Host access to swapchain must be externally synchronized

  • Host access to semaphore must be externally synchronized

  • Host access to fence must be externally synchronized

After queueing all rendering commands and transitioning the image to the correct layout, to queue an image for presentation, call:

VkResult vkQueuePresentKHR(
    VkQueue                                     queue,
    const VkPresentInfoKHR*                     pPresentInfo);
  • queue is a queue that is capable of presentation to the target surface’s platform on the same device as the image’s swapchain.

  • pPresentInfo is a pointer to an instance of the VkPresentInfoKHR structure specifying the parameters of the presentation.

Note

There is no requirement for an application to present images in the same order that they were acquired - applications can arbitrarily present any image that is currently acquired.

Valid Usage
  • Each element of pSwapchains member of pPresentInfo must be a swapchain that is created for a surface for which presentation is supported from queue as determined using a call to vkGetPhysicalDeviceSurfaceSupportKHR

  • If more than one member of pSwapchains was created from a display surface, all display surfaces referenced that refer to the same display must use the same display mode

  • When a semaphore unsignal operation defined by the elements of the pWaitSemaphores member of pPresentInfo executes on queue, no other queue must be waiting on the same semaphore.

  • All elements of the pWaitSemaphores member of pPresentInfo must be semaphores that are signaled, or have semaphore signal operations previously submitted for execution.

Any writes to memory backing the images referenced by the pImageIndices and pSwapchains members of pPresentInfo, that are available before vkQueuePresentKHR is executed, are automatically made visible to the read access performed by the presentation engine. This automatic visibility operation for an image happens-after the semaphore signal operation, and happens-before the presentation engine accesses the image.

Queueing an image for presentation defines a set of queue operations, including waiting on the semaphores and submitting a presentation request to the presentation engine. However, the scope of this set of queue operations does not include the actual processing of the image by the presentation engine.

If vkQueuePresentKHR fails to enqueue the corresponding set of queue operations, it may return VK_ERROR_OUT_OF_HOST_MEMORY or VK_ERROR_OUT_OF_DEVICE_MEMORY. If it does, the implementation must ensure that the state and contents of any resources or synchronization primitives referenced is unaffected by the call or its failure.

If vkQueuePresentKHR fails in such a way that the implementation is unable to make that guarantee, the implementation must return VK_ERROR_DEVICE_LOST.

However, if the presentation request is rejected by the presentation engine with an error VK_ERROR_OUT_OF_DATE_KHR or VK_ERROR_SURFACE_LOST_KHR, the set of queue operations are still considered to be enqueued and thus any semaphore to be waited on gets unsignaled when the corresponding queue operation is complete.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

  • pPresentInfo must be a valid pointer to a valid VkPresentInfoKHR structure

Host Synchronization
  • Host access to queue must be externally synchronized

  • Host access to pPresentInfo.pWaitSemaphores[] must be externally synchronized

  • Host access to pPresentInfo.pSwapchains[] must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

  • VK_SUBOPTIMAL_KHR

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_DEVICE_LOST

  • VK_ERROR_OUT_OF_DATE_KHR

  • VK_ERROR_SURFACE_LOST_KHR

The VkPresentInfoKHR structure is defined as:

typedef struct VkPresentInfoKHR {
    VkStructureType          sType;
    const void*              pNext;
    uint32_t                 waitSemaphoreCount;
    const VkSemaphore*       pWaitSemaphores;
    uint32_t                 swapchainCount;
    const VkSwapchainKHR*    pSwapchains;
    const uint32_t*          pImageIndices;
    VkResult*                pResults;
} VkPresentInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • waitSemaphoreCount is the number of semaphores to wait for before issuing the present request. The number may be zero.

  • pWaitSemaphores, if not NULL, is an array of VkSemaphore objects with waitSemaphoreCount entries, and specifies the semaphores to wait for before issuing the present request.

  • swapchainCount is the number of swapchains being presented to by this command.

  • pSwapchains is an array of VkSwapchainKHR objects with swapchainCount entries. A given swapchain must not appear in this list more than once.

  • pImageIndices is an array of indices into the array of each swapchain’s presentable images, with swapchainCount entries. Each entry in this array identifies the image to present on the corresponding entry in the pSwapchains array.

  • pResults is an array of VkResult typed elements with swapchainCount entries. Applications that do not need per-swapchain results can use NULL for pResults. If non-NULL, each entry in pResults will be set to the VkResult for presenting the swapchain corresponding to the same index in pSwapchains.

Before an application can present an image, the image’s layout must be transitioned to the VK_IMAGE_LAYOUT_PRESENT_SRC_KHR layout, or for a shared presentable image the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR layout.

Note

When transitioning the image to VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR or VK_IMAGE_LAYOUT_PRESENT_SRC_KHR, there is no need to delay subsequent processing, or perform any visibility operations (as vkQueuePresentKHR performs automatic visibility operations). To achieve this, the dstAccessMask member of the VkImageMemoryBarrier should be set to 0, and the dstStageMask parameter should be set to VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT.

Valid Usage
  • Each element of pImageIndices must be the index of a presentable image acquired from the swapchain specified by the corresponding element of the pSwapchains array, and the presented image subresource must be in the VK_IMAGE_LAYOUT_PRESENT_SRC_KHR or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR layout at the time the operation is executed on a VkDevice

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PRESENT_INFO_KHR

  • Each pNext member of any structure (including this one) in the pNext chain must be either NULL or a pointer to a valid instance of VkDeviceGroupPresentInfoKHR, VkDisplayPresentInfoKHR, VkPresentRegionsKHR, or VkPresentTimesInfoGOOGLE

  • Each sType member in the pNext chain must be unique

  • If waitSemaphoreCount is not 0, pWaitSemaphores must be a valid pointer to an array of waitSemaphoreCount valid VkSemaphore handles

  • pSwapchains must be a valid pointer to an array of swapchainCount valid VkSwapchainKHR handles

  • pImageIndices must be a valid pointer to an array of swapchainCount uint32_t values

  • If pResults is not NULL, pResults must be a valid pointer to an array of swapchainCount VkResult values

  • swapchainCount must be greater than 0

  • Both of the elements of pSwapchains, and the elements of pWaitSemaphores that are valid handles must have been created, allocated, or retrieved from the same VkInstance

When the VK_KHR_incremental_present extension is enabled, additional fields can be specified that allow an application to specify that only certain rectangular regions of the presentable images of a swapchain are changed. This is an optimization hint that a presentation engine may use to only update the region of a surface that is actually changing. The application still must ensure that all pixels of a presented image contain the desired values, in case the presentation engine ignores this hint. An application can provide this hint by including the VkPresentRegionsKHR structure in the pNext chain of the VkPresentInfoKHR structure.

The VkPresentRegionsKHR structure is defined as:

typedef struct VkPresentRegionsKHR {
    VkStructureType              sType;
    const void*                  pNext;
    uint32_t                     swapchainCount;
    const VkPresentRegionKHR*    pRegions;
} VkPresentRegionsKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchainCount is the number of swapchains being presented to by this command.

  • pRegions is NULL or a pointer to an array of VkPresentRegionKHR elements with swapchainCount entries. If not NULL, each element of pRegions contains the region that has changed since the last present to the swapchain in the corresponding entry in the VkPresentInfoKHR::pSwapchains array.

Valid Usage
  • swapchainCount must be the same value as VkPresentInfoKHR::swapchainCount, where VkPresentInfoKHR is in the pNext-chain of this VkPresentRegionsKHR structure.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PRESENT_REGIONS_KHR

  • If pRegions is not NULL, pRegions must be a valid pointer to an array of swapchainCount valid VkPresentRegionKHR structures

  • swapchainCount must be greater than 0

For a given image and swapchain, the region to present is specified by the VkPresentRegionKHR structure, which is defined as:

typedef struct VkPresentRegionKHR {
    uint32_t                 rectangleCount;
    const VkRectLayerKHR*    pRectangles;
} VkPresentRegionKHR;
  • rectangleCount is the number of rectangles in pRectangles, or zero if the entire image has changed and should be presented.

  • pRectangles is either NULL or a pointer to an array of VkRectLayerKHR structures. The VkRectLayerKHR structure is the framebuffer coordinates, plus layer, of a portion of a presentable image that has changed and must be presented. If non-NULL, each entry in pRectangles is a rectangle of the given image that has changed since the last image was presented to the given swapchain.

Valid Usage (Implicit)
  • If rectangleCount is not 0, and pRectangles is not NULL, pRectangles must be a valid pointer to an array of rectangleCount VkRectLayerKHR structures

The VkRectLayerKHR structure is defined as:

typedef struct VkRectLayerKHR {
    VkOffset2D    offset;
    VkExtent2D    extent;
    uint32_t      layer;
} VkRectLayerKHR;
  • offset is the origin of the rectangle, in pixels.

  • extent is the size of the rectangle, in pixels.

  • layer is the layer of the image. For images with only one layer, the value of layer must be 0.

Valid Usage
  • The sum of offset and extent must be no greater than the imageExtent member of the VkSwapchainCreateInfoKHR structure given to vkCreateSwapchainKHR.

  • layer must be less than imageArrayLayers member of the VkSwapchainCreateInfoKHR structure given to vkCreateSwapchainKHR.

Some platforms allow the size of a surface to change, and then scale the pixels of the image to fit the surface. VkRectLayerKHR specifies pixels of the swapchain’s image(s), which will be constant for the life of the swapchain.

When the VK_KHR_display_swapchain extension is enabled additional fields can be specified when presenting an image to a swapchain by setting VkPresentInfoKHR::pNext to point to an instance of the VkDisplayPresentInfoKHR structure.

The VkDisplayPresentInfoKHR structure is defined as:

typedef struct VkDisplayPresentInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkRect2D           srcRect;
    VkRect2D           dstRect;
    VkBool32           persistent;
} VkDisplayPresentInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • srcRect is a rectangular region of pixels to present. It must be a subset of the image being presented. If VkDisplayPresentInfoKHR is not specified, this region will be assumed to be the entire presentable image.

  • dstRect is a rectangular region within the visible region of the swapchain’s display mode. If VkDisplayPresentInfoKHR is not specified, this region will be assumed to be the entire visible region of the visible region of the swapchain’s mode. If the specified rectangle is a subset of the display mode’s visible region, content from display planes below the swapchain’s plane will be visible outside the rectangle. If there are no planes below the swapchain’s, the area outside the specified rectangle will be black. If portions of the specified rectangle are outside of the display’s visible region, pixels mapping only to those portions of the rectangle will be discarded.

  • persistent: If this is VK_TRUE, the display engine will enable buffered mode on displays that support it. This allows the display engine to stop sending content to the display until a new image is presented. The display will instead maintain a copy of the last presented image. This allows less power to be used, but may increase presentation latency. If VkDisplayPresentInfoKHR is not specified, persistent mode will not be used.

If the extent of the srcRect and dstRect are not equal, the presented pixels will be scaled accordingly.

Valid Usage
  • srcRect must specify a rectangular region that is a subset of the image being presented

  • dstRect must specify a rectangular region that is a subset of the visibleRegion parameter of the display mode the swapchain being presented uses

  • If the persistentContent member of the VkDisplayPropertiesKHR structure returned by vkGetPhysicalDeviceDisplayPropertiesKHR for the display the present operation targets then persistent must be VK_FALSE

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DISPLAY_PRESENT_INFO_KHR

If the pNext chain of VkPresentInfoKHR includes a VkDeviceGroupPresentInfoKHR structure, then that structure includes an array of device masks and a device group present mode.

The VkDeviceGroupPresentInfoKHR structure is defined as:

typedef struct VkDeviceGroupPresentInfoKHR {
    VkStructureType                        sType;
    const void*                            pNext;
    uint32_t                               swapchainCount;
    const uint32_t*                        pDeviceMasks;
    VkDeviceGroupPresentModeFlagBitsKHR    mode;
} VkDeviceGroupPresentInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchainCount is zero or the number of elements in pDeviceMasks.

  • pDeviceMasks is an array of device masks, one for each element of VkPresentInfoKHR::pSwapchains.

  • mode is the device group present mode that will be used for this present.

If mode is VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR, then each element of pDeviceMasks selects which instance of the swapchain image is presented. Each element of pDeviceMasks must have exactly one bit set, and the corresponding physical device must have a presentation engine as reported by VkDeviceGroupPresentCapabilitiesKHR.

If mode is VK_DEVICE_GROUP_PRESENT_MODE_REMOTE_BIT_KHR, then each element of pDeviceMasks selects which instance of the swapchain image is presented. Each element of pDeviceMasks must have exactly one bit set, and some physical device in the logical device must include that bit in its VkDeviceGroupPresentCapabilitiesKHR::presentMask.

If mode is VK_DEVICE_GROUP_PRESENT_MODE_SUM_BIT_KHR, then each element of pDeviceMasks selects which instances of the swapchain image are component-wise summed and the sum of those images is presented. If the sum in any component is outside the representable range, the value of that component is undefined. Each element of pDeviceMasks must have a value for which all set bits are set in one of the elements of VkDeviceGroupPresentCapabilitiesKHR::presentMask.

If mode is VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_MULTI_DEVICE_BIT_KHR, then each element of pDeviceMasks selects which instance(s) of the swapchain images are presented. For each bit set in each element of pDeviceMasks, the corresponding physical device must have a presentation engine as reported by VkDeviceGroupPresentCapabilitiesKHR.

If VkDeviceGroupPresentInfoKHR is not provided or swapchainCount is zero then the masks are considered to be 1. If VkDeviceGroupPresentInfoKHR is not provided, mode is considered to be VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR.

Valid Usage
  • swapchainCount must equal 0 or VkPresentInfoKHR::swapchainCount

  • If mode is VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_BIT_KHR, then each element of pDeviceMasks must have exactly one bit set, and the corresponding element of VkDeviceGroupPresentCapabilitiesKHR::presentMask must be non-zero

  • If mode is VK_DEVICE_GROUP_PRESENT_MODE_REMOTE_BIT_KHR, then each element of pDeviceMasks must have exactly one bit set, and some physical device in the logical device must include that bit in its VkDeviceGroupPresentCapabilitiesKHR::presentMask.

  • If mode is VK_DEVICE_GROUP_PRESENT_MODE_SUM_BIT_KHR, then each element of pDeviceMasks must have a value for which all set bits are set in one of the elements of VkDeviceGroupPresentCapabilitiesKHR::presentMask

  • If mode is VK_DEVICE_GROUP_PRESENT_MODE_LOCAL_MULTI_DEVICE_BIT_KHR, then for each bit set in each element of pDeviceMasks, the corresponding element of VkDeviceGroupPresentCapabilitiesKHR::presentMask must be non-zero

  • The value of each element of pDeviceMasks must be equal to the device mask passed in VkAcquireNextImageInfoKHR::deviceMask when the image index was last acquired

  • mode must have exactly one bit set, and that bit must have been included in VkDeviceGroupSwapchainCreateInfoKHR::modes

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_INFO_KHR

  • If swapchainCount is not 0, pDeviceMasks must be a valid pointer to an array of swapchainCount uint32_t values

  • mode must be a valid VkDeviceGroupPresentModeFlagBitsKHR value

When the VK_GOOGLE_display_timing extension is enabled, additional fields can be specified that allow an application to specify the earliest time that an image should be displayed. This allows an application to avoid stutter that is caused by an image being displayed earlier than planned. Such stuttering can occur with both fixed and variable-refresh-rate displays, because stuttering occurs when the geometry is not correctly positioned for when the image is displayed. An application can instruct the presentation engine that an image should not be displayed earlier than a specified time by including the VkPresentTimesInfoGOOGLE structure in the pNext chain of the VkPresentInfoKHR structure.

The VkPresentTimesInfoGOOGLE structure is defined as:

typedef struct VkPresentTimesInfoGOOGLE {
    VkStructureType               sType;
    const void*                   pNext;
    uint32_t                      swapchainCount;
    const VkPresentTimeGOOGLE*    pTimes;
} VkPresentTimesInfoGOOGLE;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • swapchainCount is the number of swapchains being presented to by this command.

  • pTimes is NULL or a pointer to an array of VkPresentTimeGOOGLE elements with swapchainCount entries. If not NULL, each element of pTimes contains the earliest time to present the image corresponding to the entry in the VkPresentInfoKHR::pImageIndices array.

Valid Usage
  • swapchainCount must be the same value as VkPresentInfoKHR::swapchainCount, where VkPresentInfoKHR is in the pNext chain of this VkPresentTimesInfoGOOGLE structure.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PRESENT_TIMES_INFO_GOOGLE

  • If pTimes is not NULL, pTimes must be a valid pointer to an array of swapchainCount VkPresentTimeGOOGLE structures

  • swapchainCount must be greater than 0

The VkPresentTimeGOOGLE structure is defined as:

typedef struct VkPresentTimeGOOGLE {
    uint32_t    presentID;
    uint64_t    desiredPresentTime;
} VkPresentTimeGOOGLE;
  • presentID is an application-provided identification value, that can be used with the results of vkGetPastPresentationTimingGOOGLE, in order to uniquely identify this present. In order to be useful to the application, it should be unique within some period of time that is meaningful to the application.

  • desiredPresentTime specifies that the image given should not be displayed to the user any earlier than this time. desiredPresentTime is a time in nanoseconds, relative to a monotonically-increasing clock (e.g. CLOCK_MONOTONIC (see clock_gettime(2)) on Android and Linux). A value of zero specifies that the presentation engine may display the image at any time. This is useful when the application desires to provide presentID, but doesn’t need a specific desiredPresentTime.

vkQueuePresentKHR, releases the acquisition of the images referenced by imageIndices. The queue family corresponding to the queue vkQueuePresentKHR is executed on must have ownership of the presented images as defined in Resource Sharing. vkQueuePresentKHR does not alter the queue family ownership, but the presented images must not be used again before they have been reacquired using vkAcquireNextImageKHR.

The processing of the presentation happens in issue order with other queue operations, but semaphores have to be used to ensure that prior rendering and other commands in the specified queue complete before the presentation begins. The presentation command itself does not delay processing of subsequent commands on the queue, however, presentation requests sent to a particular queue are always performed in order. Exact presentation timing is controlled by the semantics of the presentation engine and native platform in use.

If an image is presented to a swapchain created from a display surface, the mode of the associated display will be updated, if necessary, to match the mode specified when creating the display surface. The mode switch and presentation of the specified image will be performed as one atomic operation.

The result codes VK_ERROR_OUT_OF_DATE_KHR and VK_SUBOPTIMAL_KHR have the same meaning when returned by vkQueuePresentKHR as they do when returned by vkAcquireNextImageKHR. If multiple swapchains are presented, the result code is determined applying the following rules in order:

  • If the device is lost, VK_ERROR_DEVICE_LOST is returned.

  • If any of the target surfaces are no longer available the error VK_ERROR_SURFACE_LOST_KHR is returned.

  • If any of the presents would have a result of VK_ERROR_OUT_OF_DATE_KHR if issued separately then VK_ERROR_OUT_OF_DATE_KHR is returned.

  • If any of the presents would have a result of VK_SUBOPTIMAL_KHR if issued separately then VK_SUBOPTIMAL_KHR is returned.

  • Otherwise VK_SUCCESS is returned.

Presentation is a read-only operation that will not affect the content of the presentable images. Upon reacquiring the image and transitioning it away from the VK_IMAGE_LAYOUT_PRESENT_SRC_KHR layout, the contents will be the same as they were prior to transitioning the image to the present source layout and presenting it. However, if a mechanism other than Vulkan is used to modify the platform window associated with the swapchain, the content of all presentable images in the swapchain becomes undefined.

Note

The application can continue to present any acquired images from a retired swapchain as long as the swapchain has not entered a state that causes vkQueuePresentKHR to return VK_ERROR_OUT_OF_DATE_KHR.

30.9. Hdr Metadata

To improve color reproduction of content it is useful to have information that can be used to better reproduce the colors as seen on the mastering display. That information can be provided to an implementation by calling vkSetHdrMetadataEXT. The metadata will be applied to the specified VkSwapchainKHR objects at the next vkQueuePresentKHR call using that VkSwapchainKHR object. The metadata will persist until a subsequent vkSetHdrMetadataEXT changes it. The definitions below are from the associated SMPTE 2086, CTA 861.3 and CIE 15:2004 specifications.

The definition of vkSetHdrMetadataEXT is:

void vkSetHdrMetadataEXT(
    VkDevice                                    device,
    uint32_t                                    swapchainCount,
    const VkSwapchainKHR*                       pSwapchains,
    const VkHdrMetadataEXT*                     pMetadata);
  • device is the logical device where the swapchain(s) were created.

  • swapchainCount is the number of swapchains included in pSwapchains.

  • pSwapchains is a pointer to the array of swapchainCount VkSwapchainKHR handles.

  • pMetadata is a pointer to the array of swapchainCount VkHdrMetadataEXT structures.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pSwapchains must be a valid pointer to an array of swapchainCount valid VkSwapchainKHR handles

  • pMetadata must be a valid pointer to an array of swapchainCount valid VkHdrMetadataEXT structures

  • swapchainCount must be greater than 0

  • Both of device, and the elements of pSwapchains must have been created, allocated, or retrieved from the same VkInstance

typedef struct VkXYColorEXT {
    float    x;
    float    y;
} VkXYColorEXT;

Chromaticity coordinates x and y are as specified in CIE 15:2004 “Calculation of chromaticity coordinates” (Section 7.3) and are limited to between 0 and 1 for real colors for the mastering display.

typedef struct VkHdrMetadataEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkXYColorEXT       displayPrimaryRed;
    VkXYColorEXT       displayPrimaryGreen;
    VkXYColorEXT       displayPrimaryBlue;
    VkXYColorEXT       whitePoint;
    float              maxLuminance;
    float              minLuminance;
    float              maxContentLightLevel;
    float              maxFrameAverageLightLevel;
} VkHdrMetadataEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • displayPrimaryRed is the mastering display’s red primary in chromaticity coordinates

  • displayPrimaryGreen is the mastering display’s green primary in chromaticity coordinates

  • displayPrimaryBlue is the mastering display’s blue primary in chromaticity coordinates

  • whitePoint is the mastering display’s white-point in chromaticity coordinates

  • maxLuminance is the maximum luminance of the mastering display in nits

  • minLuminance is the minimum luminance of the mastering display in nits

  • maxContentLightLevel is content’s maximum luminance in nits

  • maxFrameAverageLightLevel is the maximum frame average light level in nits

Note

The validity and use of this data is outside the scope of Vulkan and thus no Valid Usage is given.

31. Extended Functionality

Additional functionality may be provided by layers or extensions. A layer cannot add or modify Vulkan commands, while an extension may do so.

The set of layers to enable is specified when creating an instance, and those layers are able to intercept any Vulkan command dispatched to that instance or any of its child objects.

Extensions can operate at either the instance or device extension scope. Enabled instance extensions are able to affect the operation of the instance and any of its child objects, while device extensions may only be available on a subset of physical devices, must be individually enabled per-device, and only affect the operation of the devices where they are enabled.

Note

Examples of these might be:

  • Whole API validation is an example of a layer.

  • Debug capabilities might make a good instance extension.

  • A layer that provides implementation-specific performance telemetry and analysis could be a layer that is only active for devices created from compatible physical devices.

  • Functions to allow an application to use additional implementation features beyond the core would be a good candidate for a device extension.

31.1. Layers

When a layer is enabled, it inserts itself into the call chain for Vulkan commands the layer is interested in. A common use of layers is to validate application behavior during development. For example, the implementation will not check that Vulkan enums used by the application fall within allowed ranges. Instead, a validation layer would do those checks and flag issues. This avoids a performance penalty during production use of the application because those layers would not be enabled in production.

Vulkan layers may wrap object handles (i.e. return a different handle value to the application than that generated by the implementation). This is generally discouraged, as it increases the probability of incompatibilities with new extensions. The validation layers wrap handles in order to track the proper use and destruction of each object. See the “Vulkan Loader Specification and Architecture Overview” document for additional information.

To query the available layers, call:

VkResult vkEnumerateInstanceLayerProperties(
    uint32_t*                                   pPropertyCount,
    VkLayerProperties*                          pProperties);
  • pPropertyCount is a pointer to an integer related to the number of layer properties available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkLayerProperties structures.

If pProperties is NULL, then the number of layer properties available is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If pPropertyCount is less than the number of layer properties available, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of layers available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available layer properties were returned.

The list of available layers may change at any time due to actions outside of the Vulkan implementation, so two calls to vkEnumerateInstanceLayerProperties with the same parameters may return different results, or retrieve different pPropertyCount values or pProperties contents. Once an instance has been created, the layers enabled for that instance will continue to be enabled and valid for the lifetime of that instance, even if some of them become unavailable for future instances.

Valid Usage (Implicit)
  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkLayerProperties structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkLayerProperties structure is defined as:

typedef struct VkLayerProperties {
    char        layerName[VK_MAX_EXTENSION_NAME_SIZE];
    uint32_t    specVersion;
    uint32_t    implementationVersion;
    char        description[VK_MAX_DESCRIPTION_SIZE];
} VkLayerProperties;
  • layerName is a null-terminated UTF-8 string specifying the name of the layer. Use this name in the ppEnabledLayerNames array passed in the VkInstanceCreateInfo structure to enable this layer for an instance.

  • specVersion is the Vulkan version the layer was written to, encoded as described in the API Version Numbers and Semantics section.

  • implementationVersion is the version of this layer. It is an integer, increasing with backward compatible changes.

  • description is a null-terminated UTF-8 string providing additional details that can be used by the application to identify the layer.

To enable a layer, the name of the layer should be added to the ppEnabledLayerNames member of VkInstanceCreateInfo when creating a VkInstance.

Loader implementations may provide mechanisms outside the Vulkan API for enabling specific layers. Layers enabled through such a mechanism are implicitly enabled, while layers enabled by including the layer name in the ppEnabledLayerNames member of VkInstanceCreateInfo are explicitly enabled. Except where otherwise specified, implicitly enabled and explicitly enabled layers differ only in the way they are enabled. Explicitly enabling a layer that is implicitly enabled has no additional effect.

31.1.1. Device Layer Deprecation

Previous versions of this specification distinguished between instance and device layers. Instance layers were only able to intercept commands that operate on VkInstance and VkPhysicalDevice, except they were not able to intercept vkCreateDevice. Device layers were enabled for individual devices when they were created, and could only intercept commands operating on that device or its child objects.

Device-only layers are now deprecated, and this specification no longer distinguishes between instance and device layers. Layers are enabled during instance creation, and are able to intercept all commands operating on that instance or any of its child objects. At the time of deprecation there were no known device-only layers and no compelling reason to create one.

In order to maintain compatibility with implementations released prior to device-layer deprecation, applications should still enumerate and enable device layers. The behavior of vkEnumerateDeviceLayerProperties and valid usage of the ppEnabledLayerNames member of VkDeviceCreateInfo maximizes compatibility with applications written to work with the previous requirements.

To enumerate device layers, call:

VkResult vkEnumerateDeviceLayerProperties(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pPropertyCount,
    VkLayerProperties*                          pProperties);
  • pPropertyCount is a pointer to an integer related to the number of layer properties available or queried.

  • pProperties is either NULL or a pointer to an array of VkLayerProperties structures.

If pProperties is NULL, then the number of layer properties available is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If pPropertyCount is less than the number of layer properties available, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of layers available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available layer properties were returned.

The list of layers enumerated by vkEnumerateDeviceLayerProperties must be exactly the sequence of layers enabled for the instance. The members of VkLayerProperties for each enumerated layer must be the same as the properties when the layer was enumerated by vkEnumerateInstanceLayerProperties.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkLayerProperties structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The ppEnabledLayerNames and enabledLayerCount members of VkDeviceCreateInfo are deprecated and their values must be ignored by implementations. However, for compatibility, only an empty list of layers or a list that exactly matches the sequence enabled at instance creation time are valid, and validation layers should issue diagnostics for other cases.

Regardless of the enabled layer list provided in VkDeviceCreateInfo, the sequence of layers active for a device will be exactly the sequence of layers enabled when the parent instance was created.

31.2. Extensions

Extensions may define new Vulkan commands, structures, and enumerants. For compilation purposes, the interfaces defined by registered extensions, including new structures and enumerants as well as function pointer types for new commands, are defined in the Khronos-supplied vulkan_core.h together with the core API. However, commands defined by extensions may not be available for static linking - in which case function pointers to these commands should be queried at runtime as described in Command Function Pointers. Extensions may be provided by layers as well as by a Vulkan implementation.

Because extensions may extend or change the behavior of the Vulkan API, extension authors should add support for their extensions to the Khronos validation layers. This is especially important for new commands whose parameters have been wrapped by the validation layers. See the “Vulkan Loader Specification and Architecture Overview” document for additional information.

Note

Valid Usage sections for individual commands and structures do not currently contain which extensions have to be enabled in order to make their use valid, although it might do so in the future. It is defined only in the Valid Usage for Extensions section.

To query the available instance extensions, call:

VkResult vkEnumerateInstanceExtensionProperties(
    const char*                                 pLayerName,
    uint32_t*                                   pPropertyCount,
    VkExtensionProperties*                      pProperties);
  • pLayerName is either NULL or a pointer to a null-terminated UTF-8 string naming the layer to retrieve extensions from.

  • pPropertyCount is a pointer to an integer related to the number of extension properties available or queried, as described below.

  • pProperties is either NULL or a pointer to an array of VkExtensionProperties structures.

When pLayerName parameter is NULL, only extensions provided by the Vulkan implementation or by implicitly enabled layers are returned. When pLayerName is the name of a layer, the instance extensions provided by that layer are returned.

If pProperties is NULL, then the number of extensions properties available is returned in pPropertyCount. Otherwise, pPropertyCount must point to a variable set by the user to the number of elements in the pProperties array, and on return the variable is overwritten with the number of structures actually written to pProperties. If pPropertyCount is less than the number of extension properties available, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of extensions available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available properties were returned.

Because the list of available layers may change externally between calls to vkEnumerateInstanceExtensionProperties, two calls may retrieve different results if a pLayerName is available in one call but not in another. The extensions supported by a layer may also change between two calls, e.g. if the layer implementation is replaced by a different version between those calls.

Valid Usage (Implicit)
  • If pLayerName is not NULL, pLayerName must be a null-terminated UTF-8 string

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkExtensionProperties structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_LAYER_NOT_PRESENT

To enable an instance extension, the name of the extension should be added to the ppEnabledExtensionNames member of VkInstanceCreateInfo when creating a VkInstance.

Enabling an extension does not change behavior of functionality exposed by the core Vulkan API or any other extension, other than making valid the use of the commands, enums and structures defined by that extension.

To query the extensions available to a given physical device, call:

VkResult vkEnumerateDeviceExtensionProperties(
    VkPhysicalDevice                            physicalDevice,
    const char*                                 pLayerName,
    uint32_t*                                   pPropertyCount,
    VkExtensionProperties*                      pProperties);
  • physicalDevice is the physical device that will be queried.

  • pLayerName is either NULL or a pointer to a null-terminated UTF-8 string naming the layer to retrieve extensions from.

  • pPropertyCount is a pointer to an integer related to the number of extension properties available or queried, and is treated in the same fashion as the vkEnumerateInstanceExtensionProperties::pPropertyCount parameter.

  • pProperties is either NULL or a pointer to an array of VkExtensionProperties structures.

When pLayerName parameter is NULL, only extensions provided by the Vulkan implementation or by implicitly enabled layers are returned. When pLayerName is the name of a layer, the device extensions provided by that layer are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • If pLayerName is not NULL, pLayerName must be a null-terminated UTF-8 string

  • pPropertyCount must be a valid pointer to a uint32_t value

  • If the value referenced by pPropertyCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pPropertyCount VkExtensionProperties structures

Return Codes
Success
  • VK_SUCCESS

  • VK_INCOMPLETE

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_LAYER_NOT_PRESENT

The VkExtensionProperties structure is defined as:

typedef struct VkExtensionProperties {
    char        extensionName[VK_MAX_EXTENSION_NAME_SIZE];
    uint32_t    specVersion;
} VkExtensionProperties;
  • extensionName is a null-terminated string specifying the name of the extension.

  • specVersion is the version of this extension. It is an integer, incremented with backward compatible changes.

31.2.1. Instance Extensions and Device Extensions

This section provides some guidelines and rules for when to expose new functionality as an instance extension, as a device extension, or as both. The decision depends on the scope of the new functionality; such as whether it extends instance-level or device-level functionality. All Vulkan commands, structures, and enumerants are considered either instance-level, physical-device-level, or device-level.

New instance-level extension functionality must be structured within an instance extension. New device-level extension functionality may be structured within a device extension. Vulkan 1.0 initially required all new physical-device-level extension functionality to be structured within an instance extension. In order to avoid using an instance extension, which often requires loader support, physical-device-level extension functionality may be implemented within device extensions (which must depend on the VK_KHR_get_physical_device_properties2 extension, or on Vulkan 1.1 or later).

31.3. Extension Dependencies

Some extensions are dependent on other extensions to function. To enable extensions with dependencies, such required extensions must also be enabled through the same API mechanisms when creating an instance with vkCreateInstance or a device with vkCreateDevice. Each extension which has such dependencies documents them in the appendix summarizing that extension.

If an extension is supported (as queried by vkEnumerateInstanceExtensionProperties or vkEnumerateDeviceExtensionProperties), then required extensions of that extension must also be supported for the same instance or physical device.

Any device extension that has an instance extension dependency that is not enabled by vkCreateInstance is considered to be unsupported, hence it must not be returned by vkEnumerateDeviceExtensionProperties for any VkPhysicalDevice child of the instance.

31.4. Extension Compatibility

By default, all extensions are considered compatible with each other and any core API version, unless otherwise stated. Thus enabling such extensions does not otherwise alter the behavior of the application.

Each extension that is mutually exclusive or otherwise incompatible with another extension or set of extensions documents them in the appendix summarizing that extension and has a corresponding Valid Usage statement disallowing enabling such an incompatible combination of extensions at VkInstance creation time or VkDevice creation time, depending on the type of extensions participating in the interaction.

32. Features, Limits, and Formats

Vulkan is designed to support a wide variety of implementations, and as such there are a number of features, limits, and formats which are not supported on all implementations. Features describe functionality which is optional and which must be explicitly enabled before use. Limits describe implementation-dependent minimums, maximums, and other device characteristics that an application may need to be aware of. Supported buffer and image formats may vary across implementations. A minimum set of format features are guaranteed, but others must be explicitly queried before use to ensure they are supported by the implementation.

Note

The features and limits are reported via basic structures (that is VkPhysicalDeviceFeatures and VkPhysicalDeviceLimits), as well as extensible structures (VkPhysicalDeviceFeatures2 and VkPhysicalDeviceProperties2) which were added in VK_KHR_get_physical_device_properties2 and included in Vulkan 1.1. When new features or limits are added in future Vulkan version or extensions, each extension should introduce one new feature structure and/or limit structure (as needed). These structures can be added to the pNext chain of the VkPhysicalDeviceFeatures2 and VkPhysicalDeviceProperties2 structures, respectively.

32.1. Features

The Specification defines a set of optional features that may be supported by a Vulkan implementation. Support for features is reported and enabled on a per-feature basis. Features are properties of the physical device.

To query supported features, call:

void vkGetPhysicalDeviceFeatures(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceFeatures*                   pFeatures);
  • physicalDevice is the physical device from which to query the supported features.

  • pFeatures is a pointer to a VkPhysicalDeviceFeatures structure in which the physical device features are returned. For each feature, a value of VK_TRUE specifies that the feature is supported on this physical device, and VK_FALSE specifies that the feature is not supported.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pFeatures must be a valid pointer to a VkPhysicalDeviceFeatures structure

Fine-grained features used by a logical device must be enabled at VkDevice creation time. If a feature is enabled that the physical device does not support, VkDevice creation will fail. If an application uses a feature without enabling it at VkDevice creation time, the device behavior is undefined. The validation layer will warn if features are used without being enabled.

The fine-grained features are enabled by passing a pointer to the VkPhysicalDeviceFeatures structure via the pEnabledFeatures member of the VkDeviceCreateInfo structure that is passed into the vkCreateDevice call. If a member of pEnabledFeatures is set to VK_TRUE or VK_FALSE, then the device will be created with the indicated feature enabled or disabled, respectively. Features can also be enabled by using the VkPhysicalDeviceFeatures2 structure.

If an application wishes to enable all features supported by a device, it can simply pass in the VkPhysicalDeviceFeatures structure that was previously returned by vkGetPhysicalDeviceFeatures. To disable an individual feature, the application can set the desired member to VK_FALSE in the same structure. Setting pEnabledFeatures to NULL and not including a VkPhysicalDeviceFeatures2 in the pNext member of VkDeviceCreateInfo is equivalent to setting all members of the structure to VK_FALSE.

Note

Some features, such as robustBufferAccess, may incur a run-time performance cost. Application writers should carefully consider the implications of enabling all supported features.

To query supported features defined by the core or extensions, call:

void vkGetPhysicalDeviceFeatures2(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceFeatures2*                  pFeatures);

or the equivalent command

void vkGetPhysicalDeviceFeatures2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceFeatures2*                  pFeatures);
  • physicalDevice is the physical device from which to query the supported features.

  • pFeatures is a pointer to a VkPhysicalDeviceFeatures2 structure in which the physical device features are returned.

Each structure in pFeatures and its pNext chain contain members corresponding to fine-grained features. vkGetPhysicalDeviceFeatures2 writes each member to a boolean value indicating whether that feature is supported.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pFeatures must be a valid pointer to a VkPhysicalDeviceFeatures2 structure

The VkPhysicalDeviceFeatures2 structure is defined as:

typedef struct VkPhysicalDeviceFeatures2 {
    VkStructureType             sType;
    void*                       pNext;
    VkPhysicalDeviceFeatures    features;
} VkPhysicalDeviceFeatures2;

or the equivalent

typedef VkPhysicalDeviceFeatures2 VkPhysicalDeviceFeatures2KHR;

The VkPhysicalDeviceFeatures2 structure is defined as:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • features is a structure of type VkPhysicalDeviceFeatures describing the fine-grained features of the Vulkan 1.0 API.

The pNext chain of this structure is used to extend the structure with features defined by extensions. This structure can be used in vkGetPhysicalDeviceFeatures2 or can be in the pNext chain of a VkDeviceCreateInfo structure, in which case it controls which features are enabled in the device in lieu of pEnabledFeatures.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2

The VkPhysicalDeviceFeatures structure is defined as:

typedef struct VkPhysicalDeviceFeatures {
    VkBool32    robustBufferAccess;
    VkBool32    fullDrawIndexUint32;
    VkBool32    imageCubeArray;
    VkBool32    independentBlend;
    VkBool32    geometryShader;
    VkBool32    tessellationShader;
    VkBool32    sampleRateShading;
    VkBool32    dualSrcBlend;
    VkBool32    logicOp;
    VkBool32    multiDrawIndirect;
    VkBool32    drawIndirectFirstInstance;
    VkBool32    depthClamp;
    VkBool32    depthBiasClamp;
    VkBool32    fillModeNonSolid;
    VkBool32    depthBounds;
    VkBool32    wideLines;
    VkBool32    largePoints;
    VkBool32    alphaToOne;
    VkBool32    multiViewport;
    VkBool32    samplerAnisotropy;
    VkBool32    textureCompressionETC2;
    VkBool32    textureCompressionASTC_LDR;
    VkBool32    textureCompressionBC;
    VkBool32    occlusionQueryPrecise;
    VkBool32    pipelineStatisticsQuery;
    VkBool32    vertexPipelineStoresAndAtomics;
    VkBool32    fragmentStoresAndAtomics;
    VkBool32    shaderTessellationAndGeometryPointSize;
    VkBool32    shaderImageGatherExtended;
    VkBool32    shaderStorageImageExtendedFormats;
    VkBool32    shaderStorageImageMultisample;
    VkBool32    shaderStorageImageReadWithoutFormat;
    VkBool32    shaderStorageImageWriteWithoutFormat;
    VkBool32    shaderUniformBufferArrayDynamicIndexing;
    VkBool32    shaderSampledImageArrayDynamicIndexing;
    VkBool32    shaderStorageBufferArrayDynamicIndexing;
    VkBool32    shaderStorageImageArrayDynamicIndexing;
    VkBool32    shaderClipDistance;
    VkBool32    shaderCullDistance;
    VkBool32    shaderFloat64;
    VkBool32    shaderInt64;
    VkBool32    shaderInt16;
    VkBool32    shaderResourceResidency;
    VkBool32    shaderResourceMinLod;
    VkBool32    sparseBinding;
    VkBool32    sparseResidencyBuffer;
    VkBool32    sparseResidencyImage2D;
    VkBool32    sparseResidencyImage3D;
    VkBool32    sparseResidency2Samples;
    VkBool32    sparseResidency4Samples;
    VkBool32    sparseResidency8Samples;
    VkBool32    sparseResidency16Samples;
    VkBool32    sparseResidencyAliased;
    VkBool32    variableMultisampleRate;
    VkBool32    inheritedQueries;
} VkPhysicalDeviceFeatures;

The members of the VkPhysicalDeviceFeatures structure describe the following features:

  • robustBufferAccess specifies that accesses to buffers are bounds-checked against the range of the buffer descriptor (as determined by VkDescriptorBufferInfo::range, VkBufferViewCreateInfo::range, or the size of the buffer). Out of bounds accesses must not cause application termination, and the effects of shader loads, stores, and atomics must conform to an implementation-dependent behavior as described below.

    • A buffer access is considered to be out of bounds if any of the following are true:

      • The pointer was formed by OpImageTexelPointer and the coordinate is less than zero or greater than or equal to the number of whole elements in the bound range.

      • The pointer was not formed by OpImageTexelPointer and the object pointed to is not wholly contained within the bound range. This includes accesses performed via variable pointers where the buffer descriptor being accessed cannot be statically determined. Uninitialized pointers and pointers equal to OpConstantNull are treated as pointing to a zero-sized object, so all accesses through such pointers are considered to be out of bounds.

        Note

        If a SPIR-V OpLoad instruction loads a structure and the tail end of the structure is out of bounds, then all members of the structure are considered out of bounds even if the members at the end are not statically used.

      • If any buffer access in a given SPIR-V block is determined to be out of bounds, then any other access of the same type (load, store, or atomic) in the same SPIR-V block that accesses an address less than 16 bytes away from the out of bounds address may also be considered out of bounds.

    • Out-of-bounds buffer loads will return any of the following values:

      • Values from anywhere within the memory range(s) bound to the buffer (possibly including bytes of memory past the end of the buffer, up to the end of the bound range).

      • Zero values, or (0,0,0,x) vectors for vector reads where x is a valid value represented in the type of the vector components and may be any of:

        • 0, 1, or the maximum representable positive integer value, for signed or unsigned integer components

        • 0.0 or 1.0, for floating-point components

    • Out-of-bounds writes may modify values within the memory range(s) bound to the buffer, but must not modify any other memory.

    • Out-of-bounds atomics may modify values within the memory range(s) bound to the buffer, but must not modify any other memory, and return an undefined value.

    • Vertex input attributes are considered out of bounds if the offset of the attribute in the bound vertex buffer range plus the size of the attribute is greater than either:

      • vertexBufferRangeSize, if bindingStride == 0; or

      • (vertexBufferRangeSize - (vertexBufferRangeSize % bindingStride))

      where vertexBufferRangeSize is the byte size of the memory range bound to the vertex buffer binding and bindingStride is the byte stride of the corresponding vertex input binding. Further, if any vertex input attribute using a specific vertex input binding is out of bounds, then all vertex input attributes using that vertex input binding for that vertex shader invocation are considered out of bounds.

      • If a vertex input attribute is out of bounds, it will be assigned one of the following values:

        • Values from anywhere within the memory range(s) bound to the buffer, converted according to the format of the attribute.

        • Zero values, format converted according to the format of the attribute.

        • Zero values, or (0,0,0,x) vectors, as described above.

    • If robustBufferAccess is not enabled, out of bounds accesses may corrupt any memory within the process and cause undefined behavior up to and including application termination.

  • fullDrawIndexUint32 specifies the full 32-bit range of indices is supported for indexed draw calls when using a VkIndexType of VK_INDEX_TYPE_UINT32. maxDrawIndexedIndexValue is the maximum index value that may be used (aside from the primitive restart index, which is always 232-1 when the VkIndexType is VK_INDEX_TYPE_UINT32). If this feature is supported, maxDrawIndexedIndexValue must be 232-1; otherwise it must be no smaller than 224-1. See maxDrawIndexedIndexValue.

  • imageCubeArray specifies whether image views with a VkImageViewType of VK_IMAGE_VIEW_TYPE_CUBE_ARRAY can be created, and that the corresponding SampledCubeArray and ImageCubeArray SPIR-V capabilities can be used in shader code.

  • independentBlend specifies whether the VkPipelineColorBlendAttachmentState settings are controlled independently per-attachment. If this feature is not enabled, the VkPipelineColorBlendAttachmentState settings for all color attachments must be identical. Otherwise, a different VkPipelineColorBlendAttachmentState can be provided for each bound color attachment.

  • geometryShader specifies whether geometry shaders are supported. If this feature is not enabled, the VK_SHADER_STAGE_GEOMETRY_BIT and VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT enum values must not be used. This also specifies whether shader modules can declare the Geometry capability.

  • tessellationShader specifies whether tessellation control and evaluation shaders are supported. If this feature is not enabled, the VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT, VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT, VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT, VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT, and VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_STATE_CREATE_INFO enum values must not be used. This also specifies whether shader modules can declare the Tessellation capability.

  • sampleRateShading specifies whether Sample Shading and multisample interpolation are supported. If this feature is not enabled, the sampleShadingEnable member of the VkPipelineMultisampleStateCreateInfo structure must be set to VK_FALSE and the minSampleShading member is ignored. This also specifies whether shader modules can declare the SampleRateShading capability.

  • dualSrcBlend specifies whether blend operations which take two sources are supported. If this feature is not enabled, the VK_BLEND_FACTOR_SRC1_COLOR, VK_BLEND_FACTOR_ONE_MINUS_SRC1_COLOR, VK_BLEND_FACTOR_SRC1_ALPHA, and VK_BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA enum values must not be used as source or destination blending factors. See Dual-Source Blending.

  • logicOp specifies whether logic operations are supported. If this feature is not enabled, the logicOpEnable member of the VkPipelineColorBlendStateCreateInfo structure must be set to VK_FALSE, and the logicOp member is ignored.

  • multiDrawIndirect specifies whether multiple draw indirect is supported. If this feature is not enabled, the drawCount parameter to the vkCmdDrawIndirect and vkCmdDrawIndexedIndirect commands must be 0 or 1. The maxDrawIndirectCount member of the VkPhysicalDeviceLimits structure must also be 1 if this feature is not supported. See maxDrawIndirectCount.

  • drawIndirectFirstInstance specifies whether indirect draw calls support the firstInstance parameter. If this feature is not enabled, the firstInstance member of all VkDrawIndirectCommand and VkDrawIndexedIndirectCommand structures that are provided to the vkCmdDrawIndirect and vkCmdDrawIndexedIndirect commands must be 0.

  • depthClamp specifies whether depth clamping is supported. If this feature is not enabled, the depthClampEnable member of the VkPipelineRasterizationStateCreateInfo structure must be set to VK_FALSE. Otherwise, setting depthClampEnable to VK_TRUE will enable depth clamping.

  • depthBiasClamp specifies whether depth bias clamping is supported. If this feature is not enabled, the depthBiasClamp member of the VkPipelineRasterizationStateCreateInfo structure must be set to 0.0 unless the VK_DYNAMIC_STATE_DEPTH_BIAS dynamic state is enabled, and the depthBiasClamp parameter to vkCmdSetDepthBias must be set to 0.0.

  • fillModeNonSolid specifies whether point and wireframe fill modes are supported. If this feature is not enabled, the VK_POLYGON_MODE_POINT and VK_POLYGON_MODE_LINE enum values must not be used.

  • depthBounds specifies whether depth bounds tests are supported. If this feature is not enabled, the depthBoundsTestEnable member of the VkPipelineDepthStencilStateCreateInfo structure must be set to VK_FALSE. When depthBoundsTestEnable is set to VK_FALSE, the minDepthBounds and maxDepthBounds members of the VkPipelineDepthStencilStateCreateInfo structure are ignored.

  • wideLines specifies whether lines with width other than 1.0 are supported. If this feature is not enabled, the lineWidth member of the VkPipelineRasterizationStateCreateInfo structure must be set to 1.0 unless the VK_DYNAMIC_STATE_LINE_WIDTH dynamic state is enabled, and the lineWidth parameter to vkCmdSetLineWidth must be set to 1.0. When this feature is supported, the range and granularity of supported line widths are indicated by the lineWidthRange and lineWidthGranularity members of the VkPhysicalDeviceLimits structure, respectively.

  • largePoints specifies whether points with size greater than 1.0 are supported. If this feature is not enabled, only a point size of 1.0 written by a shader is supported. The range and granularity of supported point sizes are indicated by the pointSizeRange and pointSizeGranularity members of the VkPhysicalDeviceLimits structure, respectively.

  • alphaToOne specifies whether the implementation is able to replace the alpha value of the color fragment output from the fragment shader with the maximum representable alpha value for fixed-point colors or 1.0 for floating-point colors. If this feature is not enabled, then the alphaToOneEnable member of the VkPipelineMultisampleStateCreateInfo structure must be set to VK_FALSE. Otherwise setting alphaToOneEnable to VK_TRUE will enable alpha-to-one behavior.

  • multiViewport specifies whether more than one viewport is supported. If this feature is not enabled, the viewportCount and scissorCount members of the VkPipelineViewportStateCreateInfo structure must be set to 1. Similarly, the viewportCount parameter to the vkCmdSetViewport command and the scissorCount parameter to the vkCmdSetScissor command must be 1, and the firstViewport parameter to the vkCmdSetViewport command and the firstScissor parameter to the vkCmdSetScissor command must be 0.

  • samplerAnisotropy specifies whether anisotropic filtering is supported. If this feature is not enabled, the anisotropyEnable member of the VkSamplerCreateInfo structure must be VK_FALSE.

  • textureCompressionETC2 specifies whether all of the ETC2 and EAC compressed texture formats are supported. If this feature is enabled, then the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for the following formats:

    • VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK

    • VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK

    • VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK

    • VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK

    • VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK

    • VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK

    • VK_FORMAT_EAC_R11_UNORM_BLOCK

    • VK_FORMAT_EAC_R11_SNORM_BLOCK

    • VK_FORMAT_EAC_R11G11_UNORM_BLOCK

    • VK_FORMAT_EAC_R11G11_SNORM_BLOCK

    vkGetPhysicalDeviceFormatProperties and vkGetPhysicalDeviceImageFormatProperties can be used to check for additional supported properties of individual formats.

  • textureCompressionASTC_LDR specifies whether all of the ASTC LDR compressed texture formats are supported. If this feature is enabled, then the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for the following formats:

    • VK_FORMAT_ASTC_4x4_UNORM_BLOCK

    • VK_FORMAT_ASTC_4x4_SRGB_BLOCK

    • VK_FORMAT_ASTC_5x4_UNORM_BLOCK

    • VK_FORMAT_ASTC_5x4_SRGB_BLOCK

    • VK_FORMAT_ASTC_5x5_UNORM_BLOCK

    • VK_FORMAT_ASTC_5x5_SRGB_BLOCK

    • VK_FORMAT_ASTC_6x5_UNORM_BLOCK

    • VK_FORMAT_ASTC_6x5_SRGB_BLOCK

    • VK_FORMAT_ASTC_6x6_UNORM_BLOCK

    • VK_FORMAT_ASTC_6x6_SRGB_BLOCK

    • VK_FORMAT_ASTC_8x5_UNORM_BLOCK

    • VK_FORMAT_ASTC_8x5_SRGB_BLOCK

    • VK_FORMAT_ASTC_8x6_UNORM_BLOCK

    • VK_FORMAT_ASTC_8x6_SRGB_BLOCK

    • VK_FORMAT_ASTC_8x8_UNORM_BLOCK

    • VK_FORMAT_ASTC_8x8_SRGB_BLOCK

    • VK_FORMAT_ASTC_10x5_UNORM_BLOCK

    • VK_FORMAT_ASTC_10x5_SRGB_BLOCK

    • VK_FORMAT_ASTC_10x6_UNORM_BLOCK

    • VK_FORMAT_ASTC_10x6_SRGB_BLOCK

    • VK_FORMAT_ASTC_10x8_UNORM_BLOCK

    • VK_FORMAT_ASTC_10x8_SRGB_BLOCK

    • VK_FORMAT_ASTC_10x10_UNORM_BLOCK

    • VK_FORMAT_ASTC_10x10_SRGB_BLOCK

    • VK_FORMAT_ASTC_12x10_UNORM_BLOCK

    • VK_FORMAT_ASTC_12x10_SRGB_BLOCK

    • VK_FORMAT_ASTC_12x12_UNORM_BLOCK

    • VK_FORMAT_ASTC_12x12_SRGB_BLOCK

    vkGetPhysicalDeviceFormatProperties and vkGetPhysicalDeviceImageFormatProperties can be used to check for additional supported properties of individual formats.

  • textureCompressionBC specifies whether all of the BC compressed texture formats are supported. If this feature is enabled, then the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for the following formats:

    • VK_FORMAT_BC1_RGB_UNORM_BLOCK

    • VK_FORMAT_BC1_RGB_SRGB_BLOCK

    • VK_FORMAT_BC1_RGBA_UNORM_BLOCK

    • VK_FORMAT_BC1_RGBA_SRGB_BLOCK

    • VK_FORMAT_BC2_UNORM_BLOCK

    • VK_FORMAT_BC2_SRGB_BLOCK

    • VK_FORMAT_BC3_UNORM_BLOCK

    • VK_FORMAT_BC3_SRGB_BLOCK

    • VK_FORMAT_BC4_UNORM_BLOCK

    • VK_FORMAT_BC4_SNORM_BLOCK

    • VK_FORMAT_BC5_UNORM_BLOCK

    • VK_FORMAT_BC5_SNORM_BLOCK

    • VK_FORMAT_BC6H_UFLOAT_BLOCK

    • VK_FORMAT_BC6H_SFLOAT_BLOCK

    • VK_FORMAT_BC7_UNORM_BLOCK

    • VK_FORMAT_BC7_SRGB_BLOCK

    vkGetPhysicalDeviceFormatProperties and vkGetPhysicalDeviceImageFormatProperties can be used to check for additional supported properties of individual formats.

  • occlusionQueryPrecise specifies whether occlusion queries returning actual sample counts are supported. Occlusion queries are created in a VkQueryPool by specifying the queryType of VK_QUERY_TYPE_OCCLUSION in the VkQueryPoolCreateInfo structure which is passed to vkCreateQueryPool. If this feature is enabled, queries of this type can enable VK_QUERY_CONTROL_PRECISE_BIT in the flags parameter to vkCmdBeginQuery. If this feature is not supported, the implementation supports only boolean occlusion queries. When any samples are passed, boolean queries will return a non-zero result value, otherwise a result value of zero is returned. When this feature is enabled and VK_QUERY_CONTROL_PRECISE_BIT is set, occlusion queries will report the actual number of samples passed.

  • pipelineStatisticsQuery specifies whether the pipeline statistics queries are supported. If this feature is not enabled, queries of type VK_QUERY_TYPE_PIPELINE_STATISTICS cannot be created, and none of the VkQueryPipelineStatisticFlagBits bits can be set in the pipelineStatistics member of the VkQueryPoolCreateInfo structure.

  • vertexPipelineStoresAndAtomics specifies whether storage buffers and images support stores and atomic operations in the vertex, tessellation, and geometry shader stages. If this feature is not enabled, all storage image, storage texel buffers, and storage buffer variables used by these stages in shader modules must be decorated with the NonWriteable decoration (or the readonly memory qualifier in GLSL).

  • fragmentStoresAndAtomics specifies whether storage buffers and images support stores and atomic operations in the fragment shader stage. If this feature is not enabled, all storage image, storage texel buffers, and storage buffer variables used by the fragment stage in shader modules must be decorated with the NonWriteable decoration (or the readonly memory qualifier in GLSL).

  • shaderTessellationAndGeometryPointSize specifies whether the PointSize built-in decoration is available in the tessellation control, tessellation evaluation, and geometry shader stages. If this feature is not enabled, members decorated with the PointSize built-in decoration must not be read from or written to and all points written from a tessellation or geometry shader will have a size of 1.0. This also specifies whether shader modules can declare the TessellationPointSize capability for tessellation control and evaluation shaders, or if the shader modules can declare the GeometryPointSize capability for geometry shaders. An implementation supporting this feature must also support one or both of the tessellationShader or geometryShader features.

  • shaderImageGatherExtended specifies whether the extended set of image gather instructions are available in shader code. If this feature is not enabled, the OpImage*Gather instructions do not support the Offset and ConstOffsets operands. This also specifies whether shader modules can declare the ImageGatherExtended capability.

  • shaderStorageImageExtendedFormats specifies whether the extended storage image formats are available in shader code. If this feature is not enabled, the formats requiring the StorageImageExtendedFormats capability are not supported for storage images. This also specifies whether shader modules can declare the StorageImageExtendedFormats capability.

  • shaderStorageImageMultisample specifies whether multisampled storage images are supported. If this feature is not enabled, images that are created with a usage that includes VK_IMAGE_USAGE_STORAGE_BIT must be created with samples equal to VK_SAMPLE_COUNT_1_BIT. This also specifies whether shader modules can declare the StorageImageMultisample capability.

  • shaderStorageImageReadWithoutFormat specifies whether storage images require a format qualifier to be specified when reading from storage images. If this feature is not enabled, the OpImageRead instruction must not have an OpTypeImage of Unknown. This also specifies whether shader modules can declare the StorageImageReadWithoutFormat capability.

  • shaderStorageImageWriteWithoutFormat specifies whether storage images require a format qualifier to be specified when writing to storage images. If this feature is not enabled, the OpImageWrite instruction must not have an OpTypeImage of Unknown. This also specifies whether shader modules can declare the StorageImageWriteWithoutFormat capability.

  • shaderUniformBufferArrayDynamicIndexing specifies whether arrays of uniform buffers can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also specifies whether shader modules can declare the UniformBufferArrayDynamicIndexing capability.

  • shaderSampledImageArrayDynamicIndexing specifies whether arrays of samplers or sampled images can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, or VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also specifies whether shader modules can declare the SampledImageArrayDynamicIndexing capability.

  • shaderStorageBufferArrayDynamicIndexing specifies whether arrays of storage buffers can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also specifies whether shader modules can declare the StorageBufferArrayDynamicIndexing capability.

  • shaderStorageImageArrayDynamicIndexing specifies whether arrays of storage images can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also specifies whether shader modules can declare the StorageImageArrayDynamicIndexing capability.

  • shaderClipDistance specifies whether clip distances are supported in shader code. If this feature is not enabled, any members decorated with the ClipDistance built-in decoration must not be read from or written to in shader modules. This also specifies whether shader modules can declare the ClipDistance capability.

  • shaderCullDistance specifies whether cull distances are supported in shader code. If this feature is not enabled, any members decorated with the CullDistance built-in decoration must not be read from or written to in shader modules. This also specifies whether shader modules can declare the CullDistance capability.

  • shaderFloat64 specifies whether 64-bit floats (doubles) are supported in shader code. If this feature is not enabled, 64-bit floating-point types must not be used in shader code. This also specifies whether shader modules can declare the Float64 capability.

  • shaderInt64 specifies whether 64-bit integers (signed and unsigned) are supported in shader code. If this feature is not enabled, 64-bit integer types must not be used in shader code. This also specifies whether shader modules can declare the Int64 capability.

  • shaderInt16 specifies whether 16-bit integers (signed and unsigned) are supported in shader code. If this feature is not enabled, 16-bit integer types must not be used in shader code. This also specifies whether shader modules can declare the Int16 capability.

  • shaderResourceResidency specifies whether image operations that return resource residency information are supported in shader code. If this feature is not enabled, the OpImageSparse* instructions must not be used in shader code. This also specifies whether shader modules can declare the SparseResidency capability. The feature requires at least one of the sparseResidency* features to be supported.

  • shaderResourceMinLod specifies whether image operations that specify the minimum resource LOD are supported in shader code. If this feature is not enabled, the MinLod image operand must not be used in shader code. This also specifies whether shader modules can declare the MinLod capability.

  • sparseBinding specifies whether resource memory can be managed at opaque sparse block level instead of at the object level. If this feature is not enabled, resource memory must be bound only on a per-object basis using the vkBindBufferMemory and vkBindImageMemory commands. In this case, buffers and images must not be created with VK_BUFFER_CREATE_SPARSE_BINDING_BIT and VK_IMAGE_CREATE_SPARSE_BINDING_BIT set in the flags member of the VkBufferCreateInfo and VkImageCreateInfo structures, respectively. Otherwise resource memory can be managed as described in Sparse Resource Features.

  • sparseResidencyBuffer specifies whether the device can access partially resident buffers. If this feature is not enabled, buffers must not be created with VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkBufferCreateInfo structure.

  • sparseResidencyImage2D specifies whether the device can access partially resident 2D images with 1 sample per pixel. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_2D and samples set to VK_SAMPLE_COUNT_1_BIT must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidencyImage3D specifies whether the device can access partially resident 3D images. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_3D must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidency2Samples specifies whether the physical device can access partially resident 2D images with 2 samples per pixel. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_2D and samples set to VK_SAMPLE_COUNT_2_BIT must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidency4Samples specifies whether the physical device can access partially resident 2D images with 4 samples per pixel. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_2D and samples set to VK_SAMPLE_COUNT_4_BIT must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidency8Samples specifies whether the physical device can access partially resident 2D images with 8 samples per pixel. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_2D and samples set to VK_SAMPLE_COUNT_8_BIT must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidency16Samples specifies whether the physical device can access partially resident 2D images with 16 samples per pixel. If this feature is not enabled, images with an imageType of VK_IMAGE_TYPE_2D and samples set to VK_SAMPLE_COUNT_16_BIT must not be created with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT set in the flags member of the VkImageCreateInfo structure.

  • sparseResidencyAliased specifies whether the physical device can correctly access data aliased into multiple locations. If this feature is not enabled, the VK_BUFFER_CREATE_SPARSE_ALIASED_BIT and VK_IMAGE_CREATE_SPARSE_ALIASED_BIT enum values must not be used in flags members of the VkBufferCreateInfo and VkImageCreateInfo structures, respectively.

  • variableMultisampleRate specifies whether all pipelines that will be bound to a command buffer during a subpass with no attachments must have the same value for VkPipelineMultisampleStateCreateInfo::rasterizationSamples. If set to VK_TRUE, the implementation supports variable multisample rates in a subpass with no attachments. If set to VK_FALSE, then all pipelines bound in such a subpass must have the same multisample rate. This has no effect in situations where a subpass uses any attachments.

  • inheritedQueries specifies whether a secondary command buffer may be executed while a query is active.

The VkPhysicalDeviceVariablePointerFeatures structure is defined as:

typedef struct VkPhysicalDeviceVariablePointerFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           variablePointersStorageBuffer;
    VkBool32           variablePointers;
} VkPhysicalDeviceVariablePointerFeatures;

or the equivalent

typedef VkPhysicalDeviceVariablePointerFeatures VkPhysicalDeviceVariablePointerFeaturesKHR;

The members of the VkPhysicalDeviceVariablePointerFeatures structure describe the following features:

  • variablePointersStorageBuffer specifies whether the implementation supports the SPIR-V VariablePointersStorageBuffer capability. When this feature is not enabled, shader modules must not declare the SPV_KHR_variable_pointers extension or the VariablePointersStorageBuffer capability.

  • variablePointers specifies whether the implementation supports the SPIR-V VariablePointers capability. When this feature is not enabled, shader modules must not declare the VariablePointers capability.

If the VkPhysicalDeviceVariablePointerFeatures structure is included in the pNext chain of VkPhysicalDeviceFeatures2, it is filled with values indicating whether each feature is supported. VkPhysicalDeviceVariablePointerFeatures can also be used in the pNext chain of VkDeviceCreateInfo to enable the features.

Valid Usage
  • If variablePointers is enabled then variablePointersStorageBuffer must also be enabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES

The VkPhysicalDeviceMultiviewFeatures structure is defined as:

typedef struct VkPhysicalDeviceMultiviewFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           multiview;
    VkBool32           multiviewGeometryShader;
    VkBool32           multiviewTessellationShader;
} VkPhysicalDeviceMultiviewFeatures;

or the equivalent

typedef VkPhysicalDeviceMultiviewFeatures VkPhysicalDeviceMultiviewFeaturesKHR;

The members of the VkPhysicalDeviceMultiviewFeatures structure describe the following features:

  • multiview specifies whether the implementation supports multiview rendering within a render pass. If this feature is not enabled, the view mask of each subpass must always be zero.

  • multiviewGeometryShader specifies whether the implementation supports multiview rendering within a render pass, with geometry shaders. If this feature is not enabled, then a pipeline compiled against a subpass with a non-zero view mask must not include a geometry shader.

  • multiviewTessellationShader specifies whether the implementation supports multiview rendering within a render pass, with tessellation shaders. If this feature is not enabled, then a pipeline compiled against a subpass with a non-zero view mask must not include any tessellation shaders.

If the VkPhysicalDeviceMultiviewFeatures structure is included in the pNext chain of VkPhysicalDeviceFeatures2, it is filled with values indicating whether each feature is supported. VkPhysicalDeviceMultiviewFeatures can also be used in the pNext chain of VkDeviceCreateInfo to enable the features.

Valid Usage
  • If multiviewGeometryShader is enabled then multiview must also be enabled.

  • If multiviewTessellationShader is enabled then multiview must also be enabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES

To query 16-bit storage features additionally supported call vkGetPhysicalDeviceFeatures2 with a VkPhysicalDevice16BitStorageFeatures structure included in the pNext chain of its pFeatures parameter. The VkPhysicalDevice16BitStorageFeatures structure can also be in the pNext chain of a VkDeviceCreateInfo structure, in which case it controls which additional features are enabled in the device.

The VkPhysicalDevice16BitStorageFeatures structure is defined as:

typedef struct VkPhysicalDevice16BitStorageFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           storageBuffer16BitAccess;
    VkBool32           uniformAndStorageBuffer16BitAccess;
    VkBool32           storagePushConstant16;
    VkBool32           storageInputOutput16;
} VkPhysicalDevice16BitStorageFeatures;

or the equivalent

typedef VkPhysicalDevice16BitStorageFeatures VkPhysicalDevice16BitStorageFeaturesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • storageBuffer16BitAccess specifies whether objects in the StorageBuffer storage class with the Block decoration can have 16-bit integer and 16-bit floating-point members. If this feature is not enabled, 16-bit integer or 16-bit floating-point members must not be used in such objects. This also specifies whether shader modules can declare the StorageBuffer16BitAccess capability.

  • uniformAndStorageBuffer16BitAccess specifies whether objects in the Uniform storage class with the Block decoration and in the StorageBuffer storage class with the same decoration can have 16-bit integer and 16-bit floating-point members. If this feature is not enabled, 16-bit integer or 16-bit floating-point members must not be used in such objects. This also specifies whether shader modules can declare the UniformAndStorageBuffer16BitAccess capability.

  • storagePushConstant16 specifies whether objects in the PushConstant storage class can have 16-bit integer and 16-bit floating-point members. If this feature is not enabled, 16-bit integer or floating-point members must not be used in such objects. This also specifies whether shader modules can declare the StoragePushConstant16 capability.

  • storageInputOutput16 specifies whether objects in the Input and Output storage classes can have 16-bit integer and 16-bit floating-point members. If this feature is not enabled, 16-bit integer or 16-bit floating-point members must not be used in such objects. This also specifies whether shader modules can declare the StorageInputOutput16 capability.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES

The VkPhysicalDeviceSamplerYcbcrConversionFeatures structure is defined as:

typedef struct VkPhysicalDeviceSamplerYcbcrConversionFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           samplerYcbcrConversion;
} VkPhysicalDeviceSamplerYcbcrConversionFeatures;

or the equivalent

typedef VkPhysicalDeviceSamplerYcbcrConversionFeatures VkPhysicalDeviceSamplerYcbcrConversionFeaturesKHR;

The members of the VkPhysicalDeviceSamplerYcbcrConversionFeatures structure describe the following feature:

  • samplerYcbcrConversion specifies whether the implementation supports sampler Y’CBCR conversion. If samplerYcbcrConversion is VK_FALSE, sampler Y’CBCR conversion is not supported, and samplers using sampler Y’CBCR conversion must not be used.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES

The VkPhysicalDeviceProtectedMemoryFeatures structure is defined as:

typedef struct VkPhysicalDeviceProtectedMemoryFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           protectedMemory;
} VkPhysicalDeviceProtectedMemoryFeatures;
  • protectedMemory specifies whether protected memory is supported.

If the VkPhysicalDeviceProtectedMemoryFeatures structure is included in the pNext chain of VkPhysicalDeviceFeatures2, it is filled with a value indicating whether the feature is supported.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_FEATURES

The VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT structure is defined as:

typedef struct VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           advancedBlendCoherentOperations;
} VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT;

The members of the VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT structure describe the following features:

  • advancedBlendCoherentOperations specifies whether blending using advanced blend operations is guaranteed to execute atomically and in primitive order. If this is VK_TRUE, VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT is treated the same as VK_ACCESS_COLOR_ATTACHMENT_READ_BIT, and advanced blending needs no additional synchronization over basic blending. If this is VK_FALSE, then memory dependencies are required to guarantee order between two advanced blending operations that occur on the same sample.

If the VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT structure is included in the pNext chain of VkPhysicalDeviceFeatures2, it is filled with values indicating whether each feature is supported. VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT can also be used in pNext chain of VkDeviceCreateInfo to enable the features.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_FEATURES_EXT

The VkPhysicalDeviceShaderDrawParameterFeatures structure is defined as:

typedef struct VkPhysicalDeviceShaderDrawParameterFeatures {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           shaderDrawParameters;
} VkPhysicalDeviceShaderDrawParameterFeatures;
  • shaderDrawParameters specifies whether shader draw parameters are supported.

If the VkPhysicalDeviceShaderDrawParameterFeatures structure is included in the pNext chain of VkPhysicalDeviceFeatures2, it is filled with a value indicating whether the feature is supported.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_DRAW_PARAMETER_FEATURES

The VkPhysicalDeviceDescriptorIndexingFeaturesEXT structure is defined as:

typedef struct VkPhysicalDeviceDescriptorIndexingFeaturesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           shaderInputAttachmentArrayDynamicIndexing;
    VkBool32           shaderUniformTexelBufferArrayDynamicIndexing;
    VkBool32           shaderStorageTexelBufferArrayDynamicIndexing;
    VkBool32           shaderUniformBufferArrayNonUniformIndexing;
    VkBool32           shaderSampledImageArrayNonUniformIndexing;
    VkBool32           shaderStorageBufferArrayNonUniformIndexing;
    VkBool32           shaderStorageImageArrayNonUniformIndexing;
    VkBool32           shaderInputAttachmentArrayNonUniformIndexing;
    VkBool32           shaderUniformTexelBufferArrayNonUniformIndexing;
    VkBool32           shaderStorageTexelBufferArrayNonUniformIndexing;
    VkBool32           descriptorBindingUniformBufferUpdateAfterBind;
    VkBool32           descriptorBindingSampledImageUpdateAfterBind;
    VkBool32           descriptorBindingStorageImageUpdateAfterBind;
    VkBool32           descriptorBindingStorageBufferUpdateAfterBind;
    VkBool32           descriptorBindingUniformTexelBufferUpdateAfterBind;
    VkBool32           descriptorBindingStorageTexelBufferUpdateAfterBind;
    VkBool32           descriptorBindingUpdateUnusedWhilePending;
    VkBool32           descriptorBindingPartiallyBound;
    VkBool32           descriptorBindingVariableDescriptorCount;
    VkBool32           runtimeDescriptorArray;
} VkPhysicalDeviceDescriptorIndexingFeaturesEXT;

The members of the VkPhysicalDeviceDescriptorIndexingFeaturesEXT structure describe the following features:

  • shaderInputAttachmentArrayDynamicIndexing indicates whether arrays of input attachments can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the InputAttachmentArrayDynamicIndexingEXT capability.

  • shaderUniformTexelBufferArrayDynamicIndexing indicates whether arrays of uniform texel buffers can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the UniformTexelBufferArrayDynamicIndexingEXT capability.

  • shaderStorageTexelBufferArrayDynamicIndexing indicates whether arrays of storage texel buffers can be indexed by dynamically uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER must be indexed only by constant integral expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the StorageTexelBufferArrayDynamicIndexingEXT capability.

  • shaderUniformBufferArrayNonUniformIndexing indicates whether arrays of uniform buffers can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the UniformBufferArrayNonUniformIndexingEXT capability.

  • shaderSampledImageArrayNonUniformIndexing indicates whether arrays of samplers or sampled images can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, or VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the SampledImageArrayNonUniformIndexingEXT capability.

  • shaderStorageBufferArrayNonUniformIndexing indicates whether arrays of storage buffers can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the StorageBufferArrayNonUniformIndexingEXT capability.

  • shaderStorageImageArrayNonUniformIndexing indicates whether arrays of storage images can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the StorageImageArrayNonUniformIndexingEXT capability.

  • shaderInputAttachmentArrayNonUniformIndexing indicates whether arrays of input attachments can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the InputAttachmentArrayNonUniformIndexingEXT capability.

  • shaderUniformTexelBufferArrayNonUniformIndexing indicates whether arrays of uniform texel buffers can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the UniformTexelBufferArrayNonUniformIndexingEXT capability.

  • shaderStorageTexelBufferArrayNonUniformIndexing indicates whether arrays of storage texel buffers can be indexed by non-uniform integer expressions in shader code. If this feature is not enabled, resources with a descriptor type of VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER must not be indexed by non-uniform integer expressions when aggregated into arrays in shader code. This also indicates whether shader modules can declare the StorageTexelBufferArrayNonUniformIndexingEXT capability.

  • descriptorBindingUniformBufferUpdateAfterBind indicates whether the implementation supports updating uniform buffer descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER.

  • descriptorBindingSampledImageUpdateAfterBind indicates whether the implementation supports updating sampled image descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_SAMPLER, VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, or VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE.

  • descriptorBindingStorageImageUpdateAfterBind indicates whether the implementation supports updating storage image descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_STORAGE_IMAGE.

  • descriptorBindingStorageBufferUpdateAfterBind indicates whether the implementation supports updating storage buffer descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_STORAGE_BUFFER.

  • descriptorBindingUniformTexelBufferUpdateAfterBind indicates whether the implementation supports updating uniform texel buffer descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER.

  • descriptorBindingStorageTexelBufferUpdateAfterBind indicates whether the implementation supports updating storage texel buffer descriptors after a set is bound. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT must not be used with VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER.

  • descriptorBindingUpdateUnusedWhilePending indicates whether the implementation supports updating descriptors while the set is in use. If this feature is not enabled, VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT must not be used.

  • descriptorBindingPartiallyBound indicates whether the implementation supports statically using a descriptor set binding in which some descriptors are not valid. If this feature is not enabled, VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT must not be used.

  • descriptorBindingVariableDescriptorCount indicates whether the implementation supports descriptor sets with a variable-sized last binding. If this feature is not enabled, VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT must not be used.

  • runtimeDescriptorArray indicates whether the implementation supports the SPIR-V RuntimeDescriptorArrayEXT capability. If this feature is not enabled, descriptors must not be declared in runtime arrays.

If the VkPhysicalDeviceDescriptorIndexingFeaturesEXT structure is included in the pNext chain of VkPhysicalDeviceFeatures2KHR, it is filled with values indicating whether each feature is supported. VkPhysicalDeviceDescriptorIndexingFeaturesEXT can also be used in the pNext chain of VkDeviceCreateInfo to enable features.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_FEATURES_EXT

32.2. Limits

There are a variety of implementation-dependent limits.

The VkPhysicalDeviceLimits are properties of the physical device. These are available in the limits member of the VkPhysicalDeviceProperties structure which is returned from vkGetPhysicalDeviceProperties.

The VkPhysicalDeviceLimits structure is defined as:

typedef struct VkPhysicalDeviceLimits {
    uint32_t              maxImageDimension1D;
    uint32_t              maxImageDimension2D;
    uint32_t              maxImageDimension3D;
    uint32_t              maxImageDimensionCube;
    uint32_t              maxImageArrayLayers;
    uint32_t              maxTexelBufferElements;
    uint32_t              maxUniformBufferRange;
    uint32_t              maxStorageBufferRange;
    uint32_t              maxPushConstantsSize;
    uint32_t              maxMemoryAllocationCount;
    uint32_t              maxSamplerAllocationCount;
    VkDeviceSize          bufferImageGranularity;
    VkDeviceSize          sparseAddressSpaceSize;
    uint32_t              maxBoundDescriptorSets;
    uint32_t              maxPerStageDescriptorSamplers;
    uint32_t              maxPerStageDescriptorUniformBuffers;
    uint32_t              maxPerStageDescriptorStorageBuffers;
    uint32_t              maxPerStageDescriptorSampledImages;
    uint32_t              maxPerStageDescriptorStorageImages;
    uint32_t              maxPerStageDescriptorInputAttachments;
    uint32_t              maxPerStageResources;
    uint32_t              maxDescriptorSetSamplers;
    uint32_t              maxDescriptorSetUniformBuffers;
    uint32_t              maxDescriptorSetUniformBuffersDynamic;
    uint32_t              maxDescriptorSetStorageBuffers;
    uint32_t              maxDescriptorSetStorageBuffersDynamic;
    uint32_t              maxDescriptorSetSampledImages;
    uint32_t              maxDescriptorSetStorageImages;
    uint32_t              maxDescriptorSetInputAttachments;
    uint32_t              maxVertexInputAttributes;
    uint32_t              maxVertexInputBindings;
    uint32_t              maxVertexInputAttributeOffset;
    uint32_t              maxVertexInputBindingStride;
    uint32_t              maxVertexOutputComponents;
    uint32_t              maxTessellationGenerationLevel;
    uint32_t              maxTessellationPatchSize;
    uint32_t              maxTessellationControlPerVertexInputComponents;
    uint32_t              maxTessellationControlPerVertexOutputComponents;
    uint32_t              maxTessellationControlPerPatchOutputComponents;
    uint32_t              maxTessellationControlTotalOutputComponents;
    uint32_t              maxTessellationEvaluationInputComponents;
    uint32_t              maxTessellationEvaluationOutputComponents;
    uint32_t              maxGeometryShaderInvocations;
    uint32_t              maxGeometryInputComponents;
    uint32_t              maxGeometryOutputComponents;
    uint32_t              maxGeometryOutputVertices;
    uint32_t              maxGeometryTotalOutputComponents;
    uint32_t              maxFragmentInputComponents;
    uint32_t              maxFragmentOutputAttachments;
    uint32_t              maxFragmentDualSrcAttachments;
    uint32_t              maxFragmentCombinedOutputResources;
    uint32_t              maxComputeSharedMemorySize;
    uint32_t              maxComputeWorkGroupCount[3];
    uint32_t              maxComputeWorkGroupInvocations;
    uint32_t              maxComputeWorkGroupSize[3];
    uint32_t              subPixelPrecisionBits;
    uint32_t              subTexelPrecisionBits;
    uint32_t              mipmapPrecisionBits;
    uint32_t              maxDrawIndexedIndexValue;
    uint32_t              maxDrawIndirectCount;
    float                 maxSamplerLodBias;
    float                 maxSamplerAnisotropy;
    uint32_t              maxViewports;
    uint32_t              maxViewportDimensions[2];
    float                 viewportBoundsRange[2];
    uint32_t              viewportSubPixelBits;
    size_t                minMemoryMapAlignment;
    VkDeviceSize          minTexelBufferOffsetAlignment;
    VkDeviceSize          minUniformBufferOffsetAlignment;
    VkDeviceSize          minStorageBufferOffsetAlignment;
    int32_t               minTexelOffset;
    uint32_t              maxTexelOffset;
    int32_t               minTexelGatherOffset;
    uint32_t              maxTexelGatherOffset;
    float                 minInterpolationOffset;
    float                 maxInterpolationOffset;
    uint32_t              subPixelInterpolationOffsetBits;
    uint32_t              maxFramebufferWidth;
    uint32_t              maxFramebufferHeight;
    uint32_t              maxFramebufferLayers;
    VkSampleCountFlags    framebufferColorSampleCounts;
    VkSampleCountFlags    framebufferDepthSampleCounts;
    VkSampleCountFlags    framebufferStencilSampleCounts;
    VkSampleCountFlags    framebufferNoAttachmentsSampleCounts;
    uint32_t              maxColorAttachments;
    VkSampleCountFlags    sampledImageColorSampleCounts;
    VkSampleCountFlags    sampledImageIntegerSampleCounts;
    VkSampleCountFlags    sampledImageDepthSampleCounts;
    VkSampleCountFlags    sampledImageStencilSampleCounts;
    VkSampleCountFlags    storageImageSampleCounts;
    uint32_t              maxSampleMaskWords;
    VkBool32              timestampComputeAndGraphics;
    float                 timestampPeriod;
    uint32_t              maxClipDistances;
    uint32_t              maxCullDistances;
    uint32_t              maxCombinedClipAndCullDistances;
    uint32_t              discreteQueuePriorities;
    float                 pointSizeRange[2];
    float                 lineWidthRange[2];
    float                 pointSizeGranularity;
    float                 lineWidthGranularity;
    VkBool32              strictLines;
    VkBool32              standardSampleLocations;
    VkDeviceSize          optimalBufferCopyOffsetAlignment;
    VkDeviceSize          optimalBufferCopyRowPitchAlignment;
    VkDeviceSize          nonCoherentAtomSize;
} VkPhysicalDeviceLimits;
  • maxImageDimension1D is the maximum dimension (width) supported for all images created with an imageType of VK_IMAGE_TYPE_1D.

  • maxImageDimension2D is the maximum dimension (width or height) supported for all images created with an imageType of VK_IMAGE_TYPE_2D and without VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT set in flags.

  • maxImageDimension3D is the maximum dimension (width, height, or depth) supported for all images created with an imageType of VK_IMAGE_TYPE_3D.

  • maxImageDimensionCube is the maximum dimension (width or height) supported for all images created with an imageType of VK_IMAGE_TYPE_2D and with VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT set in flags.

  • maxImageArrayLayers is the maximum number of layers (arrayLayers) for an image.

  • maxTexelBufferElements is the maximum number of addressable texels for a buffer view created on a buffer which was created with the VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT set in the usage member of the VkBufferCreateInfo structure.

  • maxUniformBufferRange is the maximum value that can be specified in the range member of any VkDescriptorBufferInfo structures passed to a call to vkUpdateDescriptorSets for descriptors of type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC.

  • maxStorageBufferRange is the maximum value that can be specified in the range member of any VkDescriptorBufferInfo structures passed to a call to vkUpdateDescriptorSets for descriptors of type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC.

  • maxPushConstantsSize is the maximum size, in bytes, of the pool of push constant memory. For each of the push constant ranges indicated by the pPushConstantRanges member of the VkPipelineLayoutCreateInfo structure, (offset + size) must be less than or equal to this limit.

  • maxMemoryAllocationCount is the maximum number of device memory allocations, as created by vkAllocateMemory, which can simultaneously exist.

  • maxSamplerAllocationCount is the maximum number of sampler objects, as created by vkCreateSampler, which can simultaneously exist on a device.

  • bufferImageGranularity is the granularity, in bytes, at which buffer or linear image resources, and optimal image resources can be bound to adjacent offsets in the same VkDeviceMemory object without aliasing. See Buffer-Image Granularity for more details.

  • sparseAddressSpaceSize is the total amount of address space available, in bytes, for sparse memory resources. This is an upper bound on the sum of the size of all sparse resources, regardless of whether any memory is bound to them.

  • maxBoundDescriptorSets is the maximum number of descriptor sets that can be simultaneously used by a pipeline. All DescriptorSet decorations in shader modules must have a value less than maxBoundDescriptorSets. See Descriptor Sets.

  • maxPerStageDescriptorSamplers is the maximum number of samplers that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. See Sampler and Combined Image Sampler.

  • maxPerStageDescriptorUniformBuffers is the maximum number of uniform buffers that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. See Uniform Buffer and Dynamic Uniform Buffer.

  • maxPerStageDescriptorStorageBuffers is the maximum number of storage buffers that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a pipeline shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. See Storage Buffer and Dynamic Storage Buffer.

  • maxPerStageDescriptorSampledImages is the maximum number of sampled images that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, or VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a pipeline shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. See Combined Image Sampler, Sampled Image, and Uniform Texel Buffer.

  • maxPerStageDescriptorStorageImages is the maximum number of storage images that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a pipeline shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. See Storage Image, and Storage Texel Buffer.

  • maxPerStageDescriptorInputAttachments is the maximum number of input attachments that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. A descriptor is accessible to a pipeline shader stage when the stageFlags member of the VkDescriptorSetLayoutBinding structure has the bit for that shader stage set. These are only supported for the fragment stage. See Input Attachment.

  • maxPerStageResources is the maximum number of resources that can be accessible to a single shader stage in a pipeline layout. Descriptors with a type of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER, VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER, VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC, or VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. For the fragment shader stage the framebuffer color attachments also count against this limit.

  • maxDescriptorSetSamplers is the maximum number of samplers that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_SAMPLER or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Sampler and Combined Image Sampler.

  • maxDescriptorSetUniformBuffers is the maximum number of uniform buffers that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Uniform Buffer and Dynamic Uniform Buffer.

  • maxDescriptorSetUniformBuffersDynamic is the maximum number of dynamic uniform buffers that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Dynamic Uniform Buffer.

  • maxDescriptorSetStorageBuffers is the maximum number of storage buffers that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Storage Buffer and Dynamic Storage Buffer.

  • maxDescriptorSetStorageBuffersDynamic is the maximum number of dynamic storage buffers that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Dynamic Storage Buffer.

  • maxDescriptorSetSampledImages is the maximum number of sampled images that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, or VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Combined Image Sampler, Sampled Image, and Uniform Texel Buffer.

  • maxDescriptorSetStorageImages is the maximum number of storage images that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Storage Image, and Storage Texel Buffer.

  • maxDescriptorSetInputAttachments is the maximum number of input attachments that can be included in descriptor bindings in a pipeline layout across all pipeline shader stages and descriptor set numbers. Descriptors with a type of VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT count against this limit. Only descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set count against this limit. See Input Attachment.

  • maxVertexInputAttributes is the maximum number of vertex input attributes that can be specified for a graphics pipeline. These are described in the array of VkVertexInputAttributeDescription structures that are provided at graphics pipeline creation time via the pVertexAttributeDescriptions member of the VkPipelineVertexInputStateCreateInfo structure. See Vertex Attributes and Vertex Input Description.

  • maxVertexInputBindings is the maximum number of vertex buffers that can be specified for providing vertex attributes to a graphics pipeline. These are described in the array of VkVertexInputBindingDescription structures that are provided at graphics pipeline creation time via the pVertexBindingDescriptions member of the VkPipelineVertexInputStateCreateInfo structure. The binding member of VkVertexInputBindingDescription must be less than this limit. See Vertex Input Description.

  • maxVertexInputAttributeOffset is the maximum vertex input attribute offset that can be added to the vertex input binding stride. The offset member of the VkVertexInputAttributeDescription structure must be less than or equal to this limit. See Vertex Input Description.

  • maxVertexInputBindingStride is the maximum vertex input binding stride that can be specified in a vertex input binding. The stride member of the VkVertexInputBindingDescription structure must be less than or equal to this limit. See Vertex Input Description.

  • maxVertexOutputComponents is the maximum number of components of output variables which can be output by a vertex shader. See Vertex Shaders.

  • maxTessellationGenerationLevel is the maximum tessellation generation level supported by the fixed-function tessellation primitive generator. See Tessellation.

  • maxTessellationPatchSize is the maximum patch size, in vertices, of patches that can be processed by the tessellation control shader and tessellation primitive generator. The patchControlPoints member of the VkPipelineTessellationStateCreateInfo structure specified at pipeline creation time and the value provided in the OutputVertices execution mode of shader modules must be less than or equal to this limit. See Tessellation.

  • maxTessellationControlPerVertexInputComponents is the maximum number of components of input variables which can be provided as per-vertex inputs to the tessellation control shader stage.

  • maxTessellationControlPerVertexOutputComponents is the maximum number of components of per-vertex output variables which can be output from the tessellation control shader stage.

  • maxTessellationControlPerPatchOutputComponents is the maximum number of components of per-patch output variables which can be output from the tessellation control shader stage.

  • maxTessellationControlTotalOutputComponents is the maximum total number of components of per-vertex and per-patch output variables which can be output from the tessellation control shader stage.

  • maxTessellationEvaluationInputComponents is the maximum number of components of input variables which can be provided as per-vertex inputs to the tessellation evaluation shader stage.

  • maxTessellationEvaluationOutputComponents is the maximum number of components of per-vertex output variables which can be output from the tessellation evaluation shader stage.

  • maxGeometryShaderInvocations is the maximum invocation count supported for instanced geometry shaders. The value provided in the Invocations execution mode of shader modules must be less than or equal to this limit. See Geometry Shading.

  • maxGeometryInputComponents is the maximum number of components of input variables which can be provided as inputs to the geometry shader stage.

  • maxGeometryOutputComponents is the maximum number of components of output variables which can be output from the geometry shader stage.

  • maxGeometryOutputVertices is the maximum number of vertices which can be emitted by any geometry shader.

  • maxGeometryTotalOutputComponents is the maximum total number of components of output, across all emitted vertices, which can be output from the geometry shader stage.

  • maxFragmentInputComponents is the maximum number of components of input variables which can be provided as inputs to the fragment shader stage.

  • maxFragmentOutputAttachments is the maximum number of output attachments which can be written to by the fragment shader stage.

  • maxFragmentDualSrcAttachments is the maximum number of output attachments which can be written to by the fragment shader stage when blending is enabled and one of the dual source blend modes is in use. See Dual-Source Blending and dualSrcBlend.

  • maxFragmentCombinedOutputResources is the total number of storage buffers, storage images, and output buffers which can be used in the fragment shader stage.

  • maxComputeSharedMemorySize is the maximum total storage size, in bytes, of all variables declared with the WorkgroupLocal storage class in shader modules (or with the shared storage qualifier in GLSL) in the compute shader stage.

  • maxComputeWorkGroupCount[3] is the maximum number of local workgroups that can be dispatched by a single dispatch command. These three values represent the maximum number of local workgroups for the X, Y, and Z dimensions, respectively. The workgroup count parameters to the dispatch commands must be less than or equal to the corresponding limit. See Dispatching Commands.

  • maxComputeWorkGroupInvocations is the maximum total number of compute shader invocations in a single local workgroup. The product of the X, Y, and Z sizes as specified by the LocalSize execution mode in shader modules and by the object decorated by the WorkgroupSize decoration must be less than or equal to this limit.

  • maxComputeWorkGroupSize[3] is the maximum size of a local compute workgroup, per dimension. These three values represent the maximum local workgroup size in the X, Y, and Z dimensions, respectively. The x, y, and z sizes specified by the LocalSize execution mode and by the object decorated by the WorkgroupSize decoration in shader modules must be less than or equal to the corresponding limit.

  • subPixelPrecisionBits is the number of bits of subpixel precision in framebuffer coordinates xf and yf. See Rasterization.

  • subTexelPrecisionBits is the number of bits of precision in the division along an axis of an image used for minification and magnification filters. 2subTexelPrecisionBits is the actual number of divisions along each axis of the image represented. Sub-texel values calculated during image sampling will snap to these locations when generating the filtered results.

  • mipmapPrecisionBits is the number of bits of division that the LOD calculation for mipmap fetching get snapped to when determining the contribution from each mip level to the mip filtered results. 2mipmapPrecisionBits is the actual number of divisions.

    Note

    For example, if this value is 2 bits then when linearly filtering between two levels, each level could: contribute: 0%, 33%, 66%, or 100% (this is just an example and the amount of contribution should be covered by different equations in the spec).

  • maxDrawIndexedIndexValue is the maximum index value that can be used for indexed draw calls when using 32-bit indices. This excludes the primitive restart index value of 0xFFFFFFFF. See fullDrawIndexUint32.

  • maxDrawIndirectCount is the maximum draw count that is supported for indirect draw calls. See multiDrawIndirect.

  • maxSamplerLodBias is the maximum absolute sampler LOD bias. The sum of the mipLodBias member of the VkSamplerCreateInfo structure and the Bias operand of image sampling operations in shader modules (or 0 if no Bias operand is provided to an image sampling operation) are clamped to the range [-maxSamplerLodBias,+maxSamplerLodBias]. See [samplers-mipLodBias].

  • maxSamplerAnisotropy is the maximum degree of sampler anisotropy. The maximum degree of anisotropic filtering used for an image sampling operation is the minimum of the maxAnisotropy member of the VkSamplerCreateInfo structure and this limit. See [samplers-maxAnisotropy].

  • maxViewports is the maximum number of active viewports. The viewportCount member of the VkPipelineViewportStateCreateInfo structure that is provided at pipeline creation must be less than or equal to this limit.

  • maxViewportDimensions[2] are the maximum viewport dimensions in the X (width) and Y (height) dimensions, respectively. The maximum viewport dimensions must be greater than or equal to the largest image which can be created and used as a framebuffer attachment. See Controlling the Viewport.

  • viewportBoundsRange[2] is the [minimum, maximum] range that the corners of a viewport must be contained in. This range must be at least [-2 × size, 2 × size - 1], where size = max(maxViewportDimensions[0], maxViewportDimensions[1]). See Controlling the Viewport.

    Note

    The intent of the viewportBoundsRange limit is to allow a maximum sized viewport to be arbitrarily shifted relative to the output target as long as at least some portion intersects. This would give a bounds limit of [-size + 1, 2 × size - 1] which would allow all possible non-empty-set intersections of the output target and the viewport. Since these numbers are typically powers of two, picking the signed number range using the smallest possible number of bits ends up with the specified range.

  • viewportSubPixelBits is the number of bits of subpixel precision for viewport bounds. The subpixel precision that floating-point viewport bounds are interpreted at is given by this limit.

  • minMemoryMapAlignment is the minimum required alignment, in bytes, of host visible memory allocations within the host address space. When mapping a memory allocation with vkMapMemory, subtracting offset bytes from the returned pointer will always produce an integer multiple of this limit. See Host Access to Device Memory Objects.

  • minTexelBufferOffsetAlignment is the minimum required alignment, in bytes, for the offset member of the VkBufferViewCreateInfo structure for texel buffers. When a buffer view is created for a buffer which was created with VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT set in the usage member of the VkBufferCreateInfo structure, the offset must be an integer multiple of this limit.

  • minUniformBufferOffsetAlignment is the minimum required alignment, in bytes, for the offset member of the VkDescriptorBufferInfo structure for uniform buffers. When a descriptor of type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER or VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC is updated, the offset must be an integer multiple of this limit. Similarly, dynamic offsets for uniform buffers must be multiples of this limit.

  • minStorageBufferOffsetAlignment is the minimum required alignment, in bytes, for the offset member of the VkDescriptorBufferInfo structure for storage buffers. When a descriptor of type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC is updated, the offset must be an integer multiple of this limit. Similarly, dynamic offsets for storage buffers must be multiples of this limit.

  • minTexelOffset is the minimum offset value for the ConstOffset image operand of any of the OpImageSample* or OpImageFetch* image instructions.

  • maxTexelOffset is the maximum offset value for the ConstOffset image operand of any of the OpImageSample* or OpImageFetch* image instructions.

  • minTexelGatherOffset is the minimum offset value for the Offset or ConstOffsets image operands of any of the OpImage*Gather image instructions.

  • maxTexelGatherOffset is the maximum offset value for the Offset or ConstOffsets image operands of any of the OpImage*Gather image instructions.

  • minInterpolationOffset is the minimum negative offset value for the offset operand of the InterpolateAtOffset extended instruction.

  • maxInterpolationOffset is the maximum positive offset value for the offset operand of the InterpolateAtOffset extended instruction.

  • subPixelInterpolationOffsetBits is the number of subpixel fractional bits that the x and y offsets to the InterpolateAtOffset extended instruction may be rounded to as fixed-point values.

  • maxFramebufferWidth is the maximum width for a framebuffer. The width member of the VkFramebufferCreateInfo structure must be less than or equal to this limit.

  • maxFramebufferHeight is the maximum height for a framebuffer. The height member of the VkFramebufferCreateInfo structure must be less than or equal to this limit.

  • maxFramebufferLayers is the maximum layer count for a layered framebuffer. The layers member of the VkFramebufferCreateInfo structure must be less than or equal to this limit.

  • framebufferColorSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the color sample counts that are supported for all framebuffer color attachments with floating- or fixed-point formats. There is no limit that specifies the color sample counts that are supported for all color attachments with integer formats.

  • framebufferDepthSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the supported depth sample counts for all framebuffer depth/stencil attachments, when the format includes a depth component.

  • framebufferStencilSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the supported stencil sample counts for all framebuffer depth/stencil attachments, when the format includes a stencil component.

  • framebufferNoAttachmentsSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the supported sample counts for a framebuffer with no attachments.

  • maxColorAttachments is the maximum number of color attachments that can be used by a subpass in a render pass. The colorAttachmentCount member of the VkSubpassDescription structure must be less than or equal to this limit.

  • sampledImageColorSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the sample counts supported for all 2D images created with VK_IMAGE_TILING_OPTIMAL, usage containing VK_IMAGE_USAGE_SAMPLED_BIT, and a non-integer color format.

  • sampledImageIntegerSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the sample counts supported for all 2D images created with VK_IMAGE_TILING_OPTIMAL, usage containing VK_IMAGE_USAGE_SAMPLED_BIT, and an integer color format.

  • sampledImageDepthSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the sample counts supported for all 2D images created with VK_IMAGE_TILING_OPTIMAL, usage containing VK_IMAGE_USAGE_SAMPLED_BIT, and a depth format.

  • sampledImageStencilSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the sample supported for all 2D images created with VK_IMAGE_TILING_OPTIMAL, usage containing VK_IMAGE_USAGE_SAMPLED_BIT, and a stencil format.

  • storageImageSampleCounts is a bitmask1 of VkSampleCountFlagBits indicating the sample counts supported for all 2D images created with VK_IMAGE_TILING_OPTIMAL, and usage containing VK_IMAGE_USAGE_STORAGE_BIT.

  • maxSampleMaskWords is the maximum number of array elements of a variable decorated with the SampleMask built-in decoration.

  • timestampComputeAndGraphics specifies support for timestamps on all graphics and compute queues. If this limit is set to VK_TRUE, all queues that advertise the VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT in the VkQueueFamilyProperties::queueFlags support VkQueueFamilyProperties::timestampValidBits of at least 36. See Timestamp Queries.

  • timestampPeriod is the number of nanoseconds required for a timestamp query to be incremented by 1. See Timestamp Queries.

  • maxClipDistances is the maximum number of clip distances that can be used in a single shader stage. The size of any array declared with the ClipDistance built-in decoration in a shader module must be less than or equal to this limit.

  • maxCullDistances is the maximum number of cull distances that can be used in a single shader stage. The size of any array declared with the CullDistance built-in decoration in a shader module must be less than or equal to this limit.

  • maxCombinedClipAndCullDistances is the maximum combined number of clip and cull distances that can be used in a single shader stage. The sum of the sizes of any pair of arrays declared with the ClipDistance and CullDistance built-in decoration used by a single shader stage in a shader module must be less than or equal to this limit.

  • discreteQueuePriorities is the number of discrete priorities that can be assigned to a queue based on the value of each member of VkDeviceQueueCreateInfo::pQueuePriorities. This must be at least 2, and levels must be spread evenly over the range, with at least one level at 1.0, and another at 0.0. See Queue Priority.

  • pointSizeRange[2] is the range [minimum,maximum] of supported sizes for points. Values written to variables decorated with the PointSize built-in decoration are clamped to this range.

  • lineWidthRange[2] is the range [minimum,maximum] of supported widths for lines. Values specified by the lineWidth member of the VkPipelineRasterizationStateCreateInfo or the lineWidth parameter to vkCmdSetLineWidth are clamped to this range.

  • pointSizeGranularity is the granularity of supported point sizes. Not all point sizes in the range defined by pointSizeRange are supported. This limit specifies the granularity (or increment) between successive supported point sizes.

  • lineWidthGranularity is the granularity of supported line widths. Not all line widths in the range defined by lineWidthRange are supported. This limit specifies the granularity (or increment) between successive supported line widths.

  • strictLines specifies whether lines are rasterized according to the preferred method of rasterization. If set to VK_FALSE, lines may be rasterized under a relaxed set of rules. If set to VK_TRUE, lines are rasterized as per the strict definition. See Basic Line Segment Rasterization.

  • standardSampleLocations specifies whether rasterization uses the standard sample locations as documented in Multisampling. If set to VK_TRUE, the implementation uses the documented sample locations. If set to VK_FALSE, the implementation may use different sample locations.

  • optimalBufferCopyOffsetAlignment is the optimal buffer offset alignment in bytes for vkCmdCopyBufferToImage and vkCmdCopyImageToBuffer. The per texel alignment requirements are enforced, but applications should use the optimal alignment for optimal performance and power use.

  • optimalBufferCopyRowPitchAlignment is the optimal buffer row pitch alignment in bytes for vkCmdCopyBufferToImage and vkCmdCopyImageToBuffer. Row pitch is the number of bytes between texels with the same X coordinate in adjacent rows (Y coordinates differ by one). The per texel alignment requirements are enforced, but applications should use the optimal alignment for optimal performance and power use.

  • nonCoherentAtomSize is the size and alignment in bytes that bounds concurrent access to host-mapped device memory.

  • VkPhysicalDeviceDiscardRectanglePropertiesEXT::maxDiscardRectangles is the maximum number of active discard rectangles. This limit can be queried by setting the pNext pointer from a VkPhysicalDeviceProperties2 object to an instance of VkPhysicalDeviceDiscardRectanglePropertiesEXT and using vkGetPhysicalDeviceProperties2 to fill out the members.

  • VkPhysicalDevicePointClippingProperties::pointClippingBehavior defines the clipping behavior of points. This limit can be queried by setting the pNext pointer from a VkPhysicalDeviceProperties2 object to an instance of VkPhysicalDevicePointClippingProperties and using vkGetPhysicalDeviceProperties2 to fill out the members.

  • VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT::maxVertexAttribDivisor is the maximum value of the number of instances that will repeat the value of vertex attribute data when instanced rendering is enabled. This limit can be queried by setting the pNext pointer from a VkPhysicalDeviceProperties2 object to an instance of VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT and using vkGetPhysicalDeviceProperties2 to fill out the members.

1

For all bitmasks of VkSampleCountFlagBits, the sample count limits defined above represent the minimum supported sample counts for each image type. Individual images may support additional sample counts, which are queried using vkGetPhysicalDeviceImageFormatProperties as described in Supported Sample Counts.

Bits which may be set in the sample count limits returned by VkPhysicalDeviceLimits, as well as in other queries and structures representing image sample counts, are:

typedef enum VkSampleCountFlagBits {
    VK_SAMPLE_COUNT_1_BIT = 0x00000001,
    VK_SAMPLE_COUNT_2_BIT = 0x00000002,
    VK_SAMPLE_COUNT_4_BIT = 0x00000004,
    VK_SAMPLE_COUNT_8_BIT = 0x00000008,
    VK_SAMPLE_COUNT_16_BIT = 0x00000010,
    VK_SAMPLE_COUNT_32_BIT = 0x00000020,
    VK_SAMPLE_COUNT_64_BIT = 0x00000040,
} VkSampleCountFlagBits;
  • VK_SAMPLE_COUNT_1_BIT specifies an image with one sample per pixel.

  • VK_SAMPLE_COUNT_2_BIT specifies an image with 2 samples per pixel.

  • VK_SAMPLE_COUNT_4_BIT specifies an image with 4 samples per pixel.

  • VK_SAMPLE_COUNT_8_BIT specifies an image with 8 samples per pixel.

  • VK_SAMPLE_COUNT_16_BIT specifies an image with 16 samples per pixel.

  • VK_SAMPLE_COUNT_32_BIT specifies an image with 32 samples per pixel.

  • VK_SAMPLE_COUNT_64_BIT specifies an image with 64 samples per pixel.

typedef VkFlags VkSampleCountFlags;

VkSampleCountFlags is a bitmask type for setting a mask of zero or more VkSampleCountFlagBits.

The VkPhysicalDevicePushDescriptorPropertiesKHR structure is defined as:

typedef struct VkPhysicalDevicePushDescriptorPropertiesKHR {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxPushDescriptors;
} VkPhysicalDevicePushDescriptorPropertiesKHR;

The members of the VkPhysicalDevicePushDescriptorPropertiesKHR structure describe the following implementation-dependent limits:

  • maxPushDescriptors is the maximum number of descriptors that can be used in a descriptor set created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR set.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PUSH_DESCRIPTOR_PROPERTIES_KHR

The VkPhysicalDeviceMultiviewProperties structure is defined as:

typedef struct VkPhysicalDeviceMultiviewProperties {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxMultiviewViewCount;
    uint32_t           maxMultiviewInstanceIndex;
} VkPhysicalDeviceMultiviewProperties;

or the equivalent

typedef VkPhysicalDeviceMultiviewProperties VkPhysicalDeviceMultiviewPropertiesKHR;

The members of the VkPhysicalDeviceMultiviewProperties structure describe the following implementation-dependent limits:

  • maxMultiviewViewCount is one greater than the maximum view index that can be used in a subpass.

  • maxMultiviewInstanceIndex is the maximum valid value of instance index allowed to be generated by a drawing command recorded within a subpass of a multiview render pass instance.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES

If the VkPhysicalDeviceMultiviewProperties structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

The VkPhysicalDeviceDiscardRectanglePropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceDiscardRectanglePropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxDiscardRectangles;
} VkPhysicalDeviceDiscardRectanglePropertiesEXT;

The members of the VkPhysicalDeviceDiscardRectanglePropertiesEXT structure describe the following implementation-dependent limits:

  • maxDiscardRectangles is the maximum number of discard rectangles that can be specified.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DISCARD_RECTANGLE_PROPERTIES_EXT

If the VkPhysicalDeviceDiscardRectanglePropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

The VkPhysicalDeviceSampleLocationsPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceSampleLocationsPropertiesEXT {
    VkStructureType       sType;
    void*                 pNext;
    VkSampleCountFlags    sampleLocationSampleCounts;
    VkExtent2D            maxSampleLocationGridSize;
    float                 sampleLocationCoordinateRange[2];
    uint32_t              sampleLocationSubPixelBits;
    VkBool32              variableSampleLocations;
} VkPhysicalDeviceSampleLocationsPropertiesEXT;

The members of the VkPhysicalDeviceSampleLocationsPropertiesEXT structure describe the following implementation-dependent limits:

  • sampleLocationSampleCounts is a bitmask of VkSampleCountFlagBits indicating the sample counts supporting custom sample locations.

  • maxSampleLocationGridSize is the maximum size of the pixel grid in which sample locations can vary that is supported for all sample counts in sampleLocationSampleCounts.

  • sampleLocationCoordinateRange[2] is the range of supported sample location coordinates.

  • sampleLocationSubPixelBits is the number of bits of subpixel precision for sample locations.

  • variableSampleLocations specifies whether the sample locations used by all pipelines that will be bound to a command buffer during a subpass must match. If set to VK_TRUE, the implementation supports variable sample locations in a subpass. If set to VK_FALSE, then the sample locations must stay constant in each subpass.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLE_LOCATIONS_PROPERTIES_EXT

If the VkPhysicalDeviceSampleLocationsPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

The VkPhysicalDeviceExternalMemoryHostPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceExternalMemoryHostPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkDeviceSize       minImportedHostPointerAlignment;
} VkPhysicalDeviceExternalMemoryHostPropertiesEXT;

The members of the VkPhysicalDeviceExternalMemoryHostPropertiesEXT structure describe the following implementation-dependent limits:

  • minImportedHostPointerAlignment is the minimum required alignment, in bytes, for the base address and size of host pointers that can be imported to a Vulkan memory object.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_MEMORY_HOST_PROPERTIES_EXT

If the VkPhysicalDeviceExternalMemoryHostPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2KHR, it is filled with the implementation-dependent limits.

The VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX structure is defined as:

typedef struct VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           perViewPositionAllComponents;
} VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX;

The members of the VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX structure describe the following implementation-dependent limits:

  • perViewPositionAllComponents is VK_TRUE if the implementation supports per-view position values that differ in components other than the X component.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PER_VIEW_ATTRIBUTES_PROPERTIES_NVX

If the VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

The VkPhysicalDevicePointClippingProperties structure is defined as:

typedef struct VkPhysicalDevicePointClippingProperties {
    VkStructureType            sType;
    void*                      pNext;
    VkPointClippingBehavior    pointClippingBehavior;
} VkPhysicalDevicePointClippingProperties;

or the equivalent

typedef VkPhysicalDevicePointClippingProperties VkPhysicalDevicePointClippingPropertiesKHR;

The members of the VkPhysicalDevicePointClippingProperties structure describe the following implementation-dependent limit:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pointClippingBehavior is the point clipping behavior supported by the implementation, and is of type VkPointClippingBehavior.

If the VkPhysicalDevicePointClippingProperties structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES

The VkPhysicalDeviceSubgroupProperties structure is defined as:

typedef struct VkPhysicalDeviceSubgroupProperties {
    VkStructureType           sType;
    void*                     pNext;
    uint32_t                  subgroupSize;
    VkShaderStageFlags        supportedStages;
    VkSubgroupFeatureFlags    supportedOperations;
    VkBool32                  quadOperationsInAllStages;
} VkPhysicalDeviceSubgroupProperties;

The members of the VkPhysicalDeviceSubgroupProperties structure describe the following implementation-dependent limits:

  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • subgroupSize is the number of invocations in each subgroup. This will match any SubgroupSize decorated variable used in any shader module created on this device. subgroupSize is at least 1 if any of the physical device’s queues support VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT.

  • supportedStages is a bitfield of VkShaderStageFlagBits describing the shader stages that subgroup operations are supported in. supportedStages will have the VK_SHADER_STAGE_COMPUTE_BIT bit set if any of any of the physical device’s queues support VK_QUEUE_COMPUTE_BIT.

  • supportedOperations is a bitmask of VkSubgroupFeatureFlagBits specifying the sets of subgroup operations supported on this device. supportedOperations will have the VK_SUBGROUP_FEATURE_BASIC_BIT bit set if any of the physical device’s queues support VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT.

  • quadOperationsInAllStages is a boolean that specifies whether quad subgroup operations are available in all stages, or are restricted to fragment and compute stages.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SUBGROUP_PROPERTIES

If the VkPhysicalDeviceSubgroupProperties structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

Bits which can be set in VkPhysicalDeviceSubgroupProperties::supportedOperations to specify supported subgroup operations are:

typedef enum VkSubgroupFeatureFlagBits {
    VK_SUBGROUP_FEATURE_BASIC_BIT = 0x00000001,
    VK_SUBGROUP_FEATURE_VOTE_BIT = 0x00000002,
    VK_SUBGROUP_FEATURE_ARITHMETIC_BIT = 0x00000004,
    VK_SUBGROUP_FEATURE_BALLOT_BIT = 0x00000008,
    VK_SUBGROUP_FEATURE_SHUFFLE_BIT = 0x00000010,
    VK_SUBGROUP_FEATURE_SHUFFLE_RELATIVE_BIT = 0x00000020,
    VK_SUBGROUP_FEATURE_CLUSTERED_BIT = 0x00000040,
    VK_SUBGROUP_FEATURE_QUAD_BIT = 0x00000080,
    VK_SUBGROUP_FEATURE_PARTITIONED_BIT_NV = 0x00000100,
} VkSubgroupFeatureFlagBits;
  • VK_SUBGROUP_FEATURE_BASIC_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniform capability.

  • VK_SUBGROUP_FEATURE_VOTE_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformVote capability.

  • VK_SUBGROUP_FEATURE_ARITHMETIC_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformArithmetic capability.

  • VK_SUBGROUP_FEATURE_BALLOT_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformBallot capability.

  • VK_SUBGROUP_FEATURE_SHUFFLE_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformShuffle capability.

  • VK_SUBGROUP_FEATURE_SHUFFLE_RELATIVE_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformShuffleRelative capability.

  • VK_SUBGROUP_FEATURE_CLUSTERED_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformClustered capability.

  • VK_SUBGROUP_FEATURE_QUAD_BIT specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformQuad capability.

  • VK_SUBGROUP_FEATURE_PARTITIONED_BIT_NV specifies the device will accept SPIR-V shader modules that contain the GroupNonUniformPartitionedNV capability.

typedef VkFlags VkSubgroupFeatureFlags;

VkSubgroupFeatureFlags is a bitmask type for setting a mask of zero or more VkSubgroupFeatureFlagBits.

The VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           advancedBlendMaxColorAttachments;
    VkBool32           advancedBlendIndependentBlend;
    VkBool32           advancedBlendNonPremultipliedSrcColor;
    VkBool32           advancedBlendNonPremultipliedDstColor;
    VkBool32           advancedBlendCorrelatedOverlap;
    VkBool32           advancedBlendAllOperations;
} VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT;

The members of the VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT structure describe the following implementation-dependent limits:

  • advancedBlendMaxColorAttachments is one greater than the highest color attachment index that can be used in a subpass, for a pipeline that uses an advanced blend operation.

  • advancedBlendIndependentBlend specifies whether advanced blend operations can vary per-attachment.

  • advancedBlendNonPremultipliedSrcColor specifies whether the source color can be treated as non-premultiplied. If this is VK_FALSE, then VkPipelineColorBlendAdvancedStateCreateInfoEXT::srcPremultiplied must be VK_TRUE.

  • advancedBlendNonPremultipliedDstColor specifies whether the destination color can be treated as non-premultiplied. If this is VK_FALSE, then VkPipelineColorBlendAdvancedStateCreateInfoEXT::dstPremultiplied must be VK_TRUE.

  • advancedBlendCorrelatedOverlap specifies whether the overlap mode can be treated as correlated. If this is VK_FALSE, then VkPipelineColorBlendAdvancedStateCreateInfoEXT::blendOverlap must be VK_BLEND_OVERLAP_UNCORRELATED_EXT.

  • advancedBlendAllOperations specifies whether all advanced blend operation enums are supported. See the valid usage of VkPipelineColorBlendAttachmentState.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_PROPERTIES_EXT

If the VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

The VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxVertexAttribDivisor;
} VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT;

The members of the VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT structure describe the following implementation-dependent limits:

  • maxVertexAttribDivisor is the maximum value of the number of instances that will repeat the value of vertex attribute data when instanced rendering is enabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VERTEX_ATTRIBUTE_DIVISOR_PROPERTIES_EXT

The VkPhysicalDeviceSamplerFilterMinmaxPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceSamplerFilterMinmaxPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           filterMinmaxSingleComponentFormats;
    VkBool32           filterMinmaxImageComponentMapping;
} VkPhysicalDeviceSamplerFilterMinmaxPropertiesEXT;

The members of the VkPhysicalDeviceSamplerFilterMinmaxPropertiesEXT structure describe the following implementation-dependent limits:

  • filterMinmaxSingleComponentFormats is a boolean value indicating whether a minimum set of required formats support min/max filtering.

  • filterMinmaxImageComponentMapping is a boolean value indicating whether the implementation supports non-identity component mapping of the image when doing min/max filtering.

If the VkPhysicalDeviceSamplerFilterMinmaxPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

If filterMinmaxSingleComponentFormats is VK_TRUE, the following formats must support the VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT_EXT feature with VK_IMAGE_TILING_OPTIMAL, if they support VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT.

  • VK_FORMAT_R8_UNORM

  • VK_FORMAT_R8_SNORM

  • VK_FORMAT_R16_UNORM

  • VK_FORMAT_R16_SNORM

  • VK_FORMAT_R16_SFLOAT

  • VK_FORMAT_R32_SFLOAT

  • VK_FORMAT_D16_UNORM

  • VK_FORMAT_X8_D24_UNORM_PACK32

  • VK_FORMAT_D32_SFLOAT

  • VK_FORMAT_D16_UNORM_S8_UINT

  • VK_FORMAT_D24_UNORM_S8_UINT

  • VK_FORMAT_D32_SFLOAT_S8_UINT

If the format is a depth/stencil format, this bit only specifies that the depth aspect (not the stencil aspect) of an image of this format supports min/max filtering, and that min/max filtering of the depth aspect is supported when depth compare is disabled in the sampler.

If filterMinmaxImageComponentMapping is VK_FALSE the component mapping of the image view used with min/max filtering must have been created with the r component set to VK_COMPONENT_SWIZZLE_IDENTITY. Only the r component of the sampled image value is defined and the other component values are undefined. If filterMinmaxImageComponentMapping is VK_TRUE this restriction does not apply and image component mapping works as normal.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_FILTER_MINMAX_PROPERTIES_EXT

The VkPhysicalDeviceProtectedMemoryProperties structure is defined as:

typedef struct VkPhysicalDeviceProtectedMemoryProperties {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           protectedNoFault;
} VkPhysicalDeviceProtectedMemoryProperties;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • protectedNoFault specifies whether the undefined behavior will not include process termination or device loss. If protectedNoFault is VK_FALSE, undefined behavior may include process termination or device loss. If protectedNoFault is VK_TRUE, undefined behavior will not include process termination or device loss.

If the VkPhysicalDeviceProtectedMemoryProperties structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with a value indicating the implementation-dependent behavior.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_PROPERTIES

The VkPhysicalDeviceMaintenance3Properties structure is defined as:

typedef struct VkPhysicalDeviceMaintenance3Properties {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxPerSetDescriptors;
    VkDeviceSize       maxMemoryAllocationSize;
} VkPhysicalDeviceMaintenance3Properties;

or the equivalent

typedef VkPhysicalDeviceMaintenance3Properties VkPhysicalDeviceMaintenance3PropertiesKHR;

The members of the VkPhysicalDeviceMaintenance3Properties structure describe the following implementation-dependent limits:

  • maxPerSetDescriptors is a maximum number of descriptors (summed over all descriptor types) in a single descriptor set that is guaranteed to satisfy any implementation-dependent constraints on the size of a descriptor set itself. Applications can query whether a descriptor set that goes beyond this limit is supported using vkGetDescriptorSetLayoutSupport.

  • maxMemoryAllocationSize is the maximum size of a memory allocation that can be created, even if there is more space available in the heap.

If the VkPhysicalDeviceMaintenance3Properties structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES

The VkPhysicalDeviceDescriptorIndexingPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceDescriptorIndexingPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxUpdateAfterBindDescriptorsInAllPools;
    VkBool32           shaderUniformBufferArrayNonUniformIndexingNative;
    VkBool32           shaderSampledImageArrayNonUniformIndexingNative;
    VkBool32           shaderStorageBufferArrayNonUniformIndexingNative;
    VkBool32           shaderStorageImageArrayNonUniformIndexingNative;
    VkBool32           shaderInputAttachmentArrayNonUniformIndexingNative;
    VkBool32           robustBufferAccessUpdateAfterBind;
    VkBool32           quadDivergentImplicitLod;
    uint32_t           maxPerStageDescriptorUpdateAfterBindSamplers;
    uint32_t           maxPerStageDescriptorUpdateAfterBindUniformBuffers;
    uint32_t           maxPerStageDescriptorUpdateAfterBindStorageBuffers;
    uint32_t           maxPerStageDescriptorUpdateAfterBindSampledImages;
    uint32_t           maxPerStageDescriptorUpdateAfterBindStorageImages;
    uint32_t           maxPerStageDescriptorUpdateAfterBindInputAttachments;
    uint32_t           maxPerStageUpdateAfterBindResources;
    uint32_t           maxDescriptorSetUpdateAfterBindSamplers;
    uint32_t           maxDescriptorSetUpdateAfterBindUniformBuffers;
    uint32_t           maxDescriptorSetUpdateAfterBindUniformBuffersDynamic;
    uint32_t           maxDescriptorSetUpdateAfterBindStorageBuffers;
    uint32_t           maxDescriptorSetUpdateAfterBindStorageBuffersDynamic;
    uint32_t           maxDescriptorSetUpdateAfterBindSampledImages;
    uint32_t           maxDescriptorSetUpdateAfterBindStorageImages;
    uint32_t           maxDescriptorSetUpdateAfterBindInputAttachments;
} VkPhysicalDeviceDescriptorIndexingPropertiesEXT;

The members of the VkPhysicalDeviceDescriptorIndexingPropertiesEXT structure describe the following implementation-dependent limits:

  • maxUpdateAfterBindDescriptorsInAllPools is the maximum number of descriptors (summed over all descriptor types) that can be created across all pools that are created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT bit set. Pool creation may fail when this limit is exceeded, or when the space this limit represents can’t satisfy a pool creation due to fragmentation.

  • shaderUniformBufferArrayNonUniformIndexingNative is a boolean value indicating whether uniform buffer descriptors natively support nonuniform indexing. If this is VK_FALSE, then a single dynamic instance of an instruction that nonuniformly indexes an array of uniform buffers may execute multiple times in order to access all the descriptors.

  • shaderSampledImageArrayNonUniformIndexingNative is a boolean value indicating whether sampler and image descriptors natively support nonuniform indexing. If this is VK_FALSE, then a single dynamic instance of an instruction that nonuniformly indexes an array of samplers or images may execute multiple times in order to access all the descriptors.

  • shaderStorageBufferArrayNonUniformIndexingNative is a boolean value indicating whether storage buffer descriptors natively support nonuniform indexing. If this is VK_FALSE, then a single dynamic instance of an instruction that nonuniformly indexes an array of storage buffers may execute multiple times in order to access all the descriptors.

  • shaderStorageImageArrayNonUniformIndexingNative is a boolean value indicating whether storage image descriptors natively support nonuniform indexing. If this is VK_FALSE, then a single dynamic instance of an instruction that nonuniformly indexes an array of storage images may execute multiple times in order to access all the descriptors.

  • shaderInputAttachmentArrayNonUniformIndexingNative is a boolean value indicating whether input attachment descriptors natively support nonuniform indexing. If this is VK_FALSE, then a single dynamic instance of an instruction that nonuniformly indexes an array of input attachments may execute multiple times in order to access all the descriptors.

  • robustBufferAccessUpdateAfterBind is a boolean value indicating whether robustBufferAccess can be enabled in a device simultaneously with descriptorBindingUniformBufferUpdateAfterBind, descriptorBindingStorageBufferUpdateAfterBind, descriptorBindingUniformTexelBufferUpdateAfterBind, and/or descriptorBindingStorageTexelBufferUpdateAfterBind. If this is VK_FALSE, then either robustBufferAccess must be disabled or all of these update-after-bind features must be disabled.

  • quadDivergentImplicitLod is a boolean value indicating whether implicit level of detail calculations for image operations have well-defined results when the image and/or sampler objects used for the instruction are not uniform within a quad. See Derivative Image Operations.

  • maxPerStageDescriptorUpdateAfterBindSamplers is similar to maxPerStageDescriptorSamplers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageDescriptorUpdateAfterBindUniformBuffers is similar to maxPerStageDescriptorUniformBuffers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageDescriptorUpdateAfterBindStorageBuffers is similar to maxPerStageDescriptorStorageBuffers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageDescriptorUpdateAfterBindSampledImages is similar to maxPerStageDescriptorSampledImages but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageDescriptorUpdateAfterBindStorageImages is similar to maxPerStageDescriptorStorageImages but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageDescriptorUpdateAfterBindInputAttachments is similar to maxPerStageDescriptorInputAttachments but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxPerStageUpdateAfterBindResources is similar to maxPerStageResources but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindSamplers is similar to maxDescriptorSetSamplers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindUniformBuffers is similar to maxDescriptorSetUniformBuffers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindUniformBuffersDynamic is similar to maxDescriptorSetUniformBuffersDynamic but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindStorageBuffers is similar to maxDescriptorSetStorageBuffers but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindStorageBuffersDynamic is similar to maxDescriptorSetStorageBuffersDynamic but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindSampledImages is similar to maxDescriptorSetSampledImages but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindStorageImages is similar to maxDescriptorSetStorageImages but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

  • maxDescriptorSetUpdateAfterBindInputAttachments is similar to maxDescriptorSetInputAttachments but counts descriptors from descriptor sets created with or without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT bit set.

If the VkPhysicalDeviceDescriptorIndexingPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2KHR, it is filled with the implementation-dependent limits.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_PROPERTIES_EXT

The VkPhysicalDeviceConservativeRasterizationPropertiesEXT structure is defined as:

typedef struct VkPhysicalDeviceConservativeRasterizationPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    float              primitiveOverestimationSize;
    float              maxExtraPrimitiveOverestimationSize;
    float              extraPrimitiveOverestimationSizeGranularity;
    VkBool32           primitiveUnderestimation;
    VkBool32           conservativePointAndLineRasterization;
    VkBool32           degenerateTrianglesRasterized;
    VkBool32           degenerateLinesRasterized;
    VkBool32           fullyCoveredFragmentShaderInputVariable;
    VkBool32           conservativeRasterizationPostDepthCoverage;
} VkPhysicalDeviceConservativeRasterizationPropertiesEXT;

The members of the VkPhysicalDeviceConservativeRasterizationPropertiesEXT structure describe the following implementation-dependent limits:

  • primitiveOverestimationSize is the size in pixels the generating primitive is increased at each of its edges during conservative rasterization overestimation mode. Even with a size of 0.0, conservative rasterization overestimation rules still apply and if any part of the pixel rectangle is covered by the generating primitive, fragments are generated for the entire pixel. However implementations may make the pixel coverage area even more conservative by increasing the size of the generating primitive.

  • maxExtraPrimitiveOverestimationSize is the maximum size in pixels of extra overestimation the implementation supports in the pipeline state. A value of 0.0 means the implementation does not support any additional overestimation of the generating primitive during conservative rasterization. A value above 0.0 allows the application to further increase the size of the generating primitive during conservative rasterization overestimation.

  • extraPrimitiveOverestimationSizeGranularity is the granularity of extra overestimation that can be specified in the pipeline state between 0.0 and maxExtraPrimitiveOverestimationSize inclusive. A value of 0.0 means the implementation can use the smallest representable non-zero value in the screen space pixel fixed-point grid.

  • primitiveUnderestimation is true if the implementation supports the VK_CONSERVATIVE_RASTERIZATION_MODE_UNDERESTIMATE_EXT conservative rasterization mode in addition to VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT. Otherwise the implementation only supports VK_CONSERVATIVE_RASTERIZATION_MODE_OVERESTIMATE_EXT.

  • conservativePointAndLineRasterization is true if the implementation supports conservative rasterization of point and line primitives as well as triangle primitives. Otherwise the implementation only supports triangle primitives.

  • degenerateTrianglesRasterized is false if the implementation culls primitives generated from triangles that become zero area after they are quantized to the fixed-point rasterization pixel grid. degenerateTrianglesRasterized is true if these primitives are not culled and the provoking vertex attributes and depth value are used for the fragments. The primitive area calculation is done on the primitive generated from the clipped triangle if applicable. Zero area primitives are backfacing and the application can enable backface culling if desired.

  • degenerateLinesRasterized is false if the implementation culls lines that become zero length after they are quantized to the fixed-point rasterization pixel grid. degenerateLinesRasterized is true if zero length lines are not culled and the provoking vertex attributes and depth value are used for the fragments.

  • fullyCoveredFragmentShaderInputVariable is true if the implementation supports the SPIR-V builtin fragment shader input variable FullyCoveredEXT which specifies that conservative rasterization is enabled and the fragment pixel square is fully covered by the generating primitive.

  • conservativeRasterizationPostDepthCoverage is true if the implementation supports conservative rasterization with the PostDepthCoverage execution mode enabled. When supported the SampleMask built-in input variable will reflect the coverage after the early per-fragment depth and stencil tests are applied even when conservative rasterization is enabled. Otherwise PostDepthCoverage execution mode must not be used when conservative rasterization is enabled.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_CONSERVATIVE_RASTERIZATION_PROPERTIES_EXT

If the VkPhysicalDeviceConservativeRasterizationPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceProperties2KHR, it is filled with the implementation-dependent limits and properties.

The VkPhysicalDeviceShaderCorePropertiesAMD structure is defined as:

typedef struct VkPhysicalDeviceShaderCorePropertiesAMD {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           shaderEngineCount;
    uint32_t           shaderArraysPerEngineCount;
    uint32_t           computeUnitsPerShaderArray;
    uint32_t           simdPerComputeUnit;
    uint32_t           wavefrontsPerSimd;
    uint32_t           wavefrontSize;
    uint32_t           sgprsPerSimd;
    uint32_t           minSgprAllocation;
    uint32_t           maxSgprAllocation;
    uint32_t           sgprAllocationGranularity;
    uint32_t           vgprsPerSimd;
    uint32_t           minVgprAllocation;
    uint32_t           maxVgprAllocation;
    uint32_t           vgprAllocationGranularity;
} VkPhysicalDeviceShaderCorePropertiesAMD;

The members of the VkPhysicalDeviceShaderCorePropertiesAMD structure describe the following implementation-dependent limits:

  • shaderEngineCount is an unsigned integer value indicating the number of shader engines found inside the shader core of the physical device.

  • shaderArraysPerEngineCount is an unsigned integer value indicating the number of shader arrays inside a shader engine. Each shader array has its own scan converter, set of compute units, and a render back end (color and depth buffers). Shader arrays within a shader engine share shader processor input (wave launcher) and shader export (export buffer) units. Currently, a shader engine can have one or two shader arrays.

  • computeUnitsPerShaderArray is an unsigned integer value indicating the number of compute units within a shader array. A compute unit houses a set of SIMDs along with a sequencer module and a local data store.

  • simdPerComputeUnit is an unsigned integer value indicating the number of SIMDs inside a compute unit. Each SIMD processes a single instruction at a time.

  • wavefrontSize is an unsigned integer value indicating the number of channels (or threads) in a wavefront.

  • sgprsPerSimd is an unsigned integer value indicating the number of physical Scalar General Purpose Registers (SGPRs) per SIMD.

  • minSgprAllocation is an unsigned integer value indicating the minimum number of SGPRs allocated for a wave.

  • maxSgprAllocation is an unsigned integer value indicating the maximum number of SGPRs allocated for a wave.

  • sgprAllocationGranularity is an unsigned integer value indicating the granularity of SGPR allocation for a wave.

  • vgprsPerSimd is an unsigned integer value indicating the number of physical Vector General Purpose Registers (VGPRs) per SIMD.

  • minVgprAllocation is an unsigned integer value indicating the minimum number of VGPRs allocated for a wave.

  • maxVgprAllocation is an unsigned integer value indicating the maximum number of VGPRs allocated for a wave.

  • vgprAllocationGranularity is an unsigned integer value indicating the granularity of VGPR allocation for a wave.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_CORE_PROPERTIES_AMD

If the VkPhysicalDeviceShaderCorePropertiesAMD structure is included in the pNext chain of VkPhysicalDeviceProperties2, it is filled with the implementation-dependent limits.

32.2.1. Limit Requirements

The following table specifies the required minimum/maximum for all Vulkan graphics implementations. Where a limit corresponds to a fine-grained device feature which is optional, the feature name is listed with two required limits, one when the feature is supported and one when it is not supported. If an implementation supports a feature, the limits reported are the same whether or not the feature is enabled.

Table 45. Required Limit Types
Type Limit Feature

uint32_t

maxImageDimension1D

-

uint32_t

maxImageDimension2D

-

uint32_t

maxImageDimension3D

-

uint32_t

maxImageDimensionCube

-

uint32_t

maxImageArrayLayers

-

uint32_t

maxTexelBufferElements

-

uint32_t

maxUniformBufferRange

-

uint32_t

maxStorageBufferRange

-

uint32_t

maxPushConstantsSize

-

uint32_t

maxMemoryAllocationCount

-

uint32_t

maxSamplerAllocationCount

-

VkDeviceSize

bufferImageGranularity

-

VkDeviceSize

sparseAddressSpaceSize

sparseBinding

uint32_t

maxBoundDescriptorSets

-

uint32_t

maxPerStageDescriptorSamplers

-

uint32_t

maxPerStageDescriptorUniformBuffers

-

uint32_t

maxPerStageDescriptorStorageBuffers

-

uint32_t

maxPerStageDescriptorSampledImages

-

uint32_t

maxPerStageDescriptorStorageImages

-

uint32_t

maxPerStageDescriptorInputAttachments

-

uint32_t

maxPerStageResources

-

uint32_t

maxDescriptorSetSamplers

-

uint32_t

maxDescriptorSetUniformBuffers

-

uint32_t

maxDescriptorSetUniformBuffersDynamic

-

uint32_t

maxDescriptorSetStorageBuffers

-

uint32_t

maxDescriptorSetStorageBuffersDynamic

-

uint32_t

maxDescriptorSetSampledImages

-

uint32_t

maxDescriptorSetStorageImages

-

uint32_t

maxDescriptorSetInputAttachments

-

uint32_t

maxVertexInputAttributes

-

uint32_t

maxVertexInputBindings

-

uint32_t

maxVertexInputAttributeOffset

-

uint32_t

maxVertexInputBindingStride

-

uint32_t

maxVertexOutputComponents

-

uint32_t

maxTessellationGenerationLevel

tessellationShader

uint32_t

maxTessellationPatchSize

tessellationShader

uint32_t

maxTessellationControlPerVertexInputComponents

tessellationShader

uint32_t

maxTessellationControlPerVertexOutputComponents

tessellationShader

uint32_t

maxTessellationControlPerPatchOutputComponents

tessellationShader

uint32_t

maxTessellationControlTotalOutputComponents

tessellationShader

uint32_t

maxTessellationEvaluationInputComponents

tessellationShader

uint32_t

maxTessellationEvaluationOutputComponents

tessellationShader

uint32_t

maxGeometryShaderInvocations

geometryShader

uint32_t

maxGeometryInputComponents

geometryShader

uint32_t

maxGeometryOutputComponents

geometryShader

uint32_t

maxGeometryOutputVertices

geometryShader

uint32_t

maxGeometryTotalOutputComponents

geometryShader

uint32_t

maxFragmentInputComponents

-

uint32_t

maxFragmentOutputAttachments

-

uint32_t

maxFragmentDualSrcAttachments

dualSrcBlend

uint32_t

maxFragmentCombinedOutputResources

-

uint32_t

maxComputeSharedMemorySize

-

3 × uint32_t

maxComputeWorkGroupCount

-

uint32_t

maxComputeWorkGroupInvocations

-

3 × uint32_t

maxComputeWorkGroupSize

-

uint32_t

subPixelPrecisionBits

-

uint32_t

subTexelPrecisionBits

-

uint32_t

mipmapPrecisionBits

-

uint32_t

maxDrawIndexedIndexValue

fullDrawIndexUint32

uint32_t

maxDrawIndirectCount

multiDrawIndirect

float

maxSamplerLodBias

-

float

maxSamplerAnisotropy

samplerAnisotropy

uint32_t

maxViewports

multiViewport

2 × uint32_t

maxViewportDimensions

-

2 × float

viewportBoundsRange

-

uint32_t

viewportSubPixelBits

-

size_t

minMemoryMapAlignment

-

VkDeviceSize

minTexelBufferOffsetAlignment

-

VkDeviceSize

minUniformBufferOffsetAlignment

-

VkDeviceSize

minStorageBufferOffsetAlignment

-

int32_t

minTexelOffset

-

uint32_t

maxTexelOffset

-

int32_t

minTexelGatherOffset

shaderImageGatherExtended

uint32_t

maxTexelGatherOffset

shaderImageGatherExtended

float

minInterpolationOffset

sampleRateShading

float

maxInterpolationOffset

sampleRateShading

uint32_t

subPixelInterpolationOffsetBits

sampleRateShading

uint32_t

maxFramebufferWidth

-

uint32_t

maxFramebufferHeight

-

uint32_t

maxFramebufferLayers

-

VkSampleCountFlags

framebufferColorSampleCounts

-

VkSampleCountFlags

framebufferDepthSampleCounts

-

VkSampleCountFlags

framebufferStencilSampleCounts

-

VkSampleCountFlags

framebufferNoAttachmentsSampleCounts

-

uint32_t

maxColorAttachments

-

VkSampleCountFlags

sampledImageColorSampleCounts

-

VkSampleCountFlags

sampledImageIntegerSampleCounts

-

VkSampleCountFlags

sampledImageDepthSampleCounts

-

VkSampleCountFlags

sampledImageStencilSampleCounts

-

VkSampleCountFlags

storageImageSampleCounts

shaderStorageImageMultisample

uint32_t

maxSampleMaskWords

-

VkBool32

timestampComputeAndGraphics

-

float

timestampPeriod

-

uint32_t

maxClipDistances

shaderClipDistance

uint32_t

maxCullDistances

shaderCullDistance

uint32_t

maxCombinedClipAndCullDistances

shaderCullDistance

uint32_t

discreteQueuePriorities

-

2 × float

pointSizeRange

largePoints

2 × float

lineWidthRange

wideLines

float

pointSizeGranularity

largePoints

float

lineWidthGranularity

wideLines

VkBool32

strictLines

-

VkBool32

standardSampleLocations

-

VkDeviceSize

optimalBufferCopyOffsetAlignment

-

VkDeviceSize

optimalBufferCopyRowPitchAlignment

-

VkDeviceSize

nonCoherentAtomSize

-

uint32_t

maxDiscardRectangles

VK_EXT_discard_rectangles

VkBool32

filterMinmaxSingleComponentFormats

VK_EXT_sampler_filter_minmax

VkBool32

filterMinmaxImageComponentMapping

VK_EXT_sampler_filter_minmax

float

primitiveOverestimationSize

VK_EXT_conservative_rasterization

VkBool32

maxExtraPrimitiveOverestimationSize

VK_EXT_conservative_rasterization

float

extraPrimitiveOverestimationSizeGranularity

VK_EXT_conservative_rasterization

VkBool32

degenerateTriangleRasterized

VK_EXT_conservative_rasterization

float

degenerateLinesRasterized

VK_EXT_conservative_rasterization

VkBool32

fullyCoveredFragmentShaderInputVariable

VK_EXT_conservative_rasterization

VkBool32

conservativeRasterizationPostDepthCoverage

VK_EXT_conservative_rasterization

uint32_t

maxVertexAttribDivisor

VK_EXT_vertex_attribute_divisor

Table 46. Required Limits
Limit Unsupported Limit Supported Limit Limit Type1

maxImageDimension1D

-

4096

min

maxImageDimension2D

-

4096

min

maxImageDimension3D

-

256

min

maxImageDimensionCube

-

4096

min

maxImageArrayLayers

-

256

min

maxTexelBufferElements

-

65536

min

maxUniformBufferRange

-

16384

min

maxStorageBufferRange

-

227

min

maxPushConstantsSize

-

128

min

maxMemoryAllocationCount

-

4096

min

maxSamplerAllocationCount

-

4000

min

bufferImageGranularity

-

131072

max

sparseAddressSpaceSize

0

231

min

maxBoundDescriptorSets

-

4

min

maxPerStageDescriptorSamplers

-

16

min

maxPerStageDescriptorUniformBuffers

-

12

min

maxPerStageDescriptorStorageBuffers

-

4

min

maxPerStageDescriptorSampledImages

-

16

min

maxPerStageDescriptorStorageImages

-

4

min

maxPerStageDescriptorInputAttachments

-

4

min

maxPerStageResources

-

128 2

min

maxDescriptorSetSamplers

-

96 8

min, n × PerStage

maxDescriptorSetUniformBuffers

-

72 8

min, n × PerStage

maxDescriptorSetUniformBuffersDynamic

-

8

min

maxDescriptorSetStorageBuffers

-

24 8

min, n × PerStage

maxDescriptorSetStorageBuffersDynamic

-

4

min

maxDescriptorSetSampledImages

-

96 8

min, n × PerStage

maxDescriptorSetStorageImages

-

24 8

min, n × PerStage

maxDescriptorSetInputAttachments

-

4

min

maxVertexInputAttributes

-

16

min

maxVertexInputBindings

-

16

min

maxVertexInputAttributeOffset

-

2047

min

maxVertexInputBindingStride

-

2048

min

maxVertexOutputComponents

-

64

min

maxTessellationGenerationLevel

0

64

min

maxTessellationPatchSize

0

32

min

maxTessellationControlPerVertexInputComponents

0

64

min

maxTessellationControlPerVertexOutputComponents

0

64

min

maxTessellationControlPerPatchOutputComponents

0

120

min

maxTessellationControlTotalOutputComponents

0

2048

min

maxTessellationEvaluationInputComponents

0

64

min

maxTessellationEvaluationOutputComponents

0

64

min

maxGeometryShaderInvocations

0

32

min

maxGeometryInputComponents

0

64

min

maxGeometryOutputComponents

0

64

min

maxGeometryOutputVertices

0

256

min

maxGeometryTotalOutputComponents

0

1024

min

maxFragmentInputComponents

-

64

min

maxFragmentOutputAttachments

-

4

min

maxFragmentDualSrcAttachments

0

1

min

maxFragmentCombinedOutputResources

-

4

min

maxComputeSharedMemorySize

-

16384

min

maxComputeWorkGroupCount

-

(65535,65535,65535)

min

maxComputeWorkGroupInvocations

-

128

min

maxComputeWorkGroupSize

-

(128,128,64)

min

subPixelPrecisionBits

-

4

min

subTexelPrecisionBits

-

4

min

mipmapPrecisionBits

-

4

min

maxDrawIndexedIndexValue

224-1

232-1

min

maxDrawIndirectCount

1

216-1

min

maxSamplerLodBias

-

2

min

maxSamplerAnisotropy

1

16

min

maxViewports

1

16

min

maxViewportDimensions

-

(4096,4096) 3

min

viewportBoundsRange

-

(-8192,8191) 4

(max,min)

viewportSubPixelBits

-

0

min

minMemoryMapAlignment

-

64

min

minTexelBufferOffsetAlignment

-

256

max

minUniformBufferOffsetAlignment

-

256

max

minStorageBufferOffsetAlignment

-

256

max

minTexelOffset

-

-8

max

maxTexelOffset

-

7

min

minTexelGatherOffset

0

-8

max

maxTexelGatherOffset

0

7

min

minInterpolationOffset

0.0

-0.5 5

max

maxInterpolationOffset

0.0

0.5 - (1 ULP) 5

min

subPixelInterpolationOffsetBits

0

4 5

min

maxFramebufferWidth

-

4096

min

maxFramebufferHeight

-

4096

min

maxFramebufferLayers

-

256

min

framebufferColorSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

framebufferDepthSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

framebufferStencilSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

framebufferNoAttachmentsSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

maxColorAttachments

-

4

min

sampledImageColorSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

sampledImageIntegerSampleCounts

-

VK_SAMPLE_COUNT_1_BIT

min

sampledImageDepthSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

sampledImageStencilSampleCounts

-

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

storageImageSampleCounts

VK_SAMPLE_COUNT_1_BIT

(VK_SAMPLE_COUNT_1_BIT | VK_SAMPLE_COUNT_4_BIT)

min

maxSampleMaskWords

-

1

min

timestampComputeAndGraphics

-

-

implementation dependent

timestampPeriod

-

-

duration

maxClipDistances

0

8

min

maxCullDistances

0

8

min

maxCombinedClipAndCullDistances

0

8

min

discreteQueuePriorities

-

2

min

pointSizeRange

(1.0,1.0)

(1.0,64.0 - ULP)6

(max,min)

lineWidthRange

(1.0,1.0)

(1.0,8.0 - ULP)7

(max,min)

pointSizeGranularity

0.0

1.0 6

max, fixed point increment

lineWidthGranularity

0.0

1.0 7

max, fixed point increment

strictLines

-

-

implementation dependent

standardSampleLocations

-

-

implementation dependent

optimalBufferCopyOffsetAlignment

-

-

recommendation

optimalBufferCopyRowPitchAlignment

-

-

recommendation

nonCoherentAtomSize

-

256

max

maxPushDescriptors

-

32

min

maxMultiviewViewCount

-

6

min

maxMultiviewInstanceIndex

-

227-1

min

maxDiscardRectangles

0

4

min

sampleLocationSampleCounts

-

VK_SAMPLE_COUNT_4_BIT

min

maxSampleLocationGridSize

-

(1,1)

min

sampleLocationCoordinateRange

-

(0.0, 0.9375)

(max,min)

sampleLocationSubPixelBits

-

4

min

variableSampleLocations

-

false

implementation dependent

minImportedHostPointerAlignment

-

65536

max

perViewPositionAllComponents

-

-

implementation dependent

filterMinmaxSingleComponentFormats

-

-

implementation dependent

filterMinmaxImageComponentMapping

-

-

implementation dependent

advancedBlendMaxColorAttachments

-

1

min

advancedBlendIndependentBlend

-

false

implementation dependent

advancedBlendNonPremultipliedSrcColor

-

false

implementation dependent

advancedBlendNonPremultipliedDstColor

-

false

implementation dependent

advancedBlendCorrelatedOverlap

-

false

implementation dependent

advancedBlendAllOperations

-

false

implementation dependent

maxPerSetDescriptors

-

1024

min

maxMemoryAllocationSize

-

230

min

primitiveOverestimationSize

-

0.0

min

maxExtraPrimitiveOverestimationSize

-

0.0

min

extraPrimitiveOverestimationSizeGranularity

-

0.0

min

primitiveUnderestimation

-

false

implementation dependent

conservativePointAndLineRasterization

-

false

implementation dependent

degenerateTrianglesRasterized

-

false

implementation dependent

degenerateLinesRasterized

-

false

implementation dependent

fullyCoveredFragmentShaderInputVariable

-

false

implementation dependent

conservativeRasterizationPostDepthCoverage

-

false

implementation dependent

maxUpdateAfterBindDescriptorsInAllPools

-

500000

min

shaderUniformBufferArrayNonUniformIndexingNative

-

false

implementation dependent

shaderSampledImageArrayNonUniformIndexingNative

-

false

implementation dependent

shaderStorageBufferArrayNonUniformIndexingNative

-

false

implementation dependent

shaderStorageImageArrayNonUniformIndexingNative

-

false

implementation dependent

shaderInputAttachmentArrayNonUniformIndexingNative

-

false

implementation dependent

maxPerStageDescriptorUpdateAfterBindSamplers

-

500000 9

min

maxPerStageDescriptorUpdateAfterBindUniformBuffers

-

12 9

min

maxPerStageDescriptorUpdateAfterBindStorageBuffers

-

500000 9

min

maxPerStageDescriptorUpdateAfterBindSampledImages

-

500000 9

min

maxPerStageDescriptorUpdateAfterBindStorageImages

-

500000 9

min

maxPerStageDescriptorUpdateAfterBindInputAttachments

-

500000 9

min

maxPerStageUpdateAfterBindResources

-

500000 9

min

maxDescriptorSetUpdateAfterBindSamplers

-

500000 9

min

maxDescriptorSetUpdateAfterBindUniformBuffers

-

96 8 9

min, n × PerStage

maxDescriptorSetUpdateAfterBindUniformBuffersDynamic

-

8 9

min

maxDescriptorSetUpdateAfterBindStorageBuffers

-

500000 9

min

maxDescriptorSetUpdateAfterBindStorageBuffersDynamic

-

4 9

min

maxDescriptorSetUpdateAfterBindSampledImages

-

500000 9

min

maxDescriptorSetUpdateAfterBindStorageImages

-

500000 9

min

maxDescriptorSetUpdateAfterBindInputAttachments

-

500000 9

min

maxVertexAttribDivisor

-

216-1

min

1

The Limit Type column specifies the limit is either the minimum limit all implementations must support or the maximum limit all implementations must support. For bitmasks a minimum limit is the least bits all implementations must set, but they may have additional bits set beyond this minimum.

2

The maxPerStageResources must be at least the smallest of the following:

  • the sum of the maxPerStageDescriptorUniformBuffers, maxPerStageDescriptorStorageBuffers, maxPerStageDescriptorSampledImages, maxPerStageDescriptorStorageImages, maxPerStageDescriptorInputAttachments, maxColorAttachments limits, or

  • 128.

It may not be possible to reach this limit in every stage.

3

See maxViewportDimensions for the required relationship to other limits.

4

See viewportBoundsRange for the required relationship to other limits.

5

The values minInterpolationOffset and maxInterpolationOffset describe the closed interval of supported interpolation offsets: [minInterpolationOffset, maxInterpolationOffset]. The ULP is determined by subPixelInterpolationOffsetBits. If subPixelInterpolationOffsetBits is 4, this provides increments of (1/24) = 0.0625, and thus the range of supported interpolation offsets would be [-0.5, 0.4375].

6

The point size ULP is determined by pointSizeGranularity. If the pointSizeGranularity is 0.125, the range of supported point sizes must be at least [1.0, 63.875].

7

The line width ULP is determined by lineWidthGranularity. If the lineWidthGranularity is 0.0625, the range of supported line widths must be at least [1.0, 7.9375].

8

The minimum maxDescriptorSet* limit is n times the corresponding specification minimum maxPerStageDescriptor* limit, where n is the number of shader stages supported by the VkPhysicalDevice. If all shader stages are supported, n = 6 (vertex, tessellation control, tessellation evaluation, geometry, fragment, compute).

9

The UpdateAfterBind descriptor limits must each be greater than or equal to the corresponding non-UpdateAfterBind limit.

32.3. Additional Multisampling Capabilities

In addition to the minimum capabilities described in the previous section (Limits), implementations may support additional multisampling capabilities specific to a particular sample count.

To query additional sample count specific multisampling capabilities, call:

void vkGetPhysicalDeviceMultisamplePropertiesEXT(
    VkPhysicalDevice                            physicalDevice,
    VkSampleCountFlagBits                       samples,
    VkMultisamplePropertiesEXT*                 pMultisampleProperties);
  • physicalDevice is the physical device from which to query the additional multisampling capabilities.

  • samples is the sample count to query the capabilities for.

  • pMultisampleProperties is a pointer to a structure of type VkMultisamplePropertiesEXT, in which information about the additional multisampling capabilities specific to the sample count is returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • samples must be a valid VkSampleCountFlagBits value

  • pMultisampleProperties must be a valid pointer to a VkMultisamplePropertiesEXT structure

The VkMultisamplePropertiesEXT structure is defined as

typedef struct VkMultisamplePropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkExtent2D         maxSampleLocationGridSize;
} VkMultisamplePropertiesEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • maxSampleLocationGridSize is the maximum size of the pixel grid in which sample locations can vary.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_MULTISAMPLE_PROPERTIES_EXT

  • pNext must be NULL

If the sample count for which additional multisampling capabilities are requested using vkGetPhysicalDeviceMultisamplePropertiesEXT is set in VkPhysicalDeviceSampleLocationsEXT:: sampleLocationSampleCounts the width and height members of VkMultisamplePropertiesEXT::maxSampleLocationGridSize must be greater than or equal to the corresponding members of VkPhysicalDeviceSampleLocationsEXT:: maxSampleLocationGridSize, respectively, otherwise both members must be 0.

32.4. Formats

The features for the set of formats (VkFormat) supported by the implementation are queried individually using the vkGetPhysicalDeviceFormatProperties command.

32.4.1. Format Definition

Image formats which can be passed to, and may be returned from Vulkan commands, are:

typedef enum VkFormat {
    VK_FORMAT_UNDEFINED = 0,
    VK_FORMAT_R4G4_UNORM_PACK8 = 1,
    VK_FORMAT_R4G4B4A4_UNORM_PACK16 = 2,
    VK_FORMAT_B4G4R4A4_UNORM_PACK16 = 3,
    VK_FORMAT_R5G6B5_UNORM_PACK16 = 4,
    VK_FORMAT_B5G6R5_UNORM_PACK16 = 5,
    VK_FORMAT_R5G5B5A1_UNORM_PACK16 = 6,
    VK_FORMAT_B5G5R5A1_UNORM_PACK16 = 7,
    VK_FORMAT_A1R5G5B5_UNORM_PACK16 = 8,
    VK_FORMAT_R8_UNORM = 9,
    VK_FORMAT_R8_SNORM = 10,
    VK_FORMAT_R8_USCALED = 11,
    VK_FORMAT_R8_SSCALED = 12,
    VK_FORMAT_R8_UINT = 13,
    VK_FORMAT_R8_SINT = 14,
    VK_FORMAT_R8_SRGB = 15,
    VK_FORMAT_R8G8_UNORM = 16,
    VK_FORMAT_R8G8_SNORM = 17,
    VK_FORMAT_R8G8_USCALED = 18,
    VK_FORMAT_R8G8_SSCALED = 19,
    VK_FORMAT_R8G8_UINT = 20,
    VK_FORMAT_R8G8_SINT = 21,
    VK_FORMAT_R8G8_SRGB = 22,
    VK_FORMAT_R8G8B8_UNORM = 23,
    VK_FORMAT_R8G8B8_SNORM = 24,
    VK_FORMAT_R8G8B8_USCALED = 25,
    VK_FORMAT_R8G8B8_SSCALED = 26,
    VK_FORMAT_R8G8B8_UINT = 27,
    VK_FORMAT_R8G8B8_SINT = 28,
    VK_FORMAT_R8G8B8_SRGB = 29,
    VK_FORMAT_B8G8R8_UNORM = 30,
    VK_FORMAT_B8G8R8_SNORM = 31,
    VK_FORMAT_B8G8R8_USCALED = 32,
    VK_FORMAT_B8G8R8_SSCALED = 33,
    VK_FORMAT_B8G8R8_UINT = 34,
    VK_FORMAT_B8G8R8_SINT = 35,
    VK_FORMAT_B8G8R8_SRGB = 36,
    VK_FORMAT_R8G8B8A8_UNORM = 37,
    VK_FORMAT_R8G8B8A8_SNORM = 38,
    VK_FORMAT_R8G8B8A8_USCALED = 39,
    VK_FORMAT_R8G8B8A8_SSCALED = 40,
    VK_FORMAT_R8G8B8A8_UINT = 41,
    VK_FORMAT_R8G8B8A8_SINT = 42,
    VK_FORMAT_R8G8B8A8_SRGB = 43,
    VK_FORMAT_B8G8R8A8_UNORM = 44,
    VK_FORMAT_B8G8R8A8_SNORM = 45,
    VK_FORMAT_B8G8R8A8_USCALED = 46,
    VK_FORMAT_B8G8R8A8_SSCALED = 47,
    VK_FORMAT_B8G8R8A8_UINT = 48,
    VK_FORMAT_B8G8R8A8_SINT = 49,
    VK_FORMAT_B8G8R8A8_SRGB = 50,
    VK_FORMAT_A8B8G8R8_UNORM_PACK32 = 51,
    VK_FORMAT_A8B8G8R8_SNORM_PACK32 = 52,
    VK_FORMAT_A8B8G8R8_USCALED_PACK32 = 53,
    VK_FORMAT_A8B8G8R8_SSCALED_PACK32 = 54,
    VK_FORMAT_A8B8G8R8_UINT_PACK32 = 55,
    VK_FORMAT_A8B8G8R8_SINT_PACK32 = 56,
    VK_FORMAT_A8B8G8R8_SRGB_PACK32 = 57,
    VK_FORMAT_A2R10G10B10_UNORM_PACK32 = 58,
    VK_FORMAT_A2R10G10B10_SNORM_PACK32 = 59,
    VK_FORMAT_A2R10G10B10_USCALED_PACK32 = 60,
    VK_FORMAT_A2R10G10B10_SSCALED_PACK32 = 61,
    VK_FORMAT_A2R10G10B10_UINT_PACK32 = 62,
    VK_FORMAT_A2R10G10B10_SINT_PACK32 = 63,
    VK_FORMAT_A2B10G10R10_UNORM_PACK32 = 64,
    VK_FORMAT_A2B10G10R10_SNORM_PACK32 = 65,
    VK_FORMAT_A2B10G10R10_USCALED_PACK32 = 66,
    VK_FORMAT_A2B10G10R10_SSCALED_PACK32 = 67,
    VK_FORMAT_A2B10G10R10_UINT_PACK32 = 68,
    VK_FORMAT_A2B10G10R10_SINT_PACK32 = 69,
    VK_FORMAT_R16_UNORM = 70,
    VK_FORMAT_R16_SNORM = 71,
    VK_FORMAT_R16_USCALED = 72,
    VK_FORMAT_R16_SSCALED = 73,
    VK_FORMAT_R16_UINT = 74,
    VK_FORMAT_R16_SINT = 75,
    VK_FORMAT_R16_SFLOAT = 76,
    VK_FORMAT_R16G16_UNORM = 77,
    VK_FORMAT_R16G16_SNORM = 78,
    VK_FORMAT_R16G16_USCALED = 79,
    VK_FORMAT_R16G16_SSCALED = 80,
    VK_FORMAT_R16G16_UINT = 81,
    VK_FORMAT_R16G16_SINT = 82,
    VK_FORMAT_R16G16_SFLOAT = 83,
    VK_FORMAT_R16G16B16_UNORM = 84,
    VK_FORMAT_R16G16B16_SNORM = 85,
    VK_FORMAT_R16G16B16_USCALED = 86,
    VK_FORMAT_R16G16B16_SSCALED = 87,
    VK_FORMAT_R16G16B16_UINT = 88,
    VK_FORMAT_R16G16B16_SINT = 89,
    VK_FORMAT_R16G16B16_SFLOAT = 90,
    VK_FORMAT_R16G16B16A16_UNORM = 91,
    VK_FORMAT_R16G16B16A16_SNORM = 92,
    VK_FORMAT_R16G16B16A16_USCALED = 93,
    VK_FORMAT_R16G16B16A16_SSCALED = 94,
    VK_FORMAT_R16G16B16A16_UINT = 95,
    VK_FORMAT_R16G16B16A16_SINT = 96,
    VK_FORMAT_R16G16B16A16_SFLOAT = 97,
    VK_FORMAT_R32_UINT = 98,
    VK_FORMAT_R32_SINT = 99,
    VK_FORMAT_R32_SFLOAT = 100,
    VK_FORMAT_R32G32_UINT = 101,
    VK_FORMAT_R32G32_SINT = 102,
    VK_FORMAT_R32G32_SFLOAT = 103,
    VK_FORMAT_R32G32B32_UINT = 104,
    VK_FORMAT_R32G32B32_SINT = 105,
    VK_FORMAT_R32G32B32_SFLOAT = 106,
    VK_FORMAT_R32G32B32A32_UINT = 107,
    VK_FORMAT_R32G32B32A32_SINT = 108,
    VK_FORMAT_R32G32B32A32_SFLOAT = 109,
    VK_FORMAT_R64_UINT = 110,
    VK_FORMAT_R64_SINT = 111,
    VK_FORMAT_R64_SFLOAT = 112,
    VK_FORMAT_R64G64_UINT = 113,
    VK_FORMAT_R64G64_SINT = 114,
    VK_FORMAT_R64G64_SFLOAT = 115,
    VK_FORMAT_R64G64B64_UINT = 116,
    VK_FORMAT_R64G64B64_SINT = 117,
    VK_FORMAT_R64G64B64_SFLOAT = 118,
    VK_FORMAT_R64G64B64A64_UINT = 119,
    VK_FORMAT_R64G64B64A64_SINT = 120,
    VK_FORMAT_R64G64B64A64_SFLOAT = 121,
    VK_FORMAT_B10G11R11_UFLOAT_PACK32 = 122,
    VK_FORMAT_E5B9G9R9_UFLOAT_PACK32 = 123,
    VK_FORMAT_D16_UNORM = 124,
    VK_FORMAT_X8_D24_UNORM_PACK32 = 125,
    VK_FORMAT_D32_SFLOAT = 126,
    VK_FORMAT_S8_UINT = 127,
    VK_FORMAT_D16_UNORM_S8_UINT = 128,
    VK_FORMAT_D24_UNORM_S8_UINT = 129,
    VK_FORMAT_D32_SFLOAT_S8_UINT = 130,
    VK_FORMAT_BC1_RGB_UNORM_BLOCK = 131,
    VK_FORMAT_BC1_RGB_SRGB_BLOCK = 132,
    VK_FORMAT_BC1_RGBA_UNORM_BLOCK = 133,
    VK_FORMAT_BC1_RGBA_SRGB_BLOCK = 134,
    VK_FORMAT_BC2_UNORM_BLOCK = 135,
    VK_FORMAT_BC2_SRGB_BLOCK = 136,
    VK_FORMAT_BC3_UNORM_BLOCK = 137,
    VK_FORMAT_BC3_SRGB_BLOCK = 138,
    VK_FORMAT_BC4_UNORM_BLOCK = 139,
    VK_FORMAT_BC4_SNORM_BLOCK = 140,
    VK_FORMAT_BC5_UNORM_BLOCK = 141,
    VK_FORMAT_BC5_SNORM_BLOCK = 142,
    VK_FORMAT_BC6H_UFLOAT_BLOCK = 143,
    VK_FORMAT_BC6H_SFLOAT_BLOCK = 144,
    VK_FORMAT_BC7_UNORM_BLOCK = 145,
    VK_FORMAT_BC7_SRGB_BLOCK = 146,
    VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK = 147,
    VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK = 148,
    VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK = 149,
    VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK = 150,
    VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK = 151,
    VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK = 152,
    VK_FORMAT_EAC_R11_UNORM_BLOCK = 153,
    VK_FORMAT_EAC_R11_SNORM_BLOCK = 154,
    VK_FORMAT_EAC_R11G11_UNORM_BLOCK = 155,
    VK_FORMAT_EAC_R11G11_SNORM_BLOCK = 156,
    VK_FORMAT_ASTC_4x4_UNORM_BLOCK = 157,
    VK_FORMAT_ASTC_4x4_SRGB_BLOCK = 158,
    VK_FORMAT_ASTC_5x4_UNORM_BLOCK = 159,
    VK_FORMAT_ASTC_5x4_SRGB_BLOCK = 160,
    VK_FORMAT_ASTC_5x5_UNORM_BLOCK = 161,
    VK_FORMAT_ASTC_5x5_SRGB_BLOCK = 162,
    VK_FORMAT_ASTC_6x5_UNORM_BLOCK = 163,
    VK_FORMAT_ASTC_6x5_SRGB_BLOCK = 164,
    VK_FORMAT_ASTC_6x6_UNORM_BLOCK = 165,
    VK_FORMAT_ASTC_6x6_SRGB_BLOCK = 166,
    VK_FORMAT_ASTC_8x5_UNORM_BLOCK = 167,
    VK_FORMAT_ASTC_8x5_SRGB_BLOCK = 168,
    VK_FORMAT_ASTC_8x6_UNORM_BLOCK = 169,
    VK_FORMAT_ASTC_8x6_SRGB_BLOCK = 170,
    VK_FORMAT_ASTC_8x8_UNORM_BLOCK = 171,
    VK_FORMAT_ASTC_8x8_SRGB_BLOCK = 172,
    VK_FORMAT_ASTC_10x5_UNORM_BLOCK = 173,
    VK_FORMAT_ASTC_10x5_SRGB_BLOCK = 174,
    VK_FORMAT_ASTC_10x6_UNORM_BLOCK = 175,
    VK_FORMAT_ASTC_10x6_SRGB_BLOCK = 176,
    VK_FORMAT_ASTC_10x8_UNORM_BLOCK = 177,
    VK_FORMAT_ASTC_10x8_SRGB_BLOCK = 178,
    VK_FORMAT_ASTC_10x10_UNORM_BLOCK = 179,
    VK_FORMAT_ASTC_10x10_SRGB_BLOCK = 180,
    VK_FORMAT_ASTC_12x10_UNORM_BLOCK = 181,
    VK_FORMAT_ASTC_12x10_SRGB_BLOCK = 182,
    VK_FORMAT_ASTC_12x12_UNORM_BLOCK = 183,
    VK_FORMAT_ASTC_12x12_SRGB_BLOCK = 184,
    VK_FORMAT_G8B8G8R8_422_UNORM = 1000156000,
    VK_FORMAT_B8G8R8G8_422_UNORM = 1000156001,
    VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM = 1000156002,
    VK_FORMAT_G8_B8R8_2PLANE_420_UNORM = 1000156003,
    VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM = 1000156004,
    VK_FORMAT_G8_B8R8_2PLANE_422_UNORM = 1000156005,
    VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM = 1000156006,
    VK_FORMAT_R10X6_UNORM_PACK16 = 1000156007,
    VK_FORMAT_R10X6G10X6_UNORM_2PACK16 = 1000156008,
    VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16 = 1000156009,
    VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16 = 1000156010,
    VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16 = 1000156011,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16 = 1000156012,
    VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16 = 1000156013,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16 = 1000156014,
    VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16 = 1000156015,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16 = 1000156016,
    VK_FORMAT_R12X4_UNORM_PACK16 = 1000156017,
    VK_FORMAT_R12X4G12X4_UNORM_2PACK16 = 1000156018,
    VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16 = 1000156019,
    VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16 = 1000156020,
    VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16 = 1000156021,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16 = 1000156022,
    VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16 = 1000156023,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16 = 1000156024,
    VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16 = 1000156025,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16 = 1000156026,
    VK_FORMAT_G16B16G16R16_422_UNORM = 1000156027,
    VK_FORMAT_B16G16R16G16_422_UNORM = 1000156028,
    VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM = 1000156029,
    VK_FORMAT_G16_B16R16_2PLANE_420_UNORM = 1000156030,
    VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM = 1000156031,
    VK_FORMAT_G16_B16R16_2PLANE_422_UNORM = 1000156032,
    VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM = 1000156033,
    VK_FORMAT_PVRTC1_2BPP_UNORM_BLOCK_IMG = 1000054000,
    VK_FORMAT_PVRTC1_4BPP_UNORM_BLOCK_IMG = 1000054001,
    VK_FORMAT_PVRTC2_2BPP_UNORM_BLOCK_IMG = 1000054002,
    VK_FORMAT_PVRTC2_4BPP_UNORM_BLOCK_IMG = 1000054003,
    VK_FORMAT_PVRTC1_2BPP_SRGB_BLOCK_IMG = 1000054004,
    VK_FORMAT_PVRTC1_4BPP_SRGB_BLOCK_IMG = 1000054005,
    VK_FORMAT_PVRTC2_2BPP_SRGB_BLOCK_IMG = 1000054006,
    VK_FORMAT_PVRTC2_4BPP_SRGB_BLOCK_IMG = 1000054007,
    VK_FORMAT_G8B8G8R8_422_UNORM_KHR = VK_FORMAT_G8B8G8R8_422_UNORM,
    VK_FORMAT_B8G8R8G8_422_UNORM_KHR = VK_FORMAT_B8G8R8G8_422_UNORM,
    VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM_KHR = VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM,
    VK_FORMAT_G8_B8R8_2PLANE_420_UNORM_KHR = VK_FORMAT_G8_B8R8_2PLANE_420_UNORM,
    VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM_KHR = VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM,
    VK_FORMAT_G8_B8R8_2PLANE_422_UNORM_KHR = VK_FORMAT_G8_B8R8_2PLANE_422_UNORM,
    VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM_KHR = VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM,
    VK_FORMAT_R10X6_UNORM_PACK16_KHR = VK_FORMAT_R10X6_UNORM_PACK16,
    VK_FORMAT_R10X6G10X6_UNORM_2PACK16_KHR = VK_FORMAT_R10X6G10X6_UNORM_2PACK16,
    VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16_KHR = VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16,
    VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16_KHR = VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16,
    VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16_KHR = VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16_KHR = VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16,
    VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16_KHR = VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16_KHR = VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16,
    VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16_KHR = VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16,
    VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16_KHR = VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16,
    VK_FORMAT_R12X4_UNORM_PACK16_KHR = VK_FORMAT_R12X4_UNORM_PACK16,
    VK_FORMAT_R12X4G12X4_UNORM_2PACK16_KHR = VK_FORMAT_R12X4G12X4_UNORM_2PACK16,
    VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16_KHR = VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16,
    VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16_KHR = VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16,
    VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16_KHR = VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16_KHR = VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16,
    VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16_KHR = VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16_KHR = VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16,
    VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16_KHR = VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16,
    VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16_KHR = VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16,
    VK_FORMAT_G16B16G16R16_422_UNORM_KHR = VK_FORMAT_G16B16G16R16_422_UNORM,
    VK_FORMAT_B16G16R16G16_422_UNORM_KHR = VK_FORMAT_B16G16R16G16_422_UNORM,
    VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM_KHR = VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM,
    VK_FORMAT_G16_B16R16_2PLANE_420_UNORM_KHR = VK_FORMAT_G16_B16R16_2PLANE_420_UNORM,
    VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM_KHR = VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM,
    VK_FORMAT_G16_B16R16_2PLANE_422_UNORM_KHR = VK_FORMAT_G16_B16R16_2PLANE_422_UNORM,
    VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM_KHR = VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM,
} VkFormat;
  • VK_FORMAT_UNDEFINED specifies that the format is not specified.

  • VK_FORMAT_R4G4_UNORM_PACK8 specifies a two-component, 8-bit packed unsigned normalized format that has a 4-bit R component in bits 4..7, and a 4-bit G component in bits 0..3.

  • VK_FORMAT_R4G4B4A4_UNORM_PACK16 specifies a four-component, 16-bit packed unsigned normalized format that has a 4-bit R component in bits 12..15, a 4-bit G component in bits 8..11, a 4-bit B component in bits 4..7, and a 4-bit A component in bits 0..3.

  • VK_FORMAT_B4G4R4A4_UNORM_PACK16 specifies a four-component, 16-bit packed unsigned normalized format that has a 4-bit B component in bits 12..15, a 4-bit G component in bits 8..11, a 4-bit R component in bits 4..7, and a 4-bit A component in bits 0..3.

  • VK_FORMAT_R5G6B5_UNORM_PACK16 specifies a three-component, 16-bit packed unsigned normalized format that has a 5-bit R component in bits 11..15, a 6-bit G component in bits 5..10, and a 5-bit B component in bits 0..4.

  • VK_FORMAT_B5G6R5_UNORM_PACK16 specifies a three-component, 16-bit packed unsigned normalized format that has a 5-bit B component in bits 11..15, a 6-bit G component in bits 5..10, and a 5-bit R component in bits 0..4.

  • VK_FORMAT_R5G5B5A1_UNORM_PACK16 specifies a four-component, 16-bit packed unsigned normalized format that has a 5-bit R component in bits 11..15, a 5-bit G component in bits 6..10, a 5-bit B component in bits 1..5, and a 1-bit A component in bit 0.

  • VK_FORMAT_B5G5R5A1_UNORM_PACK16 specifies a four-component, 16-bit packed unsigned normalized format that has a 5-bit B component in bits 11..15, a 5-bit G component in bits 6..10, a 5-bit R component in bits 1..5, and a 1-bit A component in bit 0.

  • VK_FORMAT_A1R5G5B5_UNORM_PACK16 specifies a four-component, 16-bit packed unsigned normalized format that has a 1-bit A component in bit 15, a 5-bit R component in bits 10..14, a 5-bit G component in bits 5..9, and a 5-bit B component in bits 0..4.

  • VK_FORMAT_R8_UNORM specifies a one-component, 8-bit unsigned normalized format that has a single 8-bit R component.

  • VK_FORMAT_R8_SNORM specifies a one-component, 8-bit signed normalized format that has a single 8-bit R component.

  • VK_FORMAT_R8_USCALED specifies a one-component, 8-bit unsigned scaled integer format that has a single 8-bit R component.

  • VK_FORMAT_R8_SSCALED specifies a one-component, 8-bit signed scaled integer format that has a single 8-bit R component.

  • VK_FORMAT_R8_UINT specifies a one-component, 8-bit unsigned integer format that has a single 8-bit R component.

  • VK_FORMAT_R8_SINT specifies a one-component, 8-bit signed integer format that has a single 8-bit R component.

  • VK_FORMAT_R8_SRGB specifies a one-component, 8-bit unsigned normalized format that has a single 8-bit R component stored with sRGB nonlinear encoding.

  • VK_FORMAT_R8G8_UNORM specifies a two-component, 16-bit unsigned normalized format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_SNORM specifies a two-component, 16-bit signed normalized format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_USCALED specifies a two-component, 16-bit unsigned scaled integer format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_SSCALED specifies a two-component, 16-bit signed scaled integer format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_UINT specifies a two-component, 16-bit unsigned integer format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_SINT specifies a two-component, 16-bit signed integer format that has an 8-bit R component in byte 0, and an 8-bit G component in byte 1.

  • VK_FORMAT_R8G8_SRGB specifies a two-component, 16-bit unsigned normalized format that has an 8-bit R component stored with sRGB nonlinear encoding in byte 0, and an 8-bit G component stored with sRGB nonlinear encoding in byte 1.

  • VK_FORMAT_R8G8B8_UNORM specifies a three-component, 24-bit unsigned normalized format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_SNORM specifies a three-component, 24-bit signed normalized format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_USCALED specifies a three-component, 24-bit unsigned scaled format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_SSCALED specifies a three-component, 24-bit signed scaled format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_UINT specifies a three-component, 24-bit unsigned integer format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_SINT specifies a three-component, 24-bit signed integer format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, and an 8-bit B component in byte 2.

  • VK_FORMAT_R8G8B8_SRGB specifies a three-component, 24-bit unsigned normalized format that has an 8-bit R component stored with sRGB nonlinear encoding in byte 0, an 8-bit G component stored with sRGB nonlinear encoding in byte 1, and an 8-bit B component stored with sRGB nonlinear encoding in byte 2.

  • VK_FORMAT_B8G8R8_UNORM specifies a three-component, 24-bit unsigned normalized format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_SNORM specifies a three-component, 24-bit signed normalized format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_USCALED specifies a three-component, 24-bit unsigned scaled format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_SSCALED specifies a three-component, 24-bit signed scaled format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_UINT specifies a three-component, 24-bit unsigned integer format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_SINT specifies a three-component, 24-bit signed integer format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, and an 8-bit R component in byte 2.

  • VK_FORMAT_B8G8R8_SRGB specifies a three-component, 24-bit unsigned normalized format that has an 8-bit B component stored with sRGB nonlinear encoding in byte 0, an 8-bit G component stored with sRGB nonlinear encoding in byte 1, and an 8-bit R component stored with sRGB nonlinear encoding in byte 2.

  • VK_FORMAT_R8G8B8A8_UNORM specifies a four-component, 32-bit unsigned normalized format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_SNORM specifies a four-component, 32-bit signed normalized format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_USCALED specifies a four-component, 32-bit unsigned scaled format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_SSCALED specifies a four-component, 32-bit signed scaled format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_UINT specifies a four-component, 32-bit unsigned integer format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_SINT specifies a four-component, 32-bit signed integer format that has an 8-bit R component in byte 0, an 8-bit G component in byte 1, an 8-bit B component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_R8G8B8A8_SRGB specifies a four-component, 32-bit unsigned normalized format that has an 8-bit R component stored with sRGB nonlinear encoding in byte 0, an 8-bit G component stored with sRGB nonlinear encoding in byte 1, an 8-bit B component stored with sRGB nonlinear encoding in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_UNORM specifies a four-component, 32-bit unsigned normalized format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_SNORM specifies a four-component, 32-bit signed normalized format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_USCALED specifies a four-component, 32-bit unsigned scaled format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_SSCALED specifies a four-component, 32-bit signed scaled format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_UINT specifies a four-component, 32-bit unsigned integer format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_SINT specifies a four-component, 32-bit signed integer format that has an 8-bit B component in byte 0, an 8-bit G component in byte 1, an 8-bit R component in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_B8G8R8A8_SRGB specifies a four-component, 32-bit unsigned normalized format that has an 8-bit B component stored with sRGB nonlinear encoding in byte 0, an 8-bit G component stored with sRGB nonlinear encoding in byte 1, an 8-bit R component stored with sRGB nonlinear encoding in byte 2, and an 8-bit A component in byte 3.

  • VK_FORMAT_A8B8G8R8_UNORM_PACK32 specifies a four-component, 32-bit packed unsigned normalized format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_SNORM_PACK32 specifies a four-component, 32-bit packed signed normalized format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_USCALED_PACK32 specifies a four-component, 32-bit packed unsigned scaled integer format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_SSCALED_PACK32 specifies a four-component, 32-bit packed signed scaled integer format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_UINT_PACK32 specifies a four-component, 32-bit packed unsigned integer format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_SINT_PACK32 specifies a four-component, 32-bit packed signed integer format that has an 8-bit A component in bits 24..31, an 8-bit B component in bits 16..23, an 8-bit G component in bits 8..15, and an 8-bit R component in bits 0..7.

  • VK_FORMAT_A8B8G8R8_SRGB_PACK32 specifies a four-component, 32-bit packed unsigned normalized format that has an 8-bit A component in bits 24..31, an 8-bit B component stored with sRGB nonlinear encoding in bits 16..23, an 8-bit G component stored with sRGB nonlinear encoding in bits 8..15, and an 8-bit R component stored with sRGB nonlinear encoding in bits 0..7.

  • VK_FORMAT_A2R10G10B10_UNORM_PACK32 specifies a four-component, 32-bit packed unsigned normalized format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2R10G10B10_SNORM_PACK32 specifies a four-component, 32-bit packed signed normalized format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2R10G10B10_USCALED_PACK32 specifies a four-component, 32-bit packed unsigned scaled integer format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2R10G10B10_SSCALED_PACK32 specifies a four-component, 32-bit packed signed scaled integer format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2R10G10B10_UINT_PACK32 specifies a four-component, 32-bit packed unsigned integer format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2R10G10B10_SINT_PACK32 specifies a four-component, 32-bit packed signed integer format that has a 2-bit A component in bits 30..31, a 10-bit R component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit B component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_UNORM_PACK32 specifies a four-component, 32-bit packed unsigned normalized format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_SNORM_PACK32 specifies a four-component, 32-bit packed signed normalized format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_USCALED_PACK32 specifies a four-component, 32-bit packed unsigned scaled integer format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_SSCALED_PACK32 specifies a four-component, 32-bit packed signed scaled integer format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_UINT_PACK32 specifies a four-component, 32-bit packed unsigned integer format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_A2B10G10R10_SINT_PACK32 specifies a four-component, 32-bit packed signed integer format that has a 2-bit A component in bits 30..31, a 10-bit B component in bits 20..29, a 10-bit G component in bits 10..19, and a 10-bit R component in bits 0..9.

  • VK_FORMAT_R16_UNORM specifies a one-component, 16-bit unsigned normalized format that has a single 16-bit R component.

  • VK_FORMAT_R16_SNORM specifies a one-component, 16-bit signed normalized format that has a single 16-bit R component.

  • VK_FORMAT_R16_USCALED specifies a one-component, 16-bit unsigned scaled integer format that has a single 16-bit R component.

  • VK_FORMAT_R16_SSCALED specifies a one-component, 16-bit signed scaled integer format that has a single 16-bit R component.

  • VK_FORMAT_R16_UINT specifies a one-component, 16-bit unsigned integer format that has a single 16-bit R component.

  • VK_FORMAT_R16_SINT specifies a one-component, 16-bit signed integer format that has a single 16-bit R component.

  • VK_FORMAT_R16_SFLOAT specifies a one-component, 16-bit signed floating-point format that has a single 16-bit R component.

  • VK_FORMAT_R16G16_UNORM specifies a two-component, 32-bit unsigned normalized format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_SNORM specifies a two-component, 32-bit signed normalized format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_USCALED specifies a two-component, 32-bit unsigned scaled integer format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_SSCALED specifies a two-component, 32-bit signed scaled integer format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_UINT specifies a two-component, 32-bit unsigned integer format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_SINT specifies a two-component, 32-bit signed integer format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16_SFLOAT specifies a two-component, 32-bit signed floating-point format that has a 16-bit R component in bytes 0..1, and a 16-bit G component in bytes 2..3.

  • VK_FORMAT_R16G16B16_UNORM specifies a three-component, 48-bit unsigned normalized format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_SNORM specifies a three-component, 48-bit signed normalized format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_USCALED specifies a three-component, 48-bit unsigned scaled integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_SSCALED specifies a three-component, 48-bit signed scaled integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_UINT specifies a three-component, 48-bit unsigned integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_SINT specifies a three-component, 48-bit signed integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16_SFLOAT specifies a three-component, 48-bit signed floating-point format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, and a 16-bit B component in bytes 4..5.

  • VK_FORMAT_R16G16B16A16_UNORM specifies a four-component, 64-bit unsigned normalized format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_SNORM specifies a four-component, 64-bit signed normalized format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_USCALED specifies a four-component, 64-bit unsigned scaled integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_SSCALED specifies a four-component, 64-bit signed scaled integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_UINT specifies a four-component, 64-bit unsigned integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_SINT specifies a four-component, 64-bit signed integer format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R16G16B16A16_SFLOAT specifies a four-component, 64-bit signed floating-point format that has a 16-bit R component in bytes 0..1, a 16-bit G component in bytes 2..3, a 16-bit B component in bytes 4..5, and a 16-bit A component in bytes 6..7.

  • VK_FORMAT_R32_UINT specifies a one-component, 32-bit unsigned integer format that has a single 32-bit R component.

  • VK_FORMAT_R32_SINT specifies a one-component, 32-bit signed integer format that has a single 32-bit R component.

  • VK_FORMAT_R32_SFLOAT specifies a one-component, 32-bit signed floating-point format that has a single 32-bit R component.

  • VK_FORMAT_R32G32_UINT specifies a two-component, 64-bit unsigned integer format that has a 32-bit R component in bytes 0..3, and a 32-bit G component in bytes 4..7.

  • VK_FORMAT_R32G32_SINT specifies a two-component, 64-bit signed integer format that has a 32-bit R component in bytes 0..3, and a 32-bit G component in bytes 4..7.

  • VK_FORMAT_R32G32_SFLOAT specifies a two-component, 64-bit signed floating-point format that has a 32-bit R component in bytes 0..3, and a 32-bit G component in bytes 4..7.

  • VK_FORMAT_R32G32B32_UINT specifies a three-component, 96-bit unsigned integer format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, and a 32-bit B component in bytes 8..11.

  • VK_FORMAT_R32G32B32_SINT specifies a three-component, 96-bit signed integer format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, and a 32-bit B component in bytes 8..11.

  • VK_FORMAT_R32G32B32_SFLOAT specifies a three-component, 96-bit signed floating-point format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, and a 32-bit B component in bytes 8..11.

  • VK_FORMAT_R32G32B32A32_UINT specifies a four-component, 128-bit unsigned integer format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, a 32-bit B component in bytes 8..11, and a 32-bit A component in bytes 12..15.

  • VK_FORMAT_R32G32B32A32_SINT specifies a four-component, 128-bit signed integer format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, a 32-bit B component in bytes 8..11, and a 32-bit A component in bytes 12..15.

  • VK_FORMAT_R32G32B32A32_SFLOAT specifies a four-component, 128-bit signed floating-point format that has a 32-bit R component in bytes 0..3, a 32-bit G component in bytes 4..7, a 32-bit B component in bytes 8..11, and a 32-bit A component in bytes 12..15.

  • VK_FORMAT_R64_UINT specifies a one-component, 64-bit unsigned integer format that has a single 64-bit R component.

  • VK_FORMAT_R64_SINT specifies a one-component, 64-bit signed integer format that has a single 64-bit R component.

  • VK_FORMAT_R64_SFLOAT specifies a one-component, 64-bit signed floating-point format that has a single 64-bit R component.

  • VK_FORMAT_R64G64_UINT specifies a two-component, 128-bit unsigned integer format that has a 64-bit R component in bytes 0..7, and a 64-bit G component in bytes 8..15.

  • VK_FORMAT_R64G64_SINT specifies a two-component, 128-bit signed integer format that has a 64-bit R component in bytes 0..7, and a 64-bit G component in bytes 8..15.

  • VK_FORMAT_R64G64_SFLOAT specifies a two-component, 128-bit signed floating-point format that has a 64-bit R component in bytes 0..7, and a 64-bit G component in bytes 8..15.

  • VK_FORMAT_R64G64B64_UINT specifies a three-component, 192-bit unsigned integer format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, and a 64-bit B component in bytes 16..23.

  • VK_FORMAT_R64G64B64_SINT specifies a three-component, 192-bit signed integer format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, and a 64-bit B component in bytes 16..23.

  • VK_FORMAT_R64G64B64_SFLOAT specifies a three-component, 192-bit signed floating-point format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, and a 64-bit B component in bytes 16..23.

  • VK_FORMAT_R64G64B64A64_UINT specifies a four-component, 256-bit unsigned integer format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, a 64-bit B component in bytes 16..23, and a 64-bit A component in bytes 24..31.

  • VK_FORMAT_R64G64B64A64_SINT specifies a four-component, 256-bit signed integer format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, a 64-bit B component in bytes 16..23, and a 64-bit A component in bytes 24..31.

  • VK_FORMAT_R64G64B64A64_SFLOAT specifies a four-component, 256-bit signed floating-point format that has a 64-bit R component in bytes 0..7, a 64-bit G component in bytes 8..15, a 64-bit B component in bytes 16..23, and a 64-bit A component in bytes 24..31.

  • VK_FORMAT_B10G11R11_UFLOAT_PACK32 specifies a three-component, 32-bit packed unsigned floating-point format that has a 10-bit B component in bits 22..31, an 11-bit G component in bits 11..21, an 11-bit R component in bits 0..10. See Unsigned 10-Bit Floating-Point Numbers and Unsigned 11-Bit Floating-Point Numbers.

  • VK_FORMAT_E5B9G9R9_UFLOAT_PACK32 specifies a three-component, 32-bit packed unsigned floating-point format that has a 5-bit shared exponent in bits 27..31, a 9-bit B component mantissa in bits 18..26, a 9-bit G component mantissa in bits 9..17, and a 9-bit R component mantissa in bits 0..8.

  • VK_FORMAT_D16_UNORM specifies a one-component, 16-bit unsigned normalized format that has a single 16-bit depth component.

  • VK_FORMAT_X8_D24_UNORM_PACK32 specifies a two-component, 32-bit format that has 24 unsigned normalized bits in the depth component and, optionally:, 8 bits that are unused.

  • VK_FORMAT_D32_SFLOAT specifies a one-component, 32-bit signed floating-point format that has 32-bits in the depth component.

  • VK_FORMAT_S8_UINT specifies a one-component, 8-bit unsigned integer format that has 8-bits in the stencil component.

  • VK_FORMAT_D16_UNORM_S8_UINT specifies a two-component, 24-bit format that has 16 unsigned normalized bits in the depth component and 8 unsigned integer bits in the stencil component.

  • VK_FORMAT_D24_UNORM_S8_UINT specifies a two-component, 32-bit packed format that has 8 unsigned integer bits in the stencil component, and 24 unsigned normalized bits in the depth component.

  • VK_FORMAT_D32_SFLOAT_S8_UINT specifies a two-component format that has 32 signed float bits in the depth component and 8 unsigned integer bits in the stencil component. There are optionally: 24-bits that are unused.

  • VK_FORMAT_BC1_RGB_UNORM_BLOCK specifies a three-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data. This format has no alpha and is considered opaque.

  • VK_FORMAT_BC1_RGB_SRGB_BLOCK specifies a three-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data with sRGB nonlinear encoding. This format has no alpha and is considered opaque.

  • VK_FORMAT_BC1_RGBA_UNORM_BLOCK specifies a four-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data, and provides 1 bit of alpha.

  • VK_FORMAT_BC1_RGBA_SRGB_BLOCK specifies a four-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data with sRGB nonlinear encoding, and provides 1 bit of alpha.

  • VK_FORMAT_BC2_UNORM_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values.

  • VK_FORMAT_BC2_SRGB_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values with sRGB nonlinear encoding.

  • VK_FORMAT_BC3_UNORM_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values.

  • VK_FORMAT_BC3_SRGB_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values with sRGB nonlinear encoding.

  • VK_FORMAT_BC4_UNORM_BLOCK specifies a one-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized red texel data.

  • VK_FORMAT_BC4_SNORM_BLOCK specifies a one-component, block-compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of signed normalized red texel data.

  • VK_FORMAT_BC5_UNORM_BLOCK specifies a two-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RG texel data with the first 64 bits encoding red values followed by 64 bits encoding green values.

  • VK_FORMAT_BC5_SNORM_BLOCK specifies a two-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of signed normalized RG texel data with the first 64 bits encoding red values followed by 64 bits encoding green values.

  • VK_FORMAT_BC6H_UFLOAT_BLOCK specifies a three-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned floating-point RGB texel data.

  • VK_FORMAT_BC6H_SFLOAT_BLOCK specifies a three-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of signed floating-point RGB texel data.

  • VK_FORMAT_BC7_UNORM_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_BC7_SRGB_BLOCK specifies a four-component, block-compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK specifies a three-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data. This format has no alpha and is considered opaque.

  • VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK specifies a three-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data with sRGB nonlinear encoding. This format has no alpha and is considered opaque.

  • VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK specifies a four-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data, and provides 1 bit of alpha.

  • VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK specifies a four-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGB texel data with sRGB nonlinear encoding, and provides 1 bit of alpha.

  • VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK specifies a four-component, ETC2 compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values.

  • VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK specifies a four-component, ETC2 compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with the first 64 bits encoding alpha values followed by 64 bits encoding RGB values with sRGB nonlinear encoding applied.

  • VK_FORMAT_EAC_R11_UNORM_BLOCK specifies a one-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized red texel data.

  • VK_FORMAT_EAC_R11_SNORM_BLOCK specifies a one-component, ETC2 compressed format where each 64-bit compressed texel block encodes a 4×4 rectangle of signed normalized red texel data.

  • VK_FORMAT_EAC_R11G11_UNORM_BLOCK specifies a two-component, ETC2 compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RG texel data with the first 64 bits encoding red values followed by 64 bits encoding green values.

  • VK_FORMAT_EAC_R11G11_SNORM_BLOCK specifies a two-component, ETC2 compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of signed normalized RG texel data with the first 64 bits encoding red values followed by 64 bits encoding green values.

  • VK_FORMAT_ASTC_4x4_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_4x4_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 4×4 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_5x4_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 5×4 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_5x4_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 5×4 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_5x5_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 5×5 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_5x5_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 5×5 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_6x5_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 6×5 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_6x5_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 6×5 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_6x6_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 6×6 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_6x6_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 6×6 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_8x5_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×5 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_8x5_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×5 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_8x6_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×6 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_8x6_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×6 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_8x8_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×8 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_8x8_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes an 8×8 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_10x5_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×5 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_10x5_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×5 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_10x6_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×6 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_10x6_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×6 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_10x8_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×8 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_10x8_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×8 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_10x10_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×10 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_10x10_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 10×10 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_12x10_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 12×10 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_12x10_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 12×10 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_ASTC_12x12_UNORM_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 12×12 rectangle of unsigned normalized RGBA texel data.

  • VK_FORMAT_ASTC_12x12_SRGB_BLOCK specifies a four-component, ASTC compressed format where each 128-bit compressed texel block encodes a 12×12 rectangle of unsigned normalized RGBA texel data with sRGB nonlinear encoding applied to the RGB components.

  • VK_FORMAT_G8B8G8R8_422_UNORM specifies a four-component, 32-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has an 8-bit G component for the even i coordinate in byte 0, an 8-bit B component in byte 1, an 8-bit G component for the odd i coordinate in byte 2, and an 8-bit R component in byte 3. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_B8G8R8G8_422_UNORM specifies a four-component, 32-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has an 8-bit B component in byte 0, an 8-bit G component for the even i coordinate in byte 1, an 8-bit R component in byte 2, and an 8-bit G component for the odd i coordinate in byte 3. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM specifies a unsigned normalized multi-planar format that has an 8-bit G component in plane 0, an 8-bit B component in plane 1, and an 8-bit R component in plane 2. The horizontal and vertical dimensions of the R and B planes are halved relative to the image dimensions, and each R and B component is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G8_B8R8_2PLANE_420_UNORM specifies a unsigned normalized multi-planar format that has an 8-bit G component in plane 0, and a two-component, 16-bit BR plane 1 consisting of an 8-bit B component in byte 0 and an 8-bit R component in byte 1. The horizontal and vertical dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM specifies a unsigned normalized multi-planar format that has an 8-bit G component in plane 0, an 8-bit B component in plane 1, and an 8-bit R component in plane 2. The horizontal dimension of the R and B plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G8_B8R8_2PLANE_422_UNORM specifies a unsigned normalized multi-planar format that has an 8-bit G component in plane 0, and a two-component, 16-bit BR plane 1 consisting of an 8-bit B component in byte 0 and an 8-bit R component in byte 1. The horizontal dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM specifies a unsigned normalized multi-planar format that has an 8-bit G component in plane 0, an 8-bit B component in plane 1, and an 8-bit R component in plane 2. Each plane has the same dimensions and each R, G and B component contributes to a single texel. The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane.

  • VK_FORMAT_R10X6_UNORM_PACK16 specifies a one-component, 16-bit unsigned normalized format that has a single 10-bit R component in the top 10 bits of a 16-bit word, with the bottom 6 bits set to 0.

  • VK_FORMAT_R10X6G10X6_UNORM_2PACK16 specifies a two-component, 32-bit unsigned normalized format that has a 10-bit R component in the top 10 bits of the word in bytes 0..1, and a 10-bit G component in the top 10 bits of the word in bytes 2..3, with the bottom 6 bits of each word set to 0.

  • VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16 specifies a four-component, 64-bit unsigned normalized format that has a 10-bit R component in the top 10 bits of the word in bytes 0..1, a 10-bit G component in the top 10 bits of the word in bytes 2..3, a 10-bit B component in the top 10 bits of the word in bytes 4..5, and a 10-bit A component in the top 10 bits of the word in bytes 6..7, with the bottom 6 bits of each word set to 0.

  • VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16 specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 10-bit G component for the even i coordinate in the top 10 bits of the word in bytes 0..1, a 10-bit B component in the top 10 bits of the word in bytes 2..3, a 10-bit G component for the odd i coordinate in the top 10 bits of the word in bytes 4..5, and a 10-bit R component in the top 10 bits of the word in bytes 6..7, with the bottom 6 bits of each word set to 0. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16 specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 10-bit B component in the top 10 bits of the word in bytes 0..1, a 10-bit G component for the even i coordinate in the top 10 bits of the word in bytes 2..3, a 10-bit R component in the top 10 bits of the word in bytes 4..5, and a 10-bit G component for the odd i coordinate in the top 10 bits of the word in bytes 6..7, with the bottom 6 bits of each word set to 0. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 10-bit G component in the top 10 bits of each 16-bit word of plane 0, a 10-bit B component in the top 10 bits of each 16-bit word of plane 1, and a 10-bit R component in the top 10 bits of each 16-bit word of plane 2, with the bottom 6 bits of each word set to 0. The horizontal and vertical dimensions of the R and B planes are halved relative to the image dimensions, and each R and B component is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 10-bit G component in the top 10 bits of each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 10-bit B component in the top 10 bits of the word in bytes 0..1, and a 10-bit R component in the top 10 bits of the word in bytes 2..3, the bottom 6 bits of each word set to 0. The horizontal and vertical dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 10-bit G component in the top 10 bits of each 16-bit word of plane 0, a 10-bit B component in the top 10 bits of each 16-bit word of plane 1, and a 10-bit R component in the top 10 bits of each 16-bit word of plane 2, with the bottom 6 bits of each word set to 0. The horizontal dimension of the R and B plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 10-bit G component in the top 10 bits of each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 10-bit B component in the top 10 bits of the word in bytes 0..1, and a 10-bit R component in the top 10 bits of the word in bytes 2..3, the bottom 6 bits of each word set to 0. The horizontal dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 10-bit G component in the top 10 bits of each 16-bit word of plane 0, a 10-bit B component in the top 10 bits of each 16-bit word of plane 1, and a 10-bit R component in the top 10 bits of each 16-bit word of plane 2, with the bottom 6 bits of each word set to 0. Each plane has the same dimensions and each R, G and B component contributes to a single texel. The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane.

  • VK_FORMAT_R12X4_UNORM_PACK16 specifies a one-component, 16-bit unsigned normalized format that has a single 12-bit R component in the top 12 bits of a 16-bit word, with the bottom 4 bits set to 0.

  • VK_FORMAT_R12X4G12X4_UNORM_2PACK16 specifies a two-component, 32-bit unsigned normalized format that has a 12-bit R component in the top 12 bits of the word in bytes 0..1, and a 12-bit G component in the top 12 bits of the word in bytes 2..3, with the bottom 4 bits of each word set to 0.

  • VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16 specifies a four-component, 64-bit unsigned normalized format that has a 12-bit R component in the top 12 bits of the word in bytes 0..1, a 12-bit G component in the top 12 bits of the word in bytes 2..3, a 12-bit B component in the top 12 bits of the word in bytes 4..5, and a 12-bit A component in the top 12 bits of the word in bytes 6..7, with the bottom 4 bits of each word set to 0.

  • VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16 specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 12-bit G component for the even i coordinate in the top 12 bits of the word in bytes 0..1, a 12-bit B component in the top 12 bits of the word in bytes 2..3, a 12-bit G component for the odd i coordinate in the top 12 bits of the word in bytes 4..5, and a 12-bit R component in the top 12 bits of the word in bytes 6..7, with the bottom 4 bits of each word set to 0. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16 specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 12-bit B component in the top 12 bits of the word in bytes 0..1, a 12-bit G component for the even i coordinate in the top 12 bits of the word in bytes 2..3, a 12-bit R component in the top 12 bits of the word in bytes 4..5, and a 12-bit G component for the odd i coordinate in the top 12 bits of the word in bytes 6..7, with the bottom 4 bits of each word set to 0. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 12-bit G component in the top 12 bits of each 16-bit word of plane 0, a 12-bit B component in the top 12 bits of each 16-bit word of plane 1, and a 12-bit R component in the top 12 bits of each 16-bit word of plane 2, with the bottom 4 bits of each word set to 0. The horizontal and vertical dimensions of the R and B planes are halved relative to the image dimensions, and each R and B component is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 12-bit G component in the top 12 bits of each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 12-bit B component in the top 12 bits of the word in bytes 0..1, and a 12-bit R component in the top 12 bits of the word in bytes 2..3, the bottom 4 bits of each word set to 0. The horizontal and vertical dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 12-bit G component in the top 12 bits of each 16-bit word of plane 0, a 12-bit B component in the top 12 bits of each 16-bit word of plane 1, and a 12-bit R component in the top 12 bits of each 16-bit word of plane 2, with the bottom 4 bits of each word set to 0. The horizontal dimension of the R and B plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 12-bit G component in the top 12 bits of each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 12-bit B component in the top 12 bits of the word in bytes 0..1, and a 12-bit R component in the top 12 bits of the word in bytes 2..3, the bottom 4 bits of each word set to 0. The horizontal dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16 specifies a unsigned normalized multi-planar format that has a 12-bit G component in the top 12 bits of each 16-bit word of plane 0, a 12-bit B component in the top 12 bits of each 16-bit word of plane 1, and a 12-bit R component in the top 12 bits of each 16-bit word of plane 2, with the bottom 4 bits of each word set to 0. Each plane has the same dimensions and each R, G and B component contributes to a single texel. The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane.

  • VK_FORMAT_G16B16G16R16_422_UNORM specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 16-bit G component for the even i coordinate in the word in bytes 0..1, a 16-bit B component in the word in bytes 2..3, a 16-bit G component for the odd i coordinate in the word in bytes 4..5, and a 16-bit R component in the word in bytes 6..7. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_B16G16R16G16_422_UNORM specifies a four-component, 64-bit format containing a pair of G components, an R component, and a B component, collectively encoding a 2×1 rectangle of unsigned normalized RGB texel data. One G value is present at each i coordinate, with the B and R values shared across both G values and thus recorded at half the horizontal resolution of the image. This format has a 16-bit B component in the word in bytes 0..1, a 16-bit G component for the even i coordinate in the word in bytes 2..3, a 16-bit R component in the word in bytes 4..5, and a 16-bit G component for the odd i coordinate in the word in bytes 6..7. Images in this format must be defined with a width that is a multiple of two. For the purposes of the constraints on copy extents, this format is treated as a compressed format with a 2×1 compressed texel block.

  • VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM specifies a unsigned normalized multi-planar format that has a 16-bit G component in each 16-bit word of plane 0, a 16-bit B component in each 16-bit word of plane 1, and a 16-bit R component in each 16-bit word of plane 2. The horizontal and vertical dimensions of the R and B planes are halved relative to the image dimensions, and each R and B component is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G16_B16R16_2PLANE_420_UNORM specifies a unsigned normalized multi-planar format that has a 16-bit G component in each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 16-bit B component in the word in bytes 0..1, and a 16-bit R component in the word in bytes 2..3. The horizontal and vertical dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\) and \(\lfloor j_G \times 0.5 \rfloor = j_B = j_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width and height that is a multiple of two.

  • VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM specifies a unsigned normalized multi-planar format that has a 16-bit G component in each 16-bit word of plane 0, a 16-bit B component in each 16-bit word of plane 1, and a 16-bit R component in each 16-bit word of plane 2. The horizontal dimension of the R and B plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G16_B16R16_2PLANE_422_UNORM specifies a unsigned normalized multi-planar format that has a 16-bit G component in each 16-bit word of plane 0, and a two-component, 32-bit BR plane 1 consisting of a 16-bit B component in the word in bytes 0..1, and a 16-bit R component in the word in bytes 2..3. The horizontal dimensions of the BR plane is halved relative to the image dimensions, and each R and B value is shared with the G components for which \(\lfloor i_G \times 0.5 \rfloor = i_B = i_R\). The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, and VK_IMAGE_ASPECT_PLANE_1_BIT for the BR plane. Images in this format must be defined with a width that is a multiple of two.

  • VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM specifies a unsigned normalized multi-planar format that has a 16-bit G component in each 16-bit word of plane 0, a 16-bit B component in each 16-bit word of plane 1, and a 16-bit R component in each 16-bit word of plane 2. Each plane has the same dimensions and each R, G and B component contributes to a single texel. The location of each plane when this image is in linear layout can be determined via vkGetImageSubresourceLayout, using VK_IMAGE_ASPECT_PLANE_0_BIT for the G plane, VK_IMAGE_ASPECT_PLANE_1_BIT for the B plane, and VK_IMAGE_ASPECT_PLANE_2_BIT for the R plane.

Compatible formats of planes of multi-planar formats

Individual planes of multi-planar formats are compatible with single-plane formats if they occupy the same number of bits per data element. In the following table, individual planes of a multi-planar format are compatible with the format listed against the relevant plane index for that multi-planar format.

Table 47. Plane Format Compatibility Table
Plane Compatible format for plane Width relative to the width w of the plane with the largest dimensions Height relative to the height h of the plane with the largest dimensions

VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM

0

VK_FORMAT_R8_UNORM

w

h

1

VK_FORMAT_R8_UNORM

w/2

h/2

2

VK_FORMAT_R8_UNORM

w/2

h/2

VK_FORMAT_G8_B8R8_2PLANE_420_UNORM

0

VK_FORMAT_R8_UNORM

w

h

1

VK_FORMAT_R8G8_UNORM

w/2

h/2

VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM

0

VK_FORMAT_R8_UNORM

w

h

1

VK_FORMAT_R8_UNORM

w/2

h

2

VK_FORMAT_R8_UNORM

w/2

h

VK_FORMAT_G8_B8R8_2PLANE_422_UNORM

0

VK_FORMAT_R8_UNORM

w

h

1

VK_FORMAT_R8G8_UNORM

w/2

h

VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM

0

VK_FORMAT_R8_UNORM

w

h

1

VK_FORMAT_R8_UNORM

w

h

2

VK_FORMAT_R8_UNORM

w

h

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16

0

VK_FORMAT_R10X6_UNORM_PACK16

w

h

1

VK_FORMAT_R10X6_UNORM_PACK16

w/2

h/2

2

VK_FORMAT_R10X6_UNORM_PACK16

w/2

h/2

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16

0

VK_FORMAT_R10X6_UNORM_PACK16

w

h

1

VK_FORMAT_R10X6G10X6_UNORM_2PACK16

w/2

h/2

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16

0

VK_FORMAT_R10X6_UNORM_PACK16

w

h

1

VK_FORMAT_R10X6_UNORM_PACK16

w/2

h

2

VK_FORMAT_R10X6_UNORM_PACK16

w/2

h

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16

0

VK_FORMAT_R10X6_UNORM_PACK16

w

h

1

VK_FORMAT_R10X6G10X6_UNORM_2PACK16

w/2

h

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16

0

VK_FORMAT_R10X6_UNORM_PACK16

w

h

1

VK_FORMAT_R10X6_UNORM_PACK16

w

h

2

VK_FORMAT_R10X6_UNORM_PACK16

w

h

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16

0

VK_FORMAT_R12X4_UNORM_PACK16

w

h

1

VK_FORMAT_R12X4_UNORM_PACK16

w/2

h/2

2

VK_FORMAT_R12X4_UNORM_PACK16

w/2

h/2

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16

0

VK_FORMAT_R12X4_UNORM_PACK16

w

h

1

VK_FORMAT_R12X4G12X4_UNORM_2PACK16

w/2

h/2

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16

0

VK_FORMAT_R12X4_UNORM_PACK16

w

h

1

VK_FORMAT_R12X4_UNORM_PACK16

w/2

h

2

VK_FORMAT_R12X4_UNORM_PACK16

w/2

h

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16

0

VK_FORMAT_R12X4_UNORM_PACK16

w

h

1

VK_FORMAT_R12X4G12X4_UNORM_2PACK16

w/2

h

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16

0

VK_FORMAT_R12X4_UNORM_PACK16

w

h

1

VK_FORMAT_R12X4_UNORM_PACK16

w

h

2

VK_FORMAT_R12X4_UNORM_PACK16

w

h

VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM

0

VK_FORMAT_R16_UNORM

w

h

1

VK_FORMAT_R16_UNORM

w/2

h/2

2

VK_FORMAT_R16_UNORM

w/2

h/2

VK_FORMAT_G16_B16R16_2PLANE_420_UNORM

0

VK_FORMAT_R16_UNORM

w

h

1

VK_FORMAT_R16G16_UNORM

w/2

h/2

VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM

0

VK_FORMAT_R16_UNORM

w

h

1

VK_FORMAT_R16_UNORM

w/2

h

2

VK_FORMAT_R16_UNORM

w/2

h

VK_FORMAT_G16_B16R16_2PLANE_422_UNORM

0

VK_FORMAT_R16_UNORM

w

h

1

VK_FORMAT_R16G16_UNORM

w/2

h

VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM

0

VK_FORMAT_R16_UNORM

w

h

1

VK_FORMAT_R16_UNORM

w

h

2

VK_FORMAT_R16_UNORM

w

h

Packed Formats

For the purposes of address alignment when accessing buffer memory containing vertex attribute or texel data, the following formats are considered packed - whole texels or attributes are stored in a single data element, rather than individual components occupying a single data element:

  • Packed into 8-bit data types:

    • VK_FORMAT_R4G4_UNORM_PACK8

  • Packed into 16-bit data types:

    • VK_FORMAT_R4G4B4A4_UNORM_PACK16

    • VK_FORMAT_B4G4R4A4_UNORM_PACK16

    • VK_FORMAT_R5G6B5_UNORM_PACK16

    • VK_FORMAT_B5G6R5_UNORM_PACK16

    • VK_FORMAT_R5G5B5A1_UNORM_PACK16

    • VK_FORMAT_B5G5R5A1_UNORM_PACK16

    • VK_FORMAT_A1R5G5B5_UNORM_PACK16

  • Packed into 32-bit data types:

    • VK_FORMAT_A8B8G8R8_UNORM_PACK32

    • VK_FORMAT_A8B8G8R8_SNORM_PACK32

    • VK_FORMAT_A8B8G8R8_USCALED_PACK32

    • VK_FORMAT_A8B8G8R8_SSCALED_PACK32

    • VK_FORMAT_A8B8G8R8_UINT_PACK32

    • VK_FORMAT_A8B8G8R8_SINT_PACK32

    • VK_FORMAT_A8B8G8R8_SRGB_PACK32

    • VK_FORMAT_A2R10G10B10_UNORM_PACK32

    • VK_FORMAT_A2R10G10B10_SNORM_PACK32

    • VK_FORMAT_A2R10G10B10_USCALED_PACK32

    • VK_FORMAT_A2R10G10B10_SSCALED_PACK32

    • VK_FORMAT_A2R10G10B10_UINT_PACK32

    • VK_FORMAT_A2R10G10B10_SINT_PACK32

    • VK_FORMAT_A2B10G10R10_UNORM_PACK32

    • VK_FORMAT_A2B10G10R10_SNORM_PACK32

    • VK_FORMAT_A2B10G10R10_USCALED_PACK32

    • VK_FORMAT_A2B10G10R10_SSCALED_PACK32

    • VK_FORMAT_A2B10G10R10_UINT_PACK32

    • VK_FORMAT_A2B10G10R10_SINT_PACK32

    • VK_FORMAT_B10G11R11_UFLOAT_PACK32

    • VK_FORMAT_E5B9G9R9_UFLOAT_PACK32

    • VK_FORMAT_X8_D24_UNORM_PACK32

Identification of Formats

A “format” is represented by a single enum value. The name of a format is usually built up by using the following pattern:

VK_FORMAT_{component-format|compression-scheme}_{numeric-format}

The component-format indicates either the size of the R, G, B, and A components (if they are present) in the case of a color format, or the size of the depth (D) and stencil (S) components (if they are present) in the case of a depth/stencil format (see below). An X indicates a component that is unused, but may be present for padding.

Table 48. Interpretation of Numeric Format
Numeric format Description

UNORM

The components are unsigned normalized values in the range [0,1]

SNORM

The components are signed normalized values in the range [-1,1]

USCALED

The components are unsigned integer values that get converted to floating-point in the range [0,2n-1]

SSCALED

The components are signed integer values that get converted to floating-point in the range [-2n-1,2n-1-1]

UINT

The components are unsigned integer values in the range [0,2n-1]

SINT

The components are signed integer values in the range [-2n-1,2n-1-1]

UFLOAT

The components are unsigned floating-point numbers (used by packed, shared exponent, and some compressed formats)

SFLOAT

The components are signed floating-point numbers

SRGB

The R, G, and B components are unsigned normalized values that represent values using sRGB nonlinear encoding, while the A component (if one exists) is a regular unsigned normalized value

The suffix _PACKnn indicates that the format is packed into an underlying type with nn bits. The suffix _mPACKnn is a short-hand that indicates that the format has several components (which may or may not be stored in separate planes) that are each packed into an underlying type with nn bits.

The suffix _BLOCK indicates that the format is a block-compressed format, with the representation of multiple pixels encoded interdependently within a region.

Table 49. Interpretation of Compression Scheme
Compression scheme Description

BC

Block Compression. See Block-Compressed Image Formats.

ETC2

Ericsson Texture Compression. See ETC Compressed Image Formats.

EAC

ETC2 Alpha Compression. See ETC Compressed Image Formats.

ASTC

Adaptive Scalable Texture Compression (LDR Profile). See ASTC Compressed Image Formats.

For multi-planar images, the components in separate planes are separated by underscores, and the number of planes is indicated by the addition of a _2PLANE or _3PLANE suffix. Similarly, the separate aspects of depth-stencil formats are separated by underscores, although these are not considered separate planes. Formats are suffixed by _422 to indicate that planes other than the first are reduced in size by a factor of two horizontally or that the R and B values appear at half the horizontal frequency of the G values, _420 to indicate that planes other than the first are reduced in size by a factor of two both horizontally and vertically, and _444 for consistency to indicate that all three planes of a three-planar image are the same size.

Note

No common format has a single plane containing both R and B channels but does not store these channels at reduced horizontal resolution.

Representation

Color formats must be represented in memory in exactly the form indicated by the format’s name. This means that promoting one format to another with more bits per component and/or additional components must not occur for color formats. Depth/stencil formats have more relaxed requirements as discussed below. Each format has an element size, the number of bytes used to stored one element or one compressed block, with the value of the element size listed in VkFormat.

The representation of non-packed formats is that the first component specified in the name of the format is in the lowest memory addresses and the last component specified is in the highest memory addresses. See Byte mappings for non-packed/compressed color formats. The in-memory ordering of bytes within a component is determined by the host endianness.

Table 50. Byte mappings for non-packed/compressed color formats
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ← Byte

R

VK_FORMAT_R8_*

R

G

VK_FORMAT_R8G8_*

R

G

B

VK_FORMAT_R8G8B8_*

B

G

R

VK_FORMAT_B8G8R8_*

R

G

B

A

VK_FORMAT_R8G8B8A8_*

B

G

R

A

VK_FORMAT_B8G8R8A8_*

G0

B

G1

R

VK_FORMAT_G8B8G8R8_422_UNORM

B

G0

R

G1

VK_FORMAT_B8G8R8G8_422_UNORM

R

VK_FORMAT_R16_*

R

G

VK_FORMAT_R16G16_*

R

G

B

VK_FORMAT_R16G16B16_*

R

G

B

A

VK_FORMAT_R16G16B16A16_*

G0

B

G1

R

VK_FORMAT_G10X6B10X6G10X6R10X6_4PACK16_422_UNORM VK_FORMAT_G12X4B12X4G12X4R12X4_4PACK16_422_UNORM VK_FORMAT_G16B16G16R16_UNORM

B

G0

R

G1

VK_FORMAT_B10X6G10X6R10X6G10X6_4PACK16_422_UNORM VK_FORMAT_B12X4G12X4R12X4G12X4_4PACK16_422_UNORM VK_FORMAT_B16G16R16G16_422_UNORM

R

VK_FORMAT_R32_*

R

G

VK_FORMAT_R32G32_*

R

G

B

VK_FORMAT_R32G32B32_*

R

G

B

A

VK_FORMAT_R32G32B32A32_*

R

VK_FORMAT_R64_*

R

G

VK_FORMAT_R64G64_*

VK_FORMAT_R64G64B64_* as VK_FORMAT_R64G64_* but with B in bytes 16-23

VK_FORMAT_R64G64B64A64_* as VK_FORMAT_R64G64B64_* but with A in bytes 24-31

Packed formats store multiple components within one underlying type. The bit representation is that the first component specified in the name of the format is in the most-significant bits and the last component specified is in the least-significant bits of the underlying type. The in-memory ordering of bytes comprising the underlying type is determined by the host endianness.

Table 51. Bit mappings for packed 8-bit formats
Bit

7

6

5

4

3

2

1

0

VK_FORMAT_R4G4_UNORM_PACK8

R

G

3

2

1

0

3

2

1

0

Table 52. Bit mappings for packed 16-bit formats
Bit

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

VK_FORMAT_R4G4B4A4_UNORM_PACK16

R

G

B

A

3

2

1

0

3

2

1

0

3

2

1

0

3

2

1

0

VK_FORMAT_B4G4R4A4_UNORM_PACK16

B

G

R

A

3

2

1

0

3

2

1

0

3

2

1

0

3

2

1

0

VK_FORMAT_R5G6B5_UNORM_PACK16

R

G

B

4

3

2

1

0

5

4

3

2

1

0

4

3

2

1

0

VK_FORMAT_B5G6R5_UNORM_PACK16

B

G

R

4

3

2

1

0

5

4

3

2

1

0

4

3

2

1

0

VK_FORMAT_R5G5B5A1_UNORM_PACK16

R

G

B

A

4

3

2

1

0

4

3

2

1

0

4

3

2

1

0

0

VK_FORMAT_B5G5R5A1_UNORM_PACK16

B

G

R

A

4

3

2

1

0

4

3

2

1

0

4

3

2

1

0

0

VK_FORMAT_A1R5G5B5_UNORM_PACK16

A

R

G

B

0

4

3

2

1

0

4

3

2

1

0

4

3

2

1

0

VK_FORMAT_R10X6_UNORM_PACK16

R

X

9

8

7

6

5

4

3

2

1

0

5

4

3

2

1

0

VK_FORMAT_R12X4_UNORM_PACK16

R

X

11

10

9

8

7

6

5

4

3

2

1

0

3

2

1

0

Table 53. Bit mappings for packed 32-bit formats
Bit

31

30

29

28

27

26

25

24

23

22

21

20

19

18

17

16

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

VK_FORMAT_A8B8G8R8_*_PACK32

A

B

G

R

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

VK_FORMAT_A2R10G10B10_*_PACK32

A

R

G

B

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

VK_FORMAT_A2B10G10R10_*_PACK32

A

B

G

R

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

VK_FORMAT_B10G11R11_UFLOAT_PACK32

B

G

R

9

8

7

6

5

4

3

2

1

0

10

9

8

7

6

5

4

3

2

1

0

10

9

8

7

6

5

4

3

2

1

0

VK_FORMAT_E5B9G9R9_UFLOAT_PACK32

E

B

G

R

4

3

2

1

0

8

7

6

5

4

3

2

1

0

8

7

6

5

4

3

2

1

0

8

7

6

5

4

3

2

1

0

VK_FORMAT_X8_D24_UNORM_PACK32

X

D

7

6

5

4

3

2

1

0

23

22

21

20

19

18

17

16

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

Depth/Stencil Formats

Depth/stencil formats are considered opaque and need not be stored in the exact number of bits per texel or component ordering indicated by the format enum. However, implementations must not substitute a different depth or stencil precision than that described in the format (e.g. D16 must not be implemented as D24 or D32).

Format Compatibility Classes

Uncompressed color formats are compatible with each other if they occupy the same number of bits per data element. Compressed color formats are compatible with each other if the only difference between them is the numerical type of the uncompressed pixels (e.g. signed vs. unsigned, or SRGB vs. UNORM encoding). Each depth/stencil format is only compatible with itself. In the following table, all the formats in the same row are compatible.

Table 54. Compatible formats
Class Formats

8-bit

VK_FORMAT_R4G4_UNORM_PACK8,
VK_FORMAT_R8_UNORM,
VK_FORMAT_R8_SNORM,
VK_FORMAT_R8_USCALED,
VK_FORMAT_R8_SSCALED,
VK_FORMAT_R8_UINT,
VK_FORMAT_R8_SINT,
VK_FORMAT_R8_SRGB

16-bit

VK_FORMAT_R4G4B4A4_UNORM_PACK16,
VK_FORMAT_B4G4R4A4_UNORM_PACK16,
VK_FORMAT_R5G6B5_UNORM_PACK16,
VK_FORMAT_B5G6R5_UNORM_PACK16,
VK_FORMAT_R5G5B5A1_UNORM_PACK16,
VK_FORMAT_B5G5R5A1_UNORM_PACK16,
VK_FORMAT_A1R5G5B5_UNORM_PACK16,
VK_FORMAT_R8G8_UNORM,
VK_FORMAT_R8G8_SNORM,
VK_FORMAT_R8G8_USCALED,
VK_FORMAT_R8G8_SSCALED,
VK_FORMAT_R8G8_UINT,
VK_FORMAT_R8G8_SINT,
VK_FORMAT_R8G8_SRGB,
VK_FORMAT_R16_UNORM,
VK_FORMAT_R16_SNORM,
VK_FORMAT_R16_USCALED,
VK_FORMAT_R16_SSCALED,
VK_FORMAT_R16_UINT,
VK_FORMAT_R16_SINT,
VK_FORMAT_R16_SFLOAT,
VK_FORMAT_R10X6_UNORM_PACK16,
VK_FORMAT_R12X4_UNORM_PACK16

24-bit

VK_FORMAT_R8G8B8_UNORM,
VK_FORMAT_R8G8B8_SNORM,
VK_FORMAT_R8G8B8_USCALED,
VK_FORMAT_R8G8B8_SSCALED,
VK_FORMAT_R8G8B8_UINT,
VK_FORMAT_R8G8B8_SINT,
VK_FORMAT_R8G8B8_SRGB,
VK_FORMAT_B8G8R8_UNORM,
VK_FORMAT_B8G8R8_SNORM,
VK_FORMAT_B8G8R8_USCALED,
VK_FORMAT_B8G8R8_SSCALED,
VK_FORMAT_B8G8R8_UINT,
VK_FORMAT_B8G8R8_SINT,
VK_FORMAT_B8G8R8_SRGB

32-bit

VK_FORMAT_R8G8B8A8_UNORM,
VK_FORMAT_R8G8B8A8_SNORM,
VK_FORMAT_R8G8B8A8_USCALED,
VK_FORMAT_R8G8B8A8_SSCALED,
VK_FORMAT_R8G8B8A8_UINT,
VK_FORMAT_R8G8B8A8_SINT,
VK_FORMAT_R8G8B8A8_SRGB,
VK_FORMAT_B8G8R8A8_UNORM,
VK_FORMAT_B8G8R8A8_SNORM,
VK_FORMAT_B8G8R8A8_USCALED,
VK_FORMAT_B8G8R8A8_SSCALED,
VK_FORMAT_B8G8R8A8_UINT,
VK_FORMAT_B8G8R8A8_SINT,
VK_FORMAT_B8G8R8A8_SRGB,
VK_FORMAT_A8B8G8R8_UNORM_PACK32,
VK_FORMAT_A8B8G8R8_SNORM_PACK32,
VK_FORMAT_A8B8G8R8_USCALED_PACK32,
VK_FORMAT_A8B8G8R8_SSCALED_PACK32,
VK_FORMAT_A8B8G8R8_UINT_PACK32,
VK_FORMAT_A8B8G8R8_SINT_PACK32,
VK_FORMAT_A8B8G8R8_SRGB_PACK32,
VK_FORMAT_A2R10G10B10_UNORM_PACK32,
VK_FORMAT_A2R10G10B10_SNORM_PACK32,
VK_FORMAT_A2R10G10B10_USCALED_PACK32,
VK_FORMAT_A2R10G10B10_SSCALED_PACK32,
VK_FORMAT_A2R10G10B10_UINT_PACK32,
VK_FORMAT_A2R10G10B10_SINT_PACK32,
VK_FORMAT_A2B10G10R10_UNORM_PACK32,
VK_FORMAT_A2B10G10R10_SNORM_PACK32,
VK_FORMAT_A2B10G10R10_USCALED_PACK32,
VK_FORMAT_A2B10G10R10_SSCALED_PACK32,
VK_FORMAT_A2B10G10R10_UINT_PACK32,
VK_FORMAT_A2B10G10R10_SINT_PACK32,
VK_FORMAT_R16G16_UNORM,
VK_FORMAT_R16G16_SNORM,
VK_FORMAT_R16G16_USCALED,
VK_FORMAT_R16G16_SSCALED,
VK_FORMAT_R16G16_UINT,
VK_FORMAT_R16G16_SINT,
VK_FORMAT_R16G16_SFLOAT,
VK_FORMAT_R32_UINT,
VK_FORMAT_R32_SINT,
VK_FORMAT_R32_SFLOAT,
VK_FORMAT_B10G11R11_UFLOAT_PACK32,
VK_FORMAT_E5B9G9R9_UFLOAT_PACK32,
VK_FORMAT_R10X6G10X6_UNORM_2PACK16,
VK_FORMAT_R12X4G12X4_UNORM_2PACK16

32-bit G8B8G8R8

VK_FORMAT_G8B8G8R8_422_UNORM

32-bit B8G8R8G8

VK_FORMAT_B8G8R8G8_422_UNORM

48-bit

VK_FORMAT_R16G16B16_UNORM,
VK_FORMAT_R16G16B16_SNORM,
VK_FORMAT_R16G16B16_USCALED,
VK_FORMAT_R16G16B16_SSCALED,
VK_FORMAT_R16G16B16_UINT,
VK_FORMAT_R16G16B16_SINT,
VK_FORMAT_R16G16B16_SFLOAT

64-bit

VK_FORMAT_R16G16B16A16_UNORM,
VK_FORMAT_R16G16B16A16_SNORM,
VK_FORMAT_R16G16B16A16_USCALED,
VK_FORMAT_R16G16B16A16_SSCALED,
VK_FORMAT_R16G16B16A16_UINT,
VK_FORMAT_R16G16B16A16_SINT,
VK_FORMAT_R16G16B16A16_SFLOAT,
VK_FORMAT_R32G32_UINT,
VK_FORMAT_R32G32_SINT,
VK_FORMAT_R32G32_SFLOAT,
VK_FORMAT_R64_UINT,
VK_FORMAT_R64_SINT,
VK_FORMAT_R64_SFLOAT

64-bit R10G10B10A10

VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16

64-bit G10B10G10R10

VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16

64-bit B10G10R10G10

VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16

64-bit R12G12B12A12

VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16

64-bit G12B12G12R12

VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16

64-bit B12G12R12G12

VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16

64-bit G16B16G16R16

VK_FORMAT_G16B16G16R16_422_UNORM

64-bit B16G16R16G16

VK_FORMAT_B16G16R16G16_422_UNORM

96-bit

VK_FORMAT_R32G32B32_UINT,
VK_FORMAT_R32G32B32_SINT,
VK_FORMAT_R32G32B32_SFLOAT

128-bit

VK_FORMAT_R32G32B32A32_UINT,
VK_FORMAT_R32G32B32A32_SINT,
VK_FORMAT_R32G32B32A32_SFLOAT,
VK_FORMAT_R64G64_UINT,
VK_FORMAT_R64G64_SINT,
VK_FORMAT_R64G64_SFLOAT

192-bit

VK_FORMAT_R64G64B64_UINT,
VK_FORMAT_R64G64B64_SINT,
VK_FORMAT_R64G64B64_SFLOAT

256-bit

VK_FORMAT_R64G64B64A64_UINT,
VK_FORMAT_R64G64B64A64_SINT,
VK_FORMAT_R64G64B64A64_SFLOAT

BC1_RGB

VK_FORMAT_BC1_RGB_UNORM_BLOCK,
VK_FORMAT_BC1_RGB_SRGB_BLOCK

BC1_RGBA

VK_FORMAT_BC1_RGBA_UNORM_BLOCK,
VK_FORMAT_BC1_RGBA_SRGB_BLOCK

BC2

VK_FORMAT_BC2_UNORM_BLOCK,
VK_FORMAT_BC2_SRGB_BLOCK

BC3

VK_FORMAT_BC3_UNORM_BLOCK,
VK_FORMAT_BC3_SRGB_BLOCK

BC4

VK_FORMAT_BC4_UNORM_BLOCK,
VK_FORMAT_BC4_SNORM_BLOCK

BC5

VK_FORMAT_BC5_UNORM_BLOCK,
VK_FORMAT_BC5_SNORM_BLOCK

BC6H

VK_FORMAT_BC6H_UFLOAT_BLOCK,
VK_FORMAT_BC6H_SFLOAT_BLOCK

BC7

VK_FORMAT_BC7_UNORM_BLOCK,
VK_FORMAT_BC7_SRGB_BLOCK

ETC2_RGB

VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK,
VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK

ETC2_RGBA

VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK,
VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK

ETC2_EAC_RGBA

VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK,
VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK

EAC_R

VK_FORMAT_EAC_R11_UNORM_BLOCK,
VK_FORMAT_EAC_R11_SNORM_BLOCK

EAC_RG

VK_FORMAT_EAC_R11G11_UNORM_BLOCK,
VK_FORMAT_EAC_R11G11_SNORM_BLOCK

ASTC_4x4

VK_FORMAT_ASTC_4x4_UNORM_BLOCK,
VK_FORMAT_ASTC_4x4_SRGB_BLOCK

ASTC_5x4

VK_FORMAT_ASTC_5x4_UNORM_BLOCK,
VK_FORMAT_ASTC_5x4_SRGB_BLOCK

ASTC_5x5

VK_FORMAT_ASTC_5x5_UNORM_BLOCK,
VK_FORMAT_ASTC_5x5_SRGB_BLOCK

ASTC_6x5

VK_FORMAT_ASTC_6x5_UNORM_BLOCK,
VK_FORMAT_ASTC_6x5_SRGB_BLOCK

ASTC_6x6

VK_FORMAT_ASTC_6x6_UNORM_BLOCK,
VK_FORMAT_ASTC_6x6_SRGB_BLOCK

ASTC_8x5

VK_FORMAT_ASTC_8x5_UNORM_BLOCK,
VK_FORMAT_ASTC_8x5_SRGB_BLOCK

ASTC_8x6

VK_FORMAT_ASTC_8x6_UNORM_BLOCK,
VK_FORMAT_ASTC_8x6_SRGB_BLOCK

ASTC_8x8

VK_FORMAT_ASTC_8x8_UNORM_BLOCK,
VK_FORMAT_ASTC_8x8_SRGB_BLOCK

ASTC_10x5

VK_FORMAT_ASTC_10x5_UNORM_BLOCK,
VK_FORMAT_ASTC_10x5_SRGB_BLOCK

ASTC_10x6

VK_FORMAT_ASTC_10x6_UNORM_BLOCK,
VK_FORMAT_ASTC_10x6_SRGB_BLOCK

ASTC_10x8

VK_FORMAT_ASTC_10x8_UNORM_BLOCK,
VK_FORMAT_ASTC_10x8_SRGB_BLOCK

ASTC_10x10

VK_FORMAT_ASTC_10x10_UNORM_BLOCK,
VK_FORMAT_ASTC_10x10_SRGB_BLOCK

ASTC_12x10

VK_FORMAT_ASTC_12x10_UNORM_BLOCK,
VK_FORMAT_ASTC_12x10_SRGB_BLOCK

ASTC_12x12

VK_FORMAT_ASTC_12x12_UNORM_BLOCK,
VK_FORMAT_ASTC_12x12_SRGB_BLOCK

D16

VK_FORMAT_D16_UNORM

D24

VK_FORMAT_X8_D24_UNORM_PACK32

D32

VK_FORMAT_D32_SFLOAT

S8

VK_FORMAT_S8_UINT

D16S8

VK_FORMAT_D16_UNORM_S8_UINT

D24S8

VK_FORMAT_D24_UNORM_S8_UINT

D32S8

VK_FORMAT_D32_SFLOAT_S8_UINT

8-bit 3-plane 420

VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM

8-bit 2-plane 420

VK_FORMAT_G8_B8R8_2PLANE_420_UNORM

8-bit 3-plane 422

VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM

8-bit 2-plane 422

VK_FORMAT_G8_B8R8_2PLANE_422_UNORM

8-bit 3-plane 444

VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM

10-bit 3-plane 420

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16

10-bit 2-plane 420

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16

10-bit 3-plane 422

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16

10-bit 2-plane 422

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16

10-bit 3-plane 444

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16

12-bit 3-plane 420

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16

12-bit 2-plane 420

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16

12-bit 3-plane 422

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16

12-bit 2-plane 422

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16

12-bit 3-plane 444

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16

16-bit 3-plane 420

VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM

16-bit 2-plane 420

VK_FORMAT_G16_B16R16_2PLANE_420_UNORM

16-bit 3-plane 422

VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM

16-bit 2-plane 422

VK_FORMAT_G16_B16R16_2PLANE_422_UNORM

16-bit 3-plane 444

VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM

32.4.2. Format Properties

To query supported format features which are properties of the physical device, call:

void vkGetPhysicalDeviceFormatProperties(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkFormatProperties*                         pFormatProperties);
  • physicalDevice is the physical device from which to query the format properties.

  • format is the format whose properties are queried.

  • pFormatProperties is a pointer to a VkFormatProperties structure in which physical device properties for format are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • format must be a valid VkFormat value

  • pFormatProperties must be a valid pointer to a VkFormatProperties structure

The VkFormatProperties structure is defined as:

typedef struct VkFormatProperties {
    VkFormatFeatureFlags    linearTilingFeatures;
    VkFormatFeatureFlags    optimalTilingFeatures;
    VkFormatFeatureFlags    bufferFeatures;
} VkFormatProperties;
  • linearTilingFeatures is a bitmask of VkFormatFeatureFlagBits specifying features supported by images created with a tiling parameter of VK_IMAGE_TILING_LINEAR.

  • optimalTilingFeatures is a bitmask of VkFormatFeatureFlagBits specifying features supported by images created with a tiling parameter of VK_IMAGE_TILING_OPTIMAL.

  • bufferFeatures is a bitmask of VkFormatFeatureFlagBits specifying features supported by buffers.

Note

If no format feature flags are supported, the format itself is not supported, and images of that format cannot be created.

If format is a block-compression format, then buffers must not support any features for the format.

Bits which can be set in the VkFormatProperties features linearTilingFeatures, optimalTilingFeatures, and bufferFeatures are:

typedef enum VkFormatFeatureFlagBits {
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT = 0x00000001,
    VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT = 0x00000002,
    VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT = 0x00000004,
    VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT = 0x00000008,
    VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT = 0x00000010,
    VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT = 0x00000020,
    VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT = 0x00000040,
    VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT = 0x00000080,
    VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT = 0x00000100,
    VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT = 0x00000200,
    VK_FORMAT_FEATURE_BLIT_SRC_BIT = 0x00000400,
    VK_FORMAT_FEATURE_BLIT_DST_BIT = 0x00000800,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT = 0x00001000,
    VK_FORMAT_FEATURE_TRANSFER_SRC_BIT = 0x00004000,
    VK_FORMAT_FEATURE_TRANSFER_DST_BIT = 0x00008000,
    VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT = 0x00020000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT = 0x00040000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT = 0x00080000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT = 0x00100000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT = 0x00200000,
    VK_FORMAT_FEATURE_DISJOINT_BIT = 0x00400000,
    VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT = 0x00800000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG = 0x00002000,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT_EXT = 0x00010000,
    VK_FORMAT_FEATURE_TRANSFER_SRC_BIT_KHR = VK_FORMAT_FEATURE_TRANSFER_SRC_BIT,
    VK_FORMAT_FEATURE_TRANSFER_DST_BIT_KHR = VK_FORMAT_FEATURE_TRANSFER_DST_BIT,
    VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT_KHR = VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT_KHR = VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT_KHR = VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT_KHR = VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT,
    VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT_KHR = VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT,
    VK_FORMAT_FEATURE_DISJOINT_BIT_KHR = VK_FORMAT_FEATURE_DISJOINT_BIT,
    VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT_KHR = VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT,
} VkFormatFeatureFlagBits;

The following bits may be set in linearTilingFeatures and optimalTilingFeatures, specifying that the features are supported by images or image views created with the queried vkGetPhysicalDeviceFormatProperties::format:

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT specifies that an image view can be sampled from.

  • VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT specifies that an image view can be used as a storage images.

  • VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT specifies that an image view can be used as storage image that supports atomic operations.

  • VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT specifies that an image view can be used as a framebuffer color attachment and as an input attachment.

  • VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT specifies that an image view can be used as a framebuffer color attachment that supports blending and as an input attachment.

  • VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT specifies that an image view can be used as a framebuffer depth/stencil attachment and as an input attachment.

  • VK_FORMAT_FEATURE_BLIT_SRC_BIT specifies that an image can be used as srcImage for the vkCmdBlitImage command.

  • VK_FORMAT_FEATURE_BLIT_DST_BIT specifies that an image can be used as dstImage for the vkCmdBlitImage command.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT specifies that if VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT is also set, an image view can be used with a sampler that has either of magFilter or minFilter set to VK_FILTER_LINEAR, or mipmapMode set to VK_SAMPLER_MIPMAP_MODE_LINEAR. If VK_FORMAT_FEATURE_BLIT_SRC_BIT is also set, an image can be used as the srcImage to vkCmdBlitImage with a filter of VK_FILTER_LINEAR. This bit must only be exposed for formats that also support the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT or VK_FORMAT_FEATURE_BLIT_SRC_BIT.

    If the format being queried is a depth/stencil format, this bit only specifies that the depth aspect (not the stencil aspect) of an image of this format supports linear filtering, and that linear filtering of the depth aspect is supported whether depth compare is enabled in the sampler or not. If this bit is not present, linear filtering with depth compare disabled is unsupported and linear filtering with depth compare enabled is supported, but may compute the filtered value in an implementation-dependent manner which differs from the normal rules of linear filtering. The resulting value must be in the range [0,1] and should be proportional to, or a weighted average of, the number of comparison passes or failures.

  • VK_FORMAT_FEATURE_TRANSFER_SRC_BIT specifies that an image can be used as a source image for copy commands.

  • VK_FORMAT_FEATURE_TRANSFER_DST_BIT specifies that an image can be used as a destination image for copy commands and clear commands.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT_EXT specifies VkImage can be used as a sampled image with a min or max VkSamplerReductionModeEXT. This bit must only be exposed for formats that also support the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG specifies that VkImage can be used with a sampler that has either of magFilter or minFilter set to VK_FILTER_CUBIC_IMG, or be the source image for a blit with filter set to VK_FILTER_CUBIC_IMG. This bit must only be exposed for formats that also support the VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT. If the format being queried is a depth/stencil format, this only specifies that the depth aspect is cubic filterable.

  • VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT specifies that an application can define a sampler Y’CBCR conversion using this format as a source, and that an image of this format can be used with a VkSamplerYcbcrConversionCreateInfo xChromaOffset and/or yChromaOffset of VK_CHROMA_LOCATION_MIDPOINT. Otherwise both xChromaOffset and yChromaOffset must be VK_CHROMA_LOCATION_COSITED_EVEN. If a format does not incorporate chroma downsampling (it is not a “422” or “420” format) but the implementation supports sampler Y’CBCR conversion for this format, the implementation must set VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT.

  • VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT specifies that an application can define a sampler Y’CBCR conversion using this format as a source, and that an image of this format can be used with a VkSamplerYcbcrConversionCreateInfo xChromaOffset and/or yChromaOffset of VK_CHROMA_LOCATION_COSITED_EVEN. Otherwise both xChromaOffset and yChromaOffset must be VK_CHROMA_LOCATION_MIDPOINT. If neither VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT nor VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT is set, the application must not define a sampler Y’CBCR conversion using this format as a source.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT specifies that the format can do linear sampler filtering (min/magFilter) whilst sampler Y’CBCR conversion is enabled.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT specifies that the format can have different chroma, min, and mag filters.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT specifies that reconstruction is explicit, as described in Chroma Reconstruction. If this bit is not present, reconstruction is implicit by default.

  • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT specifies that reconstruction can be forcibly made explicit by setting VkSamplerYcbcrConversionCreateInfo::forceExplicitReconstruction to VK_TRUE.

  • VK_FORMAT_FEATURE_DISJOINT_BIT specifies that a multi-planar image can have the VK_IMAGE_CREATE_DISJOINT_BIT set during image creation. An implementation must not set VK_FORMAT_FEATURE_DISJOINT_BIT for single-plane formats.

The following bits may be set in bufferFeatures, specifying that the features are supported by buffers or buffer views created with the queried vkGetPhysicalDeviceProperties::format:

  • VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT specifies that the format can be used to create a buffer view that can be bound to a VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER descriptor.

  • VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT specifies that the format can be used to create a buffer view that can be bound to a VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER descriptor.

  • VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT specifies that atomic operations are supported on VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER with this format.

  • VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT specifies that the format can be used as a vertex attribute format (VkVertexInputAttributeDescription::format).

typedef VkFlags VkFormatFeatureFlags;

VkFormatFeatureFlags is a bitmask type for setting a mask of zero or more VkFormatFeatureFlagBits.

To query supported format features which are properties of the physical device, call:

void vkGetPhysicalDeviceFormatProperties2(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkFormatProperties2*                        pFormatProperties);

or the equivalent command

void vkGetPhysicalDeviceFormatProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkFormatProperties2*                        pFormatProperties);
  • physicalDevice is the physical device from which to query the format properties.

  • format is the format whose properties are queried.

  • pFormatProperties is a pointer to a VkFormatProperties2 structure in which physical device properties for format are returned.

vkGetPhysicalDeviceFormatProperties2 behaves similarly to vkGetPhysicalDeviceFormatProperties, with the ability to return extended information in a pNext chain of output structures.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • format must be a valid VkFormat value

  • pFormatProperties must be a valid pointer to a VkFormatProperties2 structure

The VkFormatProperties2 structure is defined as:

typedef struct VkFormatProperties2 {
    VkStructureType       sType;
    void*                 pNext;
    VkFormatProperties    formatProperties;
} VkFormatProperties2;

or the equivalent

typedef VkFormatProperties2 VkFormatProperties2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • formatProperties is a structure of type VkFormatProperties describing features supported by the requested format.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2

  • pNext must be NULL

32.4.3. Required Format Support

Implementations must support at least the following set of features on the listed formats. For images, these features must be supported for every VkImageType (including arrayed and cube variants) unless otherwise noted. These features are supported on existing formats without needing to advertise an extension or needing to explicitly enable them. Support for additional functionality beyond the requirements listed here is queried using the vkGetPhysicalDeviceFormatProperties command.

The following tables show which feature bits must be supported for each format. Formats that are required to support VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT must also support VK_FORMAT_FEATURE_TRANSFER_SRC_BIT and VK_FORMAT_FEATURE_TRANSFER_DST_BIT.

Table 55. Key for format feature tables

This feature must be supported on the named format

This feature must be supported on at least some of the named formats, with more information in the table where the symbol appears

Table 56. Feature bits in optimalTilingFeatures

VK_FORMAT_FEATURE_TRANSFER_SRC_BIT

VK_FORMAT_FEATURE_TRANSFER_DST_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT_EXT

Table 57. Feature bits in bufferFeatures

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

Table 58. Mandatory format support: sub-byte channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_UNDEFINED

VK_FORMAT_R4G4_UNORM_PACK8

VK_FORMAT_R4G4B4A4_UNORM_PACK16

VK_FORMAT_B4G4R4A4_UNORM_PACK16

VK_FORMAT_R5G6B5_UNORM_PACK16

VK_FORMAT_B5G6R5_UNORM_PACK16

VK_FORMAT_R5G5B5A1_UNORM_PACK16

VK_FORMAT_B5G5R5A1_UNORM_PACK16

VK_FORMAT_A1R5G5B5_UNORM_PACK16

Table 59. Mandatory format support: 1-3 byte-sized channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_R8_UNORM

VK_FORMAT_R8_SNORM

VK_FORMAT_R8_USCALED

VK_FORMAT_R8_SSCALED

VK_FORMAT_R8_UINT

VK_FORMAT_R8_SINT

VK_FORMAT_R8_SRGB

VK_FORMAT_R8G8_UNORM

VK_FORMAT_R8G8_SNORM

VK_FORMAT_R8G8_USCALED

VK_FORMAT_R8G8_SSCALED

VK_FORMAT_R8G8_UINT

VK_FORMAT_R8G8_SINT

VK_FORMAT_R8G8_SRGB

VK_FORMAT_R8G8B8_UNORM

VK_FORMAT_R8G8B8_SNORM

VK_FORMAT_R8G8B8_USCALED

VK_FORMAT_R8G8B8_SSCALED

VK_FORMAT_R8G8B8_UINT

VK_FORMAT_R8G8B8_SINT

VK_FORMAT_R8G8B8_SRGB

VK_FORMAT_B8G8R8_UNORM

VK_FORMAT_B8G8R8_SNORM

VK_FORMAT_B8G8R8_USCALED

VK_FORMAT_B8G8R8_SSCALED

VK_FORMAT_B8G8R8_UINT

VK_FORMAT_B8G8R8_SINT

VK_FORMAT_B8G8R8_SRGB

Table 60. Mandatory format support: 4 byte-sized channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_R8G8B8A8_UNORM

VK_FORMAT_R8G8B8A8_SNORM

VK_FORMAT_R8G8B8A8_USCALED

VK_FORMAT_R8G8B8A8_SSCALED

VK_FORMAT_R8G8B8A8_UINT

VK_FORMAT_R8G8B8A8_SINT

VK_FORMAT_R8G8B8A8_SRGB

VK_FORMAT_B8G8R8A8_UNORM

VK_FORMAT_B8G8R8A8_SNORM

VK_FORMAT_B8G8R8A8_USCALED

VK_FORMAT_B8G8R8A8_SSCALED

VK_FORMAT_B8G8R8A8_UINT

VK_FORMAT_B8G8R8A8_SINT

VK_FORMAT_B8G8R8A8_SRGB

VK_FORMAT_A8B8G8R8_UNORM_PACK32

VK_FORMAT_A8B8G8R8_SNORM_PACK32

VK_FORMAT_A8B8G8R8_USCALED_PACK32

VK_FORMAT_A8B8G8R8_SSCALED_PACK32

VK_FORMAT_A8B8G8R8_UINT_PACK32

VK_FORMAT_A8B8G8R8_SINT_PACK32

VK_FORMAT_A8B8G8R8_SRGB_PACK32

Table 61. Mandatory format support: 10-bit channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_A2R10G10B10_UNORM_PACK32

VK_FORMAT_A2R10G10B10_SNORM_PACK32

VK_FORMAT_A2R10G10B10_USCALED_PACK32

VK_FORMAT_A2R10G10B10_SSCALED_PACK32

VK_FORMAT_A2R10G10B10_UINT_PACK32

VK_FORMAT_A2R10G10B10_SINT_PACK32

VK_FORMAT_A2B10G10R10_UNORM_PACK32

VK_FORMAT_A2B10G10R10_SNORM_PACK32

VK_FORMAT_A2B10G10R10_USCALED_PACK32

VK_FORMAT_A2B10G10R10_SSCALED_PACK32

VK_FORMAT_A2B10G10R10_UINT_PACK32

VK_FORMAT_A2B10G10R10_SINT_PACK32

Table 62. Mandatory format support: 16-bit channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_R16_UNORM

VK_FORMAT_R16_SNORM

VK_FORMAT_R16_USCALED

VK_FORMAT_R16_SSCALED

VK_FORMAT_R16_UINT

VK_FORMAT_R16_SINT

VK_FORMAT_R16_SFLOAT

VK_FORMAT_R16G16_UNORM

VK_FORMAT_R16G16_SNORM

VK_FORMAT_R16G16_USCALED

VK_FORMAT_R16G16_SSCALED

VK_FORMAT_R16G16_UINT

VK_FORMAT_R16G16_SINT

VK_FORMAT_R16G16_SFLOAT

VK_FORMAT_R16G16B16_UNORM

VK_FORMAT_R16G16B16_SNORM

VK_FORMAT_R16G16B16_USCALED

VK_FORMAT_R16G16B16_SSCALED

VK_FORMAT_R16G16B16_UINT

VK_FORMAT_R16G16B16_SINT

VK_FORMAT_R16G16B16_SFLOAT

VK_FORMAT_R16G16B16A16_UNORM

VK_FORMAT_R16G16B16A16_SNORM

VK_FORMAT_R16G16B16A16_USCALED

VK_FORMAT_R16G16B16A16_SSCALED

VK_FORMAT_R16G16B16A16_UINT

VK_FORMAT_R16G16B16A16_SINT

VK_FORMAT_R16G16B16A16_SFLOAT

Table 63. Mandatory format support: 32-bit channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_R32_UINT

VK_FORMAT_R32_SINT

VK_FORMAT_R32_SFLOAT

VK_FORMAT_R32G32_UINT

VK_FORMAT_R32G32_SINT

VK_FORMAT_R32G32_SFLOAT

VK_FORMAT_R32G32B32_UINT

VK_FORMAT_R32G32B32_SINT

VK_FORMAT_R32G32B32_SFLOAT

VK_FORMAT_R32G32B32A32_UINT

VK_FORMAT_R32G32B32A32_SINT

VK_FORMAT_R32G32B32A32_SFLOAT

Table 64. Mandatory format support: 64-bit/uneven channels

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_R64_UINT

VK_FORMAT_R64_SINT

VK_FORMAT_R64_SFLOAT

VK_FORMAT_R64G64_UINT

VK_FORMAT_R64G64_SINT

VK_FORMAT_R64G64_SFLOAT

VK_FORMAT_R64G64B64_UINT

VK_FORMAT_R64G64B64_SINT

VK_FORMAT_R64G64B64_SFLOAT

VK_FORMAT_R64G64B64A64_UINT

VK_FORMAT_R64G64B64A64_SINT

VK_FORMAT_R64G64B64A64_SFLOAT

VK_FORMAT_B10G11R11_UFLOAT_PACK32

VK_FORMAT_E5B9G9R9_UFLOAT_PACK32

Table 65. Mandatory format support: depth/stencil with VkImageType VK_IMAGE_TYPE_2D

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_D16_UNORM

VK_FORMAT_X8_D24_UNORM_PACK32

VK_FORMAT_D32_SFLOAT

VK_FORMAT_S8_UINT

VK_FORMAT_D16_UNORM_S8_UINT

VK_FORMAT_D24_UNORM_S8_UINT

VK_FORMAT_D32_SFLOAT_S8_UINT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT feature must be supported for at least one of VK_FORMAT_X8_D24_UNORM_PACK32 and VK_FORMAT_D32_SFLOAT, and must be supported for at least one of VK_FORMAT_D24_UNORM_S8_UINT and VK_FORMAT_D32_SFLOAT_S8_UINT.

Table 66. Mandatory format support: BC compressed formats with VkImageType VK_IMAGE_TYPE_2D and VK_IMAGE_TYPE_3D

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_BC1_RGB_UNORM_BLOCK

VK_FORMAT_BC1_RGB_SRGB_BLOCK

VK_FORMAT_BC1_RGBA_UNORM_BLOCK

VK_FORMAT_BC1_RGBA_SRGB_BLOCK

VK_FORMAT_BC2_UNORM_BLOCK

VK_FORMAT_BC2_SRGB_BLOCK

VK_FORMAT_BC3_UNORM_BLOCK

VK_FORMAT_BC3_SRGB_BLOCK

VK_FORMAT_BC4_UNORM_BLOCK

VK_FORMAT_BC4_SNORM_BLOCK

VK_FORMAT_BC5_UNORM_BLOCK

VK_FORMAT_BC5_SNORM_BLOCK

VK_FORMAT_BC6H_UFLOAT_BLOCK

VK_FORMAT_BC6H_SFLOAT_BLOCK

VK_FORMAT_BC7_UNORM_BLOCK

VK_FORMAT_BC7_SRGB_BLOCK

The VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for all the formats in at least one of: this table, Mandatory format support: ETC2 and EAC compressed formats with VkImageType VK_IMAGE_TYPE_2D, or Mandatory format support: ASTC LDR compressed formats with VkImageType VK_IMAGE_TYPE_2D.

Table 67. Mandatory format support: ETC2 and EAC compressed formats with VkImageType VK_IMAGE_TYPE_2D

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK

VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK

VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK

VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK

VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK

VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK

VK_FORMAT_EAC_R11_UNORM_BLOCK

VK_FORMAT_EAC_R11_SNORM_BLOCK

VK_FORMAT_EAC_R11G11_UNORM_BLOCK

VK_FORMAT_EAC_R11G11_SNORM_BLOCK

The VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for all the formats in at least one of: this table, Mandatory format support: BC compressed formats with VkImageType VK_IMAGE_TYPE_2D and VK_IMAGE_TYPE_3D, or Mandatory format support: ASTC LDR compressed formats with VkImageType VK_IMAGE_TYPE_2D.

Table 68. Mandatory format support: ASTC LDR compressed formats with VkImageType VK_IMAGE_TYPE_2D

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_UNIFORM_TEXEL_BUFFER_BIT

VK_FORMAT_FEATURE_VERTEX_BUFFER_BIT

VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT

VK_FORMAT_FEATURE_BLIT_DST_BIT

VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

VK_FORMAT_FEATURE_BLIT_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

Format

VK_FORMAT_ASTC_4x4_UNORM_BLOCK

VK_FORMAT_ASTC_4x4_SRGB_BLOCK

VK_FORMAT_ASTC_5x4_UNORM_BLOCK

VK_FORMAT_ASTC_5x4_SRGB_BLOCK

VK_FORMAT_ASTC_5x5_UNORM_BLOCK

VK_FORMAT_ASTC_5x5_SRGB_BLOCK

VK_FORMAT_ASTC_6x5_UNORM_BLOCK

VK_FORMAT_ASTC_6x5_SRGB_BLOCK

VK_FORMAT_ASTC_6x6_UNORM_BLOCK

VK_FORMAT_ASTC_6x6_SRGB_BLOCK

VK_FORMAT_ASTC_8x5_UNORM_BLOCK

VK_FORMAT_ASTC_8x5_SRGB_BLOCK

VK_FORMAT_ASTC_8x6_UNORM_BLOCK

VK_FORMAT_ASTC_8x6_SRGB_BLOCK

VK_FORMAT_ASTC_8x8_UNORM_BLOCK

VK_FORMAT_ASTC_8x8_SRGB_BLOCK

VK_FORMAT_ASTC_10x5_UNORM_BLOCK

VK_FORMAT_ASTC_10x5_SRGB_BLOCK

VK_FORMAT_ASTC_10x6_UNORM_BLOCK

VK_FORMAT_ASTC_10x6_SRGB_BLOCK

VK_FORMAT_ASTC_10x8_UNORM_BLOCK

VK_FORMAT_ASTC_10x8_SRGB_BLOCK

VK_FORMAT_ASTC_10x10_UNORM_BLOCK

VK_FORMAT_ASTC_10x10_SRGB_BLOCK

VK_FORMAT_ASTC_12x10_UNORM_BLOCK

VK_FORMAT_ASTC_12x10_SRGB_BLOCK

VK_FORMAT_ASTC_12x12_UNORM_BLOCK

VK_FORMAT_ASTC_12x12_SRGB_BLOCK

The VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT, VK_FORMAT_FEATURE_BLIT_SRC_BIT and VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT features must be supported in optimalTilingFeatures for all the formats in at least one of: this table, Mandatory format support: BC compressed formats with VkImageType VK_IMAGE_TYPE_2D and VK_IMAGE_TYPE_3D, or Mandatory format support: ETC2 and EAC compressed formats with VkImageType VK_IMAGE_TYPE_2D.

VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG must be supported for the following formats:

  • VK_FORMAT_R4G4_UNORM_PACK8

  • VK_FORMAT_R4G4B4A4_UNORM_PACK16

  • VK_FORMAT_B4G4R4A4_UNORM_PACK16

  • VK_FORMAT_R5G6B5_UNORM_PACK16

  • VK_FORMAT_B5G6R5_UNORM_PACK16

  • VK_FORMAT_R5G5B5A1_UNORM_PACK16

  • VK_FORMAT_B5G5R5A1_UNORM_PACK16

  • VK_FORMAT_A1R5G5B5_UNORM_PACK16

  • VK_FORMAT_R8_UNORM

  • VK_FORMAT_R8_SNORM

  • VK_FORMAT_R8_SRGB

  • VK_FORMAT_R8G8_UNORM

  • VK_FORMAT_R8G8_SNORM

  • VK_FORMAT_R8G8_SRGB

  • VK_FORMAT_R8G8B8_UNORM

  • VK_FORMAT_R8G8B8_SNORM

  • VK_FORMAT_R8G8B8_SRGB

  • VK_FORMAT_B8G8R8_UNORM

  • VK_FORMAT_B8G8R8_SNORM

  • VK_FORMAT_B8G8R8_SRGB

  • VK_FORMAT_R8G8B8A8_UNORM

  • VK_FORMAT_R8G8B8A8_SNORM

  • VK_FORMAT_R8G8B8A8_SRGB

  • VK_FORMAT_B8G8R8A8_UNORM

  • VK_FORMAT_B8G8R8A8_SNORM

  • VK_FORMAT_B8G8R8A8_SRGB

  • VK_FORMAT_A8B8G8R8_UNORM_PACK32

  • VK_FORMAT_A8B8G8R8_SNORM_PACK32

  • VK_FORMAT_A8B8G8R8_USCALED_PACK32

  • VK_FORMAT_A8B8G8R8_SSCALED_PACK32

  • VK_FORMAT_A8B8G8R8_UINT_PACK32

  • VK_FORMAT_A8B8G8R8_SINT_PACK32

  • VK_FORMAT_A8B8G8R8_SRGB_PACK32

If ETC2 compressed formats are supported, the following additional formats must support VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_IMG:

  • VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK

  • VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK

  • VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK

  • VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK

  • VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK

  • VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK

To be used with VkImageView with subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, sampler Y’CBCR conversion must be enabled for the following formats:

Table 69. Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views

Format must be supported if VkPhysicalDeviceSamplerYcbcrConversionFeatures is enabled

Format is treated as having 2×1 texel blocks by transfer operations

VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT

VK_FORMAT_FEATURE_TRANSFER_DST_BIT

VK_FORMAT_FEATURE_TRANSFER_SRC_BIT

VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT

VK_FORMAT_FEATURE_DISJOINT_BIT

Multi-planar format with three planes

Multi-planar format with two planes

Single-plane format

Format

VK_FORMAT_G8B8G8R8_422_UNORM

VK_FORMAT_B8G8R8G8_422_UNORM

VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM

VK_FORMAT_G8_B8R8_2PLANE_420_UNORM

VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM

VK_FORMAT_G8_B8R8_2PLANE_422_UNORM

VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM

VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16

VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16

VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16

VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16

VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16

VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16

VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16

VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16

VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16

VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16

VK_FORMAT_G16B16G16R16_422_UNORM

VK_FORMAT_B16G16R16G16_422_UNORM

VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM

VK_FORMAT_G16_B16R16_2PLANE_420_UNORM

VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM

VK_FORMAT_G16_B16R16_2PLANE_422_UNORM

VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM

Format features marked ✓ must be supported if the format is supported

Format features marked † may be supported by the format

32.5. Additional Image Capabilities

In addition to the minimum capabilities described in the previous sections (Limits and Formats), implementations may support additional capabilities for certain types of images. For example, larger dimensions or additional sample counts for certain image types, or additional capabilities for linear tiling format images.

To query additional capabilities specific to image types, call:

VkResult vkGetPhysicalDeviceImageFormatProperties(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkImageType                                 type,
    VkImageTiling                               tiling,
    VkImageUsageFlags                           usage,
    VkImageCreateFlags                          flags,
    VkImageFormatProperties*                    pImageFormatProperties);

The format, type, tiling, usage, and flags parameters correspond to parameters that would be consumed by vkCreateImage (as members of VkImageCreateInfo).

If format is not a supported image format, or if the combination of format, type, tiling, usage, and flags is not supported for images, then vkGetPhysicalDeviceImageFormatProperties returns VK_ERROR_FORMAT_NOT_SUPPORTED.

The limitations on an image format that are reported by vkGetPhysicalDeviceImageFormatProperties have the following property: if usage1 and usage2 of type VkImageUsageFlags are such that the bits set in usage1 are a subset of the bits set in usage2, and flags1 and flags2 of type VkImageCreateFlags are such that the bits set in flags1 are a subset of the bits set in flags2, then the limitations for usage1 and flags1 must be no more strict than the limitations for usage2 and flags2, for all values of format, type, and tiling.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • format must be a valid VkFormat value

  • type must be a valid VkImageType value

  • tiling must be a valid VkImageTiling value

  • usage must be a valid combination of VkImageUsageFlagBits values

  • usage must not be 0

  • flags must be a valid combination of VkImageCreateFlagBits values

  • pImageFormatProperties must be a valid pointer to a VkImageFormatProperties structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_FORMAT_NOT_SUPPORTED

The VkImageFormatProperties structure is defined as:

typedef struct VkImageFormatProperties {
    VkExtent3D            maxExtent;
    uint32_t              maxMipLevels;
    uint32_t              maxArrayLayers;
    VkSampleCountFlags    sampleCounts;
    VkDeviceSize          maxResourceSize;
} VkImageFormatProperties;
  • maxExtent are the maximum image dimensions. See the Allowed Extent Values section below for how these values are constrained by type.

  • maxMipLevels is the maximum number of mipmap levels. maxMipLevels must be equal to ⌈log2(max(width, height, depth))⌉ + 1, where width, height, and depth are taken from the corresponding members of maxExtent, except when one of the following conditions is true, in which case it may instead be 1:

  • maxArrayLayers is the maximum number of array layers. maxArrayLayers must either be equal to 1 or be greater than or equal to the maxImageArrayLayers member of VkPhysicalDeviceLimits. A value of 1 is valid only if tiling is VK_IMAGE_TILING_LINEAR or if type is VK_IMAGE_TYPE_3D.

  • sampleCounts is a bitmask of VkSampleCountFlagBits specifying all the supported sample counts for this image as described below.

  • maxResourceSize is an upper bound on the total image size in bytes, inclusive of all image subresources. Implementations may have an address space limit on total size of a resource, which is advertised by this property. maxResourceSize must be at least 231.

Note

There is no mechanism to query the size of an image before creating it, to compare that size against maxResourceSize. If an application attempts to create an image that exceeds this limit, the creation will fail and vkCreateImage will return VK_ERROR_OUT_OF_DEVICE_MEMORY. While the advertised limit must be at least 231, it may not be possible to create an image that approaches that size, particularly for VK_IMAGE_TYPE_1D.

If the combination of parameters to vkGetPhysicalDeviceImageFormatProperties is not supported by the implementation for use in vkCreateImage, then all members of VkImageFormatProperties will be filled with zero.

Note

Filling VkImageFormatProperties with zero for unsupported formats is an exception to the usual rule that output structures have undefined contents on error. This exception was unintentional, but is preserved for backwards compatibility.

To determine the image capabilities compatible with an external memory handle type, call:

VkResult vkGetPhysicalDeviceExternalImageFormatPropertiesNV(
    VkPhysicalDevice                            physicalDevice,
    VkFormat                                    format,
    VkImageType                                 type,
    VkImageTiling                               tiling,
    VkImageUsageFlags                           usage,
    VkImageCreateFlags                          flags,
    VkExternalMemoryHandleTypeFlagsNV           externalHandleType,
    VkExternalImageFormatPropertiesNV*          pExternalImageFormatProperties);

If externalHandleType is 0, pExternalImageFormatProperties::imageFormatProperties will return the same values as a call to vkGetPhysicalDeviceImageFormatProperties, and the other members of pExternalImageFormatProperties will all be 0. Otherwise, they are filled in as described for VkExternalImageFormatPropertiesNV.

Valid Usage (Implicit)
Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_FORMAT_NOT_SUPPORTED

The VkExternalImageFormatPropertiesNV structure is defined as:

typedef struct VkExternalImageFormatPropertiesNV {
    VkImageFormatProperties              imageFormatProperties;
    VkExternalMemoryFeatureFlagsNV       externalMemoryFeatures;
    VkExternalMemoryHandleTypeFlagsNV    exportFromImportedHandleTypes;
    VkExternalMemoryHandleTypeFlagsNV    compatibleHandleTypes;
} VkExternalImageFormatPropertiesNV;

Bits which can be set in VkExternalMemoryFeatureFlagBitsNV::externalMemoryFeatures, indicating properties of the external memory handle type, are:

typedef enum VkExternalMemoryFeatureFlagBitsNV {
    VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_NV = 0x00000001,
    VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT_NV = 0x00000002,
    VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT_NV = 0x00000004,
} VkExternalMemoryFeatureFlagBitsNV;
  • VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_NV specifies that external memory of the specified type must be created as a dedicated allocation when used in the manner specified.

  • VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT_NV specifies that the implementation supports exporting handles of the specified type.

  • VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT_NV specifies that the implementation supports importing handles of the specified type.

typedef VkFlags VkExternalMemoryFeatureFlagsNV;

VkExternalMemoryFeatureFlagsNV is a bitmask type for setting a mask of zero or more VkExternalMemoryFeatureFlagBitsNV.

To query additional capabilities specific to image types, call:

VkResult vkGetPhysicalDeviceImageFormatProperties2(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceImageFormatInfo2*     pImageFormatInfo,
    VkImageFormatProperties2*                   pImageFormatProperties);

or the equivalent command

VkResult vkGetPhysicalDeviceImageFormatProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceImageFormatInfo2*     pImageFormatInfo,
    VkImageFormatProperties2*                   pImageFormatProperties);
  • physicalDevice is the physical device from which to query the image capabilities.

  • pImageFormatInfo points to an instance of the VkPhysicalDeviceImageFormatInfo2 structure, describing the parameters that would be consumed by vkCreateImage.

  • pImageFormatProperties points to an instance of the VkImageFormatProperties2 structure in which capabilities are returned.

vkGetPhysicalDeviceImageFormatProperties2 behaves similarly to vkGetPhysicalDeviceImageFormatProperties, with the ability to return extended information in a pNext chain of output structures.

Valid Usage
Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pImageFormatInfo must be a valid pointer to a valid VkPhysicalDeviceImageFormatInfo2 structure

  • pImageFormatProperties must be a valid pointer to a VkImageFormatProperties2 structure

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

  • VK_ERROR_FORMAT_NOT_SUPPORTED

The VkPhysicalDeviceImageFormatInfo2 structure is defined as:

typedef struct VkPhysicalDeviceImageFormatInfo2 {
    VkStructureType       sType;
    const void*           pNext;
    VkFormat              format;
    VkImageType           type;
    VkImageTiling         tiling;
    VkImageUsageFlags     usage;
    VkImageCreateFlags    flags;
} VkPhysicalDeviceImageFormatInfo2;

or the equivalent

typedef VkPhysicalDeviceImageFormatInfo2 VkPhysicalDeviceImageFormatInfo2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure. The pNext chain of VkPhysicalDeviceImageFormatInfo2 is used to provide additional image parameters to vkGetPhysicalDeviceImageFormatProperties2.

  • format is a VkFormat value indicating the image format, corresponding to VkImageCreateInfo::format.

  • type is a VkImageType value indicating the image type, corresponding to VkImageCreateInfo::imageType.

  • tiling is a VkImageTiling value indicating the image tiling, corresponding to VkImageCreateInfo::tiling.

  • usage is a bitmask of VkImageUsageFlagBits indicating the intended usage of the image, corresponding to VkImageCreateInfo::usage.

  • flags is a bitmask of VkImageCreateFlagBits indicating additional parameters of the image, corresponding to VkImageCreateInfo::flags.

The members of VkPhysicalDeviceImageFormatInfo2 correspond to the arguments to vkGetPhysicalDeviceImageFormatProperties, with sType and pNext added for extensibility.

Valid Usage (Implicit)

The VkImageFormatProperties2 structure is defined as:

typedef struct VkImageFormatProperties2 {
    VkStructureType            sType;
    void*                      pNext;
    VkImageFormatProperties    imageFormatProperties;
} VkImageFormatProperties2;

or the equivalent

typedef VkImageFormatProperties2 VkImageFormatProperties2KHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure. The pNext chain of VkImageFormatProperties2 is used to allow the specification of additional capabilities to be returned from vkGetPhysicalDeviceImageFormatProperties2.

  • imageFormatProperties is an instance of a VkImageFormatProperties structure in which capabilities are returned.

If the combination of parameters to vkGetPhysicalDeviceImageFormatProperties2 is not supported by the implementation for use in vkCreateImage, then all members of imageFormatProperties will be filled with zero.

Note

Filling imageFormatProperties with zero for unsupported formats is an exception to the usual rule that output structures have undefined contents on error. This exception was unintentional, but is preserved for backwards compatibility. This exeption only applies to imageFormatProperties, not sType, pNext, or any structures chained from pNext.

Valid Usage (Implicit)

To determine if texture gather functions that take explicit LOD and/or bias argument values can be used with a given image format, add VkImageFormatProperties2 to the pNext chain of the VkPhysicalDeviceImageFormatInfo2 structure and VkTextureLODGatherFormatPropertiesAMD to the pNext chain of the VkImageFormatProperties2 structure.

The VkTextureLODGatherFormatPropertiesAMD structure is defined as:

typedef struct VkTextureLODGatherFormatPropertiesAMD {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           supportsTextureGatherLODBiasAMD;
} VkTextureLODGatherFormatPropertiesAMD;
  • sType is the type of this structure.

  • pNext is NULL.

  • supportsTextureGatherLODBiasAMD tells if the image format can be used with texture gather bias/LOD functions, as introduced by the VK_AMD_texture_gather_bias_lod extension. This field is set by the implementation. User-specified value is ignored.

To determine the image capabilities compatible with an external memory handle type, add VkPhysicalDeviceExternalImageFormatInfo to the pNext chain of the VkPhysicalDeviceImageFormatInfo2 structure and VkExternalImageFormatProperties to the pNext chain of the VkImageFormatProperties2 structure.

The VkPhysicalDeviceExternalImageFormatInfo structure is defined as:

typedef struct VkPhysicalDeviceExternalImageFormatInfo {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalMemoryHandleTypeFlagBits    handleType;
} VkPhysicalDeviceExternalImageFormatInfo;

or the equivalent

typedef VkPhysicalDeviceExternalImageFormatInfo VkPhysicalDeviceExternalImageFormatInfoKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType is a VkExternalMemoryHandleTypeFlagBits value specifying the memory handle type that will be used with the memory associated with the image.

If handleType is 0, vkGetPhysicalDeviceImageFormatProperties2 will behave as if VkPhysicalDeviceExternalImageFormatInfo was not present, and VkExternalImageFormatProperties will be ignored.

If handleType is not compatible with the format, type, tiling, usage, and flags specified in VkPhysicalDeviceImageFormatInfo2, then vkGetPhysicalDeviceImageFormatProperties2 returns VK_ERROR_FORMAT_NOT_SUPPORTED.

Valid Usage (Implicit)

Possible values of VkPhysicalDeviceExternalImageFormatInfo::handleType, specifying an external memory handle type, are:

typedef enum VkExternalMemoryHandleTypeFlagBits {
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT = 0x00000001,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT = 0x00000002,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT = 0x00000004,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT = 0x00000008,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT = 0x00000010,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT = 0x00000020,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT = 0x00000040,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT = 0x00000200,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID = 0x00000400,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT = 0x00000080,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT = 0x00000100,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT,
    VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT_KHR = VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT,
} VkExternalMemoryHandleTypeFlagBits;

or the equivalent

typedef VkExternalMemoryHandleTypeFlagBits VkExternalMemoryHandleTypeFlagBitsKHR;
  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT specifies a POSIX file descriptor handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the POSIX system calls dup, dup2, close, and the non-standard system call dup3. Additionally, it must be transportable over a socket using an SCM_RIGHTS control message. It owns a reference to the underlying memory resource represented by its Vulkan memory object.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT specifies an NT handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the functions DuplicateHandle, CloseHandle, CompareObjectHandles, GetHandleInformation, and SetHandleInformation. It owns a reference to the underlying memory resource represented by its Vulkan memory object.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT specifies a global share handle that has only limited valid usage outside of Vulkan and other compatible APIs. It is not compatible with any native APIs. It does not own a reference to the underlying memory resource represented its Vulkan memory object, and will therefore become invalid when all Vulkan memory objects associated with it are destroyed.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT specifies an NT handle returned by IDXGIResource1::CreateSharedHandle referring to a Direct3D 10 or 11 texture resource. It owns a reference to the memory used by the Direct3D resource.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT specifies a global share handle returned by IDXGIResource::GetSharedHandle referring to a Direct3D 10 or 11 texture resource. It does not own a reference to the underlying Direct3D resource, and will therefore become invalid when all Vulkan memory objects and Direct3D resources associated with it are destroyed.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT specifies an NT handle returned by ID3D12Device::CreateSharedHandle referring to a Direct3D 12 heap resource. It owns a reference to the resources used by the Direct3D heap.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT specifies an NT handle returned by ID3D12Device::CreateSharedHandle referring to a Direct3D 12 committed resource. It owns a reference to the memory used by the Direct3D resource.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT specifies a host pointer returned by a host memory allocation command. It does not own a reference to the underlying memory resource, and will therefore become invalid if the host memory is freed.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT specifies a host pointer to host mapped foreign memory. It does not own a reference to the underlying memory resource, and will therefore become invalid if the foreign memory is unmapped or otherwise becomes no longer available.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT is a file descriptor for a Linux dma_buf. It owns a reference to the underlying memory resource represented by its Vulkan memory object.

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID specifies an AHardwareBuffer object defined by the Android NDK. See Android Hardware Buffers for more details of this handle type.

Some external memory handle types can only be shared within the same underlying physical device and/or the same driver version, as defined in the following table:

Table 70. External memory handle types compatibility

Handle type

VkPhysicalDeviceIDProperties::driverUUID

VkPhysicalDeviceIDProperties::deviceUUID

VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT

Must match

Must match

VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT

No restriction

No restriction

VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT

No restriction

No restriction

VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT

No restriction

No restriction

VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID

No restriction

No restriction

Note

The above table does not restrict the drivers and devices with which VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT and VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT may be shared, as these handle types inherently mean memory that does not come from the same device, as they import memory from the host or a foreign device, respectively.

Note

Even though the above table does not restrict the drivers and devices with which VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT may be shared, query mechanisms exist in the Vulkan API that prevent the import of incompatible dma-bufs (such as vkGetMemoryFdPropertiesKHR) and that prevent incompatible usage of dma-bufs (such as VkPhysicalDeviceExternalBufferInfoKHR and VkPhysicalDeviceExternalImageFormatInfoKHR).

typedef VkFlags VkExternalMemoryHandleTypeFlags;

or the equivalent

typedef VkExternalMemoryHandleTypeFlags VkExternalMemoryHandleTypeFlagsKHR;

VkExternalMemoryHandleTypeFlags is a bitmask type for setting a mask of zero or more VkExternalMemoryHandleTypeFlagBits.

The VkExternalImageFormatProperties structure is defined as:

typedef struct VkExternalImageFormatProperties {
    VkStructureType               sType;
    void*                         pNext;
    VkExternalMemoryProperties    externalMemoryProperties;
} VkExternalImageFormatProperties;

or the equivalent

typedef VkExternalImageFormatProperties VkExternalImageFormatPropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • externalMemoryProperties is an instance of the VkExternalMemoryProperties structure specifying various capabilities of the external handle type when used with the specified image creation parameters.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES

The VkExternalMemoryProperties structure is defined as:

typedef struct VkExternalMemoryProperties {
    VkExternalMemoryFeatureFlags       externalMemoryFeatures;
    VkExternalMemoryHandleTypeFlags    exportFromImportedHandleTypes;
    VkExternalMemoryHandleTypeFlags    compatibleHandleTypes;
} VkExternalMemoryProperties;

or the equivalent

typedef VkExternalMemoryProperties VkExternalMemoryPropertiesKHR;

compatibleHandleTypes must include at least handleType. Inclusion of a handle type in compatibleHandleTypes does not imply the values returned in VkImageFormatProperties2 will be the same when VkPhysicalDeviceExternalImageFormatInfo::handleType is set to that type. The application is responsible for querying the capabilities of all handle types intended for concurrent use in a single image and intersecting them to obtain the compatible set of capabilities.

Bits which may be set in VkExternalMemoryProperties::externalMemoryFeatures, specifying features of an external memory handle type, are:

typedef enum VkExternalMemoryFeatureFlagBits {
    VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT = 0x00000001,
    VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT = 0x00000002,
    VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT = 0x00000004,
    VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_KHR = VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT,
    VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT_KHR = VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT,
    VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT_KHR = VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT,
} VkExternalMemoryFeatureFlagBits;

or the equivalent

typedef VkExternalMemoryFeatureFlagBits VkExternalMemoryFeatureFlagBitsKHR;
  • VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT specifies that images or buffers created with the specified parameters and handle type must use the mechanisms defined in the VK_NV_dedicated_allocation extension to create (or import) a dedicated allocation for the image or buffer.

  • VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT specifies that handles of this type can be exported from Vulkan memory objects.

  • VK_INTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT specifies that handles of this type can be imported as Vulkan memory objects.

Because their semantics in external APIs roughly align with that of an image or buffer with a dedicated allocation in Vulkan, implementations are required to report VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT for the following external handle types:

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_KMT_BIT

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID for images only

Implementations must not report VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT for buffers with external handle type VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID.

typedef VkFlags VkExternalMemoryFeatureFlags;

or the equivalent

typedef VkExternalMemoryFeatureFlags VkExternalMemoryFeatureFlagsKHR;

VkExternalMemoryFeatureFlags is a bitmask type for setting a mask of zero or more VkExternalMemoryFeatureFlagBits.

To determine the number of combined image samplers required to support a multi-planar format, add VkSamplerYcbcrConversionImageFormatProperties to the pNext chain of the VkImageFormatProperties2 structure in a call to vkGetPhysicalDeviceImageFormatProperties2.

The VkSamplerYcbcrConversionImageFormatProperties structure is defined as:

typedef struct VkSamplerYcbcrConversionImageFormatProperties {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           combinedImageSamplerDescriptorCount;
} VkSamplerYcbcrConversionImageFormatProperties;

or the equivalent

typedef VkSamplerYcbcrConversionImageFormatProperties VkSamplerYcbcrConversionImageFormatPropertiesKHR;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • combinedImageSamplerDescriptorCount is the number of combined image sampler descriptors that the implementation uses to access the format.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_IMAGE_FORMAT_PROPERTIES

combinedImageSamplerDescriptorCount affects only the count towards the maxDescriptorSetSamplers, maxDescriptorSetSampledImages, maxPerStageDescriptorSamplers, and maxPerStageDescriptorSampledImages limits, and does not affect binding numbers in the VkDescriptorSetLayoutBinding.

combinedImageSamplerDescriptorCount is a number between 1 and the number of planes in the format.

To obtain optimal Android hardware buffer usage flags for specific image creation parameters, attach an instance of VkAndroidHardwareBufferUsageANDROID to the pNext chain of a VkImageFormatProperties2 structure passed to vkGetPhysicalDeviceImageFormatProperties2. This structure is defined as:

typedef struct VkAndroidHardwareBufferUsageANDROID {
    VkStructureType    sType;
    void*              pNext;
    uint64_t           androidHardwareBufferUsage;
} VkAndroidHardwareBufferUsageANDROID;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • androidHardwareBufferUsage returns the the Android hardware buffer usage flags.

The androidHardwareBufferUsage field must include Android hardware buffer usage flags listed in the AHardwareBuffer Usage Equivalence table when the corresponding Vulkan image usage or image creation flags are included in the usage or flags fields of VkPhysicalDeviceImageFormatInfo2. It must include at least one GPU usage flag (AHARDWAREBUFFER_USAGE_GPU_*), even if none of the corresponding Vulkan usages or flags are requested.

Note

Requiring at least one GPU usage flag ensures that Android hardware buffer memory will be allocated in a memory pool accessible to the Vulkan implementation, and that specializing the memory layout based on usage flags doesn’t prevent it from being compatible with Vulkan. Implementations may avoid unnecessary restrictions caused by this requirement by using vendor usage flags to indicate that only the Vulkan uses indicated in VkImageFormatProperties2 are required.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_USAGE_ANDROID

32.5.1. Supported Sample Counts

vkGetPhysicalDeviceImageFormatProperties returns a bitmask of VkSampleCountFlagBits in sampleCounts specifying the supported sample counts for the image parameters.

sampleCounts will be set to VK_SAMPLE_COUNT_1_BIT if at least one of the following conditions is true:

  • tiling is VK_IMAGE_TILING_LINEAR

  • type is not VK_IMAGE_TYPE_2D

  • flags contains VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT

  • Neither the VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT flag nor the VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT flag in VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties is set

  • VkPhysicalDeviceExternalImageFormatInfoKHR::handleType is an external handle type for which multisampled image support is not required.

Otherwise, the bits set in sampleCounts will be the sample counts supported for the specified values of usage and format. For each bit set in usage, the supported sample counts relate to the limits in VkPhysicalDeviceLimits as follows:

  • If usage includes VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT and format is a floating- or fixed-point color format, a superset of VkPhysicalDeviceLimits::framebufferColorSampleCounts

  • If usage includes VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, and format includes a depth aspect, a superset of VkPhysicalDeviceLimits::framebufferDepthSampleCounts

  • If usage includes VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, and format includes a stencil aspect, a superset of VkPhysicalDeviceLimits::framebufferStencilSampleCounts

  • If usage includes VK_IMAGE_USAGE_SAMPLED_BIT, and format includes a color aspect, a superset of VkPhysicalDeviceLimits::sampledImageColorSampleCounts

  • If usage includes VK_IMAGE_USAGE_SAMPLED_BIT, and format includes a depth aspect, a superset of VkPhysicalDeviceLimits::sampledImageDepthSampleCounts

  • If usage includes VK_IMAGE_USAGE_SAMPLED_BIT, and format is an integer format, a superset of VkPhysicalDeviceLimits::sampledImageIntegerSampleCounts

  • If usage includes VK_IMAGE_USAGE_STORAGE_BIT, a superset of VkPhysicalDeviceLimits::storageImageSampleCounts

If multiple bits are set in usage, sampleCounts will be the intersection of the per-usage values described above.

If none of the bits described above are set in usage, then there is no corresponding limit in VkPhysicalDeviceLimits. In this case, sampleCounts must include at least VK_SAMPLE_COUNT_1_BIT.

32.5.2. Allowed Extent Values Based On Image Type

Implementations may support extent values larger than the required minimum/maximum values for certain types of images subject to the constraints below.

Note

Implementations must support images with dimensions up to the required minimum/maximum values for all types of images. It follows that the query for additional capabilities must return extent values that are at least as large as the required values.

For VK_IMAGE_TYPE_1D:

For VK_IMAGE_TYPE_2D when flags does not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT:

For VK_IMAGE_TYPE_2D when flags contains VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT:

For VK_IMAGE_TYPE_3D:

32.6. Additional Buffer Capabilities

In addition to the capabilities described in the previous sections (Limits and Formats), implementations may support additional buffer capabilities.

To query the external handle types supported by buffers, call:

void vkGetPhysicalDeviceExternalBufferProperties(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalBufferInfo*   pExternalBufferInfo,
    VkExternalBufferProperties*                 pExternalBufferProperties);

or the equivalent command

void vkGetPhysicalDeviceExternalBufferPropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalBufferInfo*   pExternalBufferInfo,
    VkExternalBufferProperties*                 pExternalBufferProperties);
  • physicalDevice is the physical device from which to query the buffer capabilities.

  • pExternalBufferInfo points to an instance of the VkPhysicalDeviceExternalBufferInfo structure, describing the parameters that would be consumed by vkCreateBuffer.

  • pExternalBufferProperties points to an instance of the VkExternalBufferProperties structure in which capabilities are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pExternalBufferInfo must be a valid pointer to a valid VkPhysicalDeviceExternalBufferInfo structure

  • pExternalBufferProperties must be a valid pointer to a VkExternalBufferProperties structure

The VkPhysicalDeviceExternalBufferInfo structure is defined as:

typedef struct VkPhysicalDeviceExternalBufferInfo {
    VkStructureType                       sType;
    const void*                           pNext;
    VkBufferCreateFlags                   flags;
    VkBufferUsageFlags                    usage;
    VkExternalMemoryHandleTypeFlagBits    handleType;
} VkPhysicalDeviceExternalBufferInfo;

or the equivalent

typedef VkPhysicalDeviceExternalBufferInfo VkPhysicalDeviceExternalBufferInfoKHR;
Valid Usage (Implicit)

The VkExternalBufferProperties structure is defined as:

typedef struct VkExternalBufferProperties {
    VkStructureType               sType;
    void*                         pNext;
    VkExternalMemoryProperties    externalMemoryProperties;
} VkExternalBufferProperties;

or the equivalent

typedef VkExternalBufferProperties VkExternalBufferPropertiesKHR;
  • sType is the type of this structure

  • pNext is NULL or a pointer to an extension-specific structure.

  • externalMemoryProperties is an instance of the VkExternalMemoryProperties structure specifying various capabilities of the external handle type when used with the specified buffer creation parameters.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES

  • pNext must be NULL

32.7. Optional Semaphore Capabilities

Semaphores may support import and export of their payload to external handles. To query the external handle types supported by semaphores, call:

void vkGetPhysicalDeviceExternalSemaphoreProperties(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalSemaphoreInfo* pExternalSemaphoreInfo,
    VkExternalSemaphoreProperties*              pExternalSemaphoreProperties);

or the equivalent command

void vkGetPhysicalDeviceExternalSemaphorePropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalSemaphoreInfo* pExternalSemaphoreInfo,
    VkExternalSemaphoreProperties*              pExternalSemaphoreProperties);
Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pExternalSemaphoreInfo must be a valid pointer to a valid VkPhysicalDeviceExternalSemaphoreInfo structure

  • pExternalSemaphoreProperties must be a valid pointer to a VkExternalSemaphoreProperties structure

The VkPhysicalDeviceExternalSemaphoreInfo structure is defined as:

typedef struct VkPhysicalDeviceExternalSemaphoreInfo {
    VkStructureType                          sType;
    const void*                              pNext;
    VkExternalSemaphoreHandleTypeFlagBits    handleType;
} VkPhysicalDeviceExternalSemaphoreInfo;

or the equivalent

typedef VkPhysicalDeviceExternalSemaphoreInfo VkPhysicalDeviceExternalSemaphoreInfoKHR;
  • sType is the type of this structure

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType is a VkExternalSemaphoreHandleTypeFlagBits value specifying the external semaphore handle type for which capabilities will be returned.

Valid Usage (Implicit)

Bits which may be set in VkPhysicalDeviceExternalSemaphoreInfo::handleType, specifying an external semaphore handle type, are:

typedef enum VkExternalSemaphoreHandleTypeFlagBits {
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT = 0x00000001,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT = 0x00000002,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT = 0x00000004,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT = 0x00000008,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT = 0x00000010,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT_KHR = VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR = VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_KHR = VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT_KHR = VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT,
    VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT_KHR = VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT,
} VkExternalSemaphoreHandleTypeFlagBits;

or the equivalent

typedef VkExternalSemaphoreHandleTypeFlagBits VkExternalSemaphoreHandleTypeFlagBitsKHR;
  • VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT specifies a POSIX file descriptor handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the POSIX system calls dup, dup2, close, and the non-standard system call dup3. Additionally, it must be transportable over a socket using an SCM_RIGHTS control message. It owns a reference to the underlying synchronization primitive represented by its Vulkan semaphore object.

  • VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT specifies an NT handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the functions DuplicateHandle, CloseHandle, CompareObjectHandles, GetHandleInformation, and SetHandleInformation. It owns a reference to the underlying synchronization primitive represented by its Vulkan semaphore object.

  • VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT specifies a global share handle that has only limited valid usage outside of Vulkan and other compatible APIs. It is not compatible with any native APIs. It does not own a reference to the underlying synchronization primitive represented its Vulkan semaphore object, and will therefore become invalid when all Vulkan semaphore objects associated with it are destroyed.

  • VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT specifies an NT handle returned by ID3D12Device::CreateSharedHandle referring to a Direct3D 12 fence. It owns a reference to the underlying synchronization primitive associated with the Direct3D fence.

  • VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT specifies a POSIX file descriptor handle to a Linux Sync File or Android Fence object. It can be used with any native API accepting a valid sync file or fence as input. It owns a reference to the underlying synchronization primitive associated with the file descriptor. Implementations which support importing this handle type must accept any type of sync or fence FD supported by the native system they are running on.

Note

Handles of type VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT generated by the implementation may represent either Linux Sync Files or Android Fences at the implementation’s discretion. Applications should only use operations defined for both types of file descriptors, unless they know via means external to Vulkan the type of the file descriptor, or are prepared to deal with the system-defined operation failures resulting from using the wrong type.

Some external semaphore handle types can only be shared within the same underlying physical device and/or the same driver version, as defined in the following table:

Table 71. External semaphore handle types compatibility

Handle type

VkPhysicalDeviceIDProperties::driverUUID

VkPhysicalDeviceIDProperties::deviceUUID

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT

Must match

Must match

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT

Must match

Must match

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT

Must match

Must match

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT

Must match

Must match

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_SYNC_FD_BIT

No restriction

No restriction

typedef VkFlags VkExternalSemaphoreHandleTypeFlags;

or the equivalent

typedef VkExternalSemaphoreHandleTypeFlags VkExternalSemaphoreHandleTypeFlagsKHR;

VkExternalSemaphoreHandleTypeFlags is a bitmask type for setting a mask of zero or more VkExternalSemaphoreHandleTypeFlagBits.

The VkExternalSemaphoreProperties structure is defined as:

typedef struct VkExternalSemaphoreProperties {
    VkStructureType                       sType;
    void*                                 pNext;
    VkExternalSemaphoreHandleTypeFlags    exportFromImportedHandleTypes;
    VkExternalSemaphoreHandleTypeFlags    compatibleHandleTypes;
    VkExternalSemaphoreFeatureFlags       externalSemaphoreFeatures;
} VkExternalSemaphoreProperties;

or the equivalent

typedef VkExternalSemaphoreProperties VkExternalSemaphorePropertiesKHR;

If handleType is not supported by the implementation, then VkExternalSemaphoreProperties::externalSemaphoreFeatures will be set to zero.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES

  • pNext must be NULL

Possible values of VkExternalSemaphoreProperties::externalSemaphoreFeatures, specifying the features of an external semaphore handle type, are:

typedef enum VkExternalSemaphoreFeatureFlagBits {
    VK_EXTERNAL_SEMAPHORE_FEATURE_EXPORTABLE_BIT = 0x00000001,
    VK_EXTERNAL_SEMAPHORE_FEATURE_IMPORTABLE_BIT = 0x00000002,
    VK_EXTERNAL_SEMAPHORE_FEATURE_EXPORTABLE_BIT_KHR = VK_EXTERNAL_SEMAPHORE_FEATURE_EXPORTABLE_BIT,
    VK_EXTERNAL_SEMAPHORE_FEATURE_IMPORTABLE_BIT_KHR = VK_EXTERNAL_SEMAPHORE_FEATURE_IMPORTABLE_BIT,
} VkExternalSemaphoreFeatureFlagBits;

or the equivalent

typedef VkExternalSemaphoreFeatureFlagBits VkExternalSemaphoreFeatureFlagBitsKHR;
  • VK_EXTERNAL_SEMAPHORE_FEATURE_EXPORTABLE_BIT specifies that handles of this type can be exported from Vulkan semaphore objects.

  • VK_EXTERNAL_SEMAPHORE_FEATURE_IMPORTABLE_BIT specifies that handles of this type can be imported as Vulkan semaphore objects.

typedef VkFlags VkExternalSemaphoreFeatureFlags;

or the equivalent

typedef VkExternalSemaphoreFeatureFlags VkExternalSemaphoreFeatureFlagsKHR;

VkExternalSemaphoreFeatureFlags is a bitmask type for setting a mask of zero or more VkExternalSemaphoreFeatureFlagBits.

32.8. Optional Fence Capabilities

Fences may support import and export of their payload to external handles. To query the external handle types supported by fences, call:

void vkGetPhysicalDeviceExternalFenceProperties(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalFenceInfo*    pExternalFenceInfo,
    VkExternalFenceProperties*                  pExternalFenceProperties);

or the equivalent command

void vkGetPhysicalDeviceExternalFencePropertiesKHR(
    VkPhysicalDevice                            physicalDevice,
    const VkPhysicalDeviceExternalFenceInfo*    pExternalFenceInfo,
    VkExternalFenceProperties*                  pExternalFenceProperties);
  • physicalDevice is the physical device from which to query the fence capabilities.

  • pExternalFenceInfo points to an instance of the VkPhysicalDeviceExternalFenceInfo structure, describing the parameters that would be consumed by vkCreateFence.

  • pExternalFenceProperties points to an instance of the VkExternalFenceProperties structure in which capabilities are returned.

Valid Usage (Implicit)
  • physicalDevice must be a valid VkPhysicalDevice handle

  • pExternalFenceInfo must be a valid pointer to a valid VkPhysicalDeviceExternalFenceInfo structure

  • pExternalFenceProperties must be a valid pointer to a VkExternalFenceProperties structure

The VkPhysicalDeviceExternalFenceInfo structure is defined as:

typedef struct VkPhysicalDeviceExternalFenceInfo {
    VkStructureType                      sType;
    const void*                          pNext;
    VkExternalFenceHandleTypeFlagBits    handleType;
} VkPhysicalDeviceExternalFenceInfo;

or the equivalent

typedef VkPhysicalDeviceExternalFenceInfo VkPhysicalDeviceExternalFenceInfoKHR;
  • sType is the type of this structure

  • pNext is NULL or a pointer to an extension-specific structure.

  • handleType is a VkExternalFenceHandleTypeFlagBits value indicating an external fence handle type for which capabilities will be returned.

Note

Handles of type VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT generated by the implementation may represent either Linux Sync Files or Android Fences at the implementation’s discretion. Applications should only use operations defined for both types of file descriptors, unless they know via means external to Vulkan the type of the file descriptor, or are prepared to deal with the system-defined operation failures resulting from using the wrong type.

Valid Usage (Implicit)

Bits which may be set in VkPhysicalDeviceExternalFenceInfo::handleType, and in the exportFromImportedHandleTypes and compatibleHandleTypes members of VkExternalFenceProperties, to indicate external fence handle types, are:

typedef enum VkExternalFenceHandleTypeFlagBits {
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT = 0x00000001,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT = 0x00000002,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT = 0x00000004,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT = 0x00000008,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT_KHR = VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR = VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT_KHR = VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT,
    VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT_KHR = VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT,
} VkExternalFenceHandleTypeFlagBits;

or the equivalent

typedef VkExternalFenceHandleTypeFlagBits VkExternalFenceHandleTypeFlagBitsKHR;
  • VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT specifies a POSIX file descriptor handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the POSIX system calls dup, dup2, close, and the non-standard system call dup3. Additionally, it must be transportable over a socket using an SCM_RIGHTS control message. It owns a reference to the underlying synchronization primitive represented by its Vulkan fence object.

  • VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT specifies an NT handle that has only limited valid usage outside of Vulkan and other compatible APIs. It must be compatible with the functions DuplicateHandle, CloseHandle, CompareObjectHandles, GetHandleInformation, and SetHandleInformation. It owns a reference to the underlying synchronization primitive represented by its Vulkan fence object.

  • VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT specifies a global share handle that has only limited valid usage outside of Vulkan and other compatible APIs. It is not compatible with any native APIs. It does not own a reference to the underlying synchronization primitive represented by its Vulkan fence object, and will therefore become invalid when all Vulkan fence objects associated with it are destroyed.

  • VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT specifies a POSIX file descriptor handle to a Linux Sync File or Android Fence. It can be used with any native API accepting a valid sync file or fence as input. It owns a reference to the underlying synchronization primitive associated with the file descriptor. Implementations which support importing this handle type must accept any type of sync or fence FD supported by the native system they are running on.

Some external fence handle types can only be shared within the same underlying physical device and/or the same driver version, as defined in the following table:

Table 72. External fence handle types compatibility

Handle type

VkPhysicalDeviceIDProperties::driverUUID

VkPhysicalDeviceIDProperties::deviceUUID

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_FD_BIT

Must match

Must match

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT

Must match

Must match

VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT

Must match

Must match

VK_EXTERNAL_FENCE_HANDLE_TYPE_SYNC_FD_BIT

No restriction

No restriction

typedef VkFlags VkExternalFenceHandleTypeFlags;

or the equivalent

typedef VkExternalFenceHandleTypeFlags VkExternalFenceHandleTypeFlagsKHR;

VkExternalFenceHandleTypeFlags is a bitmask type for setting a mask of zero or more VkExternalFenceHandleTypeFlagBits.

The VkExternalFenceProperties structure is defined as:

typedef struct VkExternalFenceProperties {
    VkStructureType                   sType;
    void*                             pNext;
    VkExternalFenceHandleTypeFlags    exportFromImportedHandleTypes;
    VkExternalFenceHandleTypeFlags    compatibleHandleTypes;
    VkExternalFenceFeatureFlags       externalFenceFeatures;
} VkExternalFenceProperties;

or the equivalent

typedef VkExternalFenceProperties VkExternalFencePropertiesKHR;

If handleType is not supported by the implementation, then VkExternalFenceProperties::externalFenceFeatures will be set to zero.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES

  • pNext must be NULL

Bits which may be set in VkExternalFenceProperties::externalFenceFeatures, indicating features of a fence external handle type, are:

typedef enum VkExternalFenceFeatureFlagBits {
    VK_EXTERNAL_FENCE_FEATURE_EXPORTABLE_BIT = 0x00000001,
    VK_EXTERNAL_FENCE_FEATURE_IMPORTABLE_BIT = 0x00000002,
    VK_EXTERNAL_FENCE_FEATURE_EXPORTABLE_BIT_KHR = VK_EXTERNAL_FENCE_FEATURE_EXPORTABLE_BIT,
    VK_EXTERNAL_FENCE_FEATURE_IMPORTABLE_BIT_KHR = VK_EXTERNAL_FENCE_FEATURE_IMPORTABLE_BIT,
} VkExternalFenceFeatureFlagBits;

or the equivalent

typedef VkExternalFenceFeatureFlagBits VkExternalFenceFeatureFlagBitsKHR;
  • VK_EXTERNAL_FENCE_FEATURE_EXPORTABLE_BIT specifies handles of this type can be exported from Vulkan fence objects.

  • VK_EXTERNAL_FENCE_FEATURE_IMPORTABLE_BIT specifies handles of this type can be imported to Vulkan fence objects.

typedef VkFlags VkExternalFenceFeatureFlags;

or the equivalent

typedef VkExternalFenceFeatureFlags VkExternalFenceFeatureFlagsKHR;

VkExternalFenceFeatureFlags is a bitmask type for setting a mask of zero or more VkExternalFenceFeatureFlagBits.

33. Debugging

To aid developers in tracking down errors in the application’s use of Vulkan, particularly in combination with an external debugger or profiler, debugging extensions may be available.

The VkObjectType enumeration defines values, each of which corresponds to a specific Vulkan handle type. These values can be used to associate debug information with a particular type of object through one or more extensions.

typedef enum VkObjectType {
    VK_OBJECT_TYPE_UNKNOWN = 0,
    VK_OBJECT_TYPE_INSTANCE = 1,
    VK_OBJECT_TYPE_PHYSICAL_DEVICE = 2,
    VK_OBJECT_TYPE_DEVICE = 3,
    VK_OBJECT_TYPE_QUEUE = 4,
    VK_OBJECT_TYPE_SEMAPHORE = 5,
    VK_OBJECT_TYPE_COMMAND_BUFFER = 6,
    VK_OBJECT_TYPE_FENCE = 7,
    VK_OBJECT_TYPE_DEVICE_MEMORY = 8,
    VK_OBJECT_TYPE_BUFFER = 9,
    VK_OBJECT_TYPE_IMAGE = 10,
    VK_OBJECT_TYPE_EVENT = 11,
    VK_OBJECT_TYPE_QUERY_POOL = 12,
    VK_OBJECT_TYPE_BUFFER_VIEW = 13,
    VK_OBJECT_TYPE_IMAGE_VIEW = 14,
    VK_OBJECT_TYPE_SHADER_MODULE = 15,
    VK_OBJECT_TYPE_PIPELINE_CACHE = 16,
    VK_OBJECT_TYPE_PIPELINE_LAYOUT = 17,
    VK_OBJECT_TYPE_RENDER_PASS = 18,
    VK_OBJECT_TYPE_PIPELINE = 19,
    VK_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT = 20,
    VK_OBJECT_TYPE_SAMPLER = 21,
    VK_OBJECT_TYPE_DESCRIPTOR_POOL = 22,
    VK_OBJECT_TYPE_DESCRIPTOR_SET = 23,
    VK_OBJECT_TYPE_FRAMEBUFFER = 24,
    VK_OBJECT_TYPE_COMMAND_POOL = 25,
    VK_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION = 1000156000,
    VK_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE = 1000085000,
    VK_OBJECT_TYPE_SURFACE_KHR = 1000000000,
    VK_OBJECT_TYPE_SWAPCHAIN_KHR = 1000001000,
    VK_OBJECT_TYPE_DISPLAY_KHR = 1000002000,
    VK_OBJECT_TYPE_DISPLAY_MODE_KHR = 1000002001,
    VK_OBJECT_TYPE_DEBUG_REPORT_CALLBACK_EXT = 1000011000,
    VK_OBJECT_TYPE_OBJECT_TABLE_NVX = 1000086000,
    VK_OBJECT_TYPE_INDIRECT_COMMANDS_LAYOUT_NVX = 1000086001,
    VK_OBJECT_TYPE_DEBUG_UTILS_MESSENGER_EXT = 1000128000,
    VK_OBJECT_TYPE_VALIDATION_CACHE_EXT = 1000160000,
    VK_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_KHR = VK_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE,
    VK_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION_KHR = VK_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION,
} VkObjectType;
Table 73. VkObjectType and Vulkan Handle Relationship
VkObjectType Vulkan Handle Type

VK_OBJECT_TYPE_UNKNOWN

Unknown/Undefined Handle

VK_OBJECT_TYPE_INSTANCE

VkInstance

VK_OBJECT_TYPE_PHYSICAL_DEVICE

VkPhysicalDevice

VK_OBJECT_TYPE_DEVICE

VkDevice

VK_OBJECT_TYPE_QUEUE

VkQueue

VK_OBJECT_TYPE_SEMAPHORE

VkSemaphore

VK_OBJECT_TYPE_COMMAND_BUFFER

VkCommandBuffer

VK_OBJECT_TYPE_FENCE

VkFence

VK_OBJECT_TYPE_DEVICE_MEMORY

VkDeviceMemory

VK_OBJECT_TYPE_BUFFER

VkBuffer

VK_OBJECT_TYPE_IMAGE

VkImage

VK_OBJECT_TYPE_EVENT

VkEvent

VK_OBJECT_TYPE_QUERY_POOL

VkQueryPool

VK_OBJECT_TYPE_BUFFER_VIEW

VkBufferView

VK_OBJECT_TYPE_IMAGE_VIEW

VkImageView

VK_OBJECT_TYPE_SHADER_MODULE

VkShaderModule

VK_OBJECT_TYPE_PIPELINE_CACHE

VkPipelineCache

VK_OBJECT_TYPE_PIPELINE_LAYOUT

VkPipelineLayout

VK_OBJECT_TYPE_RENDER_PASS

VkRenderPass

VK_OBJECT_TYPE_PIPELINE

VkPipeline

VK_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT

VkDescriptorSetLayout

VK_OBJECT_TYPE_SAMPLER

VkSampler

VK_OBJECT_TYPE_DESCRIPTOR_POOL

VkDescriptorPool

VK_OBJECT_TYPE_DESCRIPTOR_SET

VkDescriptorSet

VK_OBJECT_TYPE_FRAMEBUFFER

VkFramebuffer

VK_OBJECT_TYPE_COMMAND_POOL

VkCommandPool

VK_OBJECT_TYPE_SURFACE_KHR

VkSurfaceKHR

VK_OBJECT_TYPE_SWAPCHAIN_KHR

VkSwapchainKHR

VK_OBJECT_TYPE_DISPLAY_KHR

VkDisplayKHR

VK_OBJECT_TYPE_DISPLAY_MODE_KHR

VkDisplayModeKHR

VK_OBJECT_TYPE_DEBUG_REPORT_CALLBACK_EXT

VkDebugReportCallbackEXT

VK_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE

VkDescriptorUpdateTemplate

VK_OBJECT_TYPE_OBJECT_TABLE_NVX

VkObjectTableNVX

VK_OBJECT_TYPE_INDIRECT_COMMANDS_LAYOUT_NVX

VkIndirectCommandsLayoutNVX

VK_OBJECT_TYPE_VALIDATION_CACHE_EXT

VkValidationCacheEXT

If this Specification was generated with any such extensions included, they will be described in the remainder of this chapter.

33.1. Debug Utilities

Vulkan provides flexible debugging utilities for debugging an application.

The Object Debug Annotation section describes how to associate either a name or binary data with a specific Vulkan object.

The Queue Labels section describes how to annotate and group the work submitted to a queue.

The Command Buffer Labels section describes how to associate logical elements of the scene with commands in a VkCommandBuffer.

The Debug Messengers section describes how to create debug messenger objects associated with an application supplied callback to capture debug messages from a variety of Vulkan components.

33.1.1. Object Debug Annotation

It can be useful for an application to provide its own content relative to a specific Vulkan object. The following commands allow application developers to associate user-defined information with Vulkan objects.

Object Naming

An object can be provided a user-defined name by calling vkSetDebugUtilsObjectNameEXT as defined below.

VkResult vkSetDebugUtilsObjectNameEXT(
    VkDevice                                    device,
    const VkDebugUtilsObjectNameInfoEXT*        pNameInfo);
  • device is the device that created the object.

  • pNameInfo is a pointer to an instance of the VkDebugUtilsObjectNameInfoEXT structure specifying the parameters of the name to set on the object.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pNameInfo must be a valid pointer to a valid VkDebugUtilsObjectNameInfoEXT structure

Host Synchronization
  • Host access to pNameInfo.objectHandle must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDebugUtilsObjectNameInfoEXT structure is defined as:

typedef struct VkDebugUtilsObjectNameInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkObjectType       objectType;
    uint64_t           objectHandle;
    const char*        pObjectName;
} VkDebugUtilsObjectNameInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectType is a VkObjectType specifying the type of the object to be named.

  • objectHandle is the object to be named.

  • pObjectName is a null-terminated UTF-8 string specifying the name to apply to objectHandle.

Applications may change the name associated with an object simply by calling vkSetDebugUtilsObjectNameEXT again with a new string. If pObjectName is an empty string, then any previously set name is removed.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_NAME_INFO_EXT

  • pNext must be NULL

  • objectType must be a valid VkObjectType value

  • If pObjectName is not NULL, pObjectName must be a null-terminated UTF-8 string

Object Data Association

In addition to setting a name for an object, debugging and validation layers may have uses for additional binary data on a per-object basis that have no other place in the Vulkan API.

For example, a VkShaderModule could have additional debugging data attached to it to aid in offline shader tracing.

Additional data can be attached to an object by calling vkSetDebugUtilsObjectTagEXT as defined below.

VkResult vkSetDebugUtilsObjectTagEXT(
    VkDevice                                    device,
    const VkDebugUtilsObjectTagInfoEXT*         pTagInfo);
  • device is the device that created the object.

  • pTagInfo is a pointer to an instance of the VkDebugUtilsObjectTagInfoEXT structure specifying the parameters of the tag to attach to the object.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pTagInfo must be a valid pointer to a valid VkDebugUtilsObjectTagInfoEXT structure

Host Synchronization
  • Host access to pTagInfo.objectHandle must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDebugUtilsObjectTagInfoEXT structure is defined as:

typedef struct VkDebugUtilsObjectTagInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkObjectType       objectType;
    uint64_t           objectHandle;
    uint64_t           tagName;
    size_t             tagSize;
    const void*        pTag;
} VkDebugUtilsObjectTagInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectType is a VkObjectType specifying the type of the object to be named.

  • objectHandle is the object to be tagged.

  • tagName is a numerical identifier of the tag.

  • tagSize is the number of bytes of data to attach to the object.

  • pTag is an array of tagSize bytes containing the data to be associated with the object.

The tagName parameter gives a name or identifier to the type of data being tagged. This can be used by debugging layers to easily filter for only data that can be used by that implementation.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_TAG_INFO_EXT

  • pNext must be NULL

  • objectType must be a valid VkObjectType value

  • pTag must be a valid pointer to an array of tagSize bytes

  • tagSize must be greater than 0

33.1.2. Queue Labels

All Vulkan work must be submitted using queues. It is possible for an application to use multiple queues, each containing multiple command buffers, when performing work. It can be useful to identify which queue, or even where in a queue, something has occurred.

To begin identifying a region using a debug label inside a queue, you may use the vkQueueBeginDebugUtilsLabelEXT command.

Then, when the region of interest has passed, you may end the label region using vkQueueEndDebugUtilsLabelEXT.

Additionally, a single debug label may be inserted at any time using vkQueueInsertDebugUtilsLabelEXT.

A queue debug label region is opened by calling:

void vkQueueBeginDebugUtilsLabelEXT(
    VkQueue                                     queue,
    const VkDebugUtilsLabelEXT*                 pLabelInfo);
  • queue is the queue in which to start a debug label region.

  • pLabelInfo is a pointer to an instance of the VkDebugUtilsLabelEXT structure specifying the parameters of the label region to open.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

  • pLabelInfo must be a valid pointer to a valid VkDebugUtilsLabelEXT structure

The VkDebugUtilsLabelEXT structure is defined as:

typedef struct VkDebugUtilsLabelEXT {
    VkStructureType    sType;
    const void*        pNext;
    const char*        pLabelName;
    float              color[4];
} VkDebugUtilsLabelEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pLabelName is a pointer to a null-terminated UTF-8 string that contains the name of the label.

  • color is an optional RGBA color value that can be associated with the label. A particular implementation may choose to ignore this color value. The values contain RGBA values in order, in the range 0.0 to 1.0. If all elements in color are set to 0.0 then it is ignored.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT

  • pNext must be NULL

  • pLabelName must be a null-terminated UTF-8 string

A queue debug label region is closed by calling:

void vkQueueEndDebugUtilsLabelEXT(
    VkQueue                                     queue);
  • queue is the queue in which a debug label region should be closed.

The calls to vkQueueBeginDebugUtilsLabelEXT and vkQueueEndDebugUtilsLabelEXT must be matched and balanced.

Valid Usage
  • There must be an outstanding vkQueueBeginDebugUtilsLabelEXT command prior to the vkQueueEndDebugUtilsLabelEXT on the queue

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

A single label can be inserted into a queue by calling:

void vkQueueInsertDebugUtilsLabelEXT(
    VkQueue                                     queue,
    const VkDebugUtilsLabelEXT*                 pLabelInfo);
  • queue is the queue into which a debug label will be inserted.

  • pLabelInfo is a pointer to an instance of the VkDebugUtilsLabelEXT structure specifying the parameters of the label to insert.

Valid Usage (Implicit)
  • queue must be a valid VkQueue handle

  • pLabelInfo must be a valid pointer to a valid VkDebugUtilsLabelEXT structure

33.1.3. Command Buffer Labels

Typical Vulkan applications will submit many command buffers in each frame, with each command buffer containing a large number of individual commands. Being able to logically annotate regions of command buffers that belong together as well as hierarchically subdivide the frame is important to a developer’s ability to navigate the commands viewed holistically.

To identify the beginning of a debug label region in a command buffer, vkCmdBeginDebugUtilsLabelEXT can be used as defined below.

To indicate the end of a debug label region in a command buffer, vkCmdEndDebugUtilsLabelEXT can be used.

To insert a single command buffer debug label inside of a command buffer, vkCmdInsertDebugUtilsLabelEXT can be used as defined below.

A command buffer debug label region can be opened by calling:

void vkCmdBeginDebugUtilsLabelEXT(
    VkCommandBuffer                             commandBuffer,
    const VkDebugUtilsLabelEXT*                 pLabelInfo);
  • commandBuffer is the command buffer into which the command is recorded.

  • pLabelInfo is a pointer to an instance of the VkDebugUtilsLabelEXT structure specifying the parameters of the label region to open.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pLabelInfo must be a valid pointer to a valid VkDebugUtilsLabelEXT structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

A command buffer label region can be closed by calling:

void vkCmdEndDebugUtilsLabelEXT(
    VkCommandBuffer                             commandBuffer);
  • commandBuffer is the command buffer into which the command is recorded.

An application may open a debug label region in one command buffer and close it in another, or otherwise split debug label regions across multiple command buffers or multiple queue submissions. When viewed from the linear series of submissions to a single queue, the calls to vkCmdBeginDebugUtilsLabelEXT and vkCmdEndDebugUtilsLabelEXT must be matched and balanced.

Valid Usage
  • There must be an outstanding vkCmdBeginDebugUtilsLabelEXT command prior to the vkCmdEndDebugUtilsLabelEXT on the queue that commandBuffer is submitted to

  • If commandBuffer is a secondary command buffer, there must be an outstanding vkCmdBeginDebugUtilsLabelEXT command recorded to commandBuffer that has not previously been ended by a call to vkCmdEndDebugUtilsLabelEXT.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

A single debug label can be inserted into a command buffer by calling:

void vkCmdInsertDebugUtilsLabelEXT(
    VkCommandBuffer                             commandBuffer,
    const VkDebugUtilsLabelEXT*                 pLabelInfo);
  • commandBuffer is the command buffer into which the command is recorded.

  • pInfo is a pointer to an instance of the VkDebugUtilsLabelEXT structure specifying the parameters of the label to insert.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pLabelInfo must be a valid pointer to a valid VkDebugUtilsLabelEXT structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

33.1.4. Debug Messengers

Vulkan allows an application to register multiple callbacks with any Vulkan component wishing to report debug information. Some callbacks may log the information to a file, others may cause a debug break point or other application defined behavior. A primary producer of callback messages are the validation layers. An application can register callbacks even when no validation layers are enabled, but they will only be called for the Vulkan loader and, if implemented, other layer and driver events.

A VkDebugUtilsMessengerEXT is a messenger object which handles passing along debug messages to a provided debug callback.

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDebugUtilsMessengerEXT)

The debug messenger will provide detailed feedback on the application’s use of Vulkan when events of interest occur. When an event of interest does occur, the debug messenger will submit a debug message to the debug callback that was provided during its creation. Additionally, the debug messenger is responsible with filtering out debug messages that the callback isn’t interested in and will only provide desired debug messages.

A debug messenger triggers a debug callback with a debug message when an event of interest occurs. To create a debug messenger which will trigger a debug callback, call:

VkResult vkCreateDebugUtilsMessengerEXT(
    VkInstance                                  instance,
    const VkDebugUtilsMessengerCreateInfoEXT*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDebugUtilsMessengerEXT*                   pMessenger);
  • instance the instance the messenger will be used with.

  • pCreateInfo points to a VkDebugUtilsMessengerCreateInfoEXT structure which contains the callback pointer as well as defines the conditions under which this messenger will trigger the callback.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pMessenger is a pointer to record the VkDebugUtilsMessengerEXT object created.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkDebugUtilsMessengerCreateInfoEXT structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pMessenger must be a valid pointer to a VkDebugUtilsMessengerEXT handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

The definition of VkDebugUtilsMessengerCreateInfoEXT is:

typedef struct VkDebugUtilsMessengerCreateInfoEXT {
    VkStructureType                         sType;
    const void*                             pNext;
    VkDebugUtilsMessengerCreateFlagsEXT     flags;
    VkDebugUtilsMessageSeverityFlagsEXT     messageSeverity;
    VkDebugUtilsMessageTypeFlagsEXT         messageType;
    PFN_vkDebugUtilsMessengerCallbackEXT    pfnUserCallback;
    void*                                   pUserData;
} VkDebugUtilsMessengerCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is 0 and reserved for future use.

  • messageSeverity is a bitmask of VkDebugUtilsMessageSeverityFlagBitsEXT specifying which severity of event(s) will cause this callback to be called.

  • messageTypes is a bitmask of VkDebugUtilsMessageTypeFlagBitsEXT specifying which type of event(s) will cause this callback to be called.

  • pfnUserCallback is the application callback function to call.

  • pUserData is user data to be passed to the callback.

For each VkDebugUtilsMessengerEXT that is created the VkDebugUtilsMessengerCreateInfoEXT::messageSeverity and VkDebugUtilsMessengerCreateInfoEXT::messageTypes determine when that VkDebugUtilsMessengerCreateInfoEXT::pfnUserCallback is called. The process to determine if the user’s pfnUserCallback is triggered when an event occurs is as follows:

  1. The implementation will perform a bitwise AND of the event’s VkDebugUtilsMessageSeverityFlagBitsEXT with the messageSeverity provided during creation of the VkDebugUtilsMessengerEXT object.

    1. If the value is 0, the message is skipped.

  2. The implementation will perform bitwise AND of the event’s VkDebugUtilsMessageTypeFlagBitsEXT with the messageType provided during the creation of the VkDebugUtilsMessengerEXT object.

    1. If the value is 0, the message is skipped.

  3. The callback will trigger a debug message for the current event

The callback will come directly from the component that detected the event, unless some other layer intercepts the calls for its own purposes (filter them in a different way, log to a system error log, etc.).

An application can receive multiple callbacks if multiple VkDebugUtilsMessengerEXT objects are created. A callback will always be executed in the same thread as the originating Vulkan call.

A callback can be called from multiple threads simultaneously (if the application is making Vulkan calls from multiple threads).

Valid Usage
Valid Usage (Implicit)

Bits which can be set in VkDebugUtilsMessengerCreateInfoEXT::messageSeverity, specifying event severities which cause a debug messenger to call the callback, are:

typedef enum VkDebugUtilsMessageSeverityFlagBitsEXT {
    VK_DEBUG_UTILS_MESSAGE_SEVERITY_VERBOSE_BIT_EXT = 0x00000001,
    VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT = 0x00000010,
    VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT = 0x00000100,
    VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT = 0x00001000,
} VkDebugUtilsMessageSeverityFlagBitsEXT;
  • VK_DEBUG_UTILS_MESSAGE_SEVERITY_VERBOSE_BIT_EXT specifies the most verbose output indicating all diagnostic messages from the Vulkan loader, layers, and drivers should be captured.

  • VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT specifies an informational message such as resource details that may be handy when debugging an application.

  • VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT specifies use of Vulkan that may expose an app bug. Such cases may not be immediately harmful, such as a fragment shader outputting to a location with no attachment. Other cases may point to behavior that is almost certainly bad when unintended such as using an image whose memory has not been filled. In general if you see a warning but you know that the behavior is intended/desired, then simply ignore the warning.

  • VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT specifies that an error that may cause undefined results, including an application crash.

Note

The values of VkDebugUtilsMessageSeverityFlagBitsEXT are sorted based on severity. The higher the flag value, the more severe the message. This allows for simple boolean operation comparisons when looking at VkDebugUtilsMessageSeverityFlagBitsEXT values.

For example:

    if (messageSeverity >= VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT) {
        // Do something for warnings and errors
    }

In addition, space has been left between the enums to allow for later addition of new severities in between the existing values.

Bits which can be set in VkDebugUtilsMessengerCreateInfoEXT::messageTypes, specifying event types which cause a debug messenger to call the callback, are:

typedef enum VkDebugUtilsMessageTypeFlagBitsEXT {
    VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT = 0x00000001,
    VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT = 0x00000002,
    VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT = 0x00000004,
} VkDebugUtilsMessageTypeFlagBitsEXT;
  • VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT specifies that some general event has occurred. This is typically a non-specification, non-performance event.

  • VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT specifies that something has occurred during validation against the Vulkan specification that may indicate invalid behavior.

  • VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT specifies a potentially non-optimal use of Vulkan, e.g. using vkCmdClearColorImage when setting VkAttachmentDescription::loadOp to VK_ATTACHMENT_LOAD_OP_CLEAR would have worked.

The prototype for the VkDebugUtilsMessengerCreateInfoEXT::pfnUserCallback function implemented by the application is:

typedef VkBool32 (VKAPI_PTR *PFN_vkDebugUtilsMessengerCallbackEXT)(
    VkDebugUtilsMessageSeverityFlagBitsEXT           messageSeverity,
    VkDebugUtilsMessageTypeFlagsEXT                  messageType,
    const VkDebugUtilsMessengerCallbackDataEXT*      pCallbackData,
    void*                                            pUserData);

The callback must not call vkDestroyDebugUtilsMessengerEXT.

The callback returns a VkBool32, which is interpreted in a layer-specified manner. The application should always return VK_FALSE. The VK_TRUE value is reserved for use in layer development.

The definition of VkDebugUtilsMessengerCallbackDataEXT is:

typedef struct VkDebugUtilsMessengerCallbackDataEXT {
    VkStructureType                              sType;
    const void*                                  pNext;
    VkDebugUtilsMessengerCallbackDataFlagsEXT    flags;
    const char*                                  pMessageIdName;
    int32_t                                      messageIdNumber;
    const char*                                  pMessage;
    uint32_t                                     queueLabelCount;
    VkDebugUtilsLabelEXT*                        pQueueLabels;
    uint32_t                                     cmdBufLabelCount;
    VkDebugUtilsLabelEXT*                        pCmdBufLabels;
    uint32_t                                     objectCount;
    VkDebugUtilsObjectNameInfoEXT*               pObjects;
} VkDebugUtilsMessengerCallbackDataEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is 0 and reserved for future use.

  • pMessageIdName is a null-terminated string that identifies the the particular message ID that is associated with the provided message. If the message corresponds to a validation layer message, then this string may contain the portion of the Vulkan specification that is believed to have been violated.

  • messageIdNumber is the ID number of the triggering message. If the message corresponds to a validation layer message, then this number is related to the internal number associated with the message being triggered.

  • pMessage is a null-terminated string detailing the trigger conditions.

  • queueLabelCount is a count of items contained in the pQueueLabels array.

  • pQueueLabels is NULL or a pointer to an array of VkDebugUtilsLabelEXT active in the current VkQueue at the time the callback was triggered. Refer to Queue Labels for more information.

  • cmdBufLabelCount is a count of items contained in the pCmdBufLabels array.

  • pCmdBufLabels is NULL or a pointer to an array of VkDebugUtilsLabelEXT active in the current VkCommandBuffer at the time the callback was triggered. Refer to Command Buffer Labels for more information.

  • objectCount is a count of items contained in the pObjects array.

  • pObjects is a pointer to an array of VkDebugUtilsObjectNameInfoEXT objects related to the detected issue. The array is roughly in order or importance, but the 0th element is always guaranteed to be the most important object for this message.

Note

This structure should only be considered valid during the lifetime of the triggered callback.

Since adding queue and command buffer labels behaves like pushing and popping onto a stack, the order of both pQueueLabels and pCmdBufLabels is based on the order the labels were defined. The result is that the first label in either pQueueLabels or pCmdBufLabels will be the first defined (and therefore the oldest) while the last label in each list will be the most recent.

Note

pQueueLabels will only be non-NULL if one of the objects in pObjects can be related directly to a defined VkQueue which has had one or more labels associated with it.

Likewise, pCmdBufLabels will only be non-NULL if one of the objects in pObjects can be related directly to a defined VkCommandBuffer which has had one or more labels associated with it. Additionally, while command buffer labels allow for beginning and ending across different command buffers, the debug messaging framework cannot guarantee that labels in pCmdBufLables will contain those defined outside of the associated command buffer. This is partially due to the fact that the association of one command buffer with another may not have been defined at the time the debug message is triggered.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CALLBACK_DATA_EXT

  • pNext must be NULL

  • flags must be 0

  • If pMessageIdName is not NULL, pMessageIdName must be a null-terminated UTF-8 string

  • pMessage must be a null-terminated UTF-8 string

  • objectCount must be greater than 0

There may be times that a user wishes to intentionally submit a debug message. To do this, call:

void vkSubmitDebugUtilsMessageEXT(
    VkInstance                                  instance,
    VkDebugUtilsMessageSeverityFlagBitsEXT      messageSeverity,
    VkDebugUtilsMessageTypeFlagsEXT             messageTypes,
    const VkDebugUtilsMessengerCallbackDataEXT* pCallbackData);

The call will propagate through the layers and generate callback(s) as indicated by the message’s flags. The parameters are passed on to the callback in addition to the pUserData value that was defined at the time the messenger was registered.

Valid Usage (Implicit)

To destroy a VkDebugUtilsMessengerEXT object, call:

void vkDestroyDebugUtilsMessengerEXT(
    VkInstance                                  instance,
    VkDebugUtilsMessengerEXT                    messenger,
    const VkAllocationCallbacks*                pAllocator);
  • instance the instance where the callback was created.

  • messenger the VkDebugUtilsMessengerEXT object to destroy. messenger is an externally synchronized object and must not be used on more than one thread at a time. This means that vkDestroyDebugUtilsMessengerEXT must not be called when a callback is active.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when messenger was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when messenger was created, pAllocator must be NULL

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • messenger must be a valid VkDebugUtilsMessengerEXT handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • messenger must have been created, allocated, or retrieved from instance

Host Synchronization
  • Host access to messenger must be externally synchronized

33.2. Debug Markers

Debug markers provide a flexible way for debugging and validation layers to receive annotation and debug information.

The Object Annotation section describes how to associate a name or binary data with a Vulkan object.

The Command Buffer Markers section describes how to associate logical elements of the scene with commands in the command buffer.

33.2.1. Object Annotation

The commands in this section allow application developers to associate user-defined information with Vulkan objects at will.

An object can be given a user-friendly name by calling:

VkResult vkDebugMarkerSetObjectNameEXT(
    VkDevice                                    device,
    const VkDebugMarkerObjectNameInfoEXT*       pNameInfo);
  • device is the device that created the object.

  • pNameInfo is a pointer to an instance of the VkDebugMarkerObjectNameInfoEXT structure specifying the parameters of the name to set on the object.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pNameInfo must be a valid pointer to a valid VkDebugMarkerObjectNameInfoEXT structure

Host Synchronization
  • Host access to pNameInfo.object must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDebugMarkerObjectNameInfoEXT structure is defined as:

typedef struct VkDebugMarkerObjectNameInfoEXT {
    VkStructureType               sType;
    const void*                   pNext;
    VkDebugReportObjectTypeEXT    objectType;
    uint64_t                      object;
    const char*                   pObjectName;
} VkDebugMarkerObjectNameInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectType is a VkDebugReportObjectTypeEXT specifying the type of the object to be named.

  • object is the object to be named.

  • pObjectName is a null-terminated UTF-8 string specifying the name to apply to object.

Applications may change the name associated with an object simply by calling vkDebugMarkerSetObjectNameEXT again with a new string. To remove a previously set name, pObjectName should be set to an empty string.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_NAME_INFO_EXT

  • pNext must be NULL

  • objectType must be a valid VkDebugReportObjectTypeEXT value

  • pObjectName must be a null-terminated UTF-8 string

In addition to setting a name for an object, debugging and validation layers may have uses for additional binary data on a per-object basis that has no other place in the Vulkan API. For example, a VkShaderModule could have additional debugging data attached to it to aid in offline shader tracing. To attach data to an object, call:

VkResult vkDebugMarkerSetObjectTagEXT(
    VkDevice                                    device,
    const VkDebugMarkerObjectTagInfoEXT*        pTagInfo);
  • device is the device that created the object.

  • pTagInfo is a pointer to an instance of the VkDebugMarkerObjectTagInfoEXT structure specifying the parameters of the tag to attach to the object.

Valid Usage (Implicit)
  • device must be a valid VkDevice handle

  • pTagInfo must be a valid pointer to a valid VkDebugMarkerObjectTagInfoEXT structure

Host Synchronization
  • Host access to pTagInfo.object must be externally synchronized

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDebugMarkerObjectTagInfoEXT structure is defined as:

typedef struct VkDebugMarkerObjectTagInfoEXT {
    VkStructureType               sType;
    const void*                   pNext;
    VkDebugReportObjectTypeEXT    objectType;
    uint64_t                      object;
    uint64_t                      tagName;
    size_t                        tagSize;
    const void*                   pTag;
} VkDebugMarkerObjectTagInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • objectType is a VkDebugReportObjectTypeEXT specifying the type of the object to be named.

  • object is the object to be tagged.

  • tagName is a numerical identifier of the tag.

  • tagSize is the number of bytes of data to attach to the object.

  • pTag is an array of tagSize bytes containing the data to be associated with the object.

The tagName parameter gives a name or identifier to the type of data being tagged. This can be used by debugging layers to easily filter for only data that can be used by that implementation.

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_TAG_INFO_EXT

  • pNext must be NULL

  • objectType must be a valid VkDebugReportObjectTypeEXT value

  • pTag must be a valid pointer to an array of tagSize bytes

  • tagSize must be greater than 0

33.2.2. Command Buffer Markers

Typical Vulkan applications will submit many command buffers in each frame, with each command buffer containing a large number of individual commands. Being able to logically annotate regions of command buffers that belong together as well as hierarchically subdivide the frame is important to a developer’s ability to navigate the commands viewed holistically.

The marker commands vkCmdDebugMarkerBeginEXT and vkCmdDebugMarkerEndEXT define regions of a series of commands that are grouped together, and they can be nested to create a hierarchy. The vkCmdDebugMarkerInsertEXT command allows insertion of a single label within a command buffer.

A marker region can be opened by calling:

void vkCmdDebugMarkerBeginEXT(
    VkCommandBuffer                             commandBuffer,
    const VkDebugMarkerMarkerInfoEXT*           pMarkerInfo);
  • commandBuffer is the command buffer into which the command is recorded.

  • pMarkerInfo is a pointer to an instance of the VkDebugMarkerMarkerInfoEXT structure specifying the parameters of the marker region to open.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pMarkerInfo must be a valid pointer to a valid VkDebugMarkerMarkerInfoEXT structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

The VkDebugMarkerMarkerInfoEXT structure is defined as:

typedef struct VkDebugMarkerMarkerInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    const char*        pMarkerName;
    float              color[4];
} VkDebugMarkerMarkerInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • pMarkerName is a pointer to a null-terminated UTF-8 string that contains the name of the marker.

  • color is an optional RGBA color value that can be associated with the marker. A particular implementation may choose to ignore this color value. The values contain RGBA values in order, in the range 0.0 to 1.0. If all elements in color are set to 0.0 then it is ignored.

Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_MARKER_MARKER_INFO_EXT

  • pNext must be NULL

  • pMarkerName must be a null-terminated UTF-8 string

A marker region can be closed by calling:

void vkCmdDebugMarkerEndEXT(
    VkCommandBuffer                             commandBuffer);
  • commandBuffer is the command buffer into which the command is recorded.

An application may open a marker region in one command buffer and close it in another, or otherwise split marker regions across multiple command buffers or multiple queue submissions. When viewed from the linear series of submissions to a single queue, the calls to vkCmdDebugMarkerBeginEXT and vkCmdDebugMarkerEndEXT must be matched and balanced.

Valid Usage
  • There must be an outstanding vkCmdDebugMarkerBeginEXT command prior to the vkCmdDebugMarkerEndEXT on the queue that commandBuffer is submitted to

  • If commandBuffer is a secondary command buffer, there must be an outstanding vkCmdDebugMarkerBeginEXT command recorded to commandBuffer that has not previously been ended by a call to vkCmdDebugMarkerEndEXT.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

A single marker label can be inserted into a command buffer by calling:

void vkCmdDebugMarkerInsertEXT(
    VkCommandBuffer                             commandBuffer,
    const VkDebugMarkerMarkerInfoEXT*           pMarkerInfo);
  • commandBuffer is the command buffer into which the command is recorded.

  • pMarkerInfo is a pointer to an instance of the VkDebugMarkerMarkerInfoEXT structure specifying the parameters of the marker to insert.

Valid Usage (Implicit)
  • commandBuffer must be a valid VkCommandBuffer handle

  • pMarkerInfo must be a valid pointer to a valid VkDebugMarkerMarkerInfoEXT structure

  • commandBuffer must be in the recording state

  • The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

Host Synchronization
  • Host access to the VkCommandPool that commandBuffer was allocated from must be externally synchronized

Command Properties
Command Buffer Levels Render Pass Scope Supported Queue Types Pipeline Type

Primary
Secondary

Both

Graphics
Compute

33.3. Debug Report Callbacks

Debug report callbacks are represented by VkDebugReportCallbackEXT handles:

VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDebugReportCallbackEXT)

Debug report callbacks give more detailed feedback on the application’s use of Vulkan when events of interest occur.

To register a debug report callback, an application uses vkCreateDebugReportCallbackEXT.

VkResult vkCreateDebugReportCallbackEXT(
    VkInstance                                  instance,
    const VkDebugReportCallbackCreateInfoEXT*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDebugReportCallbackEXT*                   pCallback);
  • instance the instance the callback will be logged on.

  • pCreateInfo points to a VkDebugReportCallbackCreateInfoEXT structure which defines the conditions under which this callback will be called.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

  • pCallback is a pointer to record the VkDebugReportCallbackEXT object created.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • pCreateInfo must be a valid pointer to a valid VkDebugReportCallbackCreateInfoEXT structure

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • pCallback must be a valid pointer to a VkDebugReportCallbackEXT handle

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

The definition of VkDebugReportCallbackCreateInfoEXT is:

typedef struct VkDebugReportCallbackCreateInfoEXT {
    VkStructureType                 sType;
    const void*                     pNext;
    VkDebugReportFlagsEXT           flags;
    PFN_vkDebugReportCallbackEXT    pfnCallback;
    void*                           pUserData;
} VkDebugReportCallbackCreateInfoEXT;
  • sType is the type of this structure.

  • pNext is NULL or a pointer to an extension-specific structure.

  • flags is a bitmask of VkDebugReportFlagBitsEXT specifying which event(s) will cause this callback to be called.

  • pfnCallback is the application callback function to call.

  • pUserData is user data to be passed to the callback.

For each VkDebugReportCallbackEXT that is created the VkDebugReportCallbackCreateInfoEXT::flags determine when that VkDebugReportCallbackCreateInfoEXT::pfnCallback is called. When an event happens, the implementation will do a bitwise AND of the event’s VkDebugReportFlagBitsEXT flags to each VkDebugReportCallbackEXT object’s flags. For each non-zero result the corresponding callback will be called. The callback will come directly from the component that detected the event, unless some other layer intercepts the calls for its own purposes (filter them in a different way, log to a system error log, etc.).

An application may receive multiple callbacks if multiple VkDebugReportCallbackEXT objects were created. A callback will always be executed in the same thread as the originating Vulkan call.

A callback may be called from multiple threads simultaneously (if the application is making Vulkan calls from multiple threads).

Valid Usage
Valid Usage (Implicit)
  • sType must be VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT

  • flags must be a valid combination of VkDebugReportFlagBitsEXT values

Bits which can be set in VkDebugReportCallbackCreateInfoEXT::flags, specifying events which cause a debug report, are:

typedef enum VkDebugReportFlagBitsEXT {
    VK_DEBUG_REPORT_INFORMATION_BIT_EXT = 0x00000001,
    VK_DEBUG_REPORT_WARNING_BIT_EXT = 0x00000002,
    VK_DEBUG_REPORT_PERFORMANCE_WARNING_BIT_EXT = 0x00000004,
    VK_DEBUG_REPORT_ERROR_BIT_EXT = 0x00000008,
    VK_DEBUG_REPORT_DEBUG_BIT_EXT = 0x00000010,
} VkDebugReportFlagBitsEXT;
  • VK_DEBUG_REPORT_ERROR_BIT_EXT specifies that an error that may cause undefined results, including an application crash.

  • VK_DEBUG_REPORT_WARNING_BIT_EXT specifies use of Vulkan that may expose an app bug. Such cases may not be immediately harmful, such as a fragment shader outputting to a location with no attachment. Other cases may point to behavior that is almost certainly bad when unintended such as using an image whose memory has not been filled. In general if you see a warning but you know that the behavior is intended/desired, then simply ignore the warning.

  • VK_DEBUG_REPORT_PERFORMANCE_WARNING_BIT_EXT specifies a potentially non-optimal use of Vulkan, e.g. using vkCmdClearColorImage when setting VkAttachmentDescription::loadOp to VK_ATTACHMENT_LOAD_OP_CLEAR would have worked.

  • VK_DEBUG_REPORT_INFORMATION_BIT_EXT specifies an informational message such as resource details that may be handy when debugging an application.

  • VK_DEBUG_REPORT_DEBUG_BIT_EXT specifies diagnostic information from the implementation and layers.

typedef VkFlags VkDebugReportFlagsEXT;

VkDebugReportFlagsEXT is a bitmask type for setting a mask of zero or more VkDebugReportFlagBitsEXT.

The prototype for the VkDebugReportCallbackCreateInfoEXT::pfnCallback function implemented by the application is:

typedef VkBool32 (VKAPI_PTR *PFN_vkDebugReportCallbackEXT)(
    VkDebugReportFlagsEXT                       flags,
    VkDebugReportObjectTypeEXT                  objectType,
    uint64_t                                    object,
    size_t                                      location,
    int32_t                                     messageCode,
    const char*                                 pLayerPrefix,
    const char*                                 pMessage,
    void*                                       pUserData);
  • flags specifies the VkDebugReportFlagBitsEXT that triggered this callback.

  • objectType is a VkDebugReportObjectTypeEXT value specifying the type of object being used or created at the time the event was triggered.

  • object is the object where the issue was detected. If objectType is VK_DEBUG_REPORT_OBJECT_TYPE_UNKNOWN_EXT, object is undefined.

  • location is a component (layer, driver, loader) defined value that specifies the location of the trigger. This is an optional value.

  • messageCode is a layer-defined value indicating what test triggered this callback.

  • pLayerPrefix is a null-terminated string that is an abbreviation of the name of the component making the callback. pLayerPrefix is only valid for the duration of the callback.

  • pMessage is a null-terminated string detailing the trigger conditions. pMessage is only valid for the duration of the callback.

  • pUserData is the user data given when the VkDebugReportCallbackEXT was created.

The callback must not call vkDestroyDebugReportCallbackEXT.

The callback returns a VkBool32, which is interpreted in a layer-specified manner. The application should always return VK_FALSE. The VK_TRUE value is reserved for use in layer development.

object must be a Vulkan object or VK_NULL_HANDLE. If objectType is not VK_DEBUG_REPORT_OBJECT_TYPE_UNKNOWN_EXT and object is not VK_NULL_HANDLE, object must be a Vulkan object of the corresponding type associated with objectType as defined in VkDebugReportObjectTypeEXT and Vulkan Handle Relationship.

Possible values passed to the objectType parameter of the callback function specified by VkDebugReportCallbackCreateInfoEXT::pfnCallback, specifying the type of object handle being reported, are:

typedef enum VkDebugReportObjectTypeEXT {
    VK_DEBUG_REPORT_OBJECT_TYPE_UNKNOWN_EXT = 0,
    VK_DEBUG_REPORT_OBJECT_TYPE_INSTANCE_EXT = 1,
    VK_DEBUG_REPORT_OBJECT_TYPE_PHYSICAL_DEVICE_EXT = 2,
    VK_DEBUG_REPORT_OBJECT_TYPE_DEVICE_EXT = 3,
    VK_DEBUG_REPORT_OBJECT_TYPE_QUEUE_EXT = 4,
    VK_DEBUG_REPORT_OBJECT_TYPE_SEMAPHORE_EXT = 5,
    VK_DEBUG_REPORT_OBJECT_TYPE_COMMAND_BUFFER_EXT = 6,
    VK_DEBUG_REPORT_OBJECT_TYPE_FENCE_EXT = 7,
    VK_DEBUG_REPORT_OBJECT_TYPE_DEVICE_MEMORY_EXT = 8,
    VK_DEBUG_REPORT_OBJECT_TYPE_BUFFER_EXT = 9,
    VK_DEBUG_REPORT_OBJECT_TYPE_IMAGE_EXT = 10,
    VK_DEBUG_REPORT_OBJECT_TYPE_EVENT_EXT = 11,
    VK_DEBUG_REPORT_OBJECT_TYPE_QUERY_POOL_EXT = 12,
    VK_DEBUG_REPORT_OBJECT_TYPE_BUFFER_VIEW_EXT = 13,
    VK_DEBUG_REPORT_OBJECT_TYPE_IMAGE_VIEW_EXT = 14,
    VK_DEBUG_REPORT_OBJECT_TYPE_SHADER_MODULE_EXT = 15,
    VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_CACHE_EXT = 16,
    VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_LAYOUT_EXT = 17,
    VK_DEBUG_REPORT_OBJECT_TYPE_RENDER_PASS_EXT = 18,
    VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_EXT = 19,
    VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT_EXT = 20,
    VK_DEBUG_REPORT_OBJECT_TYPE_SAMPLER_EXT = 21,
    VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_POOL_EXT = 22,
    VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_SET_EXT = 23,
    VK_DEBUG_REPORT_OBJECT_TYPE_FRAMEBUFFER_EXT = 24,
    VK_DEBUG_REPORT_OBJECT_TYPE_COMMAND_POOL_EXT = 25,
    VK_DEBUG_REPORT_OBJECT_TYPE_SURFACE_KHR_EXT = 26,
    VK_DEBUG_REPORT_OBJECT_TYPE_SWAPCHAIN_KHR_EXT = 27,
    VK_DEBUG_REPORT_OBJECT_TYPE_DEBUG_REPORT_CALLBACK_EXT_EXT = 28,
    VK_DEBUG_REPORT_OBJECT_TYPE_DISPLAY_KHR_EXT = 29,
    VK_DEBUG_REPORT_OBJECT_TYPE_DISPLAY_MODE_KHR_EXT = 30,
    VK_DEBUG_REPORT_OBJECT_TYPE_OBJECT_TABLE_NVX_EXT = 31,
    VK_DEBUG_REPORT_OBJECT_TYPE_INDIRECT_COMMANDS_LAYOUT_NVX_EXT = 32,
    VK_DEBUG_REPORT_OBJECT_TYPE_VALIDATION_CACHE_EXT_EXT = 33,
    VK_DEBUG_REPORT_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION_EXT = 1000156000,
    VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_EXT = 1000085000,
    VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_KHR_EXT = VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_EXT,
    VK_DEBUG_REPORT_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION_KHR_EXT = VK_DEBUG_REPORT_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION_EXT,
} VkDebugReportObjectTypeEXT;
Table 74. VkDebugReportObjectTypeEXT and Vulkan Handle Relationship
VkDebugReportObjectTypeEXT Vulkan Handle Type

VK_DEBUG_REPORT_OBJECT_TYPE_UNKNOWN_EXT

Unknown/Undefined Handle

VK_DEBUG_REPORT_OBJECT_TYPE_INSTANCE_EXT

VkInstance

VK_DEBUG_REPORT_OBJECT_TYPE_PHYSICAL_DEVICE_EXT

VkPhysicalDevice

VK_DEBUG_REPORT_OBJECT_TYPE_DEVICE_EXT

VkDevice

VK_DEBUG_REPORT_OBJECT_TYPE_QUEUE_EXT

VkQueue

VK_DEBUG_REPORT_OBJECT_TYPE_SEMAPHORE_EXT

VkSemaphore

VK_DEBUG_REPORT_OBJECT_TYPE_COMMAND_BUFFER_EXT

VkCommandBuffer

VK_DEBUG_REPORT_OBJECT_TYPE_FENCE_EXT

VkFence

VK_DEBUG_REPORT_OBJECT_TYPE_DEVICE_MEMORY_EXT

VkDeviceMemory

VK_DEBUG_REPORT_OBJECT_TYPE_BUFFER_EXT

VkBuffer

VK_DEBUG_REPORT_OBJECT_TYPE_IMAGE_EXT

VkImage

VK_DEBUG_REPORT_OBJECT_TYPE_EVENT_EXT

VkEvent

VK_DEBUG_REPORT_OBJECT_TYPE_QUERY_POOL_EXT

VkQueryPool

VK_DEBUG_REPORT_OBJECT_TYPE_BUFFER_VIEW_EXT

VkBufferView

VK_DEBUG_REPORT_OBJECT_TYPE_IMAGE_VIEW_EXT

VkImageView

VK_DEBUG_REPORT_OBJECT_TYPE_SHADER_MODULE_EXT

VkShaderModule

VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_CACHE_EXT

VkPipelineCache

VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_LAYOUT_EXT

VkPipelineLayout

VK_DEBUG_REPORT_OBJECT_TYPE_RENDER_PASS_EXT

VkRenderPass

VK_DEBUG_REPORT_OBJECT_TYPE_PIPELINE_EXT

VkPipeline

VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT_EXT

VkDescriptorSetLayout

VK_DEBUG_REPORT_OBJECT_TYPE_SAMPLER_EXT

VkSampler

VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_POOL_EXT

VkDescriptorPool

VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_SET_EXT

VkDescriptorSet

VK_DEBUG_REPORT_OBJECT_TYPE_FRAMEBUFFER_EXT

VkFramebuffer

VK_DEBUG_REPORT_OBJECT_TYPE_COMMAND_POOL_EXT

VkCommandPool

VK_DEBUG_REPORT_OBJECT_TYPE_SURFACE_KHR_EXT

VkSurfaceKHR

VK_DEBUG_REPORT_OBJECT_TYPE_SWAPCHAIN_KHR_EXT

VkSwapchainKHR

VK_DEBUG_REPORT_OBJECT_TYPE_DEBUG_REPORT_CALLBACK_EXT_EXT

VkDebugReportCallbackEXT

VK_DEBUG_REPORT_OBJECT_TYPE_DISPLAY_KHR_EXT

VkDisplayKHR

VK_DEBUG_REPORT_OBJECT_TYPE_DISPLAY_MODE_KHR_EXT

VkDisplayModeKHR

VK_DEBUG_REPORT_OBJECT_TYPE_OBJECT_TABLE_NVX_EXT

VkObjectTableNVX

VK_DEBUG_REPORT_OBJECT_TYPE_INDIRECT_COMMANDS_LAYOUT_NVX_EXT

VkIndirectCommandsLayoutNVX

VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_EXT

VkDescriptorUpdateTemplate

Note

The primary expected use of VK_ERROR_VALIDATION_FAILED_EXT is for validation layer testing. It is not expected that an application would see this error code during normal use of the validation layers.

To inject its own messages into the debug stream, call:

void vkDebugReportMessageEXT(
    VkInstance                                  instance,
    VkDebugReportFlagsEXT                       flags,
    VkDebugReportObjectTypeEXT                  objectType,
    uint64_t                                    object,
    size_t                                      location,
    int32_t                                     messageCode,
    const char*                                 pLayerPrefix,
    const char*                                 pMessage);
  • instance is the debug stream’s VkInstance.

  • flags specifies the VkDebugReportFlagBitsEXT classification of this event/message.

  • objectType is a VkDebugReportObjectTypeEXT specifying the type of object being used or created at the time the event was triggered.

  • object this is the object where the issue was detected. object can be VK_NULL_HANDLE if there is no object associated with the event.

  • location is an application defined value.

  • messageCode is an application defined value.

  • pLayerPrefix is the abbreviation of the component making this event/message.

  • pMessage is a null-terminated string detailing the trigger conditions.

The call will propagate through the layers and generate callback(s) as indicated by the message’s flags. The parameters are passed on to the callback in addition to the pUserData value that was defined at the time the callback was registered.

Valid Usage
  • object must be a Vulkan object or VK_NULL_HANDLE

  • If objectType is not VK_DEBUG_REPORT_OBJECT_TYPE_UNKNOWN_EXT and object is not VK_NULL_HANDLE, object must be a Vulkan object of the corresponding type associated with objectType as defined in VkDebugReportObjectTypeEXT and Vulkan Handle Relationship.

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • flags must be a valid combination of VkDebugReportFlagBitsEXT values

  • flags must not be 0

  • objectType must be a valid VkDebugReportObjectTypeEXT value

  • pLayerPrefix must be a null-terminated UTF-8 string

  • pMessage must be a null-terminated UTF-8 string

To destroy a VkDebugReportCallbackEXT object, call:

void vkDestroyDebugReportCallbackEXT(
    VkInstance                                  instance,
    VkDebugReportCallbackEXT                    callback,
    const VkAllocationCallbacks*                pAllocator);
  • instance the instance where the callback was created.

  • callback the VkDebugReportCallbackEXT object to destroy. callback is an externally synchronized object and must not be used on more than one thread at a time. This means that vkDestroyDebugReportCallbackEXT must not be called when a callback is active.

  • pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage
  • If VkAllocationCallbacks were provided when callback was created, a compatible set of callbacks must be provided here

  • If no VkAllocationCallbacks were provided when callback was created, pAllocator must be NULL

Valid Usage (Implicit)
  • instance must be a valid VkInstance handle

  • callback must be a valid VkDebugReportCallbackEXT handle

  • If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • callback must have been created, allocated, or retrieved from instance

Host Synchronization
  • Host access to callback must be externally synchronized

Appendix A: Vulkan Environment for SPIR-V

Shaders for Vulkan are defined by the Khronos SPIR-V Specification as well as the Khronos SPIR-V Extended Instructions for GLSL Specification. This appendix defines additional SPIR-V requirements applying to Vulkan shaders.

Versions and Formats

A Vulkan 1.1 implementation must support the 1.0, 1.1, 1.2, and 1.3 versions of SPIR-V and the 1.0 version of the SPIR-V Extended Instructions for GLSL.

A SPIR-V module passed into vkCreateShaderModule is interpreted as a series of 32-bit words in host endianness, with literal strings packed as described in section 2.2 of the SPIR-V Specification. The first few words of the SPIR-V module must be a magic number and a SPIR-V version number, as described in section 2.3 of the SPIR-V Specification.

Capabilities

Implementations must support the following capability operands declared by OpCapability:

  • Matrix

  • Shader

  • InputAttachment

  • Sampled1D

  • Image1D

  • SampledBuffer

  • ImageBuffer

  • ImageQuery

  • DerivativeControl

  • DeviceGroup

  • MultiView

If the implementation supports any of the optional features described in the Features chapter, then the capability operand(s) corresponding to that feature must also be supported.

Table 75. List of optional SPIR-V Capabilities and corresponding feature or extension names
SPIR-V OpCapability Vulkan feature or extension name

Geometry

geometryShader

Tessellation

tessellationShader

Float64

shaderFloat64

Int64

shaderInt64

Int16

shaderInt16

TessellationPointSize

shaderTessellationAndGeometryPointSize

GeometryPointSize

shaderTessellationAndGeometryPointSize

ImageGatherExtended

shaderImageGatherExtended

StorageImageMultisample

shaderStorageImageMultisample

UniformBufferArrayDynamicIndexing

shaderUniformBufferArrayDynamicIndexing

SampledImageArrayDynamicIndexing

shaderSampledImageArrayDynamicIndexing

StorageBufferArrayDynamicIndexing

shaderStorageBufferArrayDynamicIndexing

StorageImageArrayDynamicIndexing

shaderStorageImageArrayDynamicIndexing

ClipDistance

shaderClipDistance

CullDistance

shaderCullDistance

ImageCubeArray

imageCubeArray

SampleRateShading

sampleRateShading

SparseResidency

shaderResourceResidency

MinLod

shaderResourceMinLod

SampledCubeArray

imageCubeArray

ImageMSArray

shaderStorageImageMultisample

StorageImageExtendedFormats

shaderStorageImageExtendedFormats

InterpolationFunction

sampleRateShading

StorageImageReadWithoutFormat

shaderStorageImageReadWithoutFormat

StorageImageWriteWithoutFormat

shaderStorageImageWriteWithoutFormat

MultiViewport

multiViewport

DrawParameters

VK_KHR_shader_draw_parameters or shaderDrawParameters

VariablePointersStorageBuffer

variablePointersStorageBuffer

VariablePointers

variablePointers

StencilExportEXT

VK_EXT_shader_stencil_export

SubgroupBallotKHR

VK_EXT_shader_subgroup_ballot

SubgroupVoteKHR

VK_EXT_shader_subgroup_vote

ImageReadWriteLodAMD

VK_AMD_shader_image_load_store_lod

ImageGatherBiasLodAMD

VK_AMD_texture_gather_bias_lod

FragmentMaskAMD

VK_AMD_shader_fragment_mask

SampleMaskOverrideCoverageNV

VK_NV_sample_mask_override_coverage

GeometryShaderPassthroughNV

VK_NV_geometry_shader_passthrough

ShaderViewportIndexLayerEXT

VK_EXT_shader_viewport_index_layer

ShaderViewportIndexLayerNV

VK_NV_viewport_array2

ShaderViewportMaskNV

VK_NV_viewport_array2

PerViewAttributesNV

VK_NVX_multiview_per_view_attributes

StorageBuffer16BitAccess

StorageBuffer16BitAccess

UniformAndStorageBuffer16BitAccess

UniformAndStorageBuffer16BitAccess

StoragePushConstant16

storagePushConstant16

StorageInputOutput16

storageInputOutput16

GroupNonUniform

VK_SUBGROUP_FEATURE_BASIC_BIT

GroupNonUniformVote

VK_SUBGROUP_FEATURE_VOTE_BIT

GroupNonUniformArithmetic

VK_SUBGROUP_FEATURE_ARITHMETIC_BIT

GroupNonUniformBallot

VK_SUBGROUP_FEATURE_BALLOT_BIT

GroupNonUniformShuffle

VK_SUBGROUP_FEATURE_SHUFFLE_BIT

GroupNonUniformShuffleRelative

VK_SUBGROUP_FEATURE_SHUFFLE_RELATIVE_BIT

GroupNonUniformClustered

VK_SUBGROUP_FEATURE_CLUSTERED_BIT

GroupNonUniformQuad

VK_SUBGROUP_FEATURE_QUAD_BIT

GroupNonUniformPartitionedNV

VK_SUBGROUP_FEATURE_PARTITIONED_BIT_NV

SampleMaskPostDepthCoverage

VK_EXT_post_depth_coverage

ShaderNonUniformEXT

VK_EXT_descriptor_indexing

RuntimeDescriptorArrayEXT

runtimeDescriptorArray

InputAttachmentArrayDynamicIndexingEXT

shaderInputAttachmentArrayDynamicIndexing

UniformTexelBufferArrayDynamicIndexingEXT

shaderUniformTexelBufferArrayDynamicIndexing

StorageTexelBufferArrayDynamicIndexingEXT

shaderStorageTexelBufferArrayDynamicIndexing

UniformBufferArrayNonUniformIndexingEXT

shaderUniformBufferArrayNonUniformIndexing

SampledImageArrayNonUniformIndexingEXT

shaderSampledImageArrayNonUniformIndexing

StorageBufferArrayNonUniformIndexingEXT

shaderStorageBufferArrayNonUniformIndexing

StorageImageArrayNonUniformIndexingEXT

shaderStorageImageArrayNonUniformIndexing

InputAttachmentArrayNonUniformIndexingEXT

shaderInputAttachmentArrayNonUniformIndexing

UniformTexelBufferArrayNonUniformIndexingEXT

shaderUniformTexelBufferArrayNonUniformIndexing

StorageTexelBufferArrayNonUniformIndexingEXT

shaderStorageTexelBufferArrayNonUniformIndexing

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_variable_pointers SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_shader_explicit_vertex_parameter SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_gcn_shader SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_gpu_shader_half_float SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_gpu_shader_int16 SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_shader_ballot SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_shader_fragment_mask SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_shader_image_load_store_lod SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_shader_trinary_minmax SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_AMD_texture_gather_bias_lod SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_shader_draw_parameters SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_16bit_storage SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_storage_buffer_storage_class SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_post_depth_coverage SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_EXT_shader_stencil_export SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_shader_ballot SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_KHR_subgroup_vote SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_NV_sample_mask_override_coverage SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_NV_geometry_shader_passthrough SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_NV_viewport_array2 SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_EXT_shader_viewport_index_layer SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_NVX_multiview_per_view_attributes SPIR-V extension.

The application can pass a SPIR-V module to vkCreateShaderModule that uses the SPV_EXT_descriptor_indexing SPIR-V extension.

The application must not pass a SPIR-V module containing any of the following to vkCreateShaderModule:

  • any OpCapability not listed above,

  • an unsupported capability, or

  • a capability which corresponds to a Vulkan feature or extension which has not been enabled.

Validation Rules within a Module

A SPIR-V module passed to vkCreateShaderModule must conform to the following rules:

  • Every entry point must have no return value and accept no arguments.

  • Recursion: The static function-call graph for an entry point must not contain cycles.

  • The Logical addressing model must be selected.

  • Scope for execution must be limited to:

    • Workgroup

    • Subgroup

  • Scope for memory must be limited to:

    • Device

    • Workgroup

    • Subgroup

    • Invocation

  • Scope for Non Uniform Group Operations must be limited to:

    • Subgroup

  • Storage Class must be limited to:

    • UniformConstant

    • Input

    • Uniform

    • Output

    • Workgroup

    • Private

    • Function

    • PushConstant

    • Image

    • StorageBuffer

  • Memory semantics must obey the following rules:

    • Acquire must not be used with OpAtomicStore.

    • Release must not be used with OpAtomicLoad.

    • AcquireRelease must not be used with OpAtomicStore or OpAtomicLoad.

    • Sequentially consistent atomics and barriers are not supported and SequentiallyConsistent is treated as AcquireRelease. SequentiallyConsistent should not be used.

    • OpMemoryBarrier must use one of Acquire, Release, AcquireRelease, or SequentiallyConsistent and must include at least one storage class.

    • If the semantics for OpControlBarrier includes one of Acquire, Release, AcquireRelease, or SequentiallyConsistent, then it must include at least one storage class.

    • SubgroupMemory, CrossWorkgroupMemory, and AtomicCounterMemory are ignored.

  • Any OpVariable with an Initializer operand must have one of the following as its Storage Class operand:

    • Output

    • Private

    • Function

  • The OriginLowerLeft execution mode must not be used; fragment entry points must declare OriginUpperLeft.

  • The PixelCenterInteger execution mode must not be used. Pixels are always centered at half-integer coordinates.

  • Images

    • OpTypeImage must declare a scalar 32-bit float or 32-bit integer type for the “Sampled Type”. (RelaxedPrecision can be applied to a sampling instruction and to the variable holding the result of a sampling instruction.)

    • OpTypeImage must have a “Sampled” operand of 1 (sampled image) or 2 (storage image).

    • OpImageQuerySizeLod, and OpImageQueryLevels must only consume an “Image” operand whose type has its “Sampled” operand set to 1.

    • The (u,v) coordinates used for a SubpassData must be the <id> of a constant vector (0,0), or if a layer coordinate is used, must be a vector that was formed with constant 0 for the u and v components.

    • The “Depth” operand of OpTypeImage is ignored.

  • Decorations

    • The GLSLShared and GLSLPacked decorations must not be used.

    • The Flat, NoPerspective, Sample, and Centroid decorations must not be used on variables with storage class other than Input or on variables used in the interface of non-fragment shader entry points.

    • The Patch decoration must not be used on variables in the interface of a vertex, geometry, or fragment shader stage’s entry point.

    • The ViewportRelativeNV decoration must only be used on a variable decorated with Layer in the vertex, tessellation evaluation, or geometry shader stages.

    • The ViewportRelativeNV decoration must not be used unless a variable decorated with one of ViewportIndex or ViewportMaskNV is also statically used by the same OpEntryPoint.

    • The ViewportMaskNV and ViewportIndex decorations must not both be statically used by one or more OpEntryPoint’s that form the vertex processing stages of a graphics pipeline.

    • Only the round-to-nearest-even and the round-to-zero rounding modes can be used for the FPRoundingMode decoration.

    • The FPRoundingMode decoration can only be used for the floating-point conversion instructions as described in the SPV_KHR_16bit_storage SPIR-V extension.

  • OpTypeRuntimeArray must only be used for:

    • the last member of an OpTypeStruct that is in the StorageBuffer storage class decorated as Block, or that is in the Uniform storage class decorated as BufferBlock.

    • If the RuntimeDescriptorArrayEXT capability is supported, an array of variables with storage class Uniform, StorageBuffer, or UniformConstant, or for the outermost dimension of an array of arrays of such variables.

  • Linkage: See Shader Interfaces for additional linking and validation rules.

  • If OpControlBarrier is used in fragment, vertex, tessellation evaluation, or geometry stages, the execution Scope must be Subgroup.

  • Compute Shaders

    • For each compute shader entry point, either a LocalSize execution mode or an object decorated with the WorkgroupSize decoration must be specified.

  • “Result Type” for Non Uniform Group Operations must be limited to 32-bit float, 32-bit integer, boolean, or vectors of these types. If the Float64 capability is enabled, double and vector of double types are also permitted.

  • “Mask” for OpGroupNonUniformShuffleXor must be a specialization constant or a constant, or if the dynamic instance is called within a loop construct it must be one of:

    1. A specialization constant.

    2. A constant.

    3. An arthimetic operation whose operands are 1., 2., or 4.

    4. A phi node whose operands are 1., 2., or 3.

  • If OpGroupNonUniformBallotBitCount is used, the group operation must be one of:

    • Reduce

    • InclusiveScan

    • ExclusiveScan

  • Atomic instructions must declare a scalar 32-bit integer type for the Result Type and the type of the value pointed to by Pointer.

  • If an instruction loads from or stores to a resource (including atomics and image instructions) and the resource descriptor being accessed is not dynamically uniform, then the operand corresponding to that resource (e.g. the pointer or sampled image operand) must be decorated with NonUniformEXT.

Precision and Operation of SPIR-V Instructions

The following rules apply to both single and double-precision floating point instructions:

  • Positive and negative infinities and positive and negative zeros are generated as dictated by IEEE 754, but subject to the precisions allowed in the following table.

  • Dividing a non-zero by a zero results in the appropriately signed IEEE 754 infinity.

  • Any denormalized value input into a shader or potentially generated by any instruction in a shader may be flushed to 0.

  • The rounding mode cannot be set and is undefined.

  • NaNs may not be generated. Instructions that operate on a NaN may not result in a NaN.

  • Support for signaling NaNs is optional and exceptions are never raised.

The precision of double-precision instructions is at least that of single precision. For single precision (32 bit) instructions, precisions are required to be at least as follows, unless decorated with RelaxedPrecision:

Table 76. Precision of core SPIR-V Instructions
Instruction Precision

OpFAdd

Correctly rounded.

OpFSub

Correctly rounded.

OpFMul

Correctly rounded.

OpFOrdEqual, OpFUnordEqual

Correct result.

OpFOrdLessThan, OpFUnordLessThan

Correct result.

OpFOrdGreaterThan, OpFUnordGreaterThan

Correct result.

OpFOrdLessThanEqual, OpFUnordLessThanEqual

Correct result.

OpFOrdGreaterThanEqual, OpFUnordGreaterThanEqual

Correct result.

OpFDiv

2.5 ULP for b in the range [2-126, 2126].

conversions between types

Correctly rounded.

Table 77. Precision of GLSL.std.450 Instructions
Instruction Precision

fma()

Inherited from OpFMul followed by OpFAdd.

exp(x), exp2(x)

3 + 2 × |x| ULP.

log(), log2()

3 ULP outside the range [0.5, 2.0]. Absolute error < 2-21 inside the range [0.5, 2.0].

pow(x, y)

Inherited from exp2(y × log2(x)).

sqrt()

Inherited from 1.0 / inversesqrt().

inversesqrt()

2 ULP.

GLSL.std.450 extended instructions specifically defined in terms of the above instructions inherit the above errors. GLSL.std.450 extended instructions not listed above and not defined in terms of the above have undefined precision. These include, for example, the trigonometric functions and determinant.

For the OpSRem and OpSMod instructions, if either operand is negative the result is undefined.

Note

While the OpSRem and OpSMod instructions are supported by the Vulkan environment, they require non-negative values and thus do not enable additional functionality beyond what OpUMod provides.

Compatibility Between SPIR-V Image Formats And Vulkan Formats

Images which are read from or written to by shaders must have SPIR-V image formats compatible with the Vulkan image formats backing the image under the circumstances described for texture image validation. The compatibile formats are:

Table 78. SPIR-V and Vulkan Image Format Compatibility
SPIR-V Image Format Compatible Vulkan Format

Rgba32f

VK_FORMAT_R32G32B32A32_SFLOAT

Rgba16f

VK_FORMAT_R16G16B16A16_SFLOAT

R32f

VK_FORMAT_R32_SFLOAT

Rgba8

VK_FORMAT_R8G8B8A8_UNORM

Rgba8Snorm

VK_FORMAT_R8G8B8A8_SNORM

Rg32f

VK_FORMAT_R32G32_SFLOAT

Rg16f

VK_FORMAT_R16G16_SFLOAT

R11fG11fB10f

VK_FORMAT_B10G11R11_UFLOAT_PACK32

R16f

VK_FORMAT_R16_SFLOAT

Rgba16

VK_FORMAT_R16G16B16A16_UNORM

Rgb10A2

VK_FORMAT_A2B10G10R10_UNORM_PACK32

Rg16

VK_FORMAT_R16G16_UNORM

Rg8

VK_FORMAT_R8G8_UNORM

R16

VK_FORMAT_R16_UNORM

R8

VK_FORMAT_R8_UNORM

Rgba16Snorm

VK_FORMAT_R16G16B16A16_SNORM

Rg16Snorm

VK_FORMAT_R16G16_SNORM

Rg8Snorm

VK_FORMAT_R8G8_SNORM

R16Snorm

VK_FORMAT_R16_SNORM

R8Snorm

VK_FORMAT_R8_SNORM

Rgba32i

VK_FORMAT_R32G32B32A32_SINT

Rgba16i

VK_FORMAT_R16G16B16A16_SINT

Rgba8i

VK_FORMAT_R8G8B8A8_SINT

R32i

VK_FORMAT_R32_SINT

Rg32i

VK_FORMAT_R32G32_SINT

Rg16i

VK_FORMAT_R16G16_SINT

Rg8i

VK_FORMAT_R8G8_SINT

R16i

VK_FORMAT_R16_SINT

R8i

VK_FORMAT_R8_SINT

Rgba32ui

VK_FORMAT_R32G32B32A32_UINT

Rgba16ui

VK_FORMAT_R16G16B16A16_UINT

Rgba8ui

VK_FORMAT_R8G8B8A8_UINT

R32ui

VK_FORMAT_R32_UINT

Rgb10a2ui

VK_FORMAT_A2B10G10R10_UINT_PACK32

Rg32ui

VK_FORMAT_R32G32_UINT

Rg16ui

VK_FORMAT_R16G16_UINT

Rg8ui

VK_FORMAT_R8G8_UINT

R16ui

VK_FORMAT_R16_UINT

R8ui

VK_FORMAT_R8_UINT

Appendix B: Compressed Image Formats

The compressed texture formats used by Vulkan are described in the specifically identified sections of the Khronos Data Format Specification, version 1.1.

Unless otherwise described, the quantities encoded in these compressed formats are treated as normalized, unsigned values.

Those formats listed as sRGB-encoded have in-memory representations of R, G and B components which are nonlinearly-encoded as R', G', and B'; any alpha component is unchanged. As part of filtering, the nonlinear R', G', and B' values are converted to linear R, G, and B components; any alpha component is unchanged. The conversion between linear and nonlinear encoding is performed as described in the “KHR_DF_TRANSFER_SRGB” section of the Khronos Data Format Specification.

Block-Compressed Image Formats

Table 79. Mapping of Vulkan BC formats to descriptions
VkFormat Khronos Data Format Specification description

Formats described in the “S3TC Compressed Texture Image Formats” chapter

VK_FORMAT_BC1_RGB_UNORM_BLOCK

BC1 with no alpha

VK_FORMAT_BC1_RGB_SRGB_BLOCK

BC1 with no alpha, sRGB-encoded

VK_FORMAT_BC1_RGBA_UNORM_BLOCK

BC1 with alpha

VK_FORMAT_BC1_RGBA_SRGB_BLOCK

BC1 with alpha, sRGB-encoded

VK_FORMAT_BC2_UNORM_BLOCK

BC2

VK_FORMAT_BC2_SRGB_BLOCK

BC2, sRGB-encoded

VK_FORMAT_BC3_UNORM_BLOCK

BC3

VK_FORMAT_BC3_SRGB_BLOCK

BC3, sRGB-encoded

Formats described in the “RGTC Compressed Texture Image Formats” chapter

VK_FORMAT_BC4_UNORM_BLOCK

BC4 unsigned

VK_FORMAT_BC4_SNORM_BLOCK

BC4 signed

VK_FORMAT_BC5_UNORM_BLOCK

BC5 unsigned

VK_FORMAT_BC5_SNORM_BLOCK

BC5 signed

Formats described in the “BPTC Compressed Texture Image Formats” chapter

VK_FORMAT_BC6H_UFLOAT_BLOCK

BC6H (unsigned version)

VK_FORMAT_BC6H_SFLOAT_BLOCK

BC6H (signed version)

VK_FORMAT_BC7_UNORM_BLOCK

BC7

VK_FORMAT_BC7_SRGB_BLOCK

BC7, sRGB-encoded

ETC Compressed Image Formats

The following formats are described in the “ETC2 Compressed Texture Image Formats” chapter of the Khronos Data Format Specification.

Table 80. Mapping of Vulkan ETC formats to descriptions
VkFormat Khronos Data Format Specification description

VK_FORMAT_ETC2_R8G8B8_UNORM_BLOCK

RGB ETC2

VK_FORMAT_ETC2_R8G8B8_SRGB_BLOCK

RGB ETC2 with sRGB encoding

VK_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK

RGB ETC2 with punch-through alpha

VK_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK

RGB ETC2 with punch-through alpha and sRGB

VK_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK

RGBA ETC2

VK_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK

RGBA ETC2 with sRGB encoding

VK_FORMAT_EAC_R11_UNORM_BLOCK

Unsigned R11 EAC

VK_FORMAT_EAC_R11_SNORM_BLOCK

Signed R11 EAC

VK_FORMAT_EAC_R11G11_UNORM_BLOCK

Unsigned RG11 EAC

VK_FORMAT_EAC_R11G11_SNORM_BLOCK

Signed RG11 EAC

ASTC Compressed Image Formats

ASTC formats are described in the “ASTC Compressed Texture Image Formats” chapter of the Khronos Data Format Specification.

Table 81. Mapping of Vulkan ASTC formats to descriptions
VkFormat Compressed texel block dimensions sRGB-encoded

VK_FORMAT_ASTC_4x4_UNORM_BLOCK

4 × 4

No

VK_FORMAT_ASTC_4x4_SRGB_BLOCK

4 × 4

Yes

VK_FORMAT_ASTC_5x4_UNORM_BLOCK

5 × 4

No

VK_FORMAT_ASTC_5x4_SRGB_BLOCK

5 × 4

Yes

VK_FORMAT_ASTC_5x5_UNORM_BLOCK

5 × 5

No

VK_FORMAT_ASTC_5x5_SRGB_BLOCK

5 × 5

Yes

VK_FORMAT_ASTC_6x5_UNORM_BLOCK

6 × 5

No

VK_FORMAT_ASTC_6x5_SRGB_BLOCK

6 × 5

Yes

VK_FORMAT_ASTC_6x6_UNORM_BLOCK

6 × 6

No

VK_FORMAT_ASTC_6x6_SRGB_BLOCK

6 × 6

Yes

VK_FORMAT_ASTC_8x5_UNORM_BLOCK

8 × 5

No

VK_FORMAT_ASTC_8x5_SRGB_BLOCK

8 × 5

Yes

VK_FORMAT_ASTC_8x6_UNORM_BLOCK

8 × 6

No

VK_FORMAT_ASTC_8x6_SRGB_BLOCK

8 × 6

Yes

VK_FORMAT_ASTC_8x8_UNORM_BLOCK

8 × 8

No

VK_FORMAT_ASTC_8x8_SRGB_BLOCK

8 × 8

Yes

VK_FORMAT_ASTC_10x5_UNORM_BLOCK

10 × 5

No

VK_FORMAT_ASTC_10x5_SRGB_BLOCK

10 × 5

Yes

VK_FORMAT_ASTC_10x6_UNORM_BLOCK

10 × 6

No

VK_FORMAT_ASTC_10x6_SRGB_BLOCK

10 × 6

Yes

VK_FORMAT_ASTC_10x8_UNORM_BLOCK

10 × 8

No

VK_FORMAT_ASTC_10x8_SRGB_BLOCK

10 × 8

Yes

VK_FORMAT_ASTC_10x10_UNORM_BLOCK

10 × 10

No

VK_FORMAT_ASTC_10x10_SRGB_BLOCK

10 × 10

Yes

VK_FORMAT_ASTC_12x10_UNORM_BLOCK

12 × 10

No

VK_FORMAT_ASTC_12x10_SRGB_BLOCK

12 × 10

Yes

VK_FORMAT_ASTC_12x12_UNORM_BLOCK

12 × 12

No

VK_FORMAT_ASTC_12x12_SRGB_BLOCK

12 × 12

Yes

Appendix C: Core Revisions (Informative)

New minor versions of the Vulkan API are defined periodically by the Khronos Vulkan Working Group. These consist of some amount of additional functionality added to the core API, some of which may be promoted from extensions, other parts of which may be new. Extensions that are promoted in this way typically have their functionality replicated directly in the core, but with extension suffixes dropped. The existing values with suffixes are still present in the API itself as aliases of the original extension functionality. Any differences between the core and extension version of the functionality will be documented in the extension appendix, and mentioned briefly in the version description in this appendix.

It’s possible to build the specification for earlier versions, but to aid readability of the latest versions, this appendix gives an overview of the changes as compared to earlier versions.

Version 1.1

Vulkan Version 1.1 promoted a number of key extensions into the core API:

The only changes to the functionality added by these extensions were to VK_KHR_shader_draw_parameters, which had a feature bit added to determine support in the core API, and variablePointersStorageBuffer from VK_KHR_variable_pointers was made optional.

Additionally, Vulkan 1.1 added support for subgroup operations, protected memory, and a new command to enumerate the instance version.

New Object Types

  • VkDescriptorUpdateTemplate

  • VkSamplerYcbcrConversion

New Defines

  • VK_API_VERSION_1_1

New Enum Constants

  • Extending VkBufferCreateFlagBits:

    • VK_BUFFER_CREATE_PROTECTED_BIT

  • Extending VkCommandPoolCreateFlagBits:

    • VK_COMMAND_POOL_CREATE_PROTECTED_BIT

  • Extending VkDependencyFlagBits:

    • VK_DEPENDENCY_DEVICE_GROUP_BIT

    • VK_DEPENDENCY_VIEW_LOCAL_BIT

  • Extending VkDeviceQueueCreateFlagBits:

    • VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT

  • Extending VkFormat:

    • VK_FORMAT_G8B8G8R8_422_UNORM

    • VK_FORMAT_B8G8R8G8_422_UNORM

    • VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM

    • VK_FORMAT_G8_B8R8_2PLANE_420_UNORM

    • VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM

    • VK_FORMAT_G8_B8R8_2PLANE_422_UNORM

    • VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM

    • VK_FORMAT_R10X6_UNORM_PACK16

    • VK_FORMAT_R10X6G10X6_UNORM_2PACK16

    • VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16

    • VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16

    • VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16

    • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16

    • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16

    • VK_FORMAT_R12X4_UNORM_PACK16

    • VK_FORMAT_R12X4G12X4_UNORM_2PACK16

    • VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16

    • VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16

    • VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16

    • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16

    • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16

    • VK_FORMAT_G16B16G16R16_422_UNORM

    • VK_FORMAT_B16G16R16G16_422_UNORM

    • VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM

    • VK_FORMAT_G16_B16R16_2PLANE_420_UNORM

    • VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM

    • VK_FORMAT_G16_B16R16_2PLANE_422_UNORM

    • VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM

  • Extending VkFormatFeatureFlagBits:

    • VK_FORMAT_FEATURE_TRANSFER_SRC_BIT

    • VK_FORMAT_FEATURE_TRANSFER_DST_BIT

    • VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT

    • VK_FORMAT_FEATURE_DISJOINT_BIT

    • VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT

  • Extending VkImageAspectFlagBits:

    • VK_IMAGE_ASPECT_PLANE_0_BIT

    • VK_IMAGE_ASPECT_PLANE_1_BIT

    • VK_IMAGE_ASPECT_PLANE_2_BIT

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_CREATE_ALIAS_BIT

    • VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT

    • VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT

    • VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT

    • VK_IMAGE_CREATE_EXTENDED_USAGE_BIT

    • VK_IMAGE_CREATE_PROTECTED_BIT

    • VK_IMAGE_CREATE_DISJOINT_BIT

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL

    • VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL

  • Extending VkMemoryHeapFlagBits:

    • VK_MEMORY_HEAP_MULTI_INSTANCE_BIT

  • Extending VkMemoryPropertyFlagBits:

    • VK_MEMORY_PROPERTY_PROTECTED_BIT

  • Extending VkObjectType:

    • VK_OBJECT_TYPE_SAMPLER_YCBCR_CONVERSION

    • VK_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE

  • Extending VkPipelineCreateFlagBits:

    • VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT

    • VK_PIPELINE_CREATE_DISPATCH_BASE

  • Extending VkQueueFlagBits:

    • VK_QUEUE_PROTECTED_BIT

  • Extending VkResult:

    • VK_ERROR_OUT_OF_POOL_MEMORY

    • VK_ERROR_INVALID_EXTERNAL_HANDLE

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SUBGROUP_PROPERTIES

    • VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES

    • VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS

    • VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO

    • VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO

    • VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO

    • VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2

    • VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2

    • VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2

    • VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2

    • VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2

    • VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2

    • VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2

    • VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2

    • VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SPARSE_IMAGE_FORMAT_INFO_2

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES

    • VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO

    • VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO

    • VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO

    • VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES

    • VK_STRUCTURE_TYPE_PROTECTED_SUBMIT_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_FEATURES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROTECTED_MEMORY_PROPERTIES

    • VK_STRUCTURE_TYPE_DEVICE_QUEUE_INFO_2

    • VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO

    • VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO

    • VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO

    • VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES

    • VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_IMAGE_FORMAT_PROPERTIES

    • VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_IMAGE_FORMAT_INFO

    • VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_BUFFER_INFO

    • VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES

    • VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_BUFFER_CREATE_INFO

    • VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO

    • VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_FENCE_INFO

    • VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES

    • VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO

    • VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_SEMAPHORE_INFO

    • VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES

    • VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_DRAW_PARAMETER_FEATURES

New Structures

Appendix D: Layers & Extensions (Informative)

Extensions to the Vulkan API can be defined by authors, groups of authors, and the Khronos Vulkan Working Group. In order not to compromise the readability of the Vulkan Specification, the core Specification does not incorporate most extensions. The online Registry of extensions is available at URL

and allows generating versions of the Specification incorporating different extensions.

Most of the content previously in this appendix does not specify use of specific Vulkan extensions and layers, but rather specifies the processes by which extensions and layers are created. As of version 1.0.21 of the Vulkan Specification, this content has been migrated to the Vulkan Documentation and Extensions document. Authors creating extensions and layers must follow the mandatory procedures in that document.

The remainder of this appendix documents a set of extensions chosen when this document was built. Versions of the Specification published in the Registry include:

  • Core API + mandatory extensions required of all Vulkan implementations.

  • Core API + all registered and published Khronos (KHR) extensions.

  • Core API + all registered and published extensions.

Extensions are grouped as Khronos KHR, multivendor EXT, and then alphabetically by author ID. Within each group, extensions are listed in alphabetical order by their name.

Note

As of the initial Vulkan 1.1 public release, the KHX author ID is no longer used. All KHX extensions have been promoted to KHR status. Previously, this author ID was used to indicate that an extension was experimental, and is being considered for standardization in future KHR or core Vulkan API versions. We no longer use this mechanism for exposing experimental functionality.

Some vendors may use an alternate author ID ending in X for some of their extensions. The exact meaning of such an author ID is defined by each vendor, and may not be equivalent to KHX, but it is likely to indicate a lesser degree of interface stability than a non-X extension from the same vendor.

VK_KHR_16bit_storage

Name String

VK_KHR_16bit_storage

Extension Type

Device extension

Registered Extension Number

84

Revision

1

Extension and Version Dependencies
Contact
  • Jan-Harald Fredriksen @janharaldfredriksen-arm

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Alexander Galazin, ARM

  • Jan-Harald Fredriksen, ARM

  • Joerg Wagner, ARM

  • Neil Henning, Codeplay

  • Jeff Bolz, Nvidia

  • Daniel Koch, Nvidia

  • David Neto, Google

  • John Kessenich, Google

The VK_KHR_16bit_storage extension allows use of 16-bit types in shader input and output interfaces, and push constant blocks. This extension introduces several new optional features which map to SPIR-V capabilities and allow access to 16-bit data in Block-decorated objects in the Uniform and the StorageBuffer storage classes, and objects in the PushConstant storage class. This extension allows 16-bit variables to be declared and used as user-defined shader inputs and outputs but does not change location assignment and component assignment rules.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_16BIT_STORAGE_FEATURES_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

Version History

  • Revision 1, 2017-03-23 (Alexander Galazin)

    • Initial draft

VK_KHR_android_surface

Name String

VK_KHR_android_surface

Extension Type

Instance extension

Registered Extension Number

9

Revision

6

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2016-01-14

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_android_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to an ANativeWindow, Android’s native surface type. The ANativeWindow represents the producer endpoint of any buffer queue, regardless of consumer endpoint. Common consumer endpoints for ANativeWindows are the system window compositor, video encoders, and application-specific compositors importing the images through a SurfaceTexture.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_ANDROID_SURFACE_CREATE_INFO_KHR

New Enums

None

Issues

1) Does Android need a way to query for compatibility between a particular physical device (and queue family?) and a specific Android display?

RESOLVED: No. Currently on Android, any physical device is expected to be able to present to the system compositor, and all queue families must support the necessary image layout transitions and synchronization operations.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft.

  • Revision 2, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_android_surface to VK_KHR_android_surface.

  • Revision 3, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to surface creation function.

  • Revision 4, 2015-11-10 (Jesse Hall)

    • Removed VK_ERROR_INVALID_ANDROID_WINDOW_KHR.

  • Revision 5, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

  • Revision 6, 2016-01-14 (James Jones)

    • Moved VK_ERROR_NATIVE_WINDOW_IN_USE_KHR from the VK_KHR_android_surface to the VK_KHR_surface extension.

VK_KHR_bind_memory2

Name String

VK_KHR_bind_memory2

Extension Type

Device extension

Registered Extension Number

158

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Tobias Hector @tobski

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

  • Tobias Hector, Imagination Technologies

This extension provides versions of vkBindBufferMemory and vkBindImageMemory that allow multiple bindings to be performed at once, and are extensible.

This extension also introduces VK_IMAGE_CREATE_ALIAS_BIT_KHR, which allows “identical” images that alias the same memory to interpret the contents consistently, even across image layout changes.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_INFO_KHR

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_CREATE_ALIAS_BIT_KHR

New Enums

None.

New Built-In Variables

None.

New SPIR-V Capabilities

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Version History

  • Revision 1, 2017-05-19 (Tobias Hector)

    • Pulled bind memory functions into their own extension

VK_KHR_dedicated_allocation

Name String

VK_KHR_dedicated_allocation

Extension Type

Device extension

Registered Extension Number

128

Revision

3

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

  • Jason Ekstrand, Intel

This extension enables resources to be bound to a dedicated allocation, rather than suballocated. For any particular resource, applications can query whether a dedicated allocation is recommended, in which case using a dedicated allocation may improve the performance of access to that resource. Normal device memory allocations must support multiple resources per allocation, memory aliasing and sparse binding, which could interfere with some optimizations. Applications should query the implementation for when a dedicated allocation may be beneficial by adding VkMemoryDedicatedRequirementsKHR to the pNext chain of the VkMemoryRequirements2 structure passed as the pMemoryRequirements parameter to a call to vkGetBufferMemoryRequirements2 or vkGetImageMemoryRequirements2. Certain external handle types and external images or buffers may also depend on dedicated allocations on implementations that associate image or buffer metadata with OS-level memory objects.

This extension adds a two small structures to memory requirements querying and memory allocation: a new structure that flags whether an image/buffer should have a dedicated allocation, and a structure indicating the image or buffer that an allocation will be bound to.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS_KHR

    • VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO_KHR

New Enums

None.

New Functions

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Examples

    // Create an image with a dedicated allocation based on the
    // implementation's preference

    VkImageCreateInfo imageCreateInfo =
    {
        // Image creation parameters
    };

    VkImage image;
    VkResult result = vkCreateImage(
        device,
        &imageCreateInfo,
        NULL,                               // pAllocator
        &image);

    VkMemoryDedicatedRequirementsKHR dedicatedRequirements =
    {
        VK_STRUCTURE_TYPE_MEMORY_DEDICATED_REQUIREMENTS_KHR,
        NULL,                               // pNext
    };

    VkMemoryRequirements2 memoryRequirements =
    {
        VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2,
        &dedicatedRequirements,             // pNext
    };

    const VkImageMemoryRequirementsInfo2 imageRequirementsInfo =
    {
        VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2,
        NULL,                               // pNext
        image
    };

    vkGetImageMemoryRequirements2(
        device,
        &imageRequirementsInfo,
        &memoryRequirements);

    if (dedicatedRequirements.prefersDedicatedAllocation) {
        // Allocate memory with VkMemoryDedicatedAllocateInfoKHR::image
        // pointing to the image we are allocating the memory for

        VkMemoryDedicatedAllocateInfoKHR dedicatedInfo =
        {
            VK_STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO_KHR,   // sType
            NULL,                                                   // pNext
            image,                                                  // image
            VK_NULL_HANDLE,                                         // buffer
        };

        VkMemoryAllocateInfo memoryAllocateInfo =
        {
            VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO,                 // sType
            &dedicatedInfo,                                         // pNext
            memoryRequirements.size,                                // allocationSize
            FindMemoryTypeIndex(memoryRequirements.memoryTypeBits), // memoryTypeIndex
        };

        VkDeviceMemory memory;
        vkAllocateMemory(
            device,
            &memoryAllocateInfo,
            NULL,                       // pAllocator
            &memory);

        // Bind the image to the memory

        vkBindImageMemory(
            device,
            image,
            memory,
            0);
    } else {
        // Take the normal memory sub-allocation path
    }

Version History

  • Revision 1, 2017-02-27 (James Jones)

    • Copy content from VK_NV_dedicated_allocation

    • Add some references to external object interactions to the overview.

  • Revision 2, 2017-03-27 (Jason Ekstrand)

    • Rework the extension to be query-based

  • Revision 3, 2017-07-31 (Jason Ekstrand)

    • Clarify that memory objects created with VkMemoryDedicatedAllocateInfoKHR can only have the specified resource bound and no others.

VK_KHR_descriptor_update_template

Name String

VK_KHR_descriptor_update_template

Extension Type

Device extension

Registered Extension Number

86

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Markus Tavenrath @mtavenrath

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Jeff Bolz, NVIDIA

  • Michael Worcester, Imagination Technologies

Applications may wish to update a fixed set of descriptors in a large number of descriptors sets very frequently, i.e. during initializaton phase or if it’s required to rebuild descriptor sets for each frame. For those cases it’s also not unlikely that all information required to update a single descriptor set is stored in a single struct. This extension provides a way to update a fixed set of descriptors in a single VkDescriptorSet with a pointer to a user defined data structure which describes the new descriptors.

New Enum Constants

Extending VkStructureType:

  • VK_STRUCTURE_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_CREATE_INFO_KHR

Promotion to Vulkan 1.1

vkCmdPushDescriptorSetWithTemplateKHR is included as an interaction with VK_KHR_push_descriptor. If Vulkan 1.1 and VK_KHR_push_descriptor are supported, this is included by VK_KHR_push_descriptor.

The base functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Version History

  • Revision 1, 2016-01-11 (Markus Tavenrath)

    • Initial draft

VK_KHR_device_group

Name String

VK_KHR_device_group

Extension Type

Device extension

Registered Extension Number

61

Revision

3

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-10-06

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

  • Tobias Hector, Imagination Technologies

This extension provides functionality to use a logical device that consists of multiple physical devices, as created with the VK_KHR_device_group_creation extension. A device group can allocate memory across the subdevices, bind memory from one subdevice to a resource on another subdevice, record command buffers where some work executes on an arbitrary subset of the subdevices, and potentially present a swapchain image from one or more subdevices.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_FLAGS_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_RENDER_PASS_BEGIN_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_COMMAND_BUFFER_BEGIN_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_SUBMIT_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_BIND_SPARSE_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_CAPABILITIES_KHR

    • VK_STRUCTURE_TYPE_IMAGE_SWAPCHAIN_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_SWAPCHAIN_INFO_KHR

    • VK_STRUCTURE_TYPE_ACQUIRE_NEXT_IMAGE_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_SWAPCHAIN_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_BUFFER_MEMORY_DEVICE_GROUP_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_DEVICE_GROUP_INFO_KHR

  • Extending VkImageCreateFlagBits

    • VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR

  • Extending VkPipelineCreateFlagBits

    • VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT_KHR

    • VK_PIPELINE_CREATE_DISPATCH_BASE_KHR

  • Extending VkDependencyFlagBits

    • VK_DEPENDENCY_DEVICE_GROUP_BIT_KHR

  • Extending VkSwapchainCreateFlagBitsKHR

    • VK_SWAPCHAIN_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR

New Built-In Variables

New SPIR-V Capabilities

Promotion to Vulkan 1.1

The following enums, types and commands are included as interactions with VK_KHR_swapchain:

If Vulkan 1.1 and VK_KHR_swapchain are supported, these are included by VK_KHR_swapchain.

The base functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Examples

TODO

Version History

  • Revision 1, 2016-10-19 (Jeff Bolz)

    • Internal revisions

  • Revision 2, 2017-05-19 (Tobias Hector)

    • Removed extended memory bind functions to VK_KHR_bind_memory2, added dependency on that extension, and device-group-specific structs for those functions.

  • Revision 3, 2017-10-06 (Ian Elliott)

    • Corrected Vulkan 1.1 interactions with the WSI extensions. All Vulkan 1.1 WSI interactions are with the VK_KHR_swapchain extension.

  • Revision 4, 2017-10-10 (Jeff Bolz)

    • Rename "SFR" bits and structure members to use the phrase "split instance bind regions".

VK_KHR_device_group_creation

Name String

VK_KHR_device_group_creation

Extension Type

Instance extension

Registered Extension Number

71

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2016-10-19

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

This extension provides instance-level commands to enumerate groups of physical devices, and to create a logical device from a subset of one of those groups. Such a logical device can then be used with new features in the VK_KHR_device_group extension.

New Object Types

None.

New Enum Constants

New Enums

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Examples

    VkDeviceCreateInfo devCreateInfo = { VK_STRUCTURE_TYPE_DEVICE_CREATE_INFO };
    // (not shown) fill out devCreateInfo as usual.
    uint32_t deviceGroupCount = 0;
    VkPhysicalDeviceGroupPropertiesKHR *props = NULL;

    // Query the number of device groups
    vkEnumeratePhysicalDeviceGroupsKHR(g_vkInstance, &deviceGroupCount, NULL);

    // Allocate and initialize structures to query the device groups
    props = (VkPhysicalDeviceGroupPropertiesKHR *)malloc(deviceGroupCount*sizeof(VkPhysicalDeviceGroupPropertiesKHR));
    for (i = 0; i < deviceGroupCount; ++i) {
        props[i].sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES_KHR;
        props[i].pNext = NULL;
    }
    vkEnumeratePhysicalDeviceGroupsKHR(g_vkInstance, &deviceGroupCount, props);

    // If the first device group has more than one physical device. create
    // a logical device using all of the physical devices.
    VkDeviceGroupDeviceCreateInfoKHR deviceGroupInfo = { VK_STRUCTURE_TYPE_DEVICE_GROUP_DEVICE_CREATE_INFO_KHR };
    if (props[0].physicalDeviceCount > 1) {
        deviceGroupInfo.physicalDeviceCount = props[0].physicalDeviceCount;
        deviceGroupInfo.pPhysicalDevices = props[0].physicalDevices;
        devCreateInfo.pNext = &deviceGroupInfo;
    }

    vkCreateDevice(props[0].physicalDevices[0], &devCreateInfo, NULL, &g_vkDevice);
    free(props);

Version History

  • Revision 1, 2016-10-19 (Jeff Bolz)

    • Internal revisions

VK_KHR_display

Name String

VK_KHR_display

Extension Type

Instance extension

Registered Extension Number

3

Revision

21

Extension and Version Dependencies
Contact
  • James Jones @cubanismo,Norbert Nopper @FslNopper

Last Modified Date

2017-03-13

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Norbert Nopper, Freescale

  • Jeff Vigil, Qualcomm

  • Daniel Rakos, AMD

This extension provides the API to enumerate displays and available modes on a given device.

New Object Types

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DISPLAY_MODE_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_DISPLAY_SURFACE_CREATE_INFO_KHR

Issues

1) Which properties of a mode should be fixed in the mode info vs. settable in some other function when setting the mode? E.g., do we need to double the size of the mode pool to include both stereo and non-stereo modes? YUV and RGB scanout even if they both take RGB input images? BGR vs. RGB input? etc.

PROPOSED RESOLUTION: Many modern displays support at most a handful of resolutions and timings natively. Other “modes” are expected to be supported using scaling hardware on the display engine or GPU. Other properties, such as rotation and mirroring should not require duplicating hardware modes just to express all combinations. Further, these properties may be implemented on a per-display or per-overlay granularity.

To avoid the exponential growth of modes as mutable properties are added, as was the case with EGLConfig/WGL pixel formats/GLXFBConfig, this specification should separate out hardware properties and configurable state into separate objects. Modes and overlay planes will express capabilities of the hardware, while a separate structure will allow applications to configure scaling, rotation, mirroring, color keys, LUT values, alpha masks, etc. for a given swapchain independent of the mode in use. Constraints on these settings will be established by properties of the immutable objects.

Note the resolution of this issue may affect issue 5 as well.

2) What properties of a display itself are useful?

PROPOSED RESOLUTION: This issue is too broad. It was meant to prompt general discussion, but resolving this issue amounts to completing this specification. All interesting properties should be included. The issue will remain as a placeholder since removing it would make it hard to parse existing discussion notes that refer to issues by number.

3) How are multiple overlay planes within a display or mode enumerated?

PROPOSED RESOLUTION: They are referred to by an index. Each display will report the number of overlay planes it contains.

4) Should swapchains be created relative to a mode or a display?

PROPOSED RESOLUTION: When using this extension, swapchains are created relative to a mode and a plane. The mode implies the display object the swapchain will present to. If the specified mode is not the display’s current mode, the new mode will be applied when the first image is presented to the swapchain, and the default operating system mode, if any, will be restored when the swapchain is destroyed.

5) Should users query generic ranges from displays and construct their own modes explicitly using those constraints rather than querying a fixed set of modes (Most monitors only have one real “mode” these days, even though many support relatively arbitrary scaling, either on the monitor side or in the GPU display engine, making “modes” something of a relic/compatibility construct).

PROPOSED RESOLUTION: Expose both. Display info structures will expose a set of predefined modes, as well as any attributes necessary to construct a customized mode.

6) Is it fine if we return the display and display mode handles in the structure used to query their properties?

PROPOSED RESOLUTION: Yes.

7) Is there a possibility that not all displays of a device work with all of the present queues of a device? If yes, how do we determine which displays work with which present queues?

PROPOSED RESOLUTION: No known hardware has such limitations, but determining such limitations is supported automatically using the existing VK_KHR_surface and VK_KHR_swapchain query mechanisms.

8) Should all presentation need to be done relative to an overlay plane, or can a display mode + display be used alone to target an output?

PROPOSED RESOLUTION: Require specifying a plane explicitly.

9) Should displays have an associated window system display, such as an HDC or Display*?

PROPOSED RESOLUTION: No. Displays are independent of any windowing system in use on the system. Further, neither HDC nor Display* refer to a physical display object.

10) Are displays queried from a physical GPU or from a device instance?

PROPOSED RESOLUTION: Developers prefer to query modes directly from the physical GPU so they can use display information as an input to their device selection algorithms prior to device creation. This avoids the need to create dummy device instances to enumerate displays.

This preference must be weighed against the extra initialization that must be done by driver vendors prior to device instance creation to support this usage.

11) Should displays and/or modes be dispatchable objects? If functions are to take displays, overlays, or modes as their first parameter, they must be dispatchable objects as defined in Khronos bug 13529. If they are not added to the list of dispatchable objects, functions operating on them must take some higher-level object as their first parameter. There is no performance case against making them dispatchable objects, but they would be the first extension objects to be dispatchable.

PROPOSED RESOLUTION: Do not make displays or modes dispatchable. They will dispatch based on their associated physical device.

12) Should hardware cursor capabilities be exposed?

PROPOSED RESOLUTION: Defer. This could be a separate extension on top of the base WSI specs.

editing-note

There appears to be a missing sentence for the first part of issue 13 here.

if they are one physical display device to an end user, but may internally be implemented as two side-by-side displays using the same display engine (and sometimes cabling) resources as two physically separate display devices.

RESOLVED: Tiled displays will appear as a single display object in this API.

14) Should the raw EDID data be included in the display information?

RESOLVED: No. A future extension could be added which reports the EDID if necessary. This may be complicated by the outcome of issue 13.

15) Should min and max scaling factor capabilities of overlays be exposed?

RESOLVED: Yes. This is exposed indirectly by allowing applications to query the min/max position and extent of the source and destination regions from which image contents are fetched by the display engine when using a particular mode and overlay pair.

16) Should devices be able to expose planes that can be moved between displays? If so, how?

RESOLVED: Yes. Applications can determine which displays a given plane supports using vkGetDisplayPlaneSupportedDisplaysKHR.

17) Should there be a way to destroy display modes? If so, does it support destroying “built in” modes?

RESOLVED: Not in this extension. A future extension could add this functionality.

18) What should the lifetime of display and built-in display mode objects be?

RESOLVED: The lifetime of the instance. These objects cannot be destroyed. A future extension may be added to expose a way to destroy these objects and/or support display hotplug.

19) Should persistent mode for smart panels be enabled/disabled at swapchain creation time, or on a per-present basis.

RESOLVED: On a per-present basis.

Examples

Note

The example code for the VK_KHR_display and VK_KHR_display_swapchain extensions was removed from the appendix after revision 1.0.43. The display enumeration example code was ported to the cube demo that is shipped with the official Khronos SDK, and is being kept up-to-date in that location (see: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/master/demos/cube.c).

Version History

  • Revision 1, 2015-02-24 (James Jones)

    • Initial draft

  • Revision 2, 2015-03-12 (Norbert Nopper)

    • Added overlay enumeration for a display.

  • Revision 3, 2015-03-17 (Norbert Nopper)

    • Fixed typos and namings as discussed in Bugzilla.

    • Reordered and grouped functions.

    • Added functions to query count of display, mode and overlay.

    • Added native display handle, which is maybe needed on some platforms to create a native Window.

  • Revision 4, 2015-03-18 (Norbert Nopper)

    • Removed primary and virtualPostion members (see comment of James Jones in Bugzilla).

    • Added native overlay handle to info structure.

    • Replaced , with ; in struct.

  • Revision 6, 2015-03-18 (Daniel Rakos)

    • Added WSI extension suffix to all items.

    • Made the whole API more "Vulkanish".

    • Replaced all functions with a single vkGetDisplayInfoKHR function to better match the rest of the API.

    • Made the display, display mode, and overlay objects be first class objects, not subclasses of VkBaseObject as they do not support the common functions anyways.

    • Renamed *Info structures to *Properties.

    • Removed overlayIndex field from VkOverlayProperties as there is an implicit index already as a result of moving to a "Vulkanish" API.

    • Displays are not get through device, but through physical GPU to match the rest of the Vulkan API. Also this is something ISVs explicitly requested.

    • Added issue (6) and (7).

  • Revision 7, 2015-03-25 (James Jones)

    • Added an issues section

    • Added rotation and mirroring flags

  • Revision 8, 2015-03-25 (James Jones)

    • Combined the duplicate issues sections introduced in last change.

    • Added proposed resolutions to several issues.

  • Revision 9, 2015-04-01 (Daniel Rakos)

    • Rebased extension against Vulkan 0.82.0

  • Revision 10, 2015-04-01 (James Jones)

    • Added issues (10) and (11).

    • Added more straw-man issue resolutions, and cleaned up the proposed resolution for issue (4).

    • Updated the rotation and mirroring enums to have proper bitmask semantics.

  • Revision 11, 2015-04-15 (James Jones)

    • Added proposed resolution for issues (1) and (2).

    • Added issues (12), (13), (14), and (15)

    • Removed pNativeHandle field from overlay structure.

    • Fixed small compilation errors in example code.

  • Revision 12, 2015-07-29 (James Jones)

    • Rewrote the guts of the extension against the latest WSI swapchain specifications and the latest Vulkan API.

    • Address overlay planes by their index rather than an object handle and refer to them as "planes" rather than "overlays" to make it slightly clearer that even a display with no "overlays" still has at least one base "plane" that images can be displayed on.

    • Updated most of the issues.

    • Added an "extension type" section to the specification header.

    • Re-used the VK_EXT_KHR_surface surface transform enumerations rather than redefining them here.

    • Updated the example code to use the new semantics.

  • Revision 13, 2015-08-21 (Ian Elliott)

    • Renamed this extension and all of its enumerations, types, functions, etc. This makes it compliant with the proposed standard for Vulkan extensions.

    • Switched from "revision" to "version", including use of the VK_MAKE_VERSION macro in the header file.

  • Revision 14, 2015-09-01 (James Jones)

    • Restore single-field revision number.

  • Revision 15, 2015-09-08 (James Jones)

    • Added alpha flags enum.

    • Added premultiplied alpha support.

  • Revision 16, 2015-09-08 (James Jones)

    • Added description section to the spec.

    • Added issues 16 - 18.

  • Revision 17, 2015-10-02 (James Jones)

    • Planes are now a property of the entire device rather than individual displays. This allows planes to be moved between multiple displays on devices that support it.

    • Added a function to create a VkSurfaceKHR object describing a display plane and mode to align with the new per-platform surface creation conventions.

    • Removed detailed mode timing data. It was agreed that the mode extents and refresh rate are sufficient for current use cases. Other information could be added back2 in as an extension if it is needed in the future.

    • Added support for smart/persistent/buffered display devices.

  • Revision 18, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_display to VK_KHR_display.

  • Revision 19, 2015-11-02 (James Jones)

    • Updated example code to match revision 17 changes.

  • Revision 20, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to creation functions.

  • Revision 21, 2015-11-10 (Jesse Hall)

    • Added VK_DISPLAY_PLANE_ALPHA_OPAQUE_BIT_KHR, and use VkDisplayPlaneAlphaFlagBitsKHR for VkDisplayPlanePropertiesKHR::alphaMode instead of VkDisplayPlaneAlphaFlagsKHR, since it only represents one mode.

    • Added reserved flags bitmask to VkDisplayPlanePropertiesKHR.

    • Use VkSurfaceTransformFlagBitsKHR instead of obsolete VkSurfaceTransformKHR.

    • Renamed vkGetDisplayPlaneSupportedDisplaysKHR parameters for clarity.

  • Revision 22, 2015-12-18 (James Jones)

    • Added missing "planeIndex" parameter to vkGetDisplayPlaneSupportedDisplaysKHR()

  • Revision 23, 2017-03-13 (James Jones)

    • Closed all remaining issues. The specification and implementations have been shipping with the proposed resolutions for some time now.

    • Removed the sample code and noted it has been integrated into the official Vulkan SDK cube demo.

VK_KHR_display_swapchain

Name String

VK_KHR_display_swapchain

Extension Type

Device extension

Registered Extension Number

4

Revision

9

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2017-03-13

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Jeff Vigil, Qualcomm

  • Jesse Hall, Google

This extension provides an API to create a swapchain directly on a device’s display without any underlying window system.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DISPLAY_PRESENT_INFO_KHR

  • Extending VkResult:

    • VK_ERROR_INCOMPATIBLE_DISPLAY_KHR

New Enums

None

New Structures

Issues

1) Should swapchains sharing images each hold a reference to the images, or should it be up to the application to destroy the swapchains and images in an order that avoids the need for reference counting?

RESOLVED: Take a reference. The lifetime of presentable images is already complex enough.

2) Should the srcRect/dstRect parameters be specified as part of the present command, or at swapchain creation time?

RESOLVED: As part of the presentation command. This allows moving and scaling the image on the screen without the need to respecify the mode or create a new swapchain and presentable images.

3) Should srcRect/dstRect be specified as rects, or separate offset/extent values?

RESOLVED: As rects. Specifying them separately might make it easier for hardware to expose support for one but not the other, but in such cases applications must just take care to obey the reported capabilities and not use non-zero offsets or extents that require scaling, as appropriate.

4) How can applications create multiple swapchains that use the same images?

RESOLVED: By calling vkCreateSharedSwapchainsKHR.

An earlier resolution used vkCreateSwapchainKHR, chaining multiple VkSwapchainCreateInfoKHR structures through pNext. In order to allow each swapchain to also allow other extension structs, a level of indirection was used: VkSwapchainCreateInfoKHR::pNext pointed to a different structure, which had both an sType/pNext for additional extensions, and also had a pointer to the next VkSwapchainCreateInfoKHR structure. The number of swapchains to be created could only be found by walking this linked list of alternating structures, and the pSwapchains out parameter was reinterpreted to be an array of VkSwapchainKHR handles.

Another option considered was a method to specify a “shared” swapchain when creating a new swapchain, such that groups of swapchains using the same images could be built up one at a time. This was deemed unusable because drivers need to know all of the displays an image will be used on when determining which internal formats and layouts to use for that image.

Examples

Note

The example code for the VK_KHR_display and VK_KHR_display_swapchain extensions was removed from the appendix after revision 1.0.43. The display swapchain creation example code was ported to the cube demo that is shipped with the official Khronos SDK, and is being kept up-to-date in that location (see: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/master/demos/cube.c).

Version History

  • Revision 1, 2015-07-29 (James Jones)

    • Initial draft

  • Revision 2, 2015-08-21 (Ian Elliott)

    • Renamed this extension and all of its enumerations, types, functions, etc. This makes it compliant with the proposed standard for Vulkan extensions.

    • Switched from "revision" to "version", including use of the VK_MAKE_VERSION macro in the header file.

  • Revision 3, 2015-09-01 (James Jones)

    • Restore single-field revision number.

  • Revision 4, 2015-09-08 (James Jones)

    • Allow creating multiple swap chains that share the same images using a single call to vkCreateSwapChainKHR().

  • Revision 5, 2015-09-10 (Alon Or-bach)

    • Removed underscores from SWAP_CHAIN in two enums.

  • Revision 6, 2015-10-02 (James Jones)

    • Added support for smart panels/buffered displays.

  • Revision 7, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_display_swapchain to VK_KHR_display_swapchain.

  • Revision 8, 2015-11-03 (Daniel Rakos)

    • Updated sample code based on the changes to VK_KHR_swapchain.

  • Revision 9, 2015-11-10 (Jesse Hall)

    • Replaced VkDisplaySwapchainCreateInfoKHR with vkCreateSharedSwapchainsKHR, changing resolution of issue #4.

  • Revision 10, 2017-03-13 (James Jones)

    • Closed all remaining issues. The specification and implementations have been shipping with the proposed resolutions for some time now.

    • Removed the sample code and noted it has been integrated into the official Vulkan SDK cube demo.

VK_KHR_external_fence

Name String

VK_KHR_external_fence

Extension Type

Device extension

Registered Extension Number

114

Revision

1

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2017-05-08

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors

An application using external memory may wish to synchronize access to that memory using fences. This extension enables an application to create fences from which non-Vulkan handles that reference the underlying synchronization primitive can be exported.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO_KHR

New Functions

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

This extension borrows concepts, semantics, and language from VK_KHR_external_semaphore. That extension’s issues apply equally to this extension.

VK_KHR_external_fence_capabilities

Name String

VK_KHR_external_fence_capabilities

Extension Type

Instance extension

Registered Extension Number

113

Revision

1

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2017-05-08

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors

An application may wish to reference device fences in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension provides a set of capability queries and handle definitions that allow an application to determine what types of “external” fence handles an implementation supports for a given set of use cases.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_FENCE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXTERNAL_FENCE_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES_KHR

  • VK_LUID_SIZE_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

VK_KHR_external_fence_fd

Name String

VK_KHR_external_fence_fd

Extension Type

Device extension

Registered Extension Number

116

Revision

1

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2017-05-08

IP Status

No known IP claims.

Contributors

An application using external memory may wish to synchronize access to that memory using fences. This extension enables an application to export fence payload to and import fence payload from POSIX file descriptors.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_FENCE_FD_INFO_KHR

  • VK_STRUCTURE_TYPE_FENCE_GET_FD_INFO_KHR

New Enums

None.

Issues

This extension borrows concepts, semantics, and language from VK_KHR_external_semaphore_fd. That extension’s issues apply equally to this extension.

VK_KHR_external_fence_win32

Name String

VK_KHR_external_fence_win32

Extension Type

Device extension

Registered Extension Number

115

Revision

1

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2017-05-08

IP Status

No known IP claims.

Contributors

An application using external memory may wish to synchronize access to that memory using fences. This extension enables an application to export fence payload to and import fence payload from Windows handles.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_FENCE_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXPORT_FENCE_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_FENCE_GET_WIN32_HANDLE_INFO_KHR

New Enums

None.

Issues

This extension borrows concepts, semantics, and language from VK_KHR_external_semaphore_win32. That extension’s issues apply equally to this extension.

1) Should D3D12 fence handle types be supported, like they are for semaphores?

RESOLVED: No. Doing so would require extending the fence signal and wait operations to provide values to signal / wait for, like VkD3D12FenceSubmitInfoKHR does. A D3D12 fence can be signaled by importing it into a VkSemaphore instead of a VkFence, and applications can check status or wait on the D3D12 fence using non-Vulkan APIs. The convenience of being able to do these operations on VkFence objects doesn’t justify the extra API complexity.

VK_KHR_external_memory

Name String

VK_KHR_external_memory

Extension Type

Device extension

Registered Extension Number

73

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-20

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Jason Ekstrand, Intel

  • Ian Elliot, Google

  • Jesse Hall, Google

  • Tobias Hector, Imagination Technologies

  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Matthew Netsch, Qualcomm Technologies, Inc.

  • Daniel Rakos, AMD

  • Carsten Rohde, NVIDIA

  • Ray Smith, ARM

  • Chad Versace, Google

An application may wish to reference device memory in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension enables an application to export non-Vulkan handles from Vulkan memory objects such that the underlying resources can be referenced outside the scope of the Vulkan logical device that created them.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_BUFFER_CREATE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_KHR

  • VK_QUEUE_FAMILY_EXTERNAL_KHR

  • VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR

New Enums

None.

New Functions

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

1) How do applications correlate two physical devices across process or Vulkan instance boundaries?

RESOLVED: New device ID fields have been introduced by VK_KHR_external_memory_capabilities. These fields, combined with the existing VkPhysicalDeviceProperties::driverVersion field can be used to identify compatible devices across processes, drivers, and APIs. VkPhysicalDeviceProperties::pipelineCacheUUID is not sufficient for this purpose because despite its description in the specification, it need only identify a unique pipeline cache format in practice. Multiple devices may be able to use the same pipeline cache data, and hence it would be desirable for all of them to have the same pipeline cache UUID. However, only the same concrete physical device can be used when sharing memory, so an actual unique device ID was introduced. Further, the pipeline cache UUID was specific to Vulkan, but correlation with other, non-extensible APIs is required to enable interoperation with those APIs.

2) If memory objects are shared between processes and APIs, is this considered aliasing according to the rules outlined in the Memory Aliasing section?

RESOLVED: Yes. Applications must take care to obey all restrictions imposed on aliased resources when using memory across multiple Vulkan instances or other APIs.

3) Are new image layouts or metadata required to specify image layouts and layout transitions compatible with non-Vulkan APIs, or with other instances of the same Vulkan driver?

RESOLVED: Separate instances of the same Vulkan driver running on the same GPU should have identical internal layout semantics, so applications have the tools they need to ensure views of images are consistent between the two instances. Other APIs will fall into two categories: Those that are Vulkan- compatible, and those that are Vulkan-incompatible. Vulkan-incompatible APIs will require the image to be in the GENERAL layout whenever they are accessing them.

Note this does not attempt to address cross-device transitions, nor transitions to engines on the same device which are not visible within the Vulkan API. Both of these are beyond the scope of this extension.

4) Is a new barrier flag or operation of some type needed to prepare external memory for handoff to another Vulkan instance or API and/or receive it from another instance or API?

RESOLVED: Yes. Some implementations need to perform additional cache management when transitioning memory between address spaces, and other APIs, instances, or processes may operate in a separate address space. Options for defining this transition include:

A new structure has the advantage that the type of external transition can be described in as much detail as necessary. However, there is not currently a known need for anything beyond differentiating external vs. internal accesses, so this is likely an over-engineered solution. The access flag bit has the advantage that it can be applied at buffer, image, or global granularity, and semantically it maps pretty well to the operation being described. Additionally, the API already includes VK_ACCESS_MEMORY_READ_BIT and VK_ACCESS_MEMORY_WRITE_BIT which appear to be intended for this purpose. However, there is no obvious pipeline stage that would correspond to an external access, and therefore no clear way to use VK_ACCESS_MEMORY_READ_BIT or VK_ACCESS_MEMORY_WRITE_BIT. VkDependencyFlags and VkPipelineStageFlags operate at command granularity rather than image or buffer granularity, which would make an entire pipeline barrier an internal→external or external→internal barrier. This may not be a problem in practice, but seems like the wrong scope. Another downside of VkDependencyFlags is that it lacks inherent directionality: There are not src and dst variants of it in the barrier or dependency description semantics, so two bits might need to be added to describe both internal→external and external→internal transitions. Transitioning a resource to a special queue family corresponds well with the operation of transitioning to a separate Vulkan instance, in that both operations ideally include scheduling a barrier on both sides of the transition: Both the releasing and the acquiring queue or process. Using a special queue family requires adding an additional reserved queue family index. Re-using VK_QUEUE_FAMILY_IGNORED would have left it unclear how to transition a concurrent usage resource from one process to another, since the semantics would have likely been equivalent to the currently-ignored transition of VK_QUEUE_FAMILY_IGNORED → VK_QUEUE_FAMILY_IGNORED. Fortunately, creating a new reserved queue family index is not invasive.

Based on the above analysis, the approach of transitioning to a special “external” queue family was chosen.

5) Do internal driver memory arrangements and/or other internal driver image properties need to be exported and imported when sharing images across processes or APIs.

RESOLVED: Some vendors claim this is necessary on their implementations, but it was determined that the security risks of allowing opaque meta data to be passed from applications to the driver were too high. Therefore, implementations which require metadata will need to associate it with the objects represented by the external handles, and rely on the dedicated allocation mechanism to associate the exported and imported memory objects with a single image or buffer.

6) Most prior interoperation and cross-process sharing APIs have been based on image-level sharing. Should Vulkan sharing be based on memory-object sharing or image sharing?

RESOLVED: These extensions have assumed memory-level sharing is the correct granularity. Vulkan is a lower-level API than most prior APIs, and as such attempts to closely align with to the underlying primitives of the hardware and system-level drivers it abstracts. In general, the resource that holds the backing store for both images and buffers of various types is memory. Images and buffers are merely metadata containing brief descriptions of the layout of bits within that memory.

Because memory object-based sharing is aligned with the overall Vulkan API design, it exposes the full power of Vulkan on external objects. External memory can be used as backing for sparse images, for example, whereas such usage would be awkward at best with a sharing mechanism based on higher-level primitives such as images. Further, aligning the mechanism with the API in this way provides some hope of trivial compatibility with future API enhancements. If new objects backed by memory objects are added to the API, they too can be used across processes with minimal additions to the base external memory APIs.

Earlier APIs implemented interop at a higher level, and this necessitated entirely separate sharing APIs for images and buffers. To co-exist and interoperate with those APIs, the Vulkan external sharing mechanism must accomodate their model. However, if it can be agreed that memory-based sharing is the more desirable and forward-looking design, legacy interoperation considerations can be considered another reason to favor memory-based sharing: While native and legacy driver primitives that may be used to implement sharing may not be as low-level as the API here suggests, raw memory is still the least common denominator among the types. Image-based sharing can be cleanly derived from a set of base memory- object sharing APIs with minimal effort, whereas image-based sharing does not generalize well to buffer or raw-memory sharing. Therefore, following the general Vulkan design principle of minimalism, it is better to expose even interopability with image-based native and external primitives via the memory sharing API, and place sufficient limits on their usage to ensure they can be used only as backing for equivalent Vulkan images. This provides a consistent API for applications regardless of which platform or external API they are targeting, which makes development of multi-API and multi-platform applications simpler.

7) Should Vulkan define a common external handle type and provide Vulkan functions to facilitate cross-process sharing of such handles rather than relying on native handles to define the external objects?

RESOLVED: No. Cross-process sharing of resources is best left to native platforms. There are myriad security and extensibility issues with such a mechanism, and attempting to re-solve all those issues within Vulkan does not align with Vulkan’s purpose as a graphics API. If desired, such a mechanism could be built as a layer or helper library on top of the opaque native handle defined in this family of extensions.

8) Must implementations provide additional guarantees about state implicitly included in memory objects for those memory objects that may be exported?

RESOLVED: Implementations must ensure that sharing memory objects does not transfer any information between the exporting and importing instances and APIs other than that required to share the data contained in the memory objects explicitly shared. As specific examples, data from previously freed memory objects that used the same underlying physical memory, and data from memory obects using adjacent physical memory must not be visible to applications importing an exported memory object.

9) Must implementations validate external handles the application provides as input to memory import operations?

RESOLVED: Implementations must return an error to the application if the provided memory handle cannot be used to complete the requested import operation. However, implementations need not validate handles are of the exact type specified by the application.

VK_KHR_external_memory_capabilities

Name String

VK_KHR_external_memory_capabilities

Extension Type

Instance extension

Registered Extension Number

72

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-17

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Ian Elliot, Google

  • Jesse Hall, Google

  • James Jones, NVIDIA

An application may wish to reference device memory in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension provides a set of capability queries and handle definitions that allow an application to determine what types of “external” memory handles an implementation supports for a given set of use cases.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_IMAGE_FORMAT_INFO_KHR

  • VK_STRUCTURE_TYPE_EXTERNAL_IMAGE_FORMAT_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_BUFFER_INFO_KHR

  • VK_STRUCTURE_TYPE_EXTERNAL_BUFFER_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES_KHR

  • VK_LUID_SIZE_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

1) Why do so many external memory capabilities need to be queried on a per-memory-handle-type basis?

PROPOSED RESOLUTION: This is because some handle types are based on OS-native objects that have far more limited capabilities than the very generic Vulkan memory objects. Not all memory handle types can name memory objects that support 3D images, for example. Some handle types cannot even support the deferred image and memory binding behavior of Vulkan and require specifying the image when allocating or importing the memory object.

2) Do the VkExternalImageFormatPropertiesKHR and VkExternalBufferPropertiesKHR structs need to include a list of memory type bits that support the given handle type?

PROPOSED RESOLUTION: No. The memory types that don’t support the handle types will simply be filtered out of the results returned by vkGetImageMemoryRequirements and vkGetBufferMemoryRequirements when a set of handle types was specified at image or buffer creation time.

3) Should the non-opaque handle types be moved to their own extension?

PROPOSED RESOLUTION: Perhaps. However, defining the handle type bits does very little and does not require any platform-specific types on its own, and it’s easier to maintain the bitfield values in a single extension for now. Presumably more handle types could be added by separate extensions though, and it would be midly weird to have some platform-specific ones defined in the core spec and some in extensions

4) Do we need a D3D11_TILEPOOL type?

PROPOSED RESOLUTION: No. This is technically possible, but the synchronization is awkward. D3D11 surfaces must be synchronized using shared mutexes, and these synchronization primitives are shared by the entire memory object, so D3D11 shared allocations divided among multiple buffer and image bindings may be difficult to synchronize.

5) Should the Windows 7-compatible handle types be named “KMT” handles or “GLOBAL_SHARE” handles?

PROPOSED RESOLUTION: KMT, simply because it is more concise.

6) How do applications identify compatible devices and drivers across instance, process, and API boundaries when sharing memory?

PROPOSED RESOLUTION: New device properties are exposed that allow applications to correctly correlate devices and drivers. A device and driver UUID that must both match to ensure sharing compatibility between two Vulkan instances, or a Vulkan instance and an extensible external API are added. To allow correlating with Direct3D devices, a device LUID is added that corresponds to a DXGI adapter LUID. A driver ID is not needed for Direct3D because mismatched driver component versions are not a currently supported configuration on the Windows OS. Should support for such configurations be introduced at the OS level, further Vulkan extensions would be needed to correlate userspace component builds.

VK_KHR_external_memory_fd

Name String

VK_KHR_external_memory_fd

Extension Type

Device extension

Registered Extension Number

75

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

An application may wish to reference device memory in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension enables an application to export POSIX file descriptor handles from Vulkan memory objects and to import Vulkan memory objects from POSIX file descriptor handles exported from other Vulkan memory objects or from similar resources in other APIs.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_MEMORY_FD_INFO_KHR

  • VK_STRUCTURE_TYPE_MEMORY_FD_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_MEMORY_GET_FD_INFO_KHR

New Enums

None.

Issues

1) Does the application need to close the file descriptor returned by vkGetMemoryFdKHR?

RESOLVED: Yes, unless it is passed back in to a driver instance to import the memory. A successful get call transfers ownership of the file descriptor to the application, and a successful import transfers it back to the driver. Destroying the original memory object will not close the file descriptor or remove its reference to the underlying memory resource associated with it.

2) Do drivers ever need to expose multiple file descriptors per memory object?

RESOLVED: No. This would indicate there are actually multiple memory objects, rather than a single memory object.

3) How should the valid size and memory type for POSIX file descriptor memory handles created outside of Vulkan be specified?

RESOLVED: The valid memory types are queried directly from the external handle. The size will be specified by future extensions that introduce such external memory handle types.

VK_KHR_external_memory_win32

Name String

VK_KHR_external_memory_win32

Extension Type

Device extension

Registered Extension Number

74

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Carsten Rohde, NVIDIA

An application may wish to reference device memory in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension enables an application to export Windows handles from Vulkan memory objects and to import Vulkan memory objects from Windows handles exported from other Vulkan memory objects or from similar resources in other APIs.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_MEMORY_WIN32_HANDLE_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_MEMORY_GET_WIN32_HANDLE_INFO_KHR

New Enums

None.

Issues

1) Do applications need to call CloseHandle() on the values returned from vkGetMemoryWin32HandleKHR when handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR?

editing-note

(Jon) This issue refers to a token from VK_KHR_external_semaphore_win32, but there’s no dependency or interaction with that extension defined above.

RESOLVED: Yes, unless it is passed back in to another driver instance to import the object. A successful get call transfers ownership of the handle to the application. Destroying the memory object will not destroy the handle or the handle’s reference to the underlying memory resource.

2) Should the language regarding KMT/Windows 7 handles be moved to a separate extension so that it can be deprecated over time?

RESOLVED: No. Support for them can be deprecated by drivers if they choose, by no longer returning them in the supported handle types of the instance level queries.

3) How should the valid size and memory type for windows memory handles created outside of Vulkan be specified?

RESOLVED: The valid memory types are queried directly from the external handle. The size is determined by the associated image or buffer memory requirements for external handle types that require dedicated allocations, and by the size specified when creating the object from which the handle was exported for other external handle types.

VK_KHR_external_semaphore

Name String

VK_KHR_external_semaphore

Extension Type

Device extension

Registered Extension Number

78

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jason Ekstrand, Intel

  • Jesse Hall, Google

  • Tobias Hector, Imagination Technologies

  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Matthew Netsch, Qualcomm Technologies, Inc.

  • Ray Smith, ARM

  • Chad Versace, Google

An application using external memory may wish to synchronize access to that memory using semaphores. This extension enables an application to create semaphores from which non-Vulkan handles that reference the underlying synchronization primitive can be exported.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO_KHR

  • VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR

New Functions

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

1) Should there be restrictions on what side effects can occur when waiting on imported semaphores that are in an invalid state?

RESOLVED: Yes. Normally, validating such state would be the responsibility of the application, and the implementation would be free to enter an undefined state if valid usage rules were violated. However, this could cause security concerns when using imported semaphores, as it would require the importing application to trust the exporting application to ensure the state is valid. Requiring this level of trust is undesireable for many potential use cases.

2) Must implementations validate external handles the application provides as input to semaphore state import operations?

RESOLVED: Implementations must return an error to the application if the provided semaphore state handle cannot be used to complete the requested import operation. However, implementations need not validate handles are of the exact type specified by the application.

VK_KHR_external_semaphore_capabilities

Name String

VK_KHR_external_semaphore_capabilities

Extension Type

Instance extension

Registered Extension Number

77

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-20

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

An application may wish to reference device semaphores in multiple Vulkan logical devices or instances, in multiple processes, and/or in multiple APIs. This extension provides a set of capability queries and handle definitions that allow an application to determine what types of “external” semaphore handles an implementation supports for a given set of use cases.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_SEMAPHORE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXTERNAL_SEMAPHORE_PROPERTIES_KHR

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_ID_PROPERTIES_KHR

  • VK_LUID_SIZE_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

VK_KHR_external_semaphore_fd

Name String

VK_KHR_external_semaphore_fd

Extension Type

Device extension

Registered Extension Number

80

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Contributors
  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Carsten Rohde, NVIDIA

An application using external memory may wish to synchronize access to that memory using semaphores. This extension enables an application to export semaphore payload to and import semaphore payload from POSIX file descriptors.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_FD_INFO_KHR

  • VK_STRUCTURE_TYPE_SEMAPHORE_GET_FD_INFO_KHR

New Enums

None.

Issues

1) Does the application need to close the file descriptor returned by vkGetSemaphoreFdKHR?

RESOLVED: Yes, unless it is passed back in to a driver instance to import the semaphore. A successful get call transfers ownership of the file descriptor to the application, and a successful import transfers it back to the driver. Destroying the original semaphore object will not close the file descriptor or remove its reference to the underlying semaphore resource associated with it.

VK_KHR_external_semaphore_win32

Name String

VK_KHR_external_semaphore_win32

Extension Type

Device extension

Registered Extension Number

79

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Carsten Rohde, NVIDIA

An application using external memory may wish to synchronize access to that memory using semaphores. This extension enables an application to export semaphore payload to and import semaphore payload from Windows handles.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR

  • VK_STRUCTURE_TYPE_D3D12_FENCE_SUBMIT_INFO_KHR

  • VK_STRUCTURE_TYPE_SEMAPHORE_GET_WIN32_HANDLE_INFO_KHR

New Enums

None.

Issues

1) Do applications need to call CloseHandle() on the values returned from vkGetSemaphoreWin32HandleKHR when handleType is VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT_KHR?

RESOLVED: Yes, unless it is passed back in to another driver instance to import the object. A successful get call transfers ownership of the handle to the application. Destroying the semaphore object will not destroy the handle or the handle’s reference to the underlying semaphore resource.

2) Should the language regarding KMT/Windows 7 handles be moved to a separate extension so that it can be deprecated over time?

RESOLVED: No. Support for them can be deprecated by drivers if they choose, by no longer returning them in the supported handle types of the instance level queries.

3) Should applications be allowed to specify additional object attributes for shared handles?

RESOLVED: Yes. Applications will be allowed to provide similar attributes to those they would to any other handle creation API.

4) How do applications communicate the desired fence values to use with D3D12_FENCE-based Vulkan semaphores?

RESOLVED: There are a couple of options. The values for the signaled and reset states could be communicated up front when creating the object and remain static for the life of the Vulkan semaphore, or they could be specified using auxiliary structures when submitting semaphore signal and wait operations, similar to what is done with the keyed mutex extensions. The latter is more flexible and consistent with the keyed mutex usage, but the former is a much simpler API.

Since Vulkan tends to favor flexibility and consistency over simplicity, a new structure specifying D3D12 fence acquire and release values is added to the vkQueueSubmit function.

VK_KHR_get_memory_requirements2

Name String

VK_KHR_get_memory_requirements2

Extension Type

Device extension

Registered Extension Number

147

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jason Ekstrand @jekstrand

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jason Ekstrand, Intel

  • Jeff Bolz, NVIDIA

  • Jesse Hall, Google

This extension provides new entry points to query memory requirements of images and buffers in a way that can be easily extended by other extensions, without introducing any further entry points. The Vulkan 1.0 VkMemoryRequirements and VkSparseImageMemoryRequirements structures do not include a sType/pNext, this extension wraps them in new structures with sType/pNext so an application can query a chain of memory requirements structures by constructing the chain and letting the implementation fill them in. A new command is added for each vkGet*MemoryRequrements command in core Vulkan 1.0.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_BUFFER_MEMORY_REQUIREMENTS_INFO_2_KHR

    • VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2_KHR

    • VK_STRUCTURE_TYPE_IMAGE_SPARSE_MEMORY_REQUIREMENTS_INFO_2_KHR

    • VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2_KHR

    • VK_STRUCTURE_TYPE_SPARSE_IMAGE_MEMORY_REQUIREMENTS_2_KHR

New Enums

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Version History

  • Revision 1, 2017-03-23 (Jason Ekstrand)

    • Internal revisions

VK_KHR_get_physical_device_properties2

Name String

VK_KHR_get_physical_device_properties2

Extension Type

Instance extension

Registered Extension Number

60

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

  • Ian Elliott, Google

This extension provides new entry points to query device features, device properties, and format properties in a way that can be easily extended by other extensions, without introducing any further entry points. The Vulkan 1.0 feature/limit/formatproperty structures do not include sType/pNext members. This extension wraps them in new structures with sType/pNext members, so an application can query a chain of feature/limit/formatproperty structures by constructing the chain and letting the implementation fill them in. A new command is added for each vkGetPhysicalDevice* command in core Vulkan 1.0. The new feature structure (and a chain of extension structures) can also be passed in to device creation to enable features.

This extension also allows applications to use the physical-device components of device extensions before vkCreateDevice is called.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_FORMAT_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2_KHR

    • VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_SPARSE_IMAGE_FORMAT_PROPERTIES_2_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SPARSE_IMAGE_FORMAT_INFO_2_KHR

New Enums

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Examples

    // Get features with a hypothetical future extension.
    VkHypotheticalExtensionFeaturesKHR hypotheticalFeatures =
    {
        VK_STRUCTURE_TYPE_HYPOTHETICAL_FEATURES_KHR,                            // sType
        NULL,                                                                   // pNext
    };

    VkPhysicalDeviceFeatures2KHR features =
    {
        VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2_KHR,                       // sType
        &hypotheticalFeatures,                                                  // pNext
    };

    // After this call, features and hypotheticalFeatures have been filled out.
    vkGetPhysicalDeviceFeatures2KHR(physicalDevice, &features);

    // Properties/limits can be chained and queried similarly.

    // Enable some features:
    VkHypotheticalExtensionFeaturesKHR enabledHypotheticalFeatures =
    {
        VK_STRUCTURE_TYPE_HYPOTHETICAL_FEATURES_KHR,                            // sType
        NULL,                                                                   // pNext
    };

    VkPhysicalDeviceFeatures2KHR enabledFeatures =
    {
        VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_FEATURES_2_KHR,                       // sType
        &enabledHypotheticalFeatures,                                           // pNext
    };

    enabledFeatures.features.xyz = VK_TRUE;
    enabledHypotheticalFeatures.abc = VK_TRUE;

    VkDeviceCreateInfo deviceCreateInfo =
    {
        VK_STRUCTURE_TYPE_DEVICE_CREATE_INFO,                                   // sType
        &enabledFeatures,                                                       // pNext
        ...
        NULL,                                                                   // pEnabledFeatures
    }

    VkDevice device;
    vkCreateDevice(physicalDevice, &deviceCreateInfo, NULL, &device);

Version History

  • Revision 1, 2016-09-12 (Jeff Bolz)

    • Internal revisions

  • Revision 2, 2016-11-02 (Ian Elliott)

    • Added ability for applications to use the physical-device components of device extensions before vkCreateDevice is called.

VK_KHR_get_surface_capabilities2

Name String

VK_KHR_get_surface_capabilities2

Extension Type

Instance extension

Registered Extension Number

120

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2017-02-27

IP Status

No known IP claims.

Contributors
  • Ian Elliott, Google

  • James Jones, NVIDIA

  • Alon Or-bach, Samsung

This extension provides new entry points to query device surface capabilities in a way that can be easily extended by other extensions, without introducing any further entry points. This extension can be considered the VK_KHR_surface equivalent of the VK_KHR_get_physical_device_properties2 extension.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SURFACE_INFO_2_KHR

    • VK_STRUCTURE_TYPE_SURFACE_CAPABILITIES_2_KHR

    • VK_STRUCTURE_TYPE_SURFACE_FORMAT_2_KHR

New Enums

None.

Issues

1) What should this extension be named?

RESOLVED: VK_KHR_get_surface_capabilities2. Other alternatives:

  • VK_KHR_surface2

  • One extension, combining a separate display-specific query extension.

2) Should additional WSI query functions be extended?

RESOLVED:

Version History

  • Revision 1, 2017-02-27 (James Jones)

    • Initial draft.

VK_KHR_image_format_list

Name String

VK_KHR_image_format_list

Extension Type

Device extension

Registered Extension Number

148

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jason Ekstrand @jekstrand

Last Modified Date

2017-03-20

IP Status

No known IP claims.

Contributors
  • Jason Ekstrand, Intel

  • Jan-Harald Fredriksen, ARM

  • Jeff Bolz, NVIDIA

  • Jeff Leger, Qualcomm

  • Neil Henning, Codeplay

On some implementations, setting the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT on image creation can cause access to that image to perform worse than an equivalent image created without VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT because the implementation does not know what view formats will be paired with the image.

This extension allows an application to provide the list of all formats that can be used with an image when it is created. The implementation may then be able to create a more efficient image that supports the subset of formats required by the application without having to support all formats in the format compatibility class of the image format.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_IMAGE_FORMAT_LIST_CREATE_INFO_KHR

New Enums

None.

New Functions

None.

Issues

VK_KHR_incremental_present

Name String

VK_KHR_incremental_present

Extension Type

Device extension

Registered Extension Number

85

Revision

1

Extension and Version Dependencies
Contact
  • Ian Elliott @ianelliottus

Last Modified Date

2016-11-02

IP Status

No known IP claims.

Contributors
  • Ian Elliott, Google

  • Jesse Hall, Google

  • Alon Or-bach, Samsung

  • James Jones, NVIDIA

  • Daniel Rakos, AMD

  • Ray Smith, ARM

  • Mika Isojarvi, Google

  • Jeff Juliano, NVIDIA

  • Jeff Bolz, NVIDIA

This device extension extends vkQueuePresentKHR, from the VK_KHR_swapchain extension, allowing an application to specify a list of rectangular, modified regions of each image to present. This should be used in situations where an application is only changing a small portion of the presentable images within a swapchain, since it enables the presentation engine to avoid wasting time presenting parts of the surface that haven’t changed.

This extension is leveraged from the EGL_KHR_swap_buffers_with_damage extension.

New Object Types

None.

New Enum Constants

New Enums

None.

New Functions

None.

Examples

None.

Issues

1) How should we handle steroescopic-3D swapchains? We need to add a layer for each rectangle. One approach is to create another struct that contains the VkRect2D plus layer, and have VkPresentRegionsKHR point to an array of that struct. Another approach is to have two parallel arrays, pRectangles and pLayers, where pRectangles[i] and pLayers[i] must be used together. Which approach should we use, and if the array of a new structure, what should that be called?

RESOLVED: Create a new structure, which is a VkRect2D plus a layer, and will be called VkRectLayerKHR.

2) Where is the origin of the VkRectLayerKHR?

RESOLVED: The upper left corner of the presentable image(s) of the swapchain, per the definition of framebuffer coordinates.

3) Does the rectangular region, VkRectLayerKHR, specify pixels of the swapchain’s image(s), or of the surface?

RESOLVED: Of the image(s). Some presentation engines may scale the pixels of a swapchain’s image(s) to the size of the surface. The size of the swapchain’s image(s) will be consistent, where the size of the surface may vary over time.

4) What if all of the rectangles for a given swapchain contain a width and/or height of zero?

RESOLVED: The application is indicating that no pixels changed since the last present. The presentation engine may use such a hint and not update any pixels for the swapchain. However, all other semantics of vkQueuePresentKHR must still be honored, including waiting for semaphores to signal.

Version History

  • Revision 1, 2016-11-02 (Ian Elliott)

    • Internal revisions

VK_KHR_maintenance1

Name String

VK_KHR_maintenance1

Extension Type

Device extension

Registered Extension Number

70

Revision

2

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2018-03-13

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Dan Ginsburg, Valve

  • Daniel Koch, NVIDIA

  • Daniel Rakos, AMD

  • Jan-Harald Fredriksen, ARM

  • Jason Ekstrand, Intel

  • Jeff Bolz, NVIDIA

  • Jesse Hall, Google

  • John Kessenich, Google

  • Michael Worcester, Imagination Technologies

  • Neil Henning, Codeplay Software Ltd.

  • Piers Daniell, NVIDIA

  • Slawomir Grajewski, Intel

  • Tobias Hector, Imagination Technologies

  • Tom Olson, ARM

VK_KHR_maintenance1 adds a collection of minor features that were intentionally left out or overlooked from the original Vulkan 1.0 release.

The new features are as follows:

  • Allow 2D and 2D array image views to be created from 3D images, which can then be used as color framebuffer attachments. This allows applications to render to slices of a 3D image.

  • Support vkCmdCopyImage between 2D array layers and 3D slices. This extension allows copying from layers of a 2D array image to slices of a 3D image and vice versa.

  • Allow negative height to be specified in the VkViewport::height field to perform y-inversion of the clip-space to framebuffer-space transform. This allows apps to avoid having to use gl_Position.y = -gl_Position.y in shaders also targeting other APIs.

  • Allow implementations to express support for doing just transfers and clears of image formats that they otherwise support no other format features for. This is done by adding new format feature flags VK_FORMAT_FEATURE_TRANSFER_SRC_BIT_KHR and VK_FORMAT_FEATURE_TRANSFER_DST_BIT_KHR.

  • Support vkCmdFillBuffer on transfer-only queues. Previously vkCmdFillBuffer was defined to only work on command buffers allocated from command pools which support graphics or compute queues. It is now allowed on queues that just support transfer operations.

  • Fix the inconsistency of how error conditions are returned between the vkCreateGraphicsPipelines and vkCreateComputePipelines functions and the vkAllocateDescriptorSets and vkAllocateCommandBuffers functions.

  • Add new VK_ERROR_OUT_OF_POOL_MEMORY_KHR error so implementations can give a more precise reason for vkAllocateDescriptorSets failures.

  • Add a new command vkTrimCommandPoolKHR which gives the implementation an opportunity to release any unused command pool memory back to the system.

New Object Types

None.

New Enum Constants

  • VK_ERROR_OUT_OF_POOL_MEMORY_KHR

  • VK_FORMAT_FEATURE_TRANSFER_SRC_BIT_KHR

  • VK_FORMAT_FEATURE_TRANSFER_DST_BIT_KHR

  • VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT_KHR

New Enums

None.

New Structures

None.

New Functions

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

  1. Are viewports with zero height allowed?

    RESOLVED: Yes, although they have low utility.

Version History

  • Revision 1, 2016-10-26 (Piers Daniell)

    • Internal revisions

  • Revision 2, 2018-03-13 (Jon Leech)

    • Add issue for zero-height viewports

VK_KHR_maintenance2

Name String

VK_KHR_maintenance2

Extension Type

Device extension

Registered Extension Number

118

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Michael Worcester @michaelworcester

Last Modified Date

2017-09-05

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Michael Worcester, Imagination Technologies

  • Stuart Smith, Imagination Technologies

  • Jeff Bolz, NVIDIA

  • Daniel Koch, NVIDIA

  • Jan-Harald Fredriksen, ARM

  • Daniel Rakos, AMD

  • Neil Henning, Codeplay

  • Piers Daniell, NVIDIA

VK_KHR_maintenance2 adds a collection of minor features that were intentionally left out or overlooked from the original Vulkan 1.0 release.

The new features are as follows:

  • Allow the application to specify which aspect of an input attachment might be read for a given subpass.

  • Allow implementations to express the clipping behavior of points.

  • Allow creating images with usage flags that may not be supported for the base image’s format, but are supported for image views of the image that have a different but compatible format.

  • Allow creating uncompressed image views of compressed images.

  • Allow the application to select between an upper-left and lower-left origin for the tessellation domain space.

  • Adds two new image layouts for depth stencil images to allow either the depth or stencil aspect to be read-only while the other aspect is writable.

Input Attachment Specification

Input attachment specification allows an application to specify which aspect of a multi-aspect image (e.g. a combined depth stencil format) will be accessed via a subpassLoad operation.

On some implementations there may be a performance penalty if the implementation does not know (at vkCreateRenderPass time) which aspect(s) of multi-aspect images can be be accessed as input attachments.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_POINT_CLIPPING_PROPERTIES_KHR

    • VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_PIPELINE_TESSELLATION_DOMAIN_ORIGIN_STATE_CREATE_INFO_KHR

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT_KHR

    • VK_IMAGE_CREATE_EXTENDED_USAGE_BIT_KHR

  • Extending VkImageLayout

    • VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL_KHR

    • VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL_KHR

  • VK_POINT_CLIPPING_BEHAVIOR_ALL_CLIP_PLANES_KHR

  • VK_POINT_CLIPPING_BEHAVIOR_USER_CLIP_PLANES_ONLY_KHR

New Functions

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Input Attachment Specification Example

Consider the case where a render pass has two subpasses and two attachments.

Attachment 0 has the format VK_FORMAT_D24_UNORM_S8_UINT, attachment 1 has some color format.

Subpass 0 writes to attachment 0, subpass 1 reads only the depth information from attachment 0 (using inputAttachmentRead) and writes to attachment 1.

    VkInputAttachmentAspectReferenceKHR references[] = {
        {
            .subpass = 1,
            .inputAttachmentIndex = 0,
            .aspectMask = VK_IMAGE_ASPECT_DEPTH_BIT
        }
    };

    VkRenderPassInputAttachmentAspectCreateInfoKHR specifyAspects = {
        .sType = VK_STRUCTURE_TYPE_RENDER_PASS_INPUT_ATTACHMENT_ASPECT_CREATE_INFO_KHR,
        .pNext = NULL,
        .aspectReferenceCount = 1,
        .pAspectReferences = references
    };


    VkRenderPassCreateInfo createInfo = {
        ...
        .pNext = &specifyAspects,
        ...
    }

    vkCreateRenderPass(...);

Issues

1) What is the default tessellation domain origin?

RESOLVED: Vulkan 1.0 originally inadvertently documented a lower-left origin, but the conformance tests and all implementations implemented an upper-left origin. This extension adds a control to select between lower-left (for compatibility with OpenGL) and upper-left, and we retroactively fix unextended Vulkan to have a default of an upper-left origin.

Version History

  • Revision 1, 2017-04-28

VK_KHR_maintenance3

Name String

VK_KHR_maintenance3

Extension Type

Device extension

Registered Extension Number

169

Revision

1

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Status

Draft

Last Modified Date

2017-09-05

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

VK_KHR_maintenance3 adds a collection of minor features that were intentionally left out or overlooked from the original Vulkan 1.0 release.

The new features are as follows:

  • A limit on the maximum number of descriptors that are supported in a single descriptor set layout. Some implementations have a limit on the total size of descriptors in a set, which can’t be expressed in terms of the limits in Vulkan 1.0.

  • A limit on the maximum size of a single memory allocation. Some platforms have kernel interfaces that limit the maximum size of an allocation.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MAINTENANCE_3_PROPERTIES_KHR

    • VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT_KHR

New Enums

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Version History

  • Revision 1, 2017-08-22

VK_KHR_mir_surface

Name String

VK_KHR_mir_surface

Extension Type

Instance extension

Registered Extension Number

8

Revision

4

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec,Ian Elliott @ianelliottus

Last Modified Date

2015-11-28

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_mir_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to a Mir surface, as well as a query to determine support for rendering to the windows desktop.

New Object Types

None

New Enum Constants

New Enums

None

New Structures

Issues

1) Does Mir need a way to query for compatibility between a particular physical device (and queue family?) and a specific Mir connection, screen, window, etc.?

RESOLVED: Yes, vkGetPhysicalDeviceMirPresentationSupportKHR was added to address this.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft, based on the previous contents of VK_EXT_KHR_swapchain (later renamed VK_EXT_KHR_surface).

  • Revision 2, 2015-10-02 (James Jones)

    • Added vkGetPhysicalDeviceMirPresentationSupportKHR to resolve issue #1.

  • Revision 3, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_mir_surface to VK_KHR_mir_surface.

  • Revision 4, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

VK_KHR_multiview

Name String

VK_KHR_multiview

Extension Type

Device extension

Registered Extension Number

54

Revision

1

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2016-10-28

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • Jeff Bolz, NVIDIA

This extension has the same goal as the OpenGL ES GL_OVR_multiview extension - it enables rendering to multiple “views” by recording a single set of commands to be executed with slightly different behavior for each view. It includes a concise way to declare a render pass with multiple views, and gives implementations freedom to render the views in the most efficient way possible.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_RENDER_PASS_MULTIVIEW_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_FEATURES_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PROPERTIES_KHR

  • Extending VkDependencyFlagBits

    • VK_DEPENDENCY_VIEW_LOCAL_BIT_KHR

New Enums

None.

New Functions

None.

New Built-In Variables

New SPIR-V Capabilities

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Issues

None.

Examples

None.

Version History

  • Revision 1, 2016-10-28 (Jeff Bolz)

    • Internal revisions

VK_KHR_push_descriptor

Name String

VK_KHR_push_descriptor

Extension Type

Device extension

Registered Extension Number

81

Revision

2

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-09-12

IP Status

No known IP claims.

Contributors
  • Jeff Bolz, NVIDIA

  • Michael Worcester, Imagination Technologies

This extension allows descriptors to be written into the command buffer, while the implementation is responsible for managing their memory. Push descriptors may enable easier porting from older APIs and in some cases can be more efficient than writing descriptors into descriptor sets.

New Object Types

None.

New Enum Constants

None.

Issues

None.

Examples

None.

Version History

  • Revision 1, 2016-10-15 (Jeff Bolz)

    • Internal revisions

  • Revision 2, 2017-09-12 (Tobias Hector)

    • Added interactions with Vulkan 1.1

VK_KHR_relaxed_block_layout

Name String

VK_KHR_relaxed_block_layout

Extension Type

Device extension

Registered Extension Number

145

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • John Kessenich @johnkslang

Last Modified Date

2017-03-26

IP Status

No known IP claims.

Interactions and External Dependencies
  • Promoted to Vulkan 1.1 Core

Contributors
  • John Kessenich, Google

The VK_KHR_relaxed_block_layout extension allows implementations to indicate they can support more variation in block Offset decorations. For example, placing a vector of three floats at an offset of 16*N + 4.

See Offset and Stride Assignment for details.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Version History

  • Revision 1, 2017-03-26 (JohnK)

VK_KHR_sampler_mirror_clamp_to_edge

Name String

VK_KHR_sampler_mirror_clamp_to_edge

Extension Type

Device extension

Registered Extension Number

15

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Tobias Hector @tobski

Last Modified Date

2016-02-16

Contributors
  • Tobias Hector, Imagination Technologies

VK_KHR_sampler_mirror_clamp_to_edge extends the set of sampler address modes to include an additional mode (VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE) that effectively uses a texture map twice as large as the original image in which the additional half of the new image is a mirror image of the original image.

This new mode relaxes the need to generate images whose opposite edges match by using the original image to generate a matching “mirror image”. This mode allows the texture to be mirrored only once in the negative s, t, and r directions.

New Enum Constants

Example

Creating a sampler with the new address mode in each dimension

    VkSamplerCreateInfo createInfo =
    {
        VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO // sType
        // Other members set to application-desired values
    };

    createInfo.addressModeU = VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE;
    createInfo.addressModeV = VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE;
    createInfo.addressModeW = VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE;

    VkSampler sampler;
    VkResult result = vkCreateSampler(
        device,
        &createInfo,
        &sampler);

Version History

  • Revision 1, 2016-02-16 (Tobias Hector)

    • Initial draft

VK_KHR_sampler_ycbcr_conversion

Name String

VK_KHR_sampler_ycbcr_conversion

Extension Type

Device extension

Registered Extension Number

157

Revision

1

Extension and Version Dependencies
Contact
  • Andrew Garrard @fluppeteer

Last Modified Date

2017-08-11

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Andrew Garrard, Samsung Electronics

  • Tobias Hector, Imagination Technologies

  • James Jones, NVIDIA

  • Daniel Koch, NVIDIA

  • Daniel Rakos, AMD

  • Romain Guy, Google

  • Jesse Hall, Google

  • Tom Cooksey, ARM Ltd

  • Jeff Leger, Qualcomm Technologies, Inc

  • Jan-Harald Fredriksen, ARM Ltd

  • Jan Outters, Samsung Electronics

  • Alon Or-bach, Samsung Electronics

  • Michael Worcester, Imagination Technologies

  • Jeff Bolz, NVIDIA

  • Tony Zlatinski, NVIDIA

  • Matthew Netsch, Qualcomm Technologies, Inc

This extension provides the ability to perform specified color space conversions during texture sampling operations. It also adds a selection of multi-planar formats, including the ability to bind memory to the planes of an image collectively or separately.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_IMAGE_PLANE_MEMORY_INFO_KHR

    • VK_STRUCTURE_TYPE_IMAGE_PLANE_MEMORY_REQUIREMENTS_INFO_KHR

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_YCBCR_CONVERSION_FEATURES_KHR

  • Extending VkFormat:

    • VK_FORMAT_G8B8G8R8_422_UNORM_KHR

    • VK_FORMAT_B8G8R8G8_422_UNORM_KHR

    • VK_FORMAT_G8_B8_R8_3PLANE_420_UNORM_KHR

    • VK_FORMAT_G8_B8R8_2PLANE_420_UNORM_KHR

    • VK_FORMAT_G8_B8_R8_3PLANE_422_UNORM_KHR

    • VK_FORMAT_G8_B8R8_2PLANE_422_UNORM_KHR

    • VK_FORMAT_G8_B8_R8_3PLANE_444_UNORM_KHR

    • VK_FORMAT_R10X6_UNORM_PACK16_KHR

    • VK_FORMAT_R10X6G10X6_UNORM_2PACK16_KHR

    • VK_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16_KHR

    • VK_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16_KHR

    • VK_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16_KHR

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16_KHR

    • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16_KHR

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16_KHR

    • VK_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16_KHR

    • VK_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16_KHR

    • VK_FORMAT_R12X4_UNORM_PACK16_KHR

    • VK_FORMAT_R12X4G12X4_UNORM_2PACK16_KHR

    • VK_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16_KHR

    • VK_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16_KHR

    • VK_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16_KHR

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16_KHR

    • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16_KHR

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16_KHR

    • VK_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16_KHR

    • VK_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16_KHR

    • VK_FORMAT_G16B16G16R16_422_UNORM_KHR

    • VK_FORMAT_B16G16R16G16_422_UNORM_KHR

    • VK_FORMAT_G16_B16_R16_3PLANE_420_UNORM_KHR

    • VK_FORMAT_G16_B16R16_2PLANE_420_UNORM_KHR

    • VK_FORMAT_G16_B16_R16_3PLANE_422_UNORM_KHR

    • VK_FORMAT_G16_B16R16_2PLANE_422_UNORM_KHR

    • VK_FORMAT_G16_B16_R16_3PLANE_444_UNORM_KHR

  • Extending VkImageAspectFlagBits:

    • VK_IMAGE_ASPECT_PLANE_0_BIT_KHR

    • VK_IMAGE_ASPECT_PLANE_1_BIT_KHR

    • VK_IMAGE_ASPECT_PLANE_2_BIT_KHR

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_CREATE_DISJOINT_BIT_KHR

  • Extending VkFormatFeatureFlagBits:

    • VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT_KHR

    • VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT_KHR

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT_KHR

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT_KHR

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT_KHR

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT_KHR

    • VK_FORMAT_FEATURE_DISJOINT_BIT_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted. The original type, enum and command names are still available as aliases of the core functionality.

Version History

  • Revision 1, 2017-01-24 (Andrew Garrard)

    • Initial draft

  • Revision 2, 2017-01-25 (Andrew Garrard)

    • After initial feedback

  • Revision 3, 2017-01-27 (Andrew Garrard)

    • Higher bit depth formats, renaming, swizzle

  • Revision 4, 2017-02-22 (Andrew Garrard)

    • Added query function, formats as RGB, clarifications

  • Revision 5, 2017-04 (Andrew Garrard)

    • Simplified query and removed output conversions

  • Version 6, 2017-4-24 (Andrew Garrard)

    • Tidying, incorporated new image query, restored transfer functions

  • Version 7, 2017-04-25 (Andrew Garrard)

    • Added cosited option/midpoint requirement for formats, "bypassConversion"

  • Version 8, 2017-04-25 (Andrew Garrard)

    • Simplified further

  • Version 9, 2017-04-27 (Andrew Garrard)

    • Disjoint no more

  • Version 10, 2017-04-28 (Andrew Garrard)

    • Restored disjoint

  • Version 11, 2017-04-29 (Andrew Garrard)

    • Now Ycbcr conversion, and KHR

  • Version 12, 2017-06-06 (Andrew Garrard)

    • Added conversion to image view creation

  • Version 13, 2017-07-13 (Andrew Garrard)

    • Allowed cosited-only chroma samples for formats

  • Version 14, 2017-08-11 (Andrew Garrard)

    • Reflected quantization changes in BT.2100-1

VK_KHR_shader_draw_parameters

Name String

VK_KHR_shader_draw_parameters

Extension Type

Device extension

Registered Extension Number

64

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Daniel Koch, NVIDIA Corporation

  • Jeff Bolz, NVIDIA

  • Daniel Rakos, AMD

  • Jan-Harald Fredriksen, ARM

  • John Kessenich, Google

  • Stuart Smith, IMG

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_KHR_shader_draw_parameters

The extension provides access to three additional built-in shader variables in Vulkan:

  • BaseInstance, which contains the firstInstance parameter passed to draw commands,

  • BaseVertex, which contains the firstVertex/vertexOffset parameter passed to draw commands, and

  • DrawIndex, which contains the index of the draw call currently being processed from an indirect draw call.

When using GLSL source-based shader languages, the following variables from GL_ARB_shader_draw_parameters can map to these SPIR-V built-in decorations:

  • in int gl_BaseInstanceARB;BaseInstance,

  • in int gl_BaseVertexARB;BaseVertex, and

  • in int gl_DrawIDARB;DrawIndex.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New Built-In Variables

New SPIR-V Capabilities

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, however a feature bit was added to distinguish whether it’s actually available or not.

Issues

1) Is this the same functionality as GL_ARB_shader_draw_parameters?

RESOLVED: It’s actually a superset as it also adds in support for arrayed drawing commands.

In GL for GL_ARB_shader_draw_parameters, gl_BaseVertexARB holds the integer value passed to the parameter to the command that resulted in the current shader invocation. In the case where the command has no baseVertex parameter, the value of gl_BaseVertexARB is zero. This means that gl_BaseVertexARB = baseVertex (for glDrawElements commands with baseVertex) or 0. In particular there are no glDrawArrays commands that take a baseVertex parameter.

Now in Vulkan, we have BaseVertex = vertexOffset (for indexed drawing commands) or firstVertex (for arrayed drawing commands), and so Vulkan’s version is really a superset of GL functionality.

Version History

  • Revision 1, 2016-10-05 (Daniel Koch)

    • Internal revisions

VK_KHR_shared_presentable_image

Name String

VK_KHR_shared_presentable_image

Extension Type

Device extension

Registered Extension Number

112

Revision

1

Extension and Version Dependencies
Contact
  • Alon Or-bach @alonorbach

Last Modified Date

2017-03-20

IP Status

No known IP claims.

Contributors
  • Alon Or-bach, Samsung Electronics

  • Ian Elliott, Google

  • Jesse Hall, Google

  • Pablo Ceballos, Google

  • Chris Forbes, Google

  • Jeff Juliano, NVIDIA

  • James Jones, NVIDIA

  • Daniel Rakos, AMD

  • Tobias Hector, Imagination Technologies

  • Graham Connor, Imagination Technologies

  • Michael Worcester, Imagination Technologies

  • Cass Everitt, Oculus

  • Johannes Van Waveren, Oculus

This extension extends VK_KHR_swapchain to enable creation of a shared presentable image. This allows the application to use the image while the presention engine is accessing it, in order to reduce the latency between rendering and presentation.

New Object Types

None.

New Enum Constants

  • Extending VkPresentModeKHR:

    • VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR

    • VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR

  • Extending VkImageLayout:

    • VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_SHARED_PRESENT_SURFACE_CAPABILITIES_KHR

New Enums

None.

New Functions

Issues

1) Should we allow a Vulkan WSI swapchain to toggle between normal usage and shared presentation usage?

RESOLVED: No. WSI swapchains are typically recreated with new properties instead of having their properties changed. This can also save resources, assuming that fewer images are needed for shared presentation, and assuming that most VR applications do not need to switch between normal and shared usage.

2) Should we have a query for determining how the presentation engine refresh is triggered?

RESOLVED: Yes. This is done via which presentation modes a surface supports.

3) Should the object representing a shared presentable image be an extension of a VkSwapchainKHR or a separate object?

RESOLVED: Extension of a swapchain due to overlap in creation properties and to allow common functionality between shared and normal presentable images and swapchains.

4) What should we call the extension and the new structures it creates?

RESOLVED: Shared presentable image / shared present.

5) Should the minImageCount and presentMode values of the VkSwapchainCreateInfoKHR be ignored, or required to be compatible values?

RESOLVED: minImageCount must be set to 1, and presentMode should be set to either VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR.

6) What should the layout of the shared presentable image be?

RESOLVED: After acquiring the shared presentable image, the application must transition it to the VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR layout prior to it being used. After this intial transition, any image usage that was requested during swapchain creation can be performed on the image without layout transitions being performed.

7) Do we need a new API for the trigger to refresh new content?

RESOLVED: vkQueuePresentKHR to act as API to trigger a refresh, as will allow combination with other compatible extensions to vkQueuePresentKHR.

8) How should an application detect a VK_ERROR_OUT_OF_DATE_KHR error on a swapchain using the VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR present mode?

RESOLVED: Introduce vkGetSwapchainStatusKHR to allow applications to query the status of a swapchain using a shared presentation mode.

9) What should subsequent calls to vkQueuePresentKHR for VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR swapchains be defined to do?

RESOLVED: State that implementations may use it as a hint for updated content.

10) Can the ownership of a shared presentable image be transferred to a different queue?

RESOLVED: No. It is not possible to transfer ownership of a shared presentable image obtained from a swapchain created using VK_SHARING_MODE_EXCLUSIVE after it has been presented.

11) How should vkQueueSubmit behave if a command buffer uses an image from an VK_ERROR_OUT_OF_DATE_KHR swapchain?

RESOLVED: vkQueueSubmit is expected to return the VK_ERROR_DEVICE_LOST error.

12) Can Vulkan provide any guarantee on the order of rendering, to enable beam chasing?

RESOLVED: This could be achieved via use of render passes to ensure strip rendering.

Version History

  • Revision 1, 2017-03-20 (Alon Or-bach)

    • Internal revisions

VK_KHR_storage_buffer_storage_class

Name String

VK_KHR_storage_buffer_storage_class

Extension Type

Device extension

Registered Extension Number

132

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Alexander Galazin @alegal-arm

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Alexander Galazin, ARM

  • David Neto, Google

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_KHR_storage_buffer_storage_class

This extension provides a new SPIR-V StorageBuffer storage class. A Block-decorated object in this class is equivalent to a BufferBlock-decorated object in the Uniform storage class.

New Enum Constants

None.

New Structures

None.

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1.

Issues

None.

Version History

  • Revision 1, 2017-03-23 (Alexander Galazin)

    • Initial draft

VK_KHR_surface

Name String

VK_KHR_surface

Extension Type

Instance extension

Registered Extension Number

1

Revision

25

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • James Jones @cubanismo,Ian Elliott @ianelliottus

Last Modified Date

2016-08-25

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Ian Elliott, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

  • Jason Ekstrand, Intel

The VK_KHR_surface extension is an instance extension. It introduces VkSurfaceKHR objects, which abstract native platform surface or window objects for use with Vulkan. It also provides a way to determine whether a queue family in a physical device supports presenting to particular surface.

Separate extensions for each platform provide the mechanisms for creating VkSurfaceKHR objects, but once created they may be used in this and other platform-independent extensions, in particular the VK_KHR_swapchain extension.

New Object Types

New Enum Constants

  • Extending VkResult:

    • VK_ERROR_SURFACE_LOST_KHR

    • VK_ERROR_NATIVE_WINDOW_IN_USE_KHR

Examples

Note

The example code for the VK_KHR_surface and VK_KHR_swapchain extensions was removed from the appendix after revision 1.0.29. This WSI example code was ported to the cube demo that is shipped with the official Khronos SDK, and is being kept up-to-date in that location (see: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/master/demos/cube.c).

Issues

1) Should this extension include a method to query whether a physical device supports presenting to a specific window or native surface on a given platform?

RESOLVED: Yes. Without this, applications would need to create a device instance to determine whether a particular window can be presented to. Knowing that a device supports presentation to a platform in general is not sufficient, as a single machine might support multiple seats, or instances of the platform that each use different underlying physical devices. Additionally, on some platforms, such as the X Window System, different drivers and devices might be used for different windows depending on which section of the desktop they exist on.

2) Should the vkGetPhysicalDeviceSurfaceCapabilitiesKHR, vkGetPhysicalDeviceSurfaceFormatsKHR, and vkGetPhysicalDeviceSurfacePresentModesKHR functions from VK_KHR_swapchain be modified to operate on physical devices and moved to this extension to implement the resolution of issue 1?

RESOLVED: No, separate query functions are needed, as the purposes served are similar but incompatible. The vkGetPhysicalDeviceSurface*KHR functions return information that could potentially depend on an initialized device. For example, the formats supported for presentation to the surface might vary depending on which device extensions are enabled. The query introduced to resolve issue 1 should be used only to query generic driver or platform properties. The physical device parameter is intended to serve only as an identifier rather than a stateful object.

3) Should Vulkan include support Xlib or XCB as the API for accessing the X Window System platform?

RESOLVED: Both. XCB is a more modern and efficient API, but Xlib usage is deeply ingrained in many applications and likely will remain in use for the foreseeable future. Not all drivers necessarily need to support both, but including both as options in the core specification will probably encourage support, which should in turn ease adoption of the Vulkan API in older codebases. Additionally, the performance improvements possible with XCB likely will not have a measurable impact on the performance of Vulkan presentation and other minimal window system interactions defined here.

4) Should the GBM platform be included in the list of platform enums?

RESOLVED: Deferred, and will be addressed with a platform-specific extension to be written in the future.

Version History

  • Revision 1, 2015-05-20 (James Jones)

    • Initial draft, based on LunarG KHR spec, other KHR specs, patches attached to bugs.

  • Revision 2, 2015-05-22 (Ian Elliott)

    • Created initial Description section.

    • Removed query for whether a platform requires the use of a queue for presentation, since it was decided that presentation will always be modeled as being part of the queue.

    • Fixed typos and other minor mistakes.

  • Revision 3, 2015-05-26 (Ian Elliott)

    • Improved the Description section.

  • Revision 4, 2015-05-27 (James Jones)

    • Fixed compilation errors in example code.

  • Revision 5, 2015-06-01 (James Jones)

    • Added issues 1 and 2 and made related spec updates.

  • Revision 6, 2015-06-01 (James Jones)

    • Merged the platform type mappings table previously removed from VK_KHR_swapchain with the platform description table in this spec.

    • Added issues 3 and 4 documenting choices made when building the initial list of native platforms supported.

  • Revision 7, 2015-06-11 (Ian Elliott)

    • Updated table 1 per input from the KHR TSG.

    • Updated issue 4 (GBM) per discussion with Daniel Stone. He will create a platform-specific extension sometime in the future.

  • Revision 8, 2015-06-17 (James Jones)

    • Updated enum-extending values using new convention.

    • Fixed the value of VK_SURFACE_PLATFORM_INFO_TYPE_SUPPORTED_KHR.

  • Revision 9, 2015-06-17 (James Jones)

    • Rebased on Vulkan API version 126.

  • Revision 10, 2015-06-18 (James Jones)

    • Marked issues 2 and 3 resolved.

  • Revision 11, 2015-06-23 (Ian Elliott)

    • Examples now show use of function pointers for extension functions.

    • Eliminated extraneous whitespace.

  • Revision 12, 2015-07-07 (Daniel Rakos)

    • Added error section describing when each error is expected to be reported.

    • Replaced the term "queue node index" with "queue family index" in the spec as that is the agreed term to be used in the latest version of the core header and spec.

    • Replaced bool32_t with VkBool32.

  • Revision 13, 2015-08-06 (Daniel Rakos)

    • Updated spec against latest core API header version.

  • Revision 14, 2015-08-20 (Ian Elliott)

    • Renamed this extension and all of its enumerations, types, functions, etc. This makes it compliant with the proposed standard for Vulkan extensions.

    • Switched from "revision" to "version", including use of the VK_MAKE_VERSION macro in the header file.

    • Did miscellaneous cleanup, etc.

  • Revision 15, 2015-08-20 (Ian Elliott—​porting a 2015-07-29 change from James Jones)

    • Moved the surface transform enums here from VK_WSI_swapchain so they could be re-used by VK_WSI_display.

  • Revision 16, 2015-09-01 (James Jones)

    • Restore single-field revision number.

  • Revision 17, 2015-09-01 (James Jones)

    • Fix example code compilation errors.

  • Revision 18, 2015-09-26 (Jesse Hall)

    • Replaced VkSurfaceDescriptionKHR with the VkSurfaceKHR object, which is created via layered extensions. Added VkDestroySurfaceKHR.

  • Revision 19, 2015-09-28 (Jesse Hall)

    • Renamed from VK_EXT_KHR_swapchain to VK_EXT_KHR_surface.

  • Revision 20, 2015-09-30 (Jeff Vigil)

    • Add error result VK_ERROR_SURFACE_LOST_KHR.

  • Revision 21, 2015-10-15 (Daniel Rakos)

    • Updated the resolution of issue #2 and include the surface capability queries in this extension.

    • Renamed SurfaceProperties to SurfaceCapabilities as it better reflects that the values returned are the capabilities of the surface on a particular device.

    • Other minor cleanup and consistency changes.

  • Revision 22, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_surface to VK_KHR_surface.

  • Revision 23, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to vkDestroySurfaceKHR.

  • Revision 24, 2015-11-10 (Jesse Hall)

    • Removed VkSurfaceTransformKHR. Use VkSurfaceTransformFlagBitsKHR instead.

    • Rename VkSurfaceCapabilitiesKHR member maxImageArraySize to maxImageArrayLayers.

  • Revision 25, 2016-01-14 (James Jones)

    • Moved VK_ERROR_NATIVE_WINDOW_IN_USE_KHR from the VK_KHR_android_surface to the VK_KHR_surface extension.

  • 2016-08-23 (Ian Elliott)

    • Update the example code, to not have so many characters per line, and to split out a new example to show how to obtain function pointers.

  • 2016-08-25 (Ian Elliott)

    • A note was added at the beginning of the example code, stating that it will be removed from future versions of the appendix.

VK_KHR_swapchain

Name String

VK_KHR_swapchain

Extension Type

Device extension

Registered Extension Number

2

Revision

70

Extension and Version Dependencies
Contact
  • James Jones @cubanismo,Ian Elliott @ianelliottus

Last Modified Date

2017-10-06

IP Status

No known IP claims.

Interactions and External Dependencies
  • Interacts with Vulkan 1.1

Contributors
  • Patrick Doane, Blizzard

  • Ian Elliott, LunarG

  • Jesse Hall, Google

  • Mathias Heyer, NVIDIA

  • James Jones, NVIDIA

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

  • Jason Ekstrand, Intel

  • Matthaeus G. Chajdas, AMD

  • Ray Smith, ARM

The VK_KHR_swapchain extension is the device-level companion to the VK_KHR_surface extension. It introduces VkSwapchainKHR objects, which provide the ability to present rendering results to a surface.

New Object Types

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_SWAPCHAIN_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_PRESENT_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_CAPABILITIES_KHR

    • VK_STRUCTURE_TYPE_IMAGE_SWAPCHAIN_CREATE_INFO_KHR

    • VK_STRUCTURE_TYPE_BIND_IMAGE_MEMORY_SWAPCHAIN_INFO_KHR

    • VK_STRUCTURE_TYPE_ACQUIRE_NEXT_IMAGE_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_PRESENT_INFO_KHR

    • VK_STRUCTURE_TYPE_DEVICE_GROUP_SWAPCHAIN_CREATE_INFO_KHR

  • Extending VkImageLayout:

    • VK_IMAGE_LAYOUT_PRESENT_SRC_KHR

  • Extending VkResult:

    • VK_SUBOPTIMAL_KHR

    • VK_ERROR_OUT_OF_DATE_KHR

Issues

1) Does this extension allow the application to specify the memory backing of the presentable images?

RESOLVED: No. Unlike standard images, the implementation will allocate the memory backing of the presentable image.

2) What operations are allowed on presentable images?

RESOLVED: This is determined by the image usage flags specified when creating the presentable image’s swapchain.

3) Does this extension support MSAA presentable images?

RESOLVED: No. Presentable images are always single-sampled. Multi-sampled rendering must use regular images. To present the rendering results the application must manually resolve the multi- sampled image to a single-sampled presentable image prior to presentation.

4) Does this extension support stereo/multi-view presentable images?

RESOLVED: Yes. The number of views associated with a presentable image is determined by the imageArrayLayers specified when creating a swapchain. All presentable images in a given swapchain use the same array size.

5) Are the layers of stereo presentable images half-sized?

RESOLVED: No. The image extents always match those requested by the application.

6) Do the “present” and “acquire next image” commands operate on a queue? If not, do they need to include explicit semaphore objects to interlock them with queue operations?

RESOLVED: The present command operates on a queue. The image ownership operation it represents happens in order with other operations on the queue, so no explicit semaphore object is required to synchronize its actions.

Applications may want to acquire the next image in separate threads from those in which they manage their queue, or in multiple threads. To make such usage easier, the acquire next image command takes a semaphore to signal as a method of explicit synchronization. The application must later queue a wait for this semaphore before queuing execution of any commands using the image.

7) Does vkAcquireNextImageKHR block if no images are available?

RESOLVED: The command takes a timeout parameter. Special values for the timeout are 0, which makes the call a non-blocking operation, and UINT64_MAX, which blocks indefinitely. Values in between will block for up to the specified time. The call will return when an image becomes available or an error occurs. It may, but is not required to, return before the specified timeout expires if the swapchain becomes out of date.

8) Can multiple presents be queued using one vkQueuePresentKHR call?

RESOLVED: Yes. VkPresentInfoKHR contains a list of swapchains and corresponding image indices that will be presented. When supported, all presentations queued with a single vkQueuePresentKHR call will be applied atomically as one operation. The same swapchain must not appear in the list more than once. Later extensions may provide applications stronger guarantees of atomicity for such present operations, and/or allow them to query whether atomic presentation of a particular group of swapchains is possible.

9) How do the presentation and acquire next image functions notify the application the targeted surface has changed?

RESOLVED: Two new result codes are introduced for this purpose:

  • VK_SUBOPTIMAL_KHR - Presentation will still succeed, subject to the window resize behavior, but the swapchain is no longer configured optimally for the surface it targets. Applications should query updated surface information and recreate their swapchain at the next convenient opportunity.

  • VK_ERROR_OUT_OF_DATE_KHR - Failure. The swapchain is no longer compatible with the surface it targets. The application must query updated surface information and recreate the swapchain before presentation will succeed.

These can be returned by both vkAcquireNextImageKHR and vkQueuePresentKHR.

10) Does the vkAcquireNextImageKHR command return a semaphore to the application via an output parameter, or accept a semaphore to signal from the application as an object handle parameter?

RESOLVED: Accept a semaphore to signal as an object handle. This avoids the need to specify whether the application must destroy the semaphore or whether it is owned by the swapchain, and if the latter, what its lifetime is and whether it can be re-used for other operations once it is received from vkAcquireNextImageKHR.

11) What types of swapchain queuing behavior should be exposed? Options include swap interval specification, mailbox/most recent vs. FIFO queue management, targeting specific vertical blank intervals or absolute times for a given present operation, and probably others. For some of these, whether they are specified at swapchain creation time or as per-present parameters needs to be decided as well.

RESOLVED: The base swapchain extension will expose 3 possible behaviors (of which, FIFO will always be supported):

  • Immediate present: Does not wait for vertical blanking period to update the current image, likely resulting in visible tearing. No internal queue is used. Present requests are applied immediately.

  • Mailbox queue: Waits for the next vertical blanking period to update the current image. No tearing should be observed. An internal single-entry queue is used to hold pending presentation requests. If the queue is full when a new presentation request is received, the new request replaces the existing entry, and any images associated with the prior entry become available for re-use by the application.

  • FIFO queue: Waits for the next vertical blanking period to update the current image. No tearing should be observed. An internal queue containing numSwapchainImages - 1 entries is used to hold pending presentation requests. New requests are appended to the end of the queue, and one request is removed from the beginning of the queue and processed during each vertical blanking period in which the queue is non-empty

Not all surfaces will support all of these modes, so the modes supported will be returned using a surface info query. All surfaces must support the FIFO queue mode. Applications must choose one of these modes up front when creating a swapchain. Switching modes can be accomplished by recreating the swapchain.

12) Can VK_PRESENT_MODE_MAILBOX_KHR provide non-blocking guarantees for vkAcquireNextImageKHR? If so, what is the proper criteria?

RESOLVED: Yes. The difficulty is not immediately obvious here. Naively, if at least 3 images are requested, mailbox mode should always have an image available for the application if the application does not own any images when the call to vkAcquireNextImageKHR was made. However, some presentation engines may have more than one “current” image, and would still need to block in some cases. The right requirement appears to be that if the application allocates the surface’s minimum number of images + 1 then it is guaranteed non-blocking behavior when it does not currently own any images.

13) Is there a way to create and initialize a new swapchain for a surface that has generated a VK_SUBOPTIMAL_KHR return code while still using the old swapchain?

RESOLVED: Not as part of this specification. This could be useful to allow the application to create an “optimal” replacement swapchain and rebuild all its command buffers using it in a background thread at a low priority while continuing to use the “suboptimal” swapchain in the main thread. It could probably use the same “atomic replace” semantics proposed for recreating direct-to-device swapchains without incurring a mode switch. However, after discussion, it was determined some platforms probably could not support concurrent swapchains for the same surface though, so this will be left out of the base KHR extensions. A future extension could add this for platforms where it is supported.

14) Should there be a special value for VkSurfaceCapabilitiesKHR::maxImageCount to indicate there are no practical limits on the number of images in a swapchain?

RESOLVED: Yes. There where often be cases where there is no practical limit to the number of images in a swapchain other than the amount of available resources (I.e., memory) in the system. Trying to derive a hard limit from things like memory size is prone to failure. It is better in such cases to leave it to applications to figure such soft limits out via trial/failure iterations.

15) Should there be a special value for VkSurfaceCapabilitiesKHR::currentExtent to indicate the size of the platform surface is undefined?

RESOLVED: Yes. On some platforms (Wayland, for example), the surface size is defined by the images presented to it rather than the other way around.

16) Should there be a special value for VkSurfaceCapabilitiesKHR::maxImageExtent to indicate there is no practical limit on the surface size?

RESOLVED: No. It seems unlikely such a system would exist. 0 could be used to indicate the platform places no limits on the extents beyond those imposed by Vulkan for normal images, but this query could just as easily return those same limits, so a special “unlimited” value does not seem useful for this field.

17) How should surface rotation and mirroring be exposed to applications? How do they specify rotation and mirroring transforms applied prior to presentation?

RESOLVED: Applications can query both the supported and current transforms of a surface. Both are specified relative to the device’s “natural” display rotation and direction. The supported transforms indicates which orientations the presentation engine accepts images in. For example, a presentation engine that does not support transforming surfaces as part of presentation, and which is presenting to a surface that is displayed with a 90-degree rotation, would return only one supported transform bit: VK_SURFACE_TRANSFORM_ROT90_BIT_KHR. Applications must transform their rendering by the transform they specify when creating the swapchain in preTransform field.

18) Can surfaces ever not support VK_MIRROR_NONE? Can they support vertical and horizontal mirroring simultaneously? Relatedly, should VK_MIRROR_NONE[_BIT] be zero, or bit one, and should applications be allowed to specify multiple pre and current mirror transform bits, or exactly one?

RESOLVED: Since some platforms may not support presenting with a transform other than the native window’s current transform, and prerotation/mirroring are specified relative to the device’s natural rotation and direction, rather than relative to the surface’s current rotation and direction, it is necessary to express lack of support for no mirroring. To allow this, the MIRROR_NONE enum must occupy a bit in the flags. Since MIRROR_NONE must be a bit in the bitmask rather than a bitmask with no values set, allowing more than one bit to be set in the bitmask would make it possible to describe undefined transforms such as VK_MIRROR_NONE_BIT | VK_MIRROR_HORIZONTAL_BIT, or a transform that includes both “no mirroring” and “horizontal mirroring” simultaneously. Therefore, it is desirable to allow specifying all supported mirroring transforms using only one bit. The question then becomes, should there be a VK_MIRROR_HORIZONTAL_AND_VERTICAL_BIT to represent a simultaneous horizontal and vertical mirror transform? However, such a transform is equivalent to a 180 degree rotation, so presentation engines and applications that wish to support or use such a transform can express it through rotation instead. Therefore, 3 exclusive bits are sufficient to express all needed mirroring transforms.

19) Should support for sRGB be required?

RESOLVED: In the advent of UHD and HDR display devices, proper color space information is vital to the display pipeline represented by the swapchain. The app can discover the supported format/color-space pairs and select a pair most suited to its rendering needs. Currently only the sRGB color space is supported, future extensions may provide support for more color spaces. See issues 23 and 24.

20) Is there a mechanism to modify or replace an existing swapchain with one targeting the same surface?

RESOLVED: Yes. This is described above in the text.

21) Should there be a way to set prerotation and mirroring using native APIs when presenting using a Vulkan swapchain?

RESOLVED: Yes. The transforms that can be expressed in this extension are a subset of those possible on native platforms. If a platform exposes a method to specify the transform of presented images for a given surface using native methods and exposes more transforms or other properties for surfaces than Vulkan supports, it might be impossible, difficult, or inconvenient to set some of those properties using Vulkan KHR extensions and some using the native interfaces. To avoid overwriting properties set using native commands when presenting using a Vulkan swapchain, the application can set the pretransform to “inherit”, in which case the current native properties will be used, or if none are available, a platform-specific default will be used. Platforms that do not specify a reasonable default or do not provide native mechanisms to specify such transforms should not include the inherit bits in the supportedTransforms bitmask they return in VkSurfaceCapabilitiesKHR.

22) Should the content of presentable images be clipped by objects obscuring their target surface?

RESOLVED: Applications can choose which behavior they prefer. Allowing the content to be clipped could enable more optimal presentation methods on some platforms, but some applications might rely on the content of presentable images to perform techniques such as partial updates or motion blurs.

23) What is the purpose of specifying a VkColorSpaceKHR along with VkFormat when creating a swapchain?

RESOLVED: While Vulkan itself is color space agnostic (e.g. even the meaning of R, G, B and A can be freely defined by the rendering application), the swapchain eventually will have to present the images on a display device with specific color reproduction characteristics. If any color space transformations are necessary before an image can be displayed, the color space of the presented image must be known to the swapchain. A swapchain will only support a restricted set of color format and -space pairs. This set can be discovered via vkGetPhysicalDeviceSurfaceFormatsKHR. As it can be expected that most display devices support the sRGB color space, at least one format/color-space pair has to be exposed, where the color space is VK_COLOR_SPACE_SRGB_NONLINEAR.

24) How are sRGB formats and the sRGB color space related?

RESOLVED: While Vulkan exposes a number of SRGB texture formats, using such formats does not guarantee working in a specific color space. It merely means that the hardware can directly support applying the non-linear transfer functions defined by the sRGB standard color space when reading from or writing to images of that these formats. Still, it is unlikely that a swapchain will expose a *_SRGB format along with any color space other than VK_COLOR_SPACE_SRGB_NONLINEAR.

On the other hand, non-*_SRGB formats will be very likely exposed in pair with a SRGB color space. This means, the hardware will not apply any transfer function when reading from or writing to such images, yet they will still be presented on a device with sRGB display characteristics. In this case the application is responsible for applying the transfer function, for instance by using shader math.

25) How are the lifetime of surfaces and swapchains targeting them related?

RESOLVED: A surface must outlive any swapchains targeting it. A VkSurfaceKHR owns the binding of the native window to the Vulkan driver.

26) How can the client control the way the alpha channel of swapchain images is treated by the presentation engine during compositing?

RESOLVED: We should add new enum values to allow the client to negotiate with the presentation engine on how to treat image alpha values during the compositing process. Since not all platforms can practically control this through the Vulkan driver, a value of VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR is provided like for surface transforms.

27) Is vkCreateSwapchainKHR the right function to return VK_ERROR_NATIVE_WINDOW_IN_USE_KHR, or should the various platform-specific VkSurfaceKHR factory functions catch this error earlier?

RESOLVED: For most platforms, the VkSurfaceKHR structure is a simple container holding the data that identifies a native window or other object representing a surface on a particular platform. For the surface factory functions to return this error, they would likely need to register a reference on the native objects with the native display server somehow, and ensure no other such references exist. Surfaces were not intended to be that heavyweight.

Swapchains are intended to be the objects that directly manipulate native windows and communicate with the native presentation mechanisms. Swapchains will already need to communicate with the native display server to negotiate allocation and/or presentation of presentable images for a native surface. Therefore, it makes more sense for swapchain creation to be the point at which native object exclusivity is enforced. Platforms may choose to enforce further restrictions on the number of VkSurfaceKHR objects that may be created for the same native window if such a requirement makes sense on a particular platform, but a global requirement is only sensible at the swapchain level.

Examples

Note

The example code for the VK_KHR_surface and VK_KHR_swapchain extensions was removed from the appendix after revision 1.0.29. This WSI example code was ported to the cube demo that is shipped with the official Khronos SDK, and is being kept up-to-date in that location (see: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/master/demos/cube.c).

Version History

  • Revision 1, 2015-05-20 (James Jones)

    • Initial draft, based on LunarG KHR spec, other KHR specs, patches attached to bugs.

  • Revision 2, 2015-05-22 (Ian Elliott)

    • Made many agreed-upon changes from 2015-05-21 KHR TSG meeting. This includes using only a queue for presentation, and having an explicit function to acquire the next image.

    • Fixed typos and other minor mistakes.

  • Revision 3, 2015-05-26 (Ian Elliott)

    • Improved the Description section.

    • Added or resolved issues that were found in improving the Description. For example, pSurfaceDescription is used consistently, instead of sometimes using pSurface.

  • Revision 4, 2015-05-27 (James Jones)

    • Fixed some grammatical errors and typos

    • Filled in the description of imageUseFlags when creating a swapchain.

    • Added a description of swapInterval.

    • Replaced the paragraph describing the order of operations on a queue for image ownership and presentation.

  • Revision 5, 2015-05-27 (James Jones)

    • Imported relevant issues from the (abandoned) vk_wsi_persistent_swapchain_images extension.

    • Added issues 6 and 7, regarding behavior of the acquire next image and present commands with respect to queues.

    • Updated spec language and examples to align with proposed resolutions to issues 6 and 7.

  • Revision 6, 2015-05-27 (James Jones)

    • Added issue 8, regarding atomic presentation of multiple swapchains

    • Updated spec language and examples to align with proposed resolution to issue 8.

  • Revision 7, 2015-05-27 (James Jones)

    • Fixed compilation errors in example code, and made related spec fixes.

  • Revision 8, 2015-05-27 (James Jones)

    • Added issue 9, and the related VK_SUBOPTIMAL_KHR result code.

    • Renamed VK_OUT_OF_DATE_KHR to VK_ERROR_OUT_OF_DATE_KHR.

  • Revision 9, 2015-05-27 (James Jones)

    • Added inline proposed resolutions (marked with [JRJ]) to some XXX questions/issues. These should be moved to the issues section in a subsequent update if the proposals are adopted.

  • Revision 10, 2015-05-28 (James Jones)

    • Converted vkAcquireNextImageKHR back to a non-queue operation that uses a VkSemaphore object for explicit synchronization.

    • Added issue 10 to determine whether vkAcquireNextImageKHR generates or returns semaphores, or whether it operates on a semaphore provided by the application.

  • Revision 11, 2015-05-28 (James Jones)

    • Marked issues 6, 7, and 8 resolved.

    • Renamed VkSurfaceCapabilityPropertiesKHR to VkSurfacePropertiesKHR to better convey the mutable nature of the info it contains.

  • Revision 12, 2015-05-28 (James Jones)

    • Added issue 11 with a proposed resolution, and the related issue 12.

    • Updated various sections of the spec to match the proposed resolution to issue 11.

  • Revision 13, 2015-06-01 (James Jones)

    • Moved some structures to VK_EXT_KHR_swap_chain to resolve the spec’s issues 1 and 2.

  • Revision 14, 2015-06-01 (James Jones)

    • Added code for example 4 demonstrating how an application might make use of the two different present and acquire next image KHR result codes.

    • Added issue 13.

  • Revision 15, 2015-06-01 (James Jones)

    • Added issues 14 - 16 and related spec language.

    • Fixed some spelling errors.

    • Added language describing the meaningful return values for vkAcquireNextImageKHR and vkQueuePresentKHR.

  • Revision 16, 2015-06-02 (James Jones)

    • Added issues 17 and 18, as well as related spec language.

    • Removed some erroneous text added by mistake in the last update.

  • Revision 17, 2015-06-15 (Ian Elliott)

    • Changed special value from "-1" to "0" so that the data types can be unsigned.

  • Revision 18, 2015-06-15 (Ian Elliott)

    • Clarified the values of VkSurfacePropertiesKHR::minImageCount and the timeout parameter of the vkAcquireNextImageKHR function.

  • Revision 19, 2015-06-17 (James Jones)

    • Misc. cleanup. Removed resolved inline issues and fixed typos.

    • Fixed clarification of VkSurfacePropertiesKHR::minImageCount made in version 18.

    • Added a brief "Image Ownership" definition to the list of terms used in the spec.

  • Revision 20, 2015-06-17 (James Jones)

    • Updated enum-extending values using new convention.

  • Revision 21, 2015-06-17 (James Jones)

    • Added language describing how to use VK_IMAGE_LAYOUT_PRESENT_SOURCE_KHR.

    • Cleaned up an XXX comment regarding the description of which queues vkQueuePresentKHR can be used on.

  • Revision 22, 2015-06-17 (James Jones)

    • Rebased on Vulkan API version 126.

  • Revision 23, 2015-06-18 (James Jones)

    • Updated language for issue 12 to read as a proposed resolution.

    • Marked issues 11, 12, 13, 16, and 17 resolved.

    • Temporarily added links to the relevant bugs under the remaining unresolved issues.

    • Added issues 19 and 20 as well as proposed resolutions.

  • Revision 24, 2015-06-19 (Ian Elliott)

    • Changed special value for VkSurfacePropertiesKHR::currentExtent back to "-1" from "0". This value will never need to be unsigned, and "0" is actually a legal value.

  • Revision 25, 2015-06-23 (Ian Elliott)

    • Examples now show use of function pointers for extension functions.

    • Eliminated extraneous whitespace.

  • Revision 26, 2015-06-25 (Ian Elliott)

    • Resolved Issues 9 & 10 per KHR TSG meeting.

  • Revision 27, 2015-06-25 (James Jones)

    • Added oldSwapchain member to VkSwapchainCreateInfoKHR.

  • Revision 28, 2015-06-25 (James Jones)

    • Added the "inherit" bits to the rotation and mirroring flags and the associated issue 21.

  • Revision 29, 2015-06-25 (James Jones)

    • Added the "clipped" flag to VkSwapchainCreateInfoKHR, and the associated issue 22.

    • Specified that presenting an image does not modify it.

  • Revision 30, 2015-06-25 (James Jones)

    • Added language to the spec that clarifies the behavior of vkCreateSwapchainKHR() when the oldSwapchain field of VkSwapchainCreateInfoKHR is not NULL.

  • Revision 31, 2015-06-26 (Ian Elliott)

    • Example of new VkSwapchainCreateInfoKHR members, "oldSwapchain" and "clipped".

    • Example of using VkSurfacePropertiesKHR::{min|max}ImageCount to set VkSwapchainCreateInfoKHR::minImageCount.

    • Rename vkGetSurfaceInfoKHR()'s 4th parameter to "pDataSize", for consistency with other functions.

    • Add macro with C-string name of extension (just to header file).

  • Revision 32, 2015-06-26 (James Jones)

    • Minor adjustments to the language describing the behavior of "oldSwapchain"

    • Fixed the version date on my previous two updates.

  • Revision 33, 2015-06-26 (Jesse Hall)

    • Add usage flags to VkSwapchainCreateInfoKHR

  • Revision 34, 2015-06-26 (Ian Elliott)

    • Rename vkQueuePresentKHR()'s 2nd parameter to "pPresentInfo", for consistency with other functions.

  • Revision 35, 2015-06-26 (Jason Ekstrand)

    • Merged the VkRotationFlagBitsKHR and VkMirrorFlagBitsKHR enums into a single VkSurfaceTransformFlagBitsKHR enum.

  • Revision 36, 2015-06-26 (Jason Ekstrand)

    • Added a VkSurfaceTransformKHR enum that is not a bitmask. Each value in VkSurfaceTransformKHR corresponds directly to one of the bits in VkSurfaceTransformFlagBitsKHR so transforming from one to the other is easy. Having a separate enum means that currentTransform and preTransform are now unambiguous by definition.

  • Revision 37, 2015-06-29 (Ian Elliott)

    • Corrected one of the signatures of vkAcquireNextImageKHR, which had the last two parameters switched from what it is elsewhere in the specification and header files.

  • Revision 38, 2015-06-30 (Ian Elliott)

    • Corrected a typo in description of the vkGetSwapchainInfoKHR() function.

    • Corrected a typo in header file comment for VkPresentInfoKHR::sType.

  • Revision 39, 2015-07-07 (Daniel Rakos)

    • Added error section describing when each error is expected to be reported.

    • Replaced bool32_t with VkBool32.

  • Revision 40, 2015-07-10 (Ian Elliott)

    • Updated to work with version 138 of the "vulkan.h" header. This includes declaring the VkSwapchainKHR type using the new VK_DEFINE_NONDISP_HANDLE macro, and no longer extending VkObjectType (which was eliminated).

  • Revision 41 2015-07-09 (Mathias Heyer)

    • Added color space language.

  • Revision 42, 2015-07-10 (Daniel Rakos)

    • Updated query mechanism to reflect the convention changes done in the core spec.

    • Removed "queue" from the name of VK_STRUCTURE_TYPE_QUEUE_PRESENT_INFO_KHR to be consistent with the established naming convention.

    • Removed reference to the no longer existing VkObjectType enum.

  • Revision 43, 2015-07-17 (Daniel Rakos)

    • Added support for concurrent sharing of swapchain images across queue families.

    • Updated sample code based on recent changes

  • Revision 44, 2015-07-27 (Ian Elliott)

    • Noted that support for VK_PRESENT_MODE_FIFO_KHR is required. That is ICDs may optionally support IMMEDIATE and MAILBOX, but must support FIFO.

  • Revision 45, 2015-08-07 (Ian Elliott)

    • Corrected a typo in spec file (type and variable name had wrong case for the imageColorSpace member of the VkSwapchainCreateInfoKHR struct).

    • Corrected a typo in header file (last parameter in PFN_vkGetSurfacePropertiesKHR was missing "KHR" at the end of type: VkSurfacePropertiesKHR).

  • Revision 46, 2015-08-20 (Ian Elliott)

    • Renamed this extension and all of its enumerations, types, functions, etc. This makes it compliant with the proposed standard for Vulkan extensions.

    • Switched from "revision" to "version", including use of the VK_MAKE_VERSION macro in the header file.

    • Made improvements to several descriptions.

    • Changed the status of several issues from PROPOSED to RESOLVED, leaving no unresolved issues.

    • Resolved several TODOs, did miscellaneous cleanup, etc.

  • Revision 47, 2015-08-20 (Ian Elliott—​porting a 2015-07-29 change from James Jones)

    • Moved the surface transform enums to VK_WSI_swapchain so they could be re-used by VK_WSI_display.

  • Revision 48, 2015-09-01 (James Jones)

    • Various minor cleanups.

  • Revision 49, 2015-09-01 (James Jones)

    • Restore single-field revision number.

  • Revision 50, 2015-09-01 (James Jones)

    • Update Example #4 to include code that illustrates how to use the oldSwapchain field.

  • Revision 51, 2015-09-01 (James Jones)

    • Fix example code compilation errors.

  • Revision 52, 2015-09-08 (Matthaeus G. Chajdas)

    • Corrected a typo.

  • Revision 53, 2015-09-10 (Alon Or-bach)

    • Removed underscore from SWAP_CHAIN left in VK_STRUCTURE_TYPE_SWAPCHAIN_CREATE_INFO_KHR.

  • Revision 54, 2015-09-11 (Jesse Hall)

    • Described the execution and memory coherence requirements for image transitions to and from VK_IMAGE_LAYOUT_PRESENT_SOURCE_KHR.

  • Revision 55, 2015-09-11 (Ray Smith)

    • Added errors for destroying and binding memory to presentable images

  • Revision 56, 2015-09-18 (James Jones)

    • Added fence argument to vkAcquireNextImageKHR

    • Added example of how to meter a host thread based on presentation rate.

  • Revision 57, 2015-09-26 (Jesse Hall)

    • Replace VkSurfaceDescriptionKHR with VkSurfaceKHR.

    • Added issue 25 with agreed resolution.

  • Revision 58, 2015-09-28 (Jesse Hall)

    • Renamed from VK_EXT_KHR_device_swapchain to VK_EXT_KHR_swapchain.

  • Revision 59, 2015-09-29 (Ian Elliott)

    • Changed vkDestroySwapchainKHR() to return void.

  • Revision 60, 2015-10-01 (Jeff Vigil)

    • Added error result VK_ERROR_SURFACE_LOST_KHR.

  • Revision 61, 2015-10-05 (Jason Ekstrand)

    • Added the VkCompositeAlpha enum and corresponding structure fields.

  • Revision 62, 2015-10-12 (Daniel Rakos)

    • Added VK_PRESENT_MODE_FIFO_RELAXED_KHR.

  • Revision 63, 2015-10-15 (Daniel Rakos)

    • Moved surface capability queries to VK_EXT_KHR_surface.

  • Revision 64, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_swapchain to VK_KHR_swapchain.

  • Revision 65, 2015-10-28 (Ian Elliott)

    • Added optional pResult member to VkPresentInfoKHR, so that per-swapchain results can be obtained from vkQueuePresentKHR().

  • Revision 66, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to create and destroy functions.

    • Updated resource transition language.

    • Updated sample code.

  • Revision 67, 2015-11-10 (Jesse Hall)

    • Add reserved flags bitmask to VkSwapchainCreateInfoKHR.

    • Modify naming and member ordering to match API style conventions, and so the VkSwapchainCreateInfoKHR image property members mirror corresponding VkImageCreateInfo members but with an 'image' prefix.

    • Make VkPresentInfoKHR::pResults non-const; it is an output array parameter.

    • Make pPresentInfo parameter to vkQueuePresentKHR const.

  • Revision 68, 2016-04-05 (Ian Elliott)

    • Moved the "validity" include for vkAcquireNextImage to be in its proper place, after the prototype and list of parameters.

    • Clarified language about presentable images, including how they are acquired, when applications can and cannot use them, etc. As part of this, removed language about "ownership" of presentable images, and replaced it with more-consistent language about presentable images being "acquired" by the application.

  • 2016-08-23 (Ian Elliott)

    • Update the example code, to use the final API command names, to not have so many characters per line, and to split out a new example to show how to obtain function pointers. This code is more similar to the LunarG "cube" demo program.

  • 2016-08-25 (Ian Elliott)

    • A note was added at the beginning of the example code, stating that it will be removed from future versions of the appendix.

  • Revision 69, 2017-09-07 (Tobias Hector)

    • Added interactions with Vulkan 1.1

  • Revision 70, 2017-10-06 (Ian Elliott)

    • Corrected interactions with Vulkan 1.1

VK_KHR_variable_pointers

Name String

VK_KHR_variable_pointers

Extension Type

Device extension

Registered Extension Number

121

Revision

1

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Last Modified Date

2017-09-05

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • John Kessenich, Google

  • Neil Henning, Codeplay

  • David Neto, Google

  • Daniel Koch, Nvidia

  • Graeme Leese, Broadcom

  • Weifeng Zhang, Qualcomm

  • Stephen Clarke, Imagination Technologies

  • Jason Ekstrand, Intel

  • Jesse Hall, Google

The VK_KHR_variable_pointers extension allows implementations to indicate their level of support for the SPV_KHR_variable_pointers SPIR-V extension. The SPIR-V extension allows shader modules to use invocation-private pointers into uniform and/or storage buffers, where the pointer values can be dynamic and non-uniform.

The SPV_KHR_variable_pointers extension introduces two capabilities. The first, VariablePointersStorageBuffer, must be supported by all implementations of this extension. The second, VariablePointers, is optional.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VARIABLE_POINTER_FEATURES_KHR

Promotion to Vulkan 1.1

All functionality in this extension is included in core Vulkan 1.1, with the KHR suffix omitted, however support for the variablePointersStorageBuffer feature is made optional. The original type, enum and command names are still available as aliases of the core functionality.

Issues

1) Do we need an optional property for the SPIR-V VariablePointersStorageBuffer capability or should it be mandatory when this extension is advertised?

RESOLVED: Add it as a distinct feature, but make support mandatory. Adding it as a feature makes the extension easier to include in a future core API version. In the extension, the feature is mandatory, so that presence of the extension guarantees some functionality. When included in a core API version, the feature would be optional.

2) Can support for these capabilities vary between shader stages?

RESOLVED: No, if the capability is supported in any stage it must be supported in all stages.

3) Should the capabilities be features or limits?

RESOLVED: Features, primarily for consistency with other similar extensions.

Version History

  • Revision 1, 2017-03-14 (Jesse Hall and John Kessenich)

    • Internal revisions

VK_KHR_wayland_surface

Name String

VK_KHR_wayland_surface

Extension Type

Instance extension

Registered Extension Number

7

Revision

6

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec,Ian Elliott @ianelliottus

Last Modified Date

2015-11-28

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_wayland_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to a Wayland wl_surface, as well as a query to determine support for rendering to a Wayland compositor.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_WAYLAND_SURFACE_CREATE_INFO_KHR

New Enums

None

Issues

1) Does Wayland need a way to query for compatibility between a particular physical device and a specific Wayland display? This would be a more general query than vkGetPhysicalDeviceSurfaceSupportKHR: if the Wayland-specific query returned VK_TRUE for a (VkPhysicalDevice, struct wl_display*) pair, then the physical device could be assumed to support presentation to any VkSurfaceKHR for surfaces on the display.

RESOLVED: Yes. vkGetPhysicalDeviceWaylandPresentationSupportKHR was added to address this issue.

2) Should we require surfaces created with vkCreateWaylandSurfaceKHR to support the VK_PRESENT_MODE_MAILBOX_KHR present mode?

RESOLVED: Yes. Wayland is an inherently mailbox window system and mailbox support is required for some Wayland compositor interactions to work as expected. While handling these interactions may be possible with VK_PRESENT_MODE_FIFO_KHR, it is much more difficult to do without deadlock and requiring all Wayland applications to be able to support implementations which only support VK_PRESENT_MODE_FIFO_KHR would be an onerous restriction on application developers.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft, based on the previous contents of VK_EXT_KHR_swapchain (later renamed VK_EXT_KHR_surface).

  • Revision 2, 2015-10-02 (James Jones)

    • Added vkGetPhysicalDeviceWaylandPresentationSupportKHR() to resolve issue #1.

    • Adjusted wording of issue #1 to match the agreed-upon solution.

    • Renamed "window" parameters to "surface" to match Wayland conventions.

  • Revision 3, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_wayland_surface to VK_KHR_wayland_surface.

  • Revision 4, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to vkCreateWaylandSurfaceKHR.

  • Revision 5, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

  • Revision 6, 2017-02-08 (Jason Ekstrand)

    • Added the requirement that implementations support VK_PRESENT_MODE_MAILBOX_KHR.

    • Added wording about interactions between vkQueuePresentKHR and the Wayland requests sent to the compositor.

VK_KHR_win32_keyed_mutex

Name String

VK_KHR_win32_keyed_mutex

Extension Type

Device extension

Registered Extension Number

76

Revision

1

Extension and Version Dependencies
Contact
  • Carsten Rohde @crohde

Last Modified Date

2016-10-21

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Jeff Juliano, NVIDIA

  • Carsten Rohde, NVIDIA

Applications that wish to import Direct3D 11 memory objects into the Vulkan API may wish to use the native keyed mutex mechanism to synchronize access to the memory between Vulkan and Direct3D. This extension provides a way for an application to access the keyed mutex associated with an imported Vulkan memory object when submitting command buffers to a queue.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_KHR

New Enums

None.

New Functions

None.

Issues

None.

VK_KHR_win32_surface

Name String

VK_KHR_win32_surface

Extension Type

Instance extension

Registered Extension Number

10

Revision

6

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec,Ian Elliott @ianelliottus

Last Modified Date

2017-04-24

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_win32_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to a Win32 HWND, as well as a query to determine support for rendering to the windows desktop.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_WIN32_SURFACE_CREATE_INFO_KHR

New Enums

None

Issues

1) Does Win32 need a way to query for compatibility between a particular physical device and a specific screen? Compatibility between a physical device and a window generally only depends on what screen the window is on. However, there is not an obvious way to identify a screen without already having a window on the screen.

RESOLVED: No. While it may be useful, there is not a clear way to do this on Win32. However, a method was added to query support for presenting to the windows desktop as a whole.

2) If a native window object (HWND) is used by one graphics API, and then is later used by a different graphics API (one of which is Vulkan), can these uses interfere with each other?

RESOLVED: Yes.

Uses of a window object by multiple graphics APIs results in undefined behavior. Such behavior may succeed when using one Vulkan implementation but fail when using a different Vulkan implementation. Potential failures include:

  • Creating then destroying a flip presentation model DXGI swapchain on a window object can prevent vkCreateSwapchainKHR from succeeding on the same window object.

  • Creating then destroying a VkSwapchainKHR on a window object can prevent creation of a bitblt model DXGI swapchain on the same window object.

  • Creating then destroying a VkSwapchainKHR on a window object can effectively SetPixelFormat to a different format than the format chosen by an OpenGL application.

  • Creating then destroying a VkSwapchainKHR on a window object on one VkPhysicalDevice can prevent vkCreateSwapchainKHR from succeeding on the same window object, but on a different VkPhysicalDevice that is associated with a different Vulkan ICD.

In all cases the problem can be worked around by creating a new window object.

Technical details include:

  • Creating a DXGI swapchain over a window object can alter the object for the remainder of its lifetime. The alteration persists even after the DXGI swapchain has been destroyed. This alteration can make it impossible for a conformant Vulkan implementation to create a VkSwapchainKHR over the same window object. Mention of this alteration can be found in the remarks section of the MSDN documentation for DXGI_SWAP_EFFECT.

  • Calling GDI’s SetPixelFormat (needed by OpenGL’s WGL layer) on a window object alters the object for the remainder of its lifetime. The MSDN documentation for SetPixelFormat explains that a window object’s pixel format can be set only one time.

  • Creating a VkSwapchainKHR over a window object can alter the object for the remaining life of its lifetime. Either of the above alterations may occur as a side-effect of VkSwapchainKHR.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft, based on the previous contents of VK_EXT_KHR_swapchain (later renamed VK_EXT_KHR_surface).

  • Revision 2, 2015-10-02 (James Jones)

    • Added presentation support query for win32 desktops.

  • Revision 3, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_win32_surface to VK_KHR_win32_surface.

  • Revision 4, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to vkCreateWin32SurfaceKHR.

  • Revision 5, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

  • Revision 6, 2017-04-24 (Jeff Juliano)

    • Add issue 2 addressing reuse of a native window object in a different Graphics API, or by a different Vulkan ICD.

VK_KHR_xcb_surface

Name String

VK_KHR_xcb_surface

Extension Type

Instance extension

Registered Extension Number

6

Revision

6

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec,Ian Elliott @ianelliottus

Last Modified Date

2015-11-28

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_xcb_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to an X11 Window, using the XCB client-side library, as well as a query to determine support for rendering via XCB.

New Object Types

None

New Enum Constants

New Enums

None

New Structures

Issues

1) Does XCB need a way to query for compatibility between a particular physical device and a specific screen? This would be a more general query than vkGetPhysicalDeviceSurfaceSupportKHR: If it returned VK_TRUE, then the physical device could be assumed to support presentation to any window on that screen.

RESOLVED: Yes, this is needed for toolkits that want to create a VkDevice before creating a window. To ensure the query is reliable, it must be made against a particular X visual rather than the screen in general.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft, based on the previous contents of VK_EXT_KHR_swapchain (later renamed VK_EXT_KHR_surface).

  • Revision 2, 2015-10-02 (James Jones)

    • Added presentation support query for an (xcb_connection_t*, xcb_visualid_t) pair.

    • Removed "root" parameter from CreateXcbSurfaceKHR(), as it is redundant when a window on the same screen is specified as well.

    • Adjusted wording of issue #1 and added agreed upon resolution.

  • Revision 3, 2015-10-14 (Ian Elliott)

    • Removed "root" parameter from CreateXcbSurfaceKHR() in one more place.

  • Revision 4, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_xcb_surface to VK_KHR_xcb_surface.

  • Revision 5, 2015-10-23 (Daniel Rakos)

    • Added allocation callbacks to vkCreateXcbSurfaceKHR.

  • Revision 6, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

VK_KHR_xlib_surface

Name String

VK_KHR_xlib_surface

Extension Type

Instance extension

Registered Extension Number

5

Revision

6

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec,Ian Elliott @ianelliottus

Last Modified Date

2015-11-28

IP Status

No known IP claims.

Contributors
  • Patrick Doane, Blizzard

  • Jason Ekstrand, Intel

  • Ian Elliott, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Jesse Hall, Google

  • James Jones, NVIDIA

  • Antoine Labour, Google

  • Jon Leech, Khronos

  • David Mao, AMD

  • Norbert Nopper, Freescale

  • Alon Or-bach, Samsung

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Ray Smith, ARM

  • Jeff Vigil, Qualcomm

  • Chia-I Wu, LunarG

The VK_KHR_xlib_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to an X11 Window, using the Xlib client-side library, as well as a query to determine support for rendering via Xlib.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_XLIB_SURFACE_CREATE_INFO_KHR

New Enums

None

Issues

1) Does X11 need a way to query for compatibility between a particular physical device and a specific screen? This would be a more general query than vkGetPhysicalDeviceSurfaceSupportKHR : if it returned VK_TRUE, then the physical device could be assumed to support presentation to any window on that screen.

RESOLVED: Yes, this is needed for toolkits that want to create a VkDevice before creating a window. To ensure the query is reliable, it must be made against a particular X visual rather than the screen in general.

Version History

  • Revision 1, 2015-09-23 (Jesse Hall)

    • Initial draft, based on the previous contents of VK_EXT_KHR_swapchain (later renamed VK_EXT_KHR_surface).

  • Revision 2, 2015-10-02 (James Jones)

    • Added presentation support query for (Display*, VisualID) pair.

    • Removed "root" parameter from CreateXlibSurfaceKHR(), as it is redundant when a window on the same screen is specified as well.

    • Added appropriate X errors.

    • Adjusted wording of issue #1 and added agreed upon resolution.

  • Revision 3, 2015-10-14 (Ian Elliott)

    • Renamed this extension from VK_EXT_KHR_x11_surface to VK_EXT_KHR_xlib_surface.

  • Revision 4, 2015-10-26 (Ian Elliott)

    • Renamed from VK_EXT_KHR_xlib_surface to VK_KHR_xlib_surface.

  • Revision 5, 2015-11-03 (Daniel Rakos)

    • Added allocation callbacks to vkCreateXlibSurfaceKHR.

  • Revision 6, 2015-11-28 (Daniel Rakos)

    • Updated the surface create function to take a pCreateInfo structure.

VK_EXT_acquire_xlib_display

Name String

VK_EXT_acquire_xlib_display

Extension Type

Instance extension

Registered Extension Number

90

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-12-13

IP Status

No known IP claims.

Contributors
  • Dave Airlie, Red Hat

  • Pierre Boudier, NVIDIA

  • James Jones, NVIDIA

  • Damien Leone, NVIDIA

  • Pierre-Loup Griffais, Valve

  • Liam Middlebrook, NVIDIA

  • Daniel Vetter, Intel

This extension allows an application to take exclusive control on a display currently associated with an X11 screen. When control is acquired, the display will be deassociated from the X11 screen until control is released or the specified display connection is closed. Essentially, the X11 screen will behave as if the monitor has been unplugged until control is released.

New Enum Constants

None.

New Enums

None.

New Structures

None.

Issues

1) Should vkAcquireXlibDisplayEXT take an RandR display ID, or a Vulkan display handle as input?

RESOLVED: A Vulkan display handle. Otherwise there would be no way to specify handles to displays that had been “blacklisted” or prevented from being included in the X11 display list by some native platform or vendor-specific mechanism.

2) How does an application figure out which RandR display corresponds to a Vulkan display?

RESOLVED: A new function, vkGetRandROutputDisplayEXT, is introduced for this purpose.

3) Should vkGetRandROutputDisplayEXT be part of this extension, or a general Vulkan + RandR or Vulkan + Xlib extension?

RESOLVED: To avoid yet another extension, include it in this extension.

Version History

  • Revision 1, 2016-12-13 (James Jones)

    • Initial draft

VK_EXT_blend_operation_advanced

Name String

VK_EXT_blend_operation_advanced

Extension Type

Device extension

Registered Extension Number

149

Revision

2

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-06-12

Contributors
  • Jeff Bolz, NVIDIA

This extension adds a number of “advanced” blending operations that can be used to perform new color blending operations, many of which are more complex than the standard blend modes provided by unextended Vulkan. This extension requires different styles of usage, depending on the level of hardware support and the enabled features:

  • If VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT::advancedBlendCoherentOperations is VK_FALSE, the new blending operations are supported, but a memory dependency must separate each advanced blend operation on a given sample. VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT is used to synchronize reads using advanced blend operations.

  • If VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT::advancedBlendCoherentOperations is VK_TRUE, advanced blend operations obey primitive order just like basic blend operations.

In unextended Vulkan, the set of blending operations is limited, and can be expressed very simply. The VK_BLEND_OP_MIN and VK_BLEND_OP_MAX blend operations simply compute component-wise minimums or maximums of source and destination color components. The VK_BLEND_OP_ADD, VK_BLEND_OP_SUBTRACT, and VK_BLEND_OP_REVERSE_SUBTRACT modes multiply the source and destination colors by source and destination factors and either add the two products together or subtract one from the other. This limited set of operations supports many common blending operations but precludes the use of more sophisticated transparency and blending operations commonly available in many dedicated imaging APIs.

This extension provides a number of new “advanced” blending operations. Unlike traditional blending operations using VK_BLEND_OP_ADD, these blending equations do not use source and destination factors specified by VkBlendFactor. Instead, each blend operation specifies a complete equation based on the source and destination colors. These new blend operations are used for both RGB and alpha components; they must not be used to perform separate RGB and alpha blending (via different values of color and alpha VkBlendOp).

These blending operations are performed using premultiplied colors, where RGB colors can be considered premultiplied or non-premultiplied by alpha, according to the srcPremultiplied and dstPremultiplied members of VkPipelineColorBlendAdvancedStateCreateInfoEXT. If a color is considered non-premultiplied, the (R,G,B) color components are multiplied by the alpha component prior to blending. For non-premultiplied color components in the range [0,1], the corresponding premultiplied color component would have values in the range [0 × A, 1 × A].

Many of these advanced blending equations are formulated where the result of blending source and destination colors with partial coverage have three separate contributions: from the portions covered by both the source and the destination, from the portion covered only by the source, and from the portion covered only by the destination. The blend parameter VkPipelineColorBlendAdvancedStateCreateInfoEXT::blendOverlap can be used to specify a correlation between source and destination pixel coverage. If set to VK_BLEND_OVERLAP_CONJOINT_EXT, the source and destination are considered to have maximal overlap, as would be the case if drawing two objects on top of each other. If set to VK_BLEND_OVERLAP_DISJOINT_EXT, the source and destination are considered to have minimal overlap, as would be the case when rendering a complex polygon tessellated into individual non-intersecting triangles. If set to VK_BLEND_OVERLAP_UNCORRELATED_EXT, the source and destination coverage are assumed to have no spatial correlation within the pixel.

In addition to the coherency issues on implementations not supporting advancedBlendCoherentOperations, this extension has several limitations worth noting. First, the new blend operations have a limit on the number of color attachments they can be used with, as indicated by VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT::advancedBlendMaxColorAttachments. Additionally, blending precision may be limited to 16-bit floating-point, which may result in a loss of precision and dynamic range for framebuffer formats with 32-bit floating-point components, and in a loss of precision for formats with 12- and 16-bit signed or unsigned normalized integer components.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_FEATURES_EXT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_BLEND_OPERATION_ADVANCED_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_ADVANCED_STATE_CREATE_INFO_EXT

  • Extending VkAccessFlagBits:

    • VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT

  • Extending VkBlendOp:

    • VK_BLEND_OP_ZERO_EXT

    • VK_BLEND_OP_SRC_EXT

    • VK_BLEND_OP_DST_EXT

    • VK_BLEND_OP_SRC_OVER_EXT

    • VK_BLEND_OP_DST_OVER_EXT

    • VK_BLEND_OP_SRC_IN_EXT

    • VK_BLEND_OP_DST_IN_EXT

    • VK_BLEND_OP_SRC_OUT_EXT

    • VK_BLEND_OP_DST_OUT_EXT

    • VK_BLEND_OP_SRC_ATOP_EXT

    • VK_BLEND_OP_DST_ATOP_EXT

    • VK_BLEND_OP_XOR_EXT

    • VK_BLEND_OP_MULTIPLY_EXT

    • VK_BLEND_OP_SCREEN_EXT

    • VK_BLEND_OP_OVERLAY_EXT

    • VK_BLEND_OP_DARKEN_EXT

    • VK_BLEND_OP_LIGHTEN_EXT

    • VK_BLEND_OP_COLORDODGE_EXT

    • VK_BLEND_OP_COLORBURN_EXT

    • VK_BLEND_OP_HARDLIGHT_EXT

    • VK_BLEND_OP_SOFTLIGHT_EXT

    • VK_BLEND_OP_DIFFERENCE_EXT

    • VK_BLEND_OP_EXCLUSION_EXT

    • VK_BLEND_OP_INVERT_EXT

    • VK_BLEND_OP_INVERT_RGB_EXT

    • VK_BLEND_OP_LINEARDODGE_EXT

    • VK_BLEND_OP_LINEARBURN_EXT

    • VK_BLEND_OP_VIVIDLIGHT_EXT

    • VK_BLEND_OP_LINEARLIGHT_EXT

    • VK_BLEND_OP_PINLIGHT_EXT

    • VK_BLEND_OP_HARDMIX_EXT

    • VK_BLEND_OP_HSL_HUE_EXT

    • VK_BLEND_OP_HSL_SATURATION_EXT

    • VK_BLEND_OP_HSL_COLOR_EXT

    • VK_BLEND_OP_HSL_LUMINOSITY_EXT

    • VK_BLEND_OP_PLUS_EXT

    • VK_BLEND_OP_PLUS_CLAMPED_EXT

    • VK_BLEND_OP_PLUS_CLAMPED_ALPHA_EXT

    • VK_BLEND_OP_PLUS_DARKER_EXT

    • VK_BLEND_OP_MINUS_EXT

    • VK_BLEND_OP_MINUS_CLAMPED_EXT

    • VK_BLEND_OP_CONTRAST_EXT

    • VK_BLEND_OP_INVERT_OVG_EXT

    • VK_BLEND_OP_RED_EXT

    • VK_BLEND_OP_GREEN_EXT

    • VK_BLEND_OP_BLUE_EXT

New Enums

New Functions

None.

Issues

None.

Version History

  • Revisions 1-2, 2017-06-12 (Jeff Bolz)

    • Internal revisions

VK_EXT_conservative_rasterization

Name String

VK_EXT_conservative_rasterization

Extension Type

Device extension

Registered Extension Number

102

Revision

1

Extension and Version Dependencies
Contact
  • Piers Daniell @pdaniell-nv

Last Modified Data

2017-08-28

Contributors
  • Daniel Koch, NVIDIA

  • Daniel Rakos, AMD

  • Jeff Bolz, NVIDIA

  • Slawomir Grajewski, Intel

  • Stu Smith, Imagination Technologies

This extension adds a new rasterization mode called conservative rasterization. There are two modes of conservative rasterization; overestimation and underestimation.

When overestimation is enabled, if any part of the primitive, including its edges, covers any part of the rectangular pixel area, including its sides, then a fragment is generated with all coverage samples turned on. This extension allows for some variation in implementations by accounting for differences in overestimation, where the generating primitive size is increased at each of its edges by some sub-pixel amount to further increase conservative pixel coverage. Implementations can allow the application to specify an extra overestimation beyond the base overestimation the implementation already does. It also allows implementations to either cull degenerate primitives or rasterize them.

When underestimation is enabled, fragments are only generated if the rectangular pixel area is fully covered by the generating primitive. If supported by the implementation, when a pixel rectangle is fully covered the fragment shader input variable builtin called FullyCoveredEXT is set to true. The shader variable works in either overestimation or underestimation mode.

Implementations can process degenerate triangles and lines by either discarding them or generating conservative fragments for them. Degenerate triangles are those that end up with zero area after the rasterizer quantizes them to the fixed-point pixel grid. Degenerate lines are those with zero length after quantization.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_CONSERVATIVE_RASTERIZATION_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_CONSERVATIVE_STATE_CREATE_INFO_EXT

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-08-28 (Piers Daniell)

    • Internal revisions

VK_EXT_debug_marker

Name String

VK_EXT_debug_marker

Extension Type

Device extension

Registered Extension Number

23

Revision

4

Extension and Version Dependencies
Contact
  • Baldur Karlsson @baldurk

Last Modified Date

2017-01-31

IP Status

No known IP claims.

Contributors
  • Baldur Karlsson

  • Dan Ginsburg, Valve

  • Jon Ashburn, LunarG

  • Kyle Spagnoli, NVIDIA

The VK_EXT_debug_marker extension is a device extension. It introduces concepts of object naming and tagging, for better tracking of Vulkan objects, as well as additional commands for recording annotations of named sections of a workload to aid organization and offline analysis in external tools.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_NAME_INFO_EXT

    • VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_TAG_INFO_EXT

    • VK_STRUCTURE_TYPE_DEBUG_MARKER_MARKER_INFO_EXT

New Enums

None

Examples

Example 1

Associate a name with an image, for easier debugging in external tools or with validation layers that can print a friendly name when referring to objects in error messages.

    extern VkDevice device;
    extern VkImage image;

    // Must call extension functions through a function pointer:
    PFN_vkDebugMarkerSetObjectNameEXT pfnDebugMarkerSetObjectNameEXT = (PFN_vkDebugMarkerSetObjectNameEXT)vkGetDeviceProcAddr(device, "vkDebugMarkerSetObjectNameEXT");

    // Set a name on the image
    const VkDebugMarkerObjectNameInfoEXT imageNameInfo =
    {
        VK_STRUCTURE_TYPE_DEBUG_MARKER_OBJECT_NAME_INFO_EXT, // sType
        NULL,                                           // pNext
        VK_DEBUG_REPORT_OBJECT_TYPE_IMAGE_EXT,          // objectType
        (uint64_t)image,                                // object
        "Brick Diffuse Texture",                        // pObjectName
    };

    pfnDebugMarkerSetObjectNameEXT(device, &imageNameInfo);

    // A subsequent error might print:
    //   Image 'Brick Diffuse Texture' (0xc0dec0dedeadbeef) is used in a
    //   command buffer with no memory bound to it.

Example 2

Annotating regions of a workload with naming information so that offline analysis tools can display a more usable visualisation of the commands submitted.

    extern VkDevice device;
    extern VkCommandBuffer commandBuffer;

    // Must call extension functions through a function pointer:
    PFN_vkCmdDebugMarkerBeginEXT pfnCmdDebugMarkerBeginEXT = (PFN_vkCmdDebugMarkerBeginEXT)vkGetDeviceProcAddr(device, "vkCmdDebugMarkerBeginEXT");
    PFN_vkCmdDebugMarkerEndEXT pfnCmdDebugMarkerEndEXT = (PFN_vkCmdDebugMarkerEndEXT)vkGetDeviceProcAddr(device, "vkCmdDebugMarkerEndEXT");
    PFN_vkCmdDebugMarkerInsertEXT pfnCmdDebugMarkerInsertEXT = (PFN_vkCmdDebugMarkerInsertEXT)vkGetDeviceProcAddr(device, "vkCmdDebugMarkerInsertEXT");

    // Describe the area being rendered
    const VkDebugMarkerMarkerInfoEXT houseMarker =
    {
        VK_STRUCTURE_TYPE_DEBUG_MARKER_MARKER_INFO_EXT, // sType
        NULL,                                           // pNext
        "Brick House",                                  // pMarkerName
        { 1.0f, 0.0f, 0.0f, 1.0f },                     // color
    };

    // Start an annotated group of calls under the 'Brick House' name
    pfnCmdDebugMarkerBeginEXT(commandBuffer, &houseMarker);
    {
        // A mutable structure for each part being rendered
        VkDebugMarkerMarkerInfoEXT housePartMarker =
        {
            VK_STRUCTURE_TYPE_DEBUG_MARKER_MARKER_INFO_EXT, // sType
            NULL,                                           // pNext
            NULL,                                           // pMarkerName
            { 0.0f, 0.0f, 0.0f, 0.0f },                     // color
        };

        // Set the name and insert the marker
        housePartMarker.pMarkerName = "Walls";
        pfnCmdDebugMarkerInsertEXT(commandBuffer, &housePartMarker);

        // Insert the drawcall for the walls
        vkCmdDrawIndexed(commandBuffer, 1000, 1, 0, 0, 0);

        // Insert a recursive region for two sets of windows
        housePartMarker.pMarkerName = "Windows";
        pfnCmdDebugMarkerBeginEXT(commandBuffer, &housePartMarker);
        {
            vkCmdDrawIndexed(commandBuffer, 75, 6, 1000, 0, 0);
            vkCmdDrawIndexed(commandBuffer, 100, 2, 1450, 0, 0);
        }
        pfnCmdDebugMarkerEndEXT(commandBuffer);

        housePartMarker.pMarkerName = "Front Door";
        pfnCmdDebugMarkerInsertEXT(commandBuffer, &housePartMarker);

        vkCmdDrawIndexed(commandBuffer, 350, 1, 1650, 0, 0);

        housePartMarker.pMarkerName = "Roof";
        pfnCmdDebugMarkerInsertEXT(commandBuffer, &housePartMarker);

        vkCmdDrawIndexed(commandBuffer, 500, 1, 2000, 0, 0);
    }
    // End the house annotation started above
    pfnCmdDebugMarkerEndEXT(commandBuffer);

Issues

1) Should the tag or name for an object be specified using the pNext parameter in the object’s Vk*CreateInfo structure?

RESOLVED: No. While this fits with other Vulkan patterns and would allow more type safety and future proofing against future objects, it has notable downsides. In particular passing the name at Vk*CreateInfo time does not allow renaming, prevents late binding of naming information, and does not allow naming of implicitly created objects such as queues and swapchain images.

2) Should the command annotation functions vkCmdDebugMarkerBeginEXT and vkCmdDebugMarkerEndEXT support the ability to specify a color?

RESOLVED: Yes. The functions have been expanded to take an optional color which can be used at will by implementations consuming the command buffer annotations in their visualisation.

3) Should the functions added in this extension accept an extensible structure as their parameter for a more flexible API, as opposed to direct function parameters? If so, which functions?

RESOLVED: Yes. All functions have been modified to take a structure type with extensible pNext pointer, to allow future extensions to add additional annotation information in the same commands.

Version History

  • Revision 1, 2016-02-24 (Baldur Karlsson)

    • Initial draft, based on LunarG marker spec

  • Revision 2, 2016-02-26 (Baldur Karlsson)

    • Renamed Dbg to DebugMarker in function names

    • Allow markers in secondary command buffers under certain circumstances

    • Minor language tweaks and edits

  • Revision 3, 2016-04-23 (Baldur Karlsson)

    • Reorganise spec layout to closer match desired organisation

    • Added optional color to markers (both regions and inserted labels)

    • Changed functions to take extensible structs instead of direct function parameters

  • Revision 4, 2017-01-31 (Baldur Karlsson)

    • Added explicit dependency on VK_EXT_debug_report

    • Moved definition of VkDebugReportObjectTypeEXT to debug report chapter.

    • Fixed typo in dates in revision history

VK_EXT_debug_report

Name String

VK_EXT_debug_report

Extension Type

Instance extension

Registered Extension Number

12

Revision

9

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Courtney Goeltzenleuchter @courtney-g

Last Modified Date

2017-09-12

IP Status

No known IP claims.

Contributors
  • Courtney Goeltzenleuchter, LunarG

  • Dan Ginsburg, Valve

  • Jon Ashburn, LunarG

  • Mark Lobodzinski, LunarG

Due to the nature of the Vulkan interface, there is very little error information available to the developer and application. By enabling optional validation layers and using the VK_EXT_debug_report extension, developers can obtain much more detailed feedback on the application’s use of Vulkan. This extension defines a way for layers and the implementation to call back to the application for events of interest to the application.

New Object Types

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT

  • Extending VkResult:

    • VK_ERROR_VALIDATION_FAILED_EXT

New Function Pointers

Examples

VK_EXT_debug_report allows an application to register multiple callbacks with the validation layers. Some callbacks may log the information to a file, others may cause a debug break point or other application defined behavior. An application can register callbacks even when no validation layers are enabled, but they will only be called for loader and, if implemented, driver events.

To capture events that occur while creating or destroying an instance an application can link a VkDebugReportCallbackCreateInfoEXT structure to the pNext element of the VkInstanceCreateInfo structure given to vkCreateInstance. This callback is only valid for the duration of the vkCreateInstance and the vkDestroyInstance call. Use vkCreateDebugReportCallbackEXT to create persistent callback objects.

Example uses: Create three callback objects. One will log errors and warnings to the debug console using Windows OutputDebugString. The second will cause the debugger to break at that callback when an error happens and the third will log warnings to stdout.

    VkResult res;
    VkDebugReportCallbackEXT cb1, cb2, cb3;

    VkDebugReportCallbackCreateInfoEXT callback1 = {
            VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT,    // sType
            NULL,                                                       // pNext
            VK_DEBUG_REPORT_ERROR_BIT_EXT |                             // flags
            VK_DEBUG_REPORT_WARNING_BIT_EXT,
            myOutputDebugString,                                        // pfnCallback
            NULL                                                        // pUserData
    };
    res = vkCreateDebugReportCallbackEXT(instance, &callback1, &cb1);
    if (res != VK_SUCCESS)
       /* Do error handling for VK_ERROR_OUT_OF_MEMORY */

    callback.flags = VK_DEBUG_REPORT_ERROR_BIT_EXT;
    callback.pfnCallback = myDebugBreak;
    callback.pUserData = NULL;
    res = vkCreateDebugReportCallbackEXT(instance, &callback, &cb2);
    if (res != VK_SUCCESS)
       /* Do error handling for VK_ERROR_OUT_OF_MEMORY */

    VkDebugReportCallbackCreateInfoEXT callback3 = {
            VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT,    // sType
            NULL,                                                       // pNext
            VK_DEBUG_REPORT_WARNING_BIT_EXT,                            // flags
            mystdOutLogger,                                             // pfnCallback
            NULL                                                        // pUserData
    };
    res = vkCreateDebugReportCallbackEXT(instance, &callback3, &cb3);
    if (res != VK_SUCCESS)
       /* Do error handling for VK_ERROR_OUT_OF_MEMORY */

    ...

    /* remove callbacks when cleaning up */
    vkDestroyDebugReportCallbackEXT(instance, cb1);
    vkDestroyDebugReportCallbackEXT(instance, cb2);
    vkDestroyDebugReportCallbackEXT(instance, cb3);
Note

In the initial release of the VK_EXT_debug_report extension, the token VK_STRUCTURE_TYPE_DEBUG_REPORT_CREATE_INFO_EXT was used. Starting in version 2 of the extension branch, VK_STRUCTURE_TYPE_DEBUG_REPORT_CALLBACK_CREATE_INFO_EXT is used instead for consistency with Vulkan naming rules. The older enum is still available for backwards compatibility.

Note

In the initial release of the VK_EXT_debug_report extension, the token VK_DEBUG_REPORT_OBJECT_TYPE_DEBUG_REPORT_EXT was used. Starting in version 8 of the extension branch, VK_DEBUG_REPORT_OBJECT_TYPE_DEBUG_REPORT_CALLBACK_EXT_EXT is used instead for consistency with Vulkan naming rules. The older enum is still available for backwards compatibility.

Issues

1) What is the hierarchy / seriousness of the message flags? E.g. ERROR > WARN > PERF_WARN …​

RESOLVED: There is no specific hierarchy. Each bit is independent and should be checked via bitwise AND. For example:

    if (localFlags & VK_DEBUG_REPORT_ERROR_BIT_EXT) {
        process error message
    }
    if (localFlags & VK_DEBUG_REPORT_DEBUG_BIT_EXT) {
        process debug message
    }

The validation layers do use them in a hierarchical way (ERROR > WARN > PERF, WARN > DEBUG > INFO) and they (at least at the time of this writing) only set one bit at a time. But it is not a requirement of this extension.

It is possible that a layer may intercept and change, or augment the flags with extension values the application’s debug report handler may not be familiar with, so it is important to treat each flag independently.

2) Should there be a VU requiring VkDebugReportCallbackCreateInfoEXT::flags to be non-zero?

RESOLVED: It may not be very useful, but we do not need VU statement requiring the VkDebugReportCallbackCreateInfoEXT::msgFlags at create-time to be non-zero. One can imagine that apps may prefer it as it allows them to set the mask as desired - including nothing - at runtime without having to check.

3) What is the difference between VK_DEBUG_REPORT_DEBUG_BIT_EXT and VK_DEBUG_REPORT_INFORMATION_BIT_EXT?

RESOLVED: VK_DEBUG_REPORT_DEBUG_BIT_EXT specifies information that could be useful debugging the Vulkan implementation itself.

Version History

  • Revision 1, 2015-05-20 (Courtney Goetzenleuchter)

    • Initial draft, based on LunarG KHR spec, other KHR specs

  • Revision 2, 2016-02-16 (Courtney Goetzenleuchter)

    • Update usage, documentation

  • Revision 3, 2016-06-14 (Courtney Goetzenleuchter)

    • Update VK_EXT_DEBUG_REPORT_SPEC_VERSION to indicate added support for vkCreateInstance and vkDestroyInstance

  • Revision 4, 2016-12-08 (Mark Lobodzinski)

    • Added Display_KHR, DisplayModeKHR extension objects

    • Added ObjectTable_NVX, IndirectCommandsLayout_NVX extension objects

    • Bumped spec revision

    • Retroactively added version history

  • Revision 5, 2017-01-31 (Baldur Karlsson)

  • Revision 6, 2017-01-31 (Baldur Karlsson)

    • Added VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_KHR_EXT

  • Revision 7, 2017-04-20 (Courtney Goeltzenleuchter)

    • Clarify wording and address questions from developers.

  • Revision 8, 2017-04-21 (Courtney Goeltzenleuchter)

    • Remove unused enum VkDebugReportErrorEXT

  • Revision 9, 2017-09-12 (Tobias Hector)

    • Added interactions with Vulkan 1.1

VK_EXT_debug_utils

Name String

VK_EXT_debug_utils

Extension Type

Instance extension

Registered Extension Number

129

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Mark Young @marky-lunarg

Last Modified Date

2017-09-14

Revision

1

IP Status

No known IP claims.

Dependencies
  • This extension is written against version 1.0 of the Vulkan API.

  • Requires VkObjectType

Contributors
  • Mark Young, LunarG

  • Baldur Karlsson

  • Ian Elliott, Google

  • Courtney Goeltzenleuchter, Google

  • Karl Schultz, LunarG

  • Mark Lobodzinski, LunarG

  • Mike Schuchardt, LunarG

  • Jaakko Konttinen, AMD

  • Dan Ginsburg, Valve Software

  • Rolando Olivares, Epic Games

  • Dan Baker, Oxide Games

  • Kyle Spagnoli, NVIDIA

  • Jon Ashburn, LunarG

Due to the nature of the Vulkan interface, there is very little error information available to the developer and application. By using the VK_EXT_debug_utils extension, developers can obtain more information. When combined with validation layers, even more detailed feedback on the application’s use of Vulkan will be provided.

This extension provides the following capabilities:

  • The ability to create a debug messenger which will pass along debug messages to an application supplied callback.

  • The ability to identify specific Vulkan objects using a name or tag to improve tracking.

  • The ability to identify specific sections within a VkQueue or VkCommandBuffer using labels to aid organization and offline analysis in external tools.

The main difference between this extension and VK_EXT_debug_report and VK_EXT_debug_marker is that those extensions use VkDebugReportObjectTypeEXT to identify objects. This extension uses the core VkObjectType in place of VkDebugReportObjectTypeEXT. The primary reason for this move is that no future object type handle enumeration values will be added to VkDebugReportObjectTypeEXT since the creation of VkObjectType.

In addition, this extension combines the functionality of both VK_EXT_debug_report and VK_EXT_debug_marker by allowing object name and debug markers (now called labels) to be returned to the application’s callback function. This should assist in clarifying the details of a debug message including: what objects are involved and potentially which location within a VkQueue or VkCommandBuffer the message occurred.

New Object Types

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_NAME_INFO_EXT

    • VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_TAG_INFO_EXT

    • VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT

    • VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CALLBACK_DATA_EXT

    • VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT

  • Extending VkResult:

    • VK_ERROR_VALIDATION_FAILED_EXT

New Function Pointers

Examples

Example 1

VK_EXT_debug_utils allows an application to register multiple callbacks with any Vulkan component wishing to report debug information. Some callbacks may log the information to a file, others may cause a debug break point or other application defined behavior. An application can register callbacks even when no validation layers are enabled, but they will only be called for loader and, if implemented, driver events.

To capture events that occur while creating or destroying an instance an application can link a VkDebugUtilsMessengerCreateInfoEXT structure to the pNext element of the VkInstanceCreateInfo structure given to vkCreateInstance. This callback is only valid for the duration of the vkCreateInstance and the vkDestroyInstance call. Use vkCreateDebugUtilsMessengerEXT to create persistent callback objects.

Example uses: Create three callback objects. One will log errors and warnings to the debug console using Windows OutputDebugString. The second will cause the debugger to break at that callback when an error happens and the third will log warnings to stdout.

    extern VkInstance instance;
    VkResult res;
    VkDebugUtilsMessengerEXT cb1, cb2, cb3;

    // Must call extension functions through a function pointer:
    PFN_vkCreateDebugUtilsMessengerEXT pfnCreateDebugUtilsMessengerEXT = (PFN_vkCreateDebugUtilsMessengerEXT)vkGetDeviceProcAddr(device, "vkCreateDebugUtilsMessengerEXT");
    PFN_vkDestroyDebugUtilsMessengerEXT pfnDestroyDebugUtilsMessengerEXT = (PFN_vkDestroyDebugUtilsMessengerEXT)vkGetDeviceProcAddr(device, "vkDestroyDebugUtilsMessengerEXT");

    VkDebugUtilsMessengeCreateInfoEXT callback1 = {
            VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT,  // sType
            NULL,                                                     // pNext
            0,                                                        // flags
            VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT |           // messageSeverity
            VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT,
            VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT |             // messageType
            VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT,
            myOutputDebugString,                                      // pfnUserCallback
            NULL                                                      // pUserData
    };
    res = pfnCreateDebugUtilsMessengerEXT(instance, &callback1, &cb1);
    if (res != VK_SUCCESS) {
       // Do error handling for VK_ERROR_OUT_OF_MEMORY
    }

    callback1.messageSeverity = VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT;
    callback1.pfnCallback = myDebugBreak;
    callback1.pUserData = NULL;
    res = pfnCreateDebugUtilsMessengerEXT(instance, &callback1, &cb2);
    if (res != VK_SUCCESS) {
       // Do error handling for VK_ERROR_OUT_OF_MEMORY
    }

    VkDebugUtilsMessengerCreateInfoEXT callback3 = {
            VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT,  // sType
            NULL,                                                     // pNext
            0,                                                        // flags
            VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT,          // messageSeverity
            VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT |             // messageType
            VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT,
            mystdOutLogger,                                           // pfnUserCallback
            NULL                                                      // pUserData
    };
    res = pfnCreateDebugUtilsMessengerEXT(instance, &callback3, &cb3);
    if (res != VK_SUCCESS) {
       // Do error handling for VK_ERROR_OUT_OF_MEMORY
    }

    ...

    // Remove callbacks when cleaning up
    pfnDestroyDebugUtilsMessengerEXT(instance, cb1);
    pfnDestroyDebugUtilsMessengerEXT(instance, cb2);
    pfnDestroyDebugUtilsMessengerEXT(instance, cb3);

Example 2

Associate a name with an image, for easier debugging in external tools or with validation layers that can print a friendly name when referring to objects in error messages.

    extern VkDevice device;
    extern VkImage image;

    // Must call extension functions through a function pointer:
    PFN_vkSetDebugUtilsObjectNameEXT pfnSetDebugUtilsObjectNameEXT = (PFN_vkSetDebugUtilsObjectNameEXT)vkGetDeviceProcAddr(device, "vkSetDebugUtilsObjectNameEXT");

    // Set a name on the image
    const VkDebugUtilsObjectNameInfoEXT imageNameInfo =
    {
        VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_NAME_INFO_EXT, // sType
        NULL,                                               // pNext
        VK_OBJECT_TYPE_IMAGE,                               // objectType
        (uint64_t)image,                                    // object
        "Brick Diffuse Texture",                            // pObjectName
    };

    pfnSetDebugUtilsObjectNameEXT(device, &imageNameInfo);

    // A subsequent error might print:
    //   Image 'Brick Diffuse Texture' (0xc0dec0dedeadbeef) is used in a
    //   command buffer with no memory bound to it.

Example 3

Annotating regions of a workload with naming information so that offline analysis tools can display a more usable visualization of the commands submitted.

    extern VkDevice device;
    extern VkCommandBuffer commandBuffer;

    // Must call extension functions through a function pointer:
    PFN_vkQueueBeginDebugUtilsLabelEXT pfnQueueBeginDebugUtilsLabelEXT = (PFN_vkQueueBeginDebugUtilsLabelEXT)vkGetDeviceProcAddr(device, "vkQueueBeginDebugUtilsLabelEXT");
    PFN_vkQueueEndDebugUtilsLabelEXT pfnQueueEndDebugUtilsLabelEXT = (PFN_vkQueueEndDebugUtilsLabelEXT)vkGetDeviceProcAddr(device, "vkQueueEndDebugUtilsLabelEXT");
    PFN_vkCmdBeginDebugUtilsLabelEXT pfnCmdBeginDebugUtilsLabelEXT = (PFN_vkCmdBeginDebugUtilsLabelEXT)vkGetDeviceProcAddr(device, "vkCmdBeginDebugUtilsLabelEXT");
    PFN_vkCmdEndDebugUtilsLabelEXT pfnCmdEndDebugUtilsLabelEXT = (PFN_vkCmdEndDebugUtilsLabelEXT)vkGetDeviceProcAddr(device, "vkCmdEndDebugUtilsLabelEXT");
    PFN_vkCmdInsertDebugUtilsLabelEXT pfnCmdInsertDebugUtilsLabelEXT = (PFN_vkCmdInsertDebugUtilsLabelEXT)vkGetDeviceProcAddr(device, "vkCmdInsertDebugUtilsLabelEXT");

    // Describe the area being rendered
    const VkDebugUtilsLabelEXT houseLabel =
    {
        VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT, // sType
        NULL,                                    // pNext
        "Brick House",                           // pLabelName
        { 1.0f, 0.0f, 0.0f, 1.0f },              // color
    };

    // Start an annotated group of calls under the 'Brick House' name
    pfnCmdBeginDebugUtilsLabelEXT(commandBuffer, &houseLabel);
    {
        // A mutable structure for each part being rendered
        VkDebugUtilsLabelEXT housePartLabel =
        {
            VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT, // sType
            NULL,                                    // pNext
            NULL,                                    // pLabelName
            { 0.0f, 0.0f, 0.0f, 0.0f },              // color
        };

        // Set the name and insert the marker
        housePartLabel.pLabelName = "Walls";
        pfnCmdInsertDebugUtilsLabelEXT(commandBuffer, &housePartLabel);

        // Insert the drawcall for the walls
        vkCmdDrawIndexed(commandBuffer, 1000, 1, 0, 0, 0);

        // Insert a recursive region for two sets of windows
        housePartLabel.pLabelName = "Windows";
        pfnCmdBeginDebugUtilsLabelEXT(commandBuffer, &housePartLabel);
        {
            vkCmdDrawIndexed(commandBuffer, 75, 6, 1000, 0, 0);
            vkCmdDrawIndexed(commandBuffer, 100, 2, 1450, 0, 0);
        }
        pfnCmdEndDebugUtilsLabelEXT(commandBuffer);

        housePartLabel.pLabelName = "Front Door";
        pfnCmdInsertDebugUtilsLabelEXT(commandBuffer, &housePartLabel);

        vkCmdDrawIndexed(commandBuffer, 350, 1, 1650, 0, 0);

        housePartLabel.pLabelName = "Roof";
        pfnCmdInsertDebugUtilsLabelEXT(commandBuffer, &housePartLabel);

        vkCmdDrawIndexed(commandBuffer, 500, 1, 2000, 0, 0);
    }
    // End the house annotation started above
    pfnCmdEndDebugUtilsLabelEXT(commandBuffer);

    // Do other work

    vkEndCommandBuffer(commandBuffer);

    // Describe the queue being used
    const VkDebugUtilsLabelEXT queueLabel =
    {
        VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT, // sType
        NULL,                                    // pNext
        "Main Render Work",                      // pLabelName
        { 0.0f, 1.0f, 0.0f, 1.0f },              // color
    };

    // Identify the queue label region
    pfnQueueBeginDebugUtilsLabelEXT(queue, &queueLabel);

    // Submit the work for the main render thread
    const VkCommandBuffer cmd_bufs[] = {commandBuffer};
    VkSubmitInfo submit_info = {.sType = VK_STRUCTURE_TYPE_SUBMIT_INFO,
                                .pNext = NULL,
                                .waitSemaphoreCount = 0,
                                .pWaitSemaphores = NULL,
                                .pWaitDstStageMask = NULL,
                                .commandBufferCount = 1,
                                .pCommandBuffers = cmd_bufs,
                                .signalSemaphoreCount = 0,
                                .pSignalSemaphores = NULL};
    vkQueueSubmit(queue, 1, &submit_info, fence);

    // End the queue label region
    pfnQueueEndDebugUtilsLabelEXT(queue);

Issues

1) Should we just name this extension VK_EXT_debug_report2

RESOLVED: No. There is enough additional changes to the structures to break backwards compatibility. So, a new name was decided that would not indicate any interaction with the previous extension.

2) Will validation layers immediately support all the new features.

RESOLVED: Not immediately. As one can imagine, there is a lot of work involved with converting the validation layer logging over to the new functionality. Basic logging, as seen in the origin VK_EXT_debug_report extension will be made available immediately. However, adding the labels and object names will take time. Since the priority for Khronos at this time is to continue focusing on Valid Usage statements, it may take a while before the new functionality is fully exposed.

3) If the validation layers won’t expose the new functionality immediately, then what’s the point of this extension?

RESOLVED: We needed a replacement for VK_EXT_debug_report because the VkDebugReportObjectTypeEXT enumeration will no longer be updated and any new objects will need to be debugged using the new functionality provided by this extension.

4) Should this extension be split into two separate parts (1 extension that is an instance extension providing the callback functionality, and another device extension providing the general debug marker and annotation functionality)?

RESOLVED: No, the functionality for this extension is too closely related. If we did split up the extension, where would the structures and enums live, and how would you define that the device behavior in the instance extension is really only valid if the device extension is enabled, and the functionality is passed in. It’s cleaner to just define this all as an instance extension, plus it allows the application to enable all debug functionality provided with one enable string during vkCreateInstance.

Version History

  • Revision 1, 2017-09-14 (Mark Young and all listed Contributors)

VK_EXT_depth_range_unrestricted

Name String

VK_EXT_depth_range_unrestricted

Extension Type

Device extension

Registered Extension Number

14

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2017-06-22

Contributors
  • Daniel Koch, NVIDIA

  • Jeff Bolz, NVIDIA

This extension removes the VkViewport minDepth and maxDepth restrictions that the values must be between 0.0 and 1.0, inclusive. It also removes the same restriction on VkPipelineDepthStencilStateCreateInfo minDepthBounds and maxDepthBounds. Finally it removes the restriction on the depth value in VkClearDepthStencilValue.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

Issues

1) How do VkViewport minDepth and maxDepth values outside of the 0.0 to 1.0 range interact with Primitive Clipping?

RESOLVED: The behavior described in Primitive Clipping still applies. If depth clamping is disabled the depth values are still clipped to 0 ≤ zc ≤ wc before the viewport transform. If depth clamping is enabled the above equation is ignored and the depth values are instead clamped to the VkViewport minDepth and maxDepth values, which in the case of this extension can be outside of the 0.0 to 1.0 range.

2) What happens if a resulting depth fragment is outside of the 0.0 to 1.0 range and the depth buffer is fixed-point rather than floating-point?

RESOLVED: The supported range of a fixed-point depth buffer is 0.0 to 1.0 and depth fragments are clamped to this range.

Version History

  • Revision 1, 2017-06-22 (Piers Daniell)

    • Internal revisions

VK_EXT_descriptor_indexing

Name String

VK_EXT_descriptor_indexing

Extension Type

Device extension

Registered Extension Number

162

Revision

2

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Status

Complete

Last Modified Data

2017-10-02

Contributors
  • Jeff Bolz, NVIDIA

  • Daniel Rakos, AMD

  • Slawomir Grajewski, Intel

  • Tobias Hector, Imagination Technologies

This extension adds several small features which together enable applications to create large descriptor sets containing substantially all of their resources, and selecting amongst those resources with dynamic (non-uniform) indexes in the shader. There are feature enables and SPIR-V capabilities for non-uniform descriptor indexing in the shader, and non-uniform indexing in the shader requires use of a new NonUniformEXT decoration defined in the SPV_EXT_descriptor_indexing SPIR-V extension. There are descriptor set layout binding creation flags enabling several features:

  • Descriptors can be updated after they are bound to a command buffer, such that the execution of the command buffer reflects the most recent update to the descriptors.

  • Descriptors that are not used by any pending command buffers can be updated, which enables writing new descriptors for frame N+1 while frame N is executing.

  • Relax the requirement that all descriptors in a binding that is "statically used" must be valid, such that descriptors that are not accessed by a submission need not be valid and can be updated while that submission is executing.

  • The final binding in a descriptor set layout can have a variable size (and unsized arrays of resources are allowed in the GL_EXT_nonuniform_qualifier and SPV_EXT_descriptor_indexing extensions).

Note that it is valid for multiple descriptor arrays in a shader to use the same set and binding number, as long as they are all compatible with the descriptor type in the pipeline layout. This means a single array binding in the descriptor set can serve multiple texture dimensionalities, or an array of buffer descriptors can be used with multiple different block layouts.

There are new descriptor set layout and descriptor pool creation flags that are required to opt in to the update-after-bind functionality, and there are separate maxPerStage* and maxDescriptorSet* limits that apply to these descriptor set layouts which may be much higher than the pre-existing limits. The old limits only count descriptors in non-updateAfterBind descriptor set layouts, and the new limits count descriptors in all descriptor set layouts in the pipeline layout.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_BINDING_FLAGS_CREATE_INFO_EXT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_FEATURES_EXT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DESCRIPTOR_INDEXING_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_ALLOCATE_INFO_EXT

    • VK_STRUCTURE_TYPE_DESCRIPTOR_SET_VARIABLE_DESCRIPTOR_COUNT_LAYOUT_SUPPORT_EXT

  • Extending VkDescriptorPoolCreateFlagBits:

    • VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT

  • Extending VkDescriptorSetLayoutCreateFlagBits:

    • VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT

  • Extending VkResult:

    • VK_ERROR_FRAGMENTATION_EXT

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-07-26 (Jeff Bolz)

    • Internal revisions

VK_EXT_direct_mode_display

Name String

VK_EXT_direct_mode_display

Extension Type

Instance extension

Registered Extension Number

89

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-12-13

IP Status

No known IP claims.

Contributors
  • Pierre Boudier, NVIDIA

  • James Jones, NVIDIA

  • Damien Leone, NVIDIA

  • Pierre-Loup Griffais, Valve

  • Liam Middlebrook, NVIDIA

This is extension, along with related platform exentions, allows applications to take exclusive control of displays associated with a native windowing system. This is especially useful for virtual reality applications that wish to hide HMDs (head mounted displays) from the native platform’s display management system, desktop, and/or other applications.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

Issues

1) Should this extension and its related platform-specific extensions leverage VK_KHR_display, or provide separate equivalent interfaces.

RESOLVED: Use VK_KHR_display concepts and objects. VK_KHR_display can be used to enumerate all displays on the system, including those attached to/in use by a window system or native platform, but VK_KHR_display_swapchain will fail to create a swapchain on in-use displays. This extension and its platform-specific children will allow applications to grab in-use displays away from window systems and/or native platforms, allowing them to be used with VK_KHR_display_swapchain.

2) Are separate calls needed to acquire displays and enable direct mode?

RESOLVED: No, these operations happen in one combined command. Acquiring a display puts it into direct mode.

Version History

  • Revision 1, 2016-12-13 (James Jones)

    • Initial draft

VK_EXT_discard_rectangles

Name String

VK_EXT_discard_rectangles

Extension Type

Device extension

Registered Extension Number

100

Revision

1

Extension and Version Dependencies
Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2016-12-22

Interactions and External Dependencies
  • Interacts with VK_KHR_device_group

  • Interacts with Vulkan 1.1

Contributors
  • Daniel Koch, NVIDIA

  • Jeff Bolz, NVIDIA

This extension provides additional orthogonally aligned “discard rectangles” specified in framebuffer-space coordinates that restrict rasterization of all points, lines and triangles.

From zero to an implementation-dependent limit (specified by maxDiscardRectangles) number of discard rectangles can be operational at once. When one or more discard rectangles are active, rasterized fragments can either survive if the fragment is within any of the operational discard rectangles (VK_DISCARD_RECTANGLE_MODE_INCLUSIVE_EXT mode) or be rejected if the fragment is within any of the operational discard rectangles (VK_DISCARD_RECTANGLE_MODE_EXCLUSIVE_EXT mode).

These discard rectangles operate orthogonally to the existing scissor test functionality. The discard rectangles can be different for each physical device in a device group by specifying the device mask and setting discard rectangle dynamic state.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DISCARD_RECTANGLE_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_PIPELINE_DISCARD_RECTANGLE_STATE_CREATE_INFO_EXT

  • Extending VkDynamicState

    • VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT

Issues

None.

Version History

  • Revision 1, 2016-12-22 (Piers Daniell)

    • Internal revisions

VK_EXT_display_control

Name String

VK_EXT_display_control

Extension Type

Device extension

Registered Extension Number

92

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-12-13

IP Status

No known IP claims.

Contributors
  • Pierre Boudier, NVIDIA

  • James Jones, NVIDIA

  • Damien Leone, NVIDIA

  • Pierre-Loup Griffais, Valve

  • Daniel Vetter, Intel

This extension defines a set of utility functions for use with the VK_KHR_display and VK_KHR_display_swapchain extensions.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DISPLAY_POWER_INFO_EXT

    • VK_STRUCTURE_TYPE_DEVICE_EVENT_INFO_EXT

    • VK_STRUCTURE_TYPE_DISPLAY_EVENT_INFO_EXT

    • VK_STRUCTURE_TYPE_SWAPCHAIN_COUNTER_CREATE_INFO_EXT

Issues

1) Should this extension add an explicit “WaitForVsync” API or a fence signaled at vsync that the application can wait on?

RESOLVED: A fence. A separate API could later be provided that allows exporting the fence to a native object that could be inserted into standard run loops on POSIX and Windows systems.

2) Should callbacks be added for a vsync event, or in general to monitor events in Vulkan?

RESOLVED: No, fences should be used. Some events are generated by interrupts which are managed in the kernel. In order to use a callback provided by the application, drivers would need to have the userspace driver spawn threads that would wait on the kernel event, and hence the callbacks could be difficult for the application to synchronize with its other work given they would arrive on a foreign thread.

3) Should vblank or scanline events be exposed?

RESOLVED: Vblank events. Scanline events could be added by a separate extension, but the latency of processing an interrupt and waking up a userspace event is high enough that the accuracy of a scanline event would be rather low. Further, per-scanline interrupts are not supported by all hardware.

Version History

  • Revision 1, 2016-12-13 (James Jones)

    • Initial draft

VK_EXT_display_surface_counter

Name String

VK_EXT_display_surface_counter

Extension Type

Instance extension

Registered Extension Number

91

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-12-13

IP Status

No known IP claims.

Contributors
  • Pierre Boudier, NVIDIA

  • James Jones, NVIDIA

  • Damien Leone, NVIDIA

  • Pierre-Loup Griffais, Valve

  • Daniel Vetter, Intel

This is extension defines a vertical blanking period counter associated with display surfaces. It provides a mechanism to query support for such a counter from a VkSurfaceKHR object.

New Enum Constants

New Structures

Issues

None.

Version History

  • Revision 1, 2016-12-13 (James Jones)

    • Initial draft

VK_EXT_external_memory_dma_buf

Name String

VK_EXT_external_memory_dma_buf

Extension Type

Device extension

Registered Extension Number

126

Revision

1

Extension and Version Dependencies
Contact
  • Chad Versace @chadversary

Last Modified Date

2017-10-10

IP Status

No known IP claims.

Contributors
  • Chad Versace, Google

  • James Jones, NVIDIA

  • Jason Ekstrand, Intel

A dma_buf is a type of file descriptor, defined by the Linux kernel, that allows sharing memory across kernel device drivers and across processes. This extension enables applications to import a dma_buf as VkDeviceMemory; to export VkDeviceMemory as a dma_buf; and to create VkBuffer objects that can be bound to that memory.

New Enum Constants

Issues

1. How does the application, when creating a VkImage that it intends to bind to dma_buf VkDeviceMemory that contains an externally produced image, specify the memory layout (such as row pitch and DRM format modifier) of the VkImage? In other words, how does the application achieve behavior comparable to that provided by EGL_EXT_image_dma_buf_import and EGL_EXT_image_dma_buf_import_modifiers?

+

RESOLVED. Features comparable to those in EGL_EXT_image_dma_buf_import and EGL_EXT_image_dma_buf_import_modifiers will be provided by an extension layered atop this one.

2. Without the ability to specify the memory layout of external dma_buf images, how is this extension useful?

+

RESOLVED. This extension provides exactly one new feature: the ability to import/export between dma_bufs and VkDeviceMemory. This feature, together with features provided by VK_KHR_external_memory_fd, is sufficient to bind a VkBuffer to dma_buf.

Version History

  • Revision 1, 2017-10-10 (Chad Versace)

    • Squashed internal revisions

VK_EXT_external_memory_host

Name String

VK_EXT_external_memory_host

Extension Type

Device extension

Registered Extension Number

179

Revision

1

Extension and Version Dependencies
Contact
  • Daniel Rakos @drakos-amd

Last Modified Date

2017-11-10

IP Status

No known IP claims.

Contributors
  • Jaakko Konttinen, AMD

  • David Mao, AMD

  • Daniel Rakos, AMD

  • Tobias Hector, Imagination Technologies

  • Jason Ekstrand, Intel

  • James Jones, NVIDIA

This extension enables an application to import host allocations and host mapped foreign device memory to Vulkan memory objects.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_IMPORT_MEMORY_HOST_POINTER_INFO_EXT

    • VK_STRUCTURE_TYPE_MEMORY_HOST_POINTER_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_EXTERNAL_MEMORY_HOST_PROPERTIES_EXT

  • Extending VkExternalMemoryHandleTypeFlagBitsKHR:

    • VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT

    • VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT

New Enums

None.

Issues

1) What memory type has to be used to import host pointers?

RESOLVED: Depends on the implementation. Applications have to use the new vkGetMemoryHostPointerPropertiesEXT command to query the supported memory types for a particular host pointer. The reported memory types may include memory types that come from a memory heap that is otherwise not usable for regular memory object allocation and thus such a heap’s size may be zero.

2) Can the application still access the contents of the host allocation after importing?

RESOLVED: Yes. However, usual synchronization requirements apply.

3) Can the application free the host allocation?

RESOLVED: No, it violates valid usage conditions. Using the memory object imported from a host allocation that’s already freed thus results in undefined behavior.

4) Is vkMapMemory expected to return the same host address which was specified when importing it to the memory object?

RESOLVED: No. Implementations are allowed to return the same address but it’s not required. Some implementations might return a different virtual mapping of the allocation, although the same physical pages will be used.

5) Is there any limitation on the alignment of the host pointer and/or size?

RESOLVED: Yes. Both the address and the size have to be an integer multiple of minImportedHostPointerAlignment. In addition, some platforms and foreign devices may have additional restrictions.

6) Can the same host allocation be imported multiple times into a given physical device?

RESOLVED: No, at least not guaranteed by this extension. Some platforms do not allow locking the same physical pages for device access multiple times, so attempting to do it may result in undefined behavior.

7) Does this extension support exporting the new handle type?

RESOLVED: No.

8) Should we include the possibility to import host mapped foreign device memory using this API?

RESOLVED: Yes, through a separate handle type. Implementations are still allowed to support only one of the handle types introduced by this extension by not returning import support for a particular handle type as returned in VkExternalMemoryPropertiesKHR.

Version History

  • Revision 1, 2017-11-10 (Daniel Rakos)

    • Internal revisions

VK_EXT_global_priority

Name String

VK_EXT_global_priority

Extension Type

Device extension

Registered Extension Number

175

Revision

2

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Andres Rodriguez @lostgoat

Last Modified Date

2017-10-06

IP Status

No known IP claims.

Contributors
  • Andres Rodriguez, Valve

  • Pierre-Loup Griffais, Valve

  • Dan Ginsburg, Valve

  • Mitch Singer, AMD

In Vulkan, users can specify device-scope queue priorities. In some cases it may be useful to extend this concept to a system-wide scope. This extension provides a mechanism for caller’s to set their system-wide priority. The default queue priority is VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT.

The driver implementation will attempt to skew hardware resource allocation in favour of the higher-priority task. Therefore, higher-priority work may retain similar latency and throughput characteristics even if the system is congested with lower priority work.

The global priority level of a queue shall take predence over the per-process queue priority (VkDeviceQueueCreateInfo::pQueuePriorities).

Abuse of this feature may result in starving the rest of the system from hardware resources. Therefore, the driver implementation may deny requests to acquire a priority above the default priority (VK_QUEUE_GLOBAL_PRIORITY_MEDIUM_EXT) if the caller does not have sufficient privileges. In this scenario VK_ERROR_NOT_PERMITTED_EXT is returned.

The driver implementation may fail the queue allocation request if resources required to complete the operation have been exhausted (either by the same process or a different process). In this scenario VK_ERROR_INITIALIZATION_FAILED is returned.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DEVICE_QUEUE_GLOBAL_PRIORITY_CREATE_INFO_EXT

  • Extending VkResult:

    • VK_ERROR_NOT_PERMITTED_EXT

New Functions

None.

Issues

None.

Version History

  • Revision 2, 2017-11-03 (Andres Rodriguez)

    • Fixed VkQueueGlobalPriorityEXT missing _EXT suffix

  • Revision 1, 2017-10-06 (Andres Rodriguez)

    • First version.

VK_EXT_hdr_metadata

Name String

VK_EXT_hdr_metadata

Extension Type

Device extension

Registered Extension Number

106

Revision

1

Extension and Version Dependencies
Contact
  • Courtney Goeltzenleuchter @courtney-g

Last Modified Date

2017-03-04

IP Status

No known IP claims.

Contributors
  • Courtney Goeltzenleuchter, Google

This extension defines two new structures and a function to assign SMPTE (the Society of Motion Picture and Television Engineers) 2086 metadata and CTA (Consumer Technology Assocation) 861.3 metadata to a swapchain. The metadata includes the color primaries, white point, and luminance range of the mastering display, which all together define the color volume that contains all the possible colors the mastering display can produce. The mastering display is the display where creative work is done and creative intent is established. To preserve such creative intent as much as possible and achieve consistent color reproduction on different viewing displays, it is useful for the display pipeline to know the color volume of the original mastering display where content was created or tuned. This avoids performing unnecessary mapping of colors that are not displayable on the original mastering display. The metadata also includes the maxContentLightLevel and maxFrameAverageLightLevel as defined by CTA 861.3.

While the general purpose of the metadata is to assist in the transformation between different color volumes of different displays and help achieve better color reproduction, it is not in the scope of this extension to define how exactly the metadata should be used in such a process. It is up to the implementation to determine how to make use of the metadata.

New Enum Constants

New Functions

Issues

1) Do we need a query function?

PROPOSED: No, Vulkan does not provide queries for state that the application can track on its own.

2) Should we specify default if not specified by the application?

PROPOSED: No, that leaves the default up to the display.

Version History

  • Revision 1, 2016-12-27 (Courtney Goeltzenleuchter)

    • Initial version

VK_EXT_post_depth_coverage

Name String

VK_EXT_post_depth_coverage

Extension Type

Device extension

Registered Extension Number

156

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2017-07-17

Interactions and External Dependencies
Contributors
  • Jeff Bolz, NVIDIA

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_KHR_post_depth_coverage

which allows the fragment shader to control whether values in the SampleMask built-in input variable reflect the coverage after the early per-fragment depth and stencil tests are applied.

This extension adds a new PostDepthCoverage execution mode under the SampleMaskPostDepthCoverage capability. When this mode is specified along with EarlyFragmentTests, the value of an input variable decorated with the SampleMask built-in reflects the coverage after the early fragment tests are applied. Otherwise, it reflects the coverage before the depth and stencil tests.

When using GLSL source-based shading languages, the post_depth_coverage layout qualifier from GL_ARB_post_depth_coverage or GL_EXT_post_depth_coverage maps to the PostDepthCoverage execution mode.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New Built-In Variables

None.

New Variable Decoration

None.

New SPIR-V Capabilities

Issues

None yet.

Version History

  • Revision 1, 2017-07-17 (Daniel Koch)

    • Internal revisions

VK_EXT_queue_family_foreign

Name String

VK_EXT_queue_family_foreign

Extension Type

Device extension

Registered Extension Number

127

Revision

1

Extension and Version Dependencies
Contact
  • Chad Versace @chadversary

Last Modified Date

2017-11-01

IP Status

No known IP claims.

Contributors
  • Chad Versace, Google

  • James Jones, NVIDIA

  • Jason Ekstrand, Intel

  • Jesse Hall, Google

  • Daniel Rakos, AMD

  • Ray Smith, ARM

This extension defines a special queue family, VK_QUEUE_FAMILY_FOREIGN_EXT, which can be used to transfer ownership of resources backed by external memory to foreign, external queues. This is similar to VK_QUEUE_FAMILY_EXTERNAL_KHR, defined in VK_KHR_external_memory. The key differences between the two are:

  • The queues represented by VK_QUEUE_FAMILY_EXTERNAL_KHR must share the same physical device and the same driver version as the current VkInstance. VK_QUEUE_FAMILY_FOREIGN_EXT has no such restrictions. It can represent devices and drivers from other vendors, and can even represent non-Vulkan-capable devices.

  • All resources backed by external memory support VK_QUEUE_FAMILY_EXTERNAL_KHR. Support for VK_QUEUE_FAMILY_FOREIGN_EXT is more restrictive.

  • Applications should expect transitions to/from VK_QUEUE_FAMILY_FOREIGN_EXT to be more expensive than transitions to/from VK_QUEUE_FAMILY_EXTERNAL_KHR.

New Enum Constants

  • Special constants:

    • VK_QUEUE_FAMILY_FOREIGN_EXT

Version History

  • Revision 1, 2017-11-01 (Chad Versace)

    • Squashed internal revisions

VK_EXT_sample_locations

Name String

VK_EXT_sample_locations

Extension Type

Device extension

Registered Extension Number

144

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Rakos @drakos-amd

Last Modified Date

2017-08-02

Contributors
  • Mais Alnasser, AMD

  • Matthaeus G. Chajdas, AMD

  • Maciej Jesionowski, AMD

  • Daniel Rakos, AMD

  • Slawomir Grajewski, Intel

  • Jeff Bolz, NVIDIA

  • Bill Licea-Kane, Qualcomm

This extension allows an application to modify the locations of samples within a pixel used in rasterization. Additionally, it allows applications to specify different sample locations for each pixel in a group of adjacent pixels, which can increase antialiasing quality (particularly if a custom resolve shader is used that takes advantage of these different locations).

It is common for implementations to optimize the storage of depth values by storing values that can be used to reconstruct depth at each sample location, rather than storing separate depth values for each sample. For example, the depth values from a single triangle may be represented using plane equations. When the depth value for a sample is needed, it is automatically evaluated at the sample location. Modifying the sample locations causes the reconstruction to no longer evaluate the same depth values as when the samples were originally generated, thus the depth aspect of a depth/stencil attachment must be cleared before rendering to it using different sample locations.

Some implementations may need to evaluate depth image values while performing image layout transitions. To accommodate this, instances of the VkSampleLocationsInfoEXT structure can be specified for each situation where an explicit or automatic layout transition has to take place. VkSampleLocationsInfoEXT can be chained from VkImageMemoryBarrier structures to provide sample locations for layout transitions performed by vkCmdWaitEvents and vkCmdPipelineBarrier calls, and VkRenderPassSampleLocationsBeginInfoEXT can be chained from VkRenderPassBeginInfo to provide sample locations for layout transitions performed implicitly by a render pass instance.

New Object Types

None.

New Enum Constants

  • Extending VkImageCreateFlagBits:

    • VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_SAMPLE_LOCATIONS_INFO_EXT

    • VK_STRUCTURE_TYPE_RENDER_PASS_SAMPLE_LOCATIONS_BEGIN_INFO_EXT

    • VK_STRUCTURE_TYPE_PIPELINE_SAMPLE_LOCATIONS_STATE_CREATE_INFO_EXT

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLE_LOCATIONS_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_MULTISAMPLE_PROPERTIES_EXT

  • Extending VkDynamicState:

    • VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT

New Enums

None.

Issues

None.

Version History

  • Revision 1, 2017-08-02 (Daniel Rakos)

    • Internal revisions

VK_EXT_sampler_filter_minmax

Name String

VK_EXT_sampler_filter_minmax

Extension Type

Device extension

Registered Extension Number

131

Revision

1

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-05-19

IP Status

No known IP claims.

Contributors
  • Jeff Bolz, NVIDIA

  • Piers Daniell, NVIDIA

In unextended Vulkan, minification and magnification filters such as LINEAR allow sampled image lookups to return a filtered texel value produced by computing a weighted average of a collection of texels in the neighborhood of the texture coordinate provided.

This extension provides a new sampler parameter which allows applications to produce a filtered texel value by computing a component-wise minimum (MIN) or maximum (MAX) of the texels that would normally be averaged. The reduction mode is orthogonal to the minification and magnification filter parameters. The filter parameters are used to identify the set of texels used to produce a final filtered value; the reduction mode identifies how these texels are combined.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SAMPLER_FILTER_MINMAX_PROPERTIES_EXT

    • VK_STRUCTURE_TYPE_SAMPLER_REDUCTION_MODE_CREATE_INFO_EXT

  • Extending VkFormatFeatureFlagBits

    • VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT_EXT

New Functions

None.

New Built-In Variables

None.

New SPIR-V Capabilities

None.

Issues

None.

Examples

None.

Version History

  • Revision 2, 2017-05-19 (Piers Daniell)

    • Renamed to EXT

  • Revision 1, 2017-03-25 (Jeff Bolz)

    • Internal revisions

VK_EXT_shader_stencil_export

Name String

VK_EXT_shader_stencil_export

Extension Type

Device extension

Registered Extension Number

141

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Dominik Witczak @dominikwitczakamd

Last Modified Date

2017-07-19

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Dominik Witczak, AMD

  • Daniel Rakos, AMD

  • Rex Xu, AMD

This extension adds support for the SPIR-V extension SPV_EXT_shader_stencil_export, providing a mechanism whereby a shader may generate the stencil reference value per invocation. When stencil testing is enabled, this allows the test to be performed against the value generated in the shader.

Version History

  • Revision 1, 2017-07-19 (Dominik Witczak)

    • Initial draft

VK_EXT_shader_subgroup_ballot

Name String

VK_EXT_shader_subgroup_ballot

Extension Type

Device extension

Registered Extension Number

65

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2016-11-28

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Jeff Bolz, NVIDIA

  • Neil Henning, Codeplay

  • Daniel Koch, NVIDIA Corporation

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_KHR_shader_ballot

This extension provides the ability for a group of invocations, which execute in parallel, to do limited forms of cross-invocation communication via a group broadcast of a invocation value, or broadcast of a bitarray representing a predicate value from each invocation in the group.

This extension provides access to a number of additional built-in shader variables in Vulkan:

  • SubgroupEqMaskKHR, which contains the subgroup mask of the current subgroup invocation,

  • SubgroupGeMaskKHR, which contains the subgroup mask of the invocations greater than or equal to the current invocation,

  • SubgroupGtMaskKHR, which contains the subgroup mask of the invocations greater than the current invocation,

  • SubgroupLeMaskKHR, which contains the subgroup mask of the invocations less than or equal to the current invocation,

  • SubgroupLtMaskKHR, which contains the subgroup mask of the invocations less than the current invocation,

  • SubgroupLocalInvocationId, which contains the index of an invocation within a subgroup, and

  • SubgroupSize, which contains the maximum number of invocations in a subgroup.

Additionally, this extension provides access to the new SPIR-V instructions:

  • OpSubgroupBallotKHR,

  • OpSubgroupFirstInvocationKHR, and

  • OpSubgroupReadInvocationKHR,

When using GLSL source-based shader languages, the following variables and shader functions from GL_ARB_shader_ballot can map to these SPIR-V built-in decorations and instructions:

  • in uint64_t gl_SubGroupEqMaskARB;SubgroupEqMaskKHR,

  • in uint64_t gl_SubGroupGeMaskARB;SubgroupGeMaskKHR,

  • in uint64_t gl_SubGroupGtMaskARB;SubgroupGtMaskKHR,

  • in uint64_t gl_SubGroupLeMaskARB;SubgroupLeMaskKHR,

  • in uint64_t gl_SubGroupLtMaskARB;SubgroupLtMaskKHR,

  • in uint gl_SubGroupInvocationARB;SubgroupLocalInvocationId,

  • uniform uint gl_SubGroupSizeARB;SubgroupSize,

  • ballotARB() → OpSubgroupBallotKHR,

  • readFirstInvocationARB() → OpSubgroupFirstInvocationKHR, and

  • readInvocationARB() → OpSubgroupReadInvocationKHR.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New SPIR-V Capabilities

Issues

None.

Version History

  • Revision 1, 2016-11-28 (Daniel Koch)

    • Initial draft

VK_EXT_shader_subgroup_vote

Name String

VK_EXT_shader_subgroup_vote

Extension Type

Device extension

Registered Extension Number

66

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2016-11-28

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Neil Henning, Codeplay

  • Daniel Koch, NVIDIA Corporation

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_KHR_subgroup_vote

This extension provides new SPIR-V instructions:

  • OpSubgroupAllKHR,

  • OpSubgroupAnyKHR, and

  • OpSubgroupAllEqualKHR.

to compute the composite of a set of boolean conditions across a group of shader invocations that are running concurrently (a subgroup). These composite results may be used to execute shaders more efficiently on a VkPhysicalDevice.

When using GLSL source-based shader languages, the following shader functions from GL_ARB_shader_group_vote can map to these SPIR-V instructions:

  • anyInvocationARB() → OpSubgroupAnyKHR,

  • allInvocationsARB() → OpSubgroupAllKHR, and

  • allInvocationsEqualARB() → OpSubgroupAllEqualKHR.

The subgroup across which the boolean conditions are evaluated is implementation-dependent, and this extension provides no guarantee over how individual shader invocations are assigned to subgroups. In particular, a subgroup has no necessary relationship with the compute shader local workgroup — any pair of shader invocations in a compute local workgroup may execute in different subgroups as used by these instructions.

Compute shaders operate on an explicitly specified group of threads (a local workgroup), but many implementations will also group non-compute shader invocations and execute them concurrently. When executing code like

if (condition) {
  result = do_fast_path();
} else {
  result = do_general_path();
}

where condition diverges between invocations, an implementation might first execute do_fast_path() for the invocations where condition is true and leave the other invocations dormant. Once do_fast_path() returns, it might call do_general_path() for invocations where condition is false and leave the other invocations dormant. In this case, the shader executes both the fast and the general path and might be better off just using the general path for all invocations.

This extension provides the ability to avoid divergent execution by evaluating a condition across an entire subgroup using code like:

if (allInvocationsARB(condition)) {
  result = do_fast_path();
} else {
  result = do_general_path();
}

The built-in function allInvocationsARB() will return the same value for all invocations in the group, so the group will either execute do_fast_path() or do_general_path(), but never both. For example, shader code might want to evaluate a complex function iteratively by starting with an approximation of the result and then refining the approximation. Some input values may require a small number of iterations to generate an accurate result (do_fast_path) while others require a larger number (do_general_path). In another example, shader code might want to evaluate a complex function (do_general_path) that can be greatly simplified when assuming a specific value for one of its inputs (do_fast_path).

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New Built-In Variables

None.

New SPIR-V Capabilities

Issues

None.

Version History

  • Revision 1, 2016-11-28 (Daniel Koch)

    • Initial draft

VK_EXT_shader_viewport_index_layer

Name String

VK_EXT_shader_viewport_index_layer

Extension Type

Device extension

Registered Extension Number

163

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2017-08-08

Interactions and External Dependencies
Contributors
  • Piers Daniell, NVIDIA

  • Jeff Bolz, NVIDIA

  • Jan-Harald Fredriksen, ARM

  • Daniel Rakos, AMD

  • Slawomir Grajeswki, Intel

This extension adds support for the ShaderViewportIndexLayerEXT capability from the SPV_EXT_shader_viewport_index_layer extension in Vulkan.

This extension allows variables decorated with the Layer and ViewportIndex built-ins to be exported from vertex or tessellation shaders, using the ShaderViewportIndexLayerEXT capability.

When using GLSL source-based shading languages, the gl_ViewportIndex and gl_Layer built-in variables map to the SPIR-V ViewportIndex and Layer built-in decorations, respectively. Behaviour of these variables is extended as described in the GL_ARB_shader_viewport_layer_array (or the precursor GL_AMD_vertex_shader_layer, GL_AMD_vertex_shader_viewport_index, and GL_NV_viewport_array2 extensions).

Note

The ShaderViewportIndexLayerEXT capability is equivalent to the ShaderViewportIndexLayerNV capability added by VK_NV_viewport_array2.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New or Modified Built-In Variables

New Variable Decoration

None.

New SPIR-V Capabilities

Issues

None yet!

Version History

  • Revision 1, 2017-08-08 (Daniel Koch)

    • Internal drafts

VK_EXT_swapchain_colorspace

Name String

VK_EXT_swapchain_colorspace

Extension Type

Instance extension

Registered Extension Number

105

Revision

3

Extension and Version Dependencies
Contact
  • Courtney Goeltzenleuchter @courtney-g

Last Modified Date

2017-03-15

IP Status

No known IP claims.

Contributors
  • Courtney Goeltzenleuchter, Google

New Enum Constants

  • Extending VkColorSpaceKHR:

    • VK_COLOR_SPACE_DISPLAY_P3_NONLINEAR_EXT - supports the Display-P3 color space and applies an sRGB-like transfer function.

    • VK_COLOR_SPACE_EXTENDED_SRGB_LINEAR_EXT - supports the extended sRGB color space and applies a linear transfer function.

    • VK_COLOR_SPACE_EXTENDED_SRGB_NONLINEAR_EXT - supports the extended sRGB color space with an sRGB nonlinear transfer function.

    • VK_COLOR_SPACE_DCI_P3_LINEAR_EXT - supports the DCI-P3 color space and applies a linear OETF.

    • VK_COLOR_SPACE_DCI_P3_NONLINEAR_EXT - supports the DCI-P3 color space and applies the Gamma 2.6 OETF.

    • VK_COLOR_SPACE_BT709_LINEAR_EXT - supports the BT709 color space and applies a linear transfer function.

    • VK_COLOR_SPACE_BT709_NONLINEAR_EXT - supports the BT709 color space and applies the SMPTE 170M OETF.

    • VK_COLOR_SPACE_BT2020_LINEAR_EXT - supports the BT2020 color space and applies a linear OETF.

    • VK_COLOR_SPACE_HDR10_ST2084_EXT - supports HDR10 (BT2020 color space and applies the SMPTE ST2084 Perceptual Quantizer (PQ) OETF).

    • VK_COLOR_SPACE_DOLBYVISION_EXT - supports Dolby Vision (BT2020 color space, proprietary encoding, and applies the SMPTE ST2084 OETF).

    • VK_COLOR_SPACE_HDR10_HLG_EXT - supports HDR10 (BT2020 color space and applies the Hybrid Log Gamma (HLG) OETF).

    • VK_COLOR_SPACE_ADOBERGB_LINEAR_EXT - supports the AdobeRGB color space and applies a linear OETF.

    • VK_COLOR_SPACE_ADOBERGB_NONLINEAR_EXT - supports the AdobeRGB color space and applies the Gamma 2.2 OETF.

    • VK_COLOR_SPACE_PASS_THROUGH_EXT - color components used “as is”. Intended to allow application to supply data for color spaces not described here.

Issues

1) Does the spec need to specify which kinds of image formats support the color spaces?

RESOLVED: Pixel format is independent of color space (though some color spaces really want / need floating point color components to be useful). Therefore, do not plan on documenting what formats support which colorspaces. An application can call vkGetPhysicalDeviceSurfaceFormatsKHR to query what a particular implementation supports.

2) How does application determine if HW supports appropriate transfer function for a colorspace?

RESOLVED: Extension indicates that implementation must not do the OETF encoding if it is not sRGB. That responsibility falls to the application shaders. Any other native OETF / EOTF functions supported by an implementation can be described by separate extension.

Version History

  • Revision 1, 2016-12-27 (Courtney Goeltzenleuchter)

    • Initial version

  • Revision 2, 2017-01-19 (Courtney Goeltzenleuchter)

    • Add pass through and multiple options for BT2020.

    • Clean up some issues with equations not displaying properly.

  • Revision 3, 2017-06-23 (Courtney Goeltzenleuchter)

    • Add extended sRGB non-linear enum.

VK_EXT_validation_flags

Name String

VK_EXT_validation_flags

Extension Type

Instance extension

Registered Extension Number

62

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Tobin Ehlis @tobine

Last Modified Date

2016-09-06

IP Status

No known IP claims.

Contributors
  • Tobin Ehlis, Google

  • Courtney Goeltzenleuchter, Google

This extension provides the VkValidationFlagsEXT struct that can be included in the pNext chain of the VkInstanceCreateInfo structure passed as the pCreateInfo parameter of vkCreateInstance. The new struct contains an array of VkValidationCheckEXT values that will be disabled by the validation layers.

New Enum Constants

New Structures

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2016-08-26 (Courtney Goeltzenleuchter)

    • Initial draft

VK_EXT_validation_cache

Name String

VK_EXT_validation_cache

Extension Type

Device extension

Registered Extension Number

161

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Cort Stratton @cdwfs

Last Modified Date

2017-08-29

IP Status

No known IP claims.

Contributors
  • Cort Stratton, Google

  • Chris Forbes, Google

This extension provides a mechanism for caching the results of potentially expensive internal validation operations across multiple runs of a Vulkan application. At the core is the VkValidationCacheEXT object type, which is managed similarly to the existing VkPipelineCache.

The new struct VkShaderModuleValidationCacheCreateInfoEXT can be included in the pNext chain at vkCreateShaderModule time. It contains a VkValidationCacheEXT to use when validating the VkShaderModule.

New Object Types

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_VALIDATION_CACHE_CREATE_INFO_EXT

    • VK_STRUCTURE_TYPE_SHADER_MODULE_VALIDATION_CACHE_CREATE_INFO_EXT

Issues

None.

Version History

  • Revision 1, 2017-08-29 (Cort Stratton)

    • Initial draft

VK_EXT_vertex_attribute_divisor

Name String

VK_EXT_vertex_attribute_divisor

Extension Type

Device extension

Registered Extension Number

191

Revision

1

Extension and Version Dependencies
Contact
  • Vikram Kushwaha @vkushwaha

Last Modified Date

2018-02-08

IP Status

No known IP claims.

Contributors
  • Vikram Kushwaha, NVIDIA

This extension allows instance-rate vertex attributes to be repeated for certain number of instances instead of advancing for every instance when instanced rendering is enabled.

New Object Types

None.

New Enum Constants

Extending VkStructureType:

  • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VERTEX_ATTRIBUTE_DIVISOR_PROPERTIES_EXT

  • VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_DIVISOR_STATE_CREATE_INFO_EXT

New Enums

None.

New Functions

None.

Issues

None.

Examples

To create a vertex binding such that the first binding uses instanced rendering and the same attribute is used for every 4 draw instances, an application could use the following set of structures:

    const VkVertexInputBindingDivisorDescriptionEXT divisorDesc =
    {
        0,
        4
    };

    const VkPipelineVertexInputDivisorStateCreateInfoEXT divisorInfo =
    {
        VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_DIVISOR_STATE_CREATE_INFO_EXT, // sType
        NULL,                                                             // pNext
        1,                                                                // vertexBindingDivisorCount
        &divisorDesc                                                      // pVertexBindingDivisors
    }

    const VkVertexInputBindingDescription binding =
    {
        0,                                                                // binding
        sizeof(Vertex),                                                   // stride
        VK_VERTEX_INPUT_RATE_INSTANCE                                     // inputRate
    };

    const VkPipelineVertexInputStateCreateInfo viInfo =
    {
        VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_CREATE_INFO,              // sType
        &divisorInfo,                                                     // pNext
        ...
    };
    //...

Version History

  • Revision 1, 2017-12-04 (Vikram Kushwaha)

    • First Version

VK_AMD_buffer_marker

Name String

VK_AMD_buffer_marker

Extension Type

Device extension

Registered Extension Number

180

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Rakos @drakos-amd

Last Modified Date

2018-01-26

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Jaakko Konttinen, AMD

  • Daniel Rakos, AMD

This extension adds a new operation to execute pipelined writes of small marker values into a VkBuffer object.

The primary purpose of these markers is to facilitate the development of debugging tools for tracking which pipelined command contributed to device loss.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

Examples

None.

Version History

  • Revision 1, 2018-01-26 (Jaakko Konttinen)

    • Initial revision

VK_AMD_draw_indirect_count

Name String

VK_AMD_draw_indirect_count

Extension Type

Device extension

Registered Extension Number

34

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Rakos @drakos-amd

Last Modified Date

2016-08-23

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Derrick Owens, AMD

  • Graham Sellers, AMD

  • Daniel Rakos, AMD

  • Dominik Witczak, AMD

This extension allows an application to source the number of draw calls for indirect draw calls from a buffer. This enables applications to generate arbitrary amounts of draw commands and execute them without host intervention.

Version History

  • Revision 2, 2016-08-23 (Dominik Witczak)

    • Minor fixes

  • Revision 1, 2016-07-21 (Matthaeus Chajdas)

    • Initial draft

VK_AMD_gcn_shader

Name String

VK_AMD_gcn_shader

Extension Type

Device extension

Registered Extension Number

26

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Dominik Witczak @dominikwitczakamd

Last Modified Date

2016-05-30

IP Status

No known IP claims.

Contributors
  • Dominik Witczak, AMD

  • Daniel Rakos, AMD

  • Rex Xu, AMD

  • Graham Sellers, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

editing-note

Shouldn’t the SPV extension be in the Interactions and External Dependencies block?

Version History

  • Revision 1, 2016-05-30 (Dominik Witczak)

    • Initial draft

VK_AMD_gpu_shader_half_float

Name String

VK_AMD_gpu_shader_half_float

Extension Type

Device extension

Registered Extension Number

37

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Dominik Witczak @dominikwitczakamd

Last Modified Date

2016-09-21

IP Status

No known IP claims.

Contributors
  • Daniel Rakos, AMD

  • Dominik Witczak, AMD

  • Donglin Wei, AMD

  • Graham Sellers, AMD

  • Qun Lin, AMD

  • Rex Xu, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

editing-note

Shouldn’t the SPV extension be in the Interactions and External Dependencies block?

Version History

  • Revision 1, 2019-09-21 (Dominik Witczak)

    • Initial draft

VK_AMD_gpu_shader_int16

Name String

VK_AMD_gpu_shader_int16

Extension Type

Device extension

Registered Extension Number

133

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Qun Lin, AMD @linqun

Last Modified Date

2017-06-08

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Daniel Rakos, AMD

  • Dominik Witczak, AMD

  • Matthaeus G. Chajdas, AMD

  • Rex Xu, AMD

  • Timothy Lottes, AMD

  • Zhi Cai, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_AMD_gpu_shader_int16

Version History

  • Revision 1, 2017-06-18 (Dominik Witczak)

    • First version.

VK_AMD_mixed_attachment_samples

Name String

VK_AMD_mixed_attachment_samples

Extension Type

Device extension

Registered Extension Number

137

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Matthaeus G. Chajdas @anteru

Last Modified Date

2017-07-24

Contributors
  • Mais Alnasser, AMD

  • Matthaeus G. Chajdas, AMD

  • Maciej Jesionowski, AMD

  • Daniel Rakos, AMD

This extension enables applications to use multisampled rendering with a depth/stencil sample count that is larger than the color sample count. Having a depth/stencil sample count larger than the color sample count allows maintaining geometry and coverage information at a higher sample rate than color information. All samples are depth/stencil tested, but only the first color sample count number of samples get a corresponding color output.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-07-24 (Daniel Rakos)

    • Internal revisions

VK_AMD_negative_viewport_height

Name String

VK_AMD_negative_viewport_height

Extension Type

Device extension

Registered Extension Number

36

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Matthaeus G. Chajdas @anteru

Last Modified Date

2016-09-02

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Graham Sellers, AMD

  • Baldur Karlsson

Interactions and External Dependencies
  • Deprecated by VK_KHR_maintenance1

  • Deprecated by Vulkan 1.1

This extension allows an application to specify a negative viewport height. The result is that the viewport transformation will flip along the y-axis.

  • Allow negative height to be specified in the slink::VkViewport::height field to perform y-inversion of the clip-space to framebuffer-space transform. This allows apps to avoid having to use gl_Position.y = -gl_Position.y in shaders also targeting other APIs.

Deprecation by VK_KHR_maintenance1 and Vulkan 1.1

Functionality in this extension is included in VK_KHR_maintenance1 and Vulkan 1.1. Due to some slight behavioral differences, this extension must not be enabled alongside VK_KHR_maintenance1, or in an instance created with version 1.1 or later requested in VkApplicationInfo::apiVersion.

Version History

  • Revision 1, 2016-09-02 (Matthaeus Chajdas)

    • Initial draft

VK_AMD_rasterization_order

Name String

VK_AMD_rasterization_order

Extension Type

Device extension

Registered Extension Number

19

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Rakos @drakos-amd

Last Modified Date

2016-04-25

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Jaakko Konttinen, AMD

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Dominik Witczak, AMD

This extension introduces the possibility for the application to control the order of primitive rasterization. In unextended Vulkan, the following stages are guaranteed to execute in API order:

  • depth bounds test

  • stencil test, stencil op, and stencil write

  • depth test and depth write

  • occlusion queries

  • blending, logic op, and color write

This extension enables applications to opt into a relaxed, implementation defined primitive rasterization order that may allow better parallel processing of primitives and thus enabling higher primitive throughput. It is applicable in cases where the primitive rasterization order is known to not affect the output of the rendering or any differences caused by a different rasterization order are not a concern from the point of view of the application’s purpose.

A few examples of cases when using the relaxed primitive rasterization order would not have an effect on the final rendering:

  • If the primitives rendered are known to not overlap in framebuffer space.

  • If depth testing is used with a comparison operator of VK_COMPARE_OP_LESS, VK_COMPARE_OP_LESS_OR_EQUAL, VK_COMPARE_OP_GREATER, or VK_COMPARE_OP_GREATER_OR_EQUAL, and the primitives rendered are known to not overlap in clip space.

  • If depth testing is not used and blending is enabled for all attachments with a commutative blend operator.

New Object Types

None

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_RASTERIZATION_ORDER_AMD

New Functions

None

Issues

1) How is this extension useful to application developers?

RESOLVED: Allows them to increase primitive throughput for cases when strict API order rasterization is not important due to the nature of the content, the configuration used, or the requirements towards the output of the rendering.

2) How does this extension interact with content optimizations aiming to reduce overdraw by appropriately ordering the input primitives?

RESOLVED: While the relaxed rasterization order might somewhat limit the effectiveness of such content optimizations, most of the benefits of it are expected to be retained even when the relaxed rasterization order is used, so applications should still apply these optimizations even if they intend to use the extension.

3) Are there any guarantees about the primitive rasterization order when using the new relaxed mode?

RESOLVED: No. In this case the rasterization order is completely implementation dependent, but in practice it is expected to partially still follow the order of incoming primitives.

4) Does the new relaxed rasterization order have any adverse effect on repeatability and other invariance rules of the API?

RESOLVED: Yes, in the sense that it extends the list of exceptions when the repeatability requirement does not apply.

Examples

None

Issues

None

Version History

  • Revision 1, 2016-04-25 (Daniel Rakos)

    • Initial draft.

VK_AMD_shader_ballot

Name String

VK_AMD_shader_ballot

Extension Type

Device extension

Registered Extension Number

38

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Dominik Witczak @dominikwitczakamd

Last Modified Date

2016-09-19

IP Status

No known IP claims.

Contributors
  • Qun Lin, AMD

  • Graham Sellers, AMD

  • Daniel Rakos, AMD

  • Rex Xu, AMD

  • Dominik Witczak, AMD

  • Matthäus G. Chajdas, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

editing-note

Shouldn’t the SPV extension be in the Interactions and External Dependencies block?

Version History

  • Revision 1, 2016-09-19 (Dominik Witczak)

    • Initial draft

VK_AMD_shader_core_properties

Name String

VK_AMD_shader_core_properties

Extension Type

Device extension

Registered Extension Number

186

Revision

1

Extension and Version Dependencies
Contact
  • Martin Dinkov @mdinkov

Last Modified Date

2018-02-15

IP Status

No known IP claims.

Contributors
  • Martin Dinkov, AMD

  • Matthaeus Chajdas, AMD

This extension exposes shader core properties for a target physical device through the VK_KHR_get_physical_device_properties2 extension. Please refer to the example below for proper usage.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_CORE_PROPERTIES_AMD

New Enums

None.

New Functions

None.

Examples

This example retrieves the shader core properties for a physical device.

extern VkInstance       instance;

PFN_vkGetPhysicalDeviceProperties2 pfnVkGetPhysicalDeviceProperties2 =
    reinterpret_cast<PFN_vkGetPhysicalDeviceProperties2>
    (vkGetInstanceProcAddr(instance, "vkGetPhysicalDeviceProperties2") );

VkPhysicalDeviceProperties2             general_props;
VkPhysicalDeviceShaderCorePropertiesAMD shader_core_properties;

shader_core_properties.pNext = nullptr;
shader_core_properties.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SHADER_CORE_PROPERTIES_AMD;

general_props.pNext = &shader_core_properties;
general_props.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2;

// After this call, shader_core_properties has been populated
pfnVkGetPhysicalDeviceProperties2(device, &general_props);

printf("Number of shader engines: %d\n",
    m_shader_core_properties.shader_engine_count =
    shader_core_properties.shaderEngineCount;
printf("Number of shader arrays: %d\n",
    m_shader_core_properties.shader_arrays_per_engine_count =
    shader_core_properties.shaderArraysPerEngineCount;
printf("Number of CUs per shader array: %d\n",
    m_shader_core_properties.compute_units_per_shader_array =
    shader_core_properties.computeUnitsPerShaderArray;
printf("Number of SIMDs per compute unit: %d\n",
    m_shader_core_properties.simd_per_compute_unit =
    shader_core_properties.simdPerComputeUnit;
printf("Number of wavefront slots in each SIMD: %d\n",
    m_shader_core_properties.wavefronts_per_simd =
    shader_core_properties.wavefrontsPerSimd;
printf("Number of threads per wavefront: %d\n",
    m_shader_core_properties.wavefront_size =
    shader_core_properties.wavefrontSize;
printf("Number of physical SGPRs per SIMD: %d\n",
    m_shader_core_properties.sgprs_per_simd =
    shader_core_properties.sgprsPerSimd;
printf("Minimum number of SGPRs that can be allocated by a wave: %d\n",
    m_shader_core_properties.min_sgpr_allocation =
    shader_core_properties.minSgprAllocation;
printf("Number of available SGPRs: %d\n",
    m_shader_core_properties.max_sgpr_allocation =
    shader_core_properties.maxSgprAllocation;
printf("SGPRs are allocated in groups of this size: %d\n",
    m_shader_core_properties.sgpr_allocation_granularity =
    shader_core_properties.sgprAllocationGranularity;
printf("Number of physical VGPRs per SIMD: %d\n",
    m_shader_core_properties.vgprs_per_simd =
    shader_core_properties.vgprsPerSimd;
printf("Minimum number of VGPRs that can be allocated by a wave: %d\n",
    m_shader_core_properties.min_vgpr_allocation =
    shader_core_properties.minVgprAllocation;
printf("Number of available VGPRs: %d\n",
    m_shader_core_properties.max_vgpr_allocation =
    shader_core_properties.maxVgprAllocation;
printf("VGPRs are allocated in groups of this size: %d\n",
    m_shader_core_properties.vgpr_allocation_granularity =
    shader_core_properties.vgprAllocationGranularity;

Version History

  • Revision 1, 2018-02-15 (Martin Dinkov)

    • Initial draft.

VK_AMD_shader_explicit_vertex_parameter

Name String

VK_AMD_shader_explicit_vertex_parameter

Extension Type

Device extension

Registered Extension Number

22

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Qun Lin, AMD @linqun

Last Modified Date

2016-05-10

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Qun Lin, AMD

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Rex Xu, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

editing-note

Shouldn’t the SPV extension be in the Interactions and External Dependencies block?

Version History

  • Revision 1, 2016-05-10 (Daniel Rakos)

    • Initial draft

VK_AMD_shader_fragment_mask

Name String

VK_AMD_shader_fragment_mask

Extension Type

Device extension

Registered Extension Number

138

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Aaron Hagan @AaronHaganAMD

Last Modified Date

2017-08-16

IP Status

No known IP claims.

Dependencies
Contributors
  • Aaron Hagan, AMD

  • Daniel Rakos, AMD

  • Timothy Lottes, AMD

This extension provides efficient read access to the fragment mask in compressed multisampled color surfaces. The fragment mask is a lookup table that associates color samples with color fragment values.

From a shader, the fragment mask can be fetched with a call to fragmentMaskFetchAMD, which returns a single uint where each subsequent four bits specify the color fragment index corresponding to the color sample, starting from the least significant bit. For example, when eight color samples are used, the color fragment index for color sample 0 will be in bits 0-3 of the fragment mask, for color sample 7 the index will be in bits 28-31.

The color fragment for a particular color sample may then be fetched with the corresponding fragment mask value using the fragmentFetchAMD shader function.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New SPIR-V Capabilities

New Structures

None.

New Functions

None.

Examples

This example shows a shader that queries the fragment mask from a multisampled compressed surface and uses it to query fragment values.

#version 450 core

#extension GL_AMD_shader_fragment_mask: enable

layout(binding = 0) uniform sampler2DMS       s2DMS;
layout(binding = 1) uniform isampler2DMSArray is2DMSArray;

layout(binding = 2, input_attachment_index = 0) uniform usubpassInputMS usubpassMS;

layout(location = 0) out vec4 fragColor;

void main()
{
    vec4 fragOne = vec4(0.0);

    uint fragMask = fragmentMaskFetchAMD(s2DMS, ivec2(2, 3));
    uint fragIndex = (fragMask & 0xF0) >> 4;
    fragOne += fragmentFetchAMD(s2DMS, ivec2(2, 3), 1);

    fragMask = fragmentMaskFetchAMD(is2DMSArray, ivec3(2, 3, 1));
    fragIndex = (fragMask & 0xF0) >> 4;
    fragOne += fragmentFetchAMD(is2DMSArray, ivec3(2, 3, 1), fragIndex);

    fragMask = fragmentMaskFetchAMD(usubpassMS);
    fragIndex = (fragMask & 0xF0) >> 4;
    fragOne += fragmentFetchAMD(usubpassMS, fragIndex);

    fragColor = fragOne;
}

Version History

  • Revision 1, 2017-08-16 (Aaron Hagan)

    • Initial draft

VK_AMD_shader_image_load_store_lod

Name String

VK_AMD_shader_image_load_store_lod

Extension Type

Device extension

Registered Extension Number

47

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Dominik Witczak @dominikwitczakamd

Last Modified Date

2017-08-21

Interactions and External Dependencies
IP Status

No known IP claims.

Contributors
  • Dominik Witczak, AMD

  • Qun Lin, AMD

  • Rex Xu, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

Version History

  • Revision 1, 2017-08-21 (Dominik Witczak)

    • Initial draft

VK_AMD_shader_info

Name String

VK_AMD_shader_info

Extension Type

Device extension

Registered Extension Number

43

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jaakko Konttinen @jaakkoamd

Last Modified Date

2017-10-09

IP Status

No known IP claims.

Contributors
  • Jaakko Konttinen, AMD

This extension adds a way to query certain information about a compiled shader which is part of a pipeline. This information may include shader disassembly, shader binary and various statistics about a shader’s resource usage.

While this extension provides a mechanism for extracting this information, the details regarding the contents or format of this information are not specified by this extension and may be provided by the vendor externally.

Furthermore, all information types are optionally supported, and users should not assume every implementation supports querying every type of information.

New Object Types

None.

New Enum Constants

None.

New Structures

New Functions

Examples

This example extracts the register usage of a fragment shader within a particular graphics pipeline:

extern VkDevice device;
extern VkPipeline gfxPipeline;

PFN_vkGetShaderInfoAMD pfnGetShaderInfoAMD = (PFN_vkGetShaderInfoAMD)vkGetDeviceProcAddr(
    device, "vkGetShaderInfoAMD");

VkShaderStatisticsInfoAMD statistics = {};

size_t dataSize = sizeof(statistics);

if (pfnGetShaderInfoAMD(device,
    gfxPipeline,
    VK_SHADER_STAGE_FRAGMENT_BIT,
    VK_SHADER_INFO_TYPE_STATISTICS_AMD,
    &dataSize,
    &statistics) == VK_SUCCESS)
{
    printf("VGPR usage: %d\n", statistics.resourceUsage.numUsedVgprs);
    printf("SGPR usage: %d\n", statistics.resourceUsage.numUsedSgprs);
}

The following example continues the previous example by subsequently attempting to query and print shader disassembly about the fragment shader:

// Query disassembly size (if available)
if (pfnGetShaderInfoAMD(device,
    gfxPipeline,
    VK_SHADER_STAGE_FRAGMENT_BIT,
    VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD,
    &dataSize,
    nullptr) == VK_SUCCESS)
{
    printf("Fragment shader disassembly:\n");

    void* disassembly = malloc(dataSize);

    // Query disassembly and print
    if (pfnGetShaderInfoAMD(device,
        gfxPipeline,
        VK_SHADER_STAGE_FRAGMENT_BIT,
        VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD,
        &dataSize,
        disassembly) == VK_SUCCESS)
    {
        printf((char*)disassembly);
    }

    free(disassembly);
}

Version History

  • Revision 1, 2017-10-09 (Jaakko Konttinen)

    • Initial revision

VK_AMD_shader_trinary_minmax

Name String

VK_AMD_shader_trinary_minmax

Extension Type

Device extension

Registered Extension Number

21

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Qun Lin, AMD @linqun

Last Modified Date

2016-05-10

IP Status

No known IP claims.

Contributors
  • Matthaeus G. Chajdas, AMD

  • Qun Lin, AMD

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Rex Xu, AMD

This extension adds support for the following SPIR-V extension in Vulkan:

editing-note

Shouldn’t the SPV extension be in the Interactions and External Dependencies block?

Version History

  • Revision 1, 2016-05-10 (Daniel Rakos)

    • Initial draft

VK_AMD_texture_gather_bias_lod

Name String

VK_AMD_texture_gather_bias_lod

Extension Type

Device extension

Registered Extension Number

42

Revision

1

Extension and Version Dependencies
Contact
  • Rex Xu @amdrexu

Last Modified Date

2017-03-21

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Dominik Witczak, AMD

  • Daniel Rakos, AMD

  • Graham Sellers, AMD

  • Matthaeus G. Chajdas, AMD

  • Qun Lin, AMD

  • Rex Xu, AMD

  • Timothy Lottes, AMD

This extension adds two related features.

Firstly, support for the following SPIR-V extension in Vulkan is added:

  • SPV_AMD_texture_gather_bias_lod

Secondly, the extension allows the application to query which formats can be used together with the new function prototypes introduced by the SPIR-V extension.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_TEXTURE_LOD_GATHER_FORMAT_PROPERTIES_AMD

New Enums

None.

New SPIR-V Capabilities

New Functions

None.

Examples

struct VkTextureLODGatherFormatPropertiesAMD
{
    VkStructureType sType;
    const void*     pNext;
    VkBool32        supportsTextureGatherLODBiasAMD;
};

// ----------------------------------------------------------------------------------------
// How to detect if an image format can be used with the new function prototypes.
VkPhysicalDeviceImageFormatInfo2   formatInfo;
VkImageFormatProperties2           formatProps;
VkTextureLODGatherFormatPropertiesAMD textureLODGatherSupport;

textureLODGatherSupport.sType = VK_STRUCTURE_TYPE_TEXTURE_LOD_GATHER_FORMAT_PROPERTIES_AMD;
textureLODGatherSupport.pNext = nullptr;

formatInfo.sType  = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_IMAGE_FORMAT_INFO_2;
formatInfo.pNext  = nullptr;
formatInfo.format = ...;
formatInfo.type   = ...;
formatInfo.tiling = ...;
formatInfo.usage  = ...;
formatInfo.flags  = ...;

formatProps.sType = VK_STRUCTURE_TYPE_IMAGE_FORMAT_PROPERTIES_2;
formatProps.pNext = &textureLODGatherSupport;

vkGetPhysicalDeviceImageFormatProperties2(physical_device, &formatInfo, &formatProps);

if (textureLODGatherSupport.supportsTextureGatherLODBiasAMD == VK_TRUE)
{
    // physical device supports SPV_AMD_texture_gather_bias_lod for the specified
    // format configuration.
}
else
{
    // physical device does not support SPV_AMD_texture_gather_bias_lod for the
    // specified format configuration.
}

Version History

  • Revision 1, 2017-03-21 (Dominik Witczak)

    • Initial draft

VK_ANDROID_external_memory_android_hardware_buffer

Name String

VK_ANDROID_external_memory_android_hardware_buffer

Extension Type

Device extension

Registered Extension Number

130

Revision

3

Extension and Version Dependencies
Contact
  • Jesse Hall @critsec

Status

Draft

Last Modified Date

2018-03-04

IP Status

No known IP claims.

Contributors
  • Ray Smith, ARM

  • Chad Versace, Google

  • Jesse Hall, Google

  • Tobias Hector, Imagination

  • James Jones, NVIDIA

  • Tony Zlatinski, NVIDIA

  • Matthew Netsch, Qualcomm

  • Andrew Garrard, Samsung

This extension enables an application to import Android AHardwareBuffer objects created outside of the Vulkan device into Vulkan memory objects, where they can be bound to images and buffers. It also allows exporting an AHardwareBuffer from a Vulkan memory object for symmetry with other operating systems. But since not all AHardwareBuffer usages and formats have Vulkan equivalents, exporting from Vulkan provides strictly less functionality than creating the AHardwareBuffer externally and importing it.

Some AHardwareBuffer images have implementation-defined external formats that may not correspond to Vulkan formats. Sampler Y’CbCr conversion can be used to sample from these images and convert them to a known colorspace.

New Object Types

None.

New Enum Constants

  • VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_USAGE_ANDROID

  • VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_PROPERTIES_ANDROID

  • VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_FORMAT_PROPERTIES_ANDROID

  • VK_STRUCTURE_TYPE_IMPORT_ANDROID_HARDWARE_BUFFER_INFO_ANDROID

  • VK_STRUCTURE_TYPE_MEMORY_GET_ANDROID_HARDWARE_BUFFER_INFO_ANDROID

  • VK_STRUCTURE_TYPE_EXTERNAL_FORMAT_ANDROID

  • VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID

New Enums

None.

Issues

1) Other external memory objects are represented as weakly-typed handles (e.g. Win32 HANDLE or POSIX file descriptor), and require a handle type parameter along with handles. AHardwareBuffer is strongly typed, so naming the handle type is redundant. Does symmetry justify adding handle type parameters/fields anyway?

RESOLVED: No. The handle type is already provided in places that treat external memory objects generically. In the places we would add it, the application code that would have to provide the handle type value is already dealing with AHardwareBuffer-specific commands/structures; the extra symmetry wouldn’t be enough to make that code generic.

2) The internal layout and therefore size of a AHardwareBuffer image may depend on native usage flags that don’t have corresponding Vulkan counterparts. Do we provide this info to vkCreateImage somehow, or allow the allocation size reported by vkGetImageMemoryRequirements to be approximate?

RESOLVED: Allow the allocation size to be unspecified when allocating the memory. It has to work this way for exported image memory anyway, since AHardwareBuffer allocation happens in vkAllocateMemory, and internally is performed by a separate HAL, not the Vulkan implementation itself. There is a similar issue with vkGetImageSubresourceLayout: the layout is determined by the allocator HAL, so it isn’t known until the image is bound to memory.

3) Should the result of sampling an external-format image with the suggested Y’CbCr conversion parameters yield the same results as using a samplerExternalOES in OpenGL ES?

RESOLVED: This would be desirable, so that apps converting from OpenGL ES to Vulkan could get the same output given the same input. But since sampling and conversion from Y’CbCr images is so loosely defined in OpenGL ES, multiple implementations do it in a way that doesn’t conform to Vulkan’s requirements. Modifying the OpenGL ES implementation would be difficult, and would change the output of existing unmodified applications. Changing the output only for applications that are being modified gives developers the chance to notice and mitigate any problems. Implementations are encouraged to minimize differences as much as possible without causing compatibility problems for existing OpenGL ES applications or violating Vulkan requirements.

4) Should AHardwareBuffers with AHARDWAREBUFFER_USAGE_CPU_* usage be mappable in Vulkan? Should it be possible to export AHardwareBuffers with such usage?

RESOLVED: Optional, and mapping in Vulkan is not the same as AHardwareBuffer_lock. The semantics of these are different: mapping in memory is persistent, just gives a raw view of the memory contents, and doesn’t involve ownership. AHardwareBuffer_lock gives the host exclusive access to the buffer, is temporary, and allows for reformatting copy-in/copy-out. Implementations aren’t required to support host-visible memory types for imported Android hardware buffers or resources backed by them. If a host-visible memory type is supported and used, the memory can be mapped in Vulkan, but doing so follows Vulkan semantics: it’s just a raw view of the data and doesn’t imply ownership (this means implementations must not internally call AHardwareBuffer_lock to implement vkMapMemory, or assume the application has done so). Implementations aren’t required to support linear-tiled images backed by Android hardware buffers, even if the AHardwareBuffer has CPU usage. There is no reliable way to allocate memory in Vulkan that can be exported to a AHardwareBuffer with CPU usage.

5) Android may add new AHardwareBuffer formats and usage flags over time. Can reference to them be added to this extension, or do they need a new extension?

RESOLVED: This extension can document the interaction between the new AHB formats/usages and existing Vulkan features. No new Vulkan features or implementation requirements can be added. The extension version number will be incremented when this additional documentation is added, but the version number doesn’t indicate that an implementaiton supports Vulkan memory or resources that map to the new AHardwareBuffer features: support for that must be queried with vkGetPhysicalDeviceImageFormatProperties2 or is implied by successfully allocating a AHardwareBuffer outside of Vulkan that uses the new feature and has a GPU usage flag.

In essence, these are new features added to a new Android API level, rather than new Vulkan features. The extension will only document how existing Vulkan features map to that new Android feature.

VK_GOOGLE_display_timing

Name String

VK_GOOGLE_display_timing

Extension Type

Device extension

Registered Extension Number

93

Revision

1

Extension and Version Dependencies
Contact
  • Ian Elliott @ianelliottus

Last Modified Date

2017-02-14

IP Status

No known IP claims.

Contributors
  • Ian Elliott, Google

  • Jesse Hall, Google

This device extension allows an application that uses the VK_KHR_swapchain extension to obtain information about the presentation engine’s display, to obtain timing information about each present, and to schedule a present to happen no earlier than a desired time. An application can use this to minimize various visual anomalies (e.g. stuttering).

Traditional game and real-time animation applications need to correctly position their geometry for when the presentable image will be presented to the user. To accomplish this, applications need various timing information about the presentation engine’s display. They need to know when presentable images were actually presented, and when they could have been presented. Applications also need to tell the presentation engine to display an image no sooner than a given time. This allows the application to avoid stuttering, so the animation looks smooth to the user.

This extension treats variable-refresh-rate (VRR) displays as if they are fixed-refresh-rate (FRR) displays.

New Object Types

None.

New Enum Constants

New Enums

None.

Issues

None.

Examples

Note

The example code for the this extension (like the VK_KHR_surface and VK_GOOGLE_display_timing extensions) is contained in the cube demo that is shipped with the official Khronos SDK, and is being kept up-to-date in that location (see: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/master/demos/cube.c ).

Version History

  • Revision 1, 2017-02-14 (Ian Elliott)

    • Internal revisions

VK_IMG_filter_cubic

Name String

VK_IMG_filter_cubic

Extension Type

Device extension

Registered Extension Number

16

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Tobias Hector @tobski

Last Modified Date

2016-02-23

Contributors
  • Tobias Hector, Imagination Technologies

VK_IMG_filter_cubic adds an additional, high quality cubic filtering mode to Vulkan, using a Catmull-Rom bicubic filter. Performing this kind of filtering can be done in a shader by using 16 samples and a number of instructions, but this can be inefficient. The cubic filter mode exposes an optimized high quality texture sampling using fixed texture sampling functionality.

New Enum Constants

Example

Creating a sampler with the new filter for both magnification and minification

    VkSamplerCreateInfo createInfo =
    {
        VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO // sType
        // Other members set to application-desired values
    };

    createInfo.magFilter = VK_FILTER_CUBIC_IMG;
    createInfo.minFilter = VK_FILTER_CUBIC_IMG;

    VkSampler sampler;
    VkResult result = vkCreateSampler(
        device,
        &createInfo,
        &sampler);

Version History

  • Revision 1, 2016-02-23 (Tobias Hector)

    • Initial version

VK_MVK_ios_surface

Name String

VK_MVK_ios_surface

Extension Type

Instance extension

Registered Extension Number

123

Revision

2

Extension and Version Dependencies
Contact
  • Bill Hollings @billhollings

Last Modified Date

2017-02-24

IP Status

No known IP claims.

Contributors
  • Bill Hollings, The Brenwill Workshop Ltd.

The VK_MVK_ios_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to a UIView, the native surface type of iOS, which is underpinned by a CAMetalLayer, to support rendering to the surface using Apple’s Metal framework.

New Object Types

None.

New Enum Constants

New Enums

None.

New Structures

New Functions

Issues

None.

Version History

  • Revision 1, 2017-02-15 (Bill Hollings)

    • Initial draft.

  • Revision 2, 2017-02-24 (Bill Hollings)

    • Minor syntax fix to emphasize firm requirement for UIView to be backed by a CAMetalLayer.

VK_MVK_macos_surface

Name String

VK_MVK_macos_surface

Extension Type

Instance extension

Registered Extension Number

124

Revision

2

Extension and Version Dependencies
Contact
  • Bill Hollings @billhollings

Last Modified Date

2017-02-24

IP Status

No known IP claims.

Contributors
  • Bill Hollings, The Brenwill Workshop Ltd.

The VK_MVK_macos_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) that refers to an NSView, the native surface type of macOS, which is underpinned by a CAMetalLayer, to support rendering to the surface using Apple’s Metal framework.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK

New Enums

None.

New Functions

Issues

None.

Version History

  • Revision 1, 2017-02-15 (Bill Hollings)

    • Initial draft.

  • Revision 2, 2017-02-24 (Bill Hollings)

    • Minor syntax fix to emphasize firm requirement for NSView to be backed by a CAMetalLayer.

VK_NN_vi_surface

Name String

VK_NN_vi_surface

Extension Type

Instance extension

Registered Extension Number

63

Revision

1

Extension and Version Dependencies
Contact
  • Mathias Heyer gitlab:@mheyer

Last Modified Date

2016-12-02

IP Status

No known IP claims.

Contributors
  • Mathias Heyer, NVIDIA

  • Michael Chock, NVIDIA

  • Yasuhiro Yoshioka, Nintendo

  • Daniel Koch, NVIDIA

The VK_NN_vi_surface extension is an instance extension. It provides a mechanism to create a VkSurfaceKHR object (defined by the VK_KHR_surface extension) associated with an nn::vi::Layer.

New Object Types

None

New Enum Constants

New Enums

None

New Structures

New Functions

Issues

1) Does VI need a way to query for compatibility between a particular physical device (and queue family?) and a specific VI display?

RESOLVED: No. It is currently always assumed that the device and display will always be compatible.

2) VkViSurfaceCreateInfoNN::pWindow is intended to store an nn::vi::NativeWindowHandle, but its declared type is a bare void* to store the window handle. Why the discrepancy?

RESOLVED: It is for C compatibility. The definition for the VI native window handle type is defined inside the nn::vi C++ namespace. This prevents its use in C source files. nn::vi::NativeWindowHandle is always defined to be void*, so this extension uses void* to match.

Version History

  • Revision 1, 2016-12-2 (Michael Chock)

    • Initial draft.

VK_NV_clip_space_w_scaling

Name String

VK_NV_clip_space_w_scaling

Extension Type

Device extension

Registered Extension Number

88

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Eric Werness @ewerness-nv

Last Modified Date

2017-02-15

Contributors
  • Eric Werness, NVIDIA

  • Kedarnath Thangudu, NVIDIA

Virtual Reality (VR) applications often involve a post-processing step to apply a “barrel” distortion to the rendered image to correct the “pincushion” distortion introduced by the optics in a VR device. The barrel distorted image has lower resolution along the edges compared to the center. Since the original image is rendered at high resolution, which is uniform across the complete image, a lot of pixels towards the edges do not make it to the final post-processed image.

This extension provides a mechanism to render VR scenes at a non-uniform resolution, in particular a resolution that falls linearly from the center towards the edges. This is achieved by scaling the w coordinate of the vertices in the clip space before perspective divide. The clip space w coordinate of the vertices can be offset as of a function of x and y coordinates as follows:

w' = w + Ax + By

In the intended use case for viewport position scaling, an application should use a set of four viewports, one for each of the four quadrants of a Cartesian coordinate system. Each viewport is set to the dimension of the image, but is scissored to the quadrant it represents. The application should specify A and B coefficients of the w-scaling equation above, that have the same value, but different signs, for each of the viewports. The signs of A and B should match the signs of x and y for the quadrant that they represent such that the value of w' will always be greater than or equal to the original w value for the entire image. Since the offset to w, (Ax + By), is always positive, and increases with the absolute values of x and y, the effective resolution will fall off linearly from the center of the image to its edges.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_W_SCALING_STATE_CREATE_INFO_NV

  • Extending VkDynamicState:

    • VK_DYANMIC_STATE_VIEWPORT_W_SCALING_NV

New Enums

None.

Issues

1) Is the pipeline struct name too long?

RESOLVED: It fits with the naming convention.

2) Separate W scaling section or fold into coordinate transformations?

RESOLVED: Leaving it as its own section for now.

Examples

VkViewport viewports[4];
VkRect2D scissors[4];
VkViewportWScalingNV scalings[4];

for (int i = 0; i < 4; i++) {
    int x = (i & 2) ? 0 : currentWindowWidth / 2;
    int y = (i & 1) ? 0 : currentWindowHeight / 2;

    viewports[i].x = 0;
    viewports[i].y = 0;
    viewports[i].width = currentWindowWidth;
    viewports[i].height = currentWindowHeight;
    viewports[i].minDepth = 0.0f;
    viewports[i].maxDepth = 1.0f;

    scissors[i].offset.x = x;
    scissors[i].offset.y = y;
    scissors[i].extent.width = currentWindowWidth/2;
    scissors[i].extent.height = currentWindowHeight/2;

    const float factor = 0.15;
    scalings[i].xcoeff = ((i & 2) ? -1.0 : 1.0) * factor;
    scalings[i].ycoeff = ((i & 1) ? -1.0 : 1.0) * factor;
}

VkPipelineViewportWScalingStateCreateInfoNV vpWScalingStateInfo = { VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_W_SCALING_STATE_CREATE_INFO_NV };

vpWScalingStateInfo.viewportWScalingEnable = VK_TRUE;
vpWScalingStateInfo.viewportCount = 4;
vpWScalingStateInfo.pViewportWScalings = &scalings[0];

VkPipelineViewportStateCreateInfo vpStateInfo = { VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO };
vpStateInfo.viewportCount = 4;
vpStateInfo.pViewports = &viewports[0];
vpStateInfo.scissorCount = 4;
vpStateInfo.pScissors = &scissors[0];
vpStateInfo.pNext = &vpWScalingStateInfo;

Example shader to read from a w-scaled texture:

// Vertex Shader
// Draw a triangle that covers the whole screen
const vec4 positions[3] = vec4[3](vec4(-1, -1, 0, 1),
                                  vec4( 3, -1, 0, 1),
                                  vec4(-1,  3, 0, 1));
out vec2 uv;
void main()
{
    vec4 pos = positions[ gl_VertexID ];
    gl_Position = pos;
    uv = pos.xy;
}

// Fragment Shader
uniform sampler2D tex;
uniform float xcoeff;
uniform float ycoeff;
out vec4 Color;
in vec2 uv;

void main()
{
    // Handle uv as if upper right quadrant
    vec2 uvabs = abs(uv);

    // unscale: transform w-scaled image into an unscaled image
    //   scale: transform unscaled image int a w-scaled image
    float unscale = 1.0 / (1 + xcoeff * uvabs.x + xcoeff * uvabs.y);
    //float scale = 1.0 / (1 - xcoeff * uvabs.x - xcoeff * uvabs.y);

    vec2 P = vec2(unscale * uvabs.x, unscale * uvabs.y);

    // Go back to the right quadrant
    P *= sign(uv);

    Color = texture(tex, P * 0.5 + 0.5);
}

Version History

  • Revision 1, 2017-02-15 (Eric Werness)

    • Internal revisions

VK_NV_dedicated_allocation

Name String

VK_NV_dedicated_allocation

Extension Type

Device extension

Registered Extension Number

27

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2016-05-31

IP Status

No known IP claims.

Contributors
  • Jeff Bolz, NVIDIA

This extension allows device memory to be allocated for a particular buffer or image resource, which on some devices can significantly improve the performance of that resource. Normal device memory allocations must support memory aliasing and sparse binding, which could interfere with optimizations like framebuffer compression or efficient page table usage. This is important for render targets and very large resources, but need not (and probably should not) be used for smaller resources that can benefit from suballocation.

This extension adds a few small structures to resource creation and memory allocation: a new structure that flags whether am image/buffer will have a dedicated allocation, and a structure indicating the image or buffer that an allocation will be bound to.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_IMAGE_CREATE_INFO_NV

    • VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_BUFFER_CREATE_INFO_NV

    • VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_MEMORY_ALLOCATE_INFO_NV

New Enums

None.

New Functions

None.

Issues

None.

Examples

    // Create an image with
    // VkDedicatedAllocationImageCreateInfoNV::dedicatedAllocation
    // set to VK_TRUE

    VkDedicatedAllocationImageCreateInfoNV dedicatedImageInfo =
    {
        VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_IMAGE_CREATE_INFO_NV,            // sType
        NULL,                                                                   // pNext
        VK_TRUE,                                                                // dedicatedAllocation
    };

    VkImageCreateInfo imageCreateInfo =
    {
        VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO,    // sType
        &dedicatedImageInfo                     // pNext
        // Other members set as usual
    };

    VkImage image;
    VkResult result = vkCreateImage(
        device,
        &imageCreateInfo,
        NULL,                       // pAllocator
        &image);

    VkMemoryRequirements memoryRequirements;
    vkGetImageMemoryRequirements(
        device,
        image,
        &memoryRequirements);

    // Allocate memory with VkDedicatedAllocationMemoryAllocateInfoNV::image
    // pointing to the image we are allocating the memory for

    VkDedicatedAllocationMemoryAllocateInfoNV dedicatedInfo =
    {
        VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_MEMORY_ALLOCATE_INFO_NV,             // sType
        NULL,                                                                       // pNext
        image,                                                                      // image
        VK_NULL_HANDLE,                                                             // buffer
    };

    VkMemoryAllocateInfo memoryAllocateInfo =
    {
        VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO,                 // sType
        &dedicatedInfo,                                         // pNext
        memoryRequirements.size,                                // allocationSize
        FindMemoryTypeIndex(memoryRequirements.memoryTypeBits), // memoryTypeIndex
    };

    VkDeviceMemory memory;
    vkAllocateMemory(
        device,
        &memoryAllocateInfo,
        NULL,                       // pAllocator
        &memory);

    // Bind the image to the memory

    vkBindImageMemory(
        device,
        image,
        memory,
        0);

Version History

  • Revision 1, 2016-05-31 (Jeff Bolz)

    • Internal revisions

VK_NV_external_memory_capabilities

Name String

VK_NV_external_memory_capabilities

Extension Type

Instance extension

Registered Extension Number

56

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • James Jones @cubanismo

Last Modified Date

2016-08-19

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • James Jones, NVIDIA

Applications may wish to import memory from the Direct 3D API, or export memory to other Vulkan instances. This extension provides a set of capability queries that allow applications determine what types of win32 memory handles an implementation supports for a given set of use cases.

New Object Types

None.

New Enum Constants

None.

Issues

1) Why do so many external memory capabilities need to be queried on a per-memory-handle-type basis?

RESOLVED: This is because some handle types are based on OS-native objects that have far more limited capabilities than the very generic Vulkan memory objects. Not all memory handle types can name memory objects that support 3D images, for example. Some handle types cannot even support the deferred image and memory binding behavior of Vulkan and require specifying the image when allocating or importing the memory object.

2) Does the VkExternalImageFormatPropertiesNV struct need to include a list of memory type bits that support the given handle type?

RESOLVED: No. The memory types that do not support the handle types will simply be filtered out of the results returned by vkGetImageMemoryRequirements when a set of handle types was specified at image creation time.

3) Should the non-opaque handle types be moved to their own extension?

RESOLVED: Perhaps. However, defining the handle type bits does very little and does not require any platform-specific types on its own, and it is easier to maintain the bitmask values in a single extension for now. Presumably more handle types could be added by separate extensions though, and it would be midly weird to have some platform-specific ones defined in the core spec and some in extensions

VK_NV_external_memory

Name String

VK_NV_external_memory

Extension Type

Device extension

Registered Extension Number

57

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-08-19

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Carsten Rohde, NVIDIA

Applications may wish to export memory to other Vulkan instances or other APIs, or import memory from other Vulkan instances or other APIs to enable Vulkan workloads to be split up across application module, process, or API boundaries. This extension enables applications to create exportable Vulkan memory objects such that the underlying resources can be referenced outside the Vulkan instance that created them.

New Object Types

None.

New Enum Constants

Extending VkStructureType:

  • VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_NV

  • VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_NV

New Enums

None.

New Functions

None.

Issues

1) If memory objects are shared between processes and APIs, is this considered aliasing according to the rules outlined in the Memory Aliasing section?

RESOLVED: Yes, but strict exceptions to the rules are added to allow some forms of aliasing in these cases. Further, other extensions may build upon these new aliasing rules to define specific support usage within Vulkan for imported native memory objects, or memory objects from other APIs.

2) Are new image layouts or metadata required to specify image layouts and layout transitions compatible with non-Vulkan APIs, or with other instances of the same Vulkan driver?

RESOLVED: No. Separate instances of the same Vulkan driver running on the same GPU should have identical internal layout semantics, so applictions have the tools they need to ensure views of images are consistent between the two instances. Other APIs will fall into two categories: Those that are Vulkan compatible (a term to be defined by subsequent interopability extensions), or Vulkan incompatible. When sharing images with Vulkan incompatible APIs, the Vulkan image must be transitioned to the VK_IMAGE_LAYOUT_GENERAL layout before handing it off to the external API.

Note this does not attempt to address cross-device transitions, nor transitions to engines on the same device which are not visible within the Vulkan API. Both of these are beyond the scope of this extension.

Examples

    // TODO: Write some sample code here.

Version History

  • Revision 1, 2016-08-19 (James Jones)

    • Initial draft

VK_NV_external_memory_win32

Name String

VK_NV_external_memory_win32

Extension Type

Device extension

Registered Extension Number

58

Revision

1

Extension and Version Dependencies
Contact
  • James Jones @cubanismo

Last Modified Date

2016-08-19

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Carsten Rohde, NVIDIA

Applications may wish to export memory to other Vulkan instances or other APIs, or import memory from other Vulkan instances or other APIs to enable Vulkan workloads to be split up across application module, process, or API boundaries. This extension enables win32 applications to export win32 handles from Vulkan memory objects such that the underlying resources can be referenced outside the Vulkan instance that created them, and import win32 handles created in the Direct3D API to Vulkan memory objects.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_NV

    • VK_STRUCTURE_TYPE_EXPORT_MEMORY_WIN32_HANDLE_INFO_NV

New Enums

None.

New Functions

Issues

1) If memory objects are shared between processes and APIs, is this considered aliasing according to the rules outlined in the Memory Aliasing section?

RESOLVED: Yes, but strict exceptions to the rules are added to allow some forms of aliasing in these cases. Further, other extensions may build upon these new aliasing rules to define specific support usage within Vulkan for imported native memory objects, or memory objects from other APIs.

2) Are new image layouts or metadata required to specify image layouts and layout transitions compatible with non-Vulkan APIs, or with other instances of the same Vulkan driver?

RESOLVED: No. Separate instances of the same Vulkan driver running on the same GPU should have identical internal layout semantics, so applictions have the tools they need to ensure views of images are consistent between the two instances. Other APIs will fall into two categories: Those that are Vulkan compatible (a term to be defined by subsequent interopability extensions), or Vulkan incompatible. When sharing images with Vulkan incompatible APIs, the Vulkan image must be transitioned to the VK_IMAGE_LAYOUT_GENERAL layout before handing it off to the external API.

Note this does not attempt to address cross-device transitions, nor transitions to engines on the same device which are not visible within the Vulkan API. Both of these are beyond the scope of this extension.

3) Do applications need to call CloseHandle() on the values returned from vkGetMemoryWin32HandleNV when handleType is VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV?

RESOLVED: Yes, unless it is passed back in to another driver instance to import the object. A successful get call transfers ownership of the handle to the application, while an import transfers ownership to the associated driver. Destroying the memory object will not destroy the handle or the handle’s reference to the underlying memory resource.

Examples

    //
    // Create an exportable memory object and export an external
    // handle from it.
    //

    // Pick an external format and handle type.
    static const VkFormat format = VK_FORMAT_R8G8B8A8_UNORM;
    static const VkExternalMemoryHandleTypeFlagsNV handleType =
        VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV;

    extern VkPhysicalDevice physicalDevice;
    extern VkDevice device;

    VkPhysicalDeviceMemoryProperties memoryProperties;
    VkExternalImageFormatPropertiesNV properties;
    VkExternalMemoryImageCreateInfoNV externalMemoryImageCreateInfo;
    VkDedicatedAllocationImageCreateInfoNV dedicatedImageCreateInfo;
    VkImageCreateInfo imageCreateInfo;
    VkImage image;
    VkMemoryRequirements imageMemoryRequirements;
    uint32_t numMemoryTypes;
    uint32_t memoryType;
    VkExportMemoryAllocateInfoNV exportMemoryAllocateInfo;
    VkDedicatedAllocationMemoryAllocateInfoNV dedicatedAllocationInfo;
    VkMemoryAllocateInfo memoryAllocateInfo;
    VkDeviceMemory memory;
    VkResult result;
    HANDLE memoryHnd;

    // Figure out how many memory types the device supports
    vkGetPhysicalDeviceMemoryProperties(physicalDevice,
                                        &memoryProperties);
    numMemoryTypes = memoryProperties.memoryTypeCount;

    // Check the external handle type capabilities for the chosen format
    // Exportable 2D image support with at least 1 mip level, 1 array
    // layer, and VK_SAMPLE_COUNT_1_BIT using optimal tiling and supporting
    // texturing and color rendering is required.
    result = vkGetPhysicalDeviceExternalImageFormatPropertiesNV(
        physicalDevice,
        format,
        VK_IMAGE_TYPE_2D,
        VK_IMAGE_TILING_OPTIMAL,
        VK_IMAGE_USAGE_SAMPLED_BIT |
        VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT,
        0,
        handleType,
        &properties);

    if ((result != VK_SUCCESS) ||
        !(properties.externalMemoryFeatures &
          VK_EXTERNAL_MEMORY_FEATURE_EXPORTABLE_BIT_NV)) {
        abort();
    }

    // Set up the external memory image creation info
    memset(&externalMemoryImageCreateInfo,
           0, sizeof(externalMemoryImageCreateInfo));
    externalMemoryImageCreateInfo.sType =
        VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_NV;
    externalMemoryImageCreateInfo.handleTypes = handleType;
    if (properties.externalMemoryFeatures &
        VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_NV) {
        memset(&dedicatedImageCreateInfo, 0, sizeof(dedicatedImageCreateInfo));
        dedicatedImageCreateInfo.sType =
            VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_IMAGE_CREATE_INFO_NV;
        dedicatedImageCreateInfo.dedicatedAllocation = VK_TRUE;
        externalMemoryImageCreateInfo.pNext = &dedicatedImageCreateInfo;
    }
    // Set up the  core image creation info
    memset(&imageCreateInfo, 0, sizeof(imageCreateInfo));
    imageCreateInfo.sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO;
    imageCreateInfo.pNext = &externalMemoryImageCreateInfo;
    imageCreateInfo.format = format;
    imageCreateInfo.extent.width = 64;
    imageCreateInfo.extent.height = 64;
    imageCreateInfo.extent.depth = 1;
    imageCreateInfo.mipLevels = 1;
    imageCreateInfo.arrayLayers = 1;
    imageCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
    imageCreateInfo.tiling = VK_IMAGE_TILING_OPTIMAL;
    imageCreateInfo.usage = VK_IMAGE_USAGE_SAMPLED_BIT |
        VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT;
    imageCreateInfo.sharingMode = VK_SHARING_MODE_EXCLUSIVE;
    imageCreateInfo.initialLayout = VK_IMAGE_LAYOUT_UNDEFINED;

    vkCreateImage(device, &imageCreateInfo, NULL, &image);

    vkGetImageMemoryRequirements(device,
                                 image,
                                 &imageMemoryRequirements);

    // For simplicity, just pick the first compatible memory type.
    for (memoryType = 0; memoryType < numMemoryTypes; memoryType++) {
        if ((1 << memoryType) & imageMemoryRequirements.memoryTypeBits) {
            break;
        }
    }

    // At least one memory type must be supported given the prior external
    // handle capability check.
    assert(memoryType < numMemoryTypes);

    // Allocate the external memory object.
    memset(&exportMemoryAllocateInfo, 0, sizeof(exportMemoryAllocateInfo));
    exportMemoryAllocateInfo.sType =
        VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_NV;
    exportMemoryAllocateInfo.handleTypes = handleType;
    if (properties.externalMemoryFeatures &
        VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_NV) {
        memset(&dedicatedAllocationInfo, 0, sizeof(dedicatedAllocationInfo));
        dedicatedAllocationInfo.sType =
            VK_STRUCTURE_TYPE_DEDICATED_ALLOCATION_MEMORY_ALLOCATE_INFO_NV;
        dedicatedAllocationInfo.image = image;
        exportMemoryAllocateInfo.pNext = &dedicatedAllocationInfo;
    }
    memset(&memoryAllocateInfo, 0, sizeof(memoryAllocateInfo));
    memoryAllocateInfo.sType = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO;
    memoryAllocateInfo.pNext = &exportMemoryAllocateInfo;
    memoryAllocateInfo.allocationSize = imageMemoryRequirements.size;
    memoryAllocateInfo.memoryTypeIndex = memoryType;

    vkAllocateMemory(device, &memoryAllocateInfo, NULL, &memory);

    if (!(properties.externalMemoryFeatures &
          VK_EXTERNAL_MEMORY_FEATURE_DEDICATED_ONLY_BIT_NV)) {
        vkBindImageMemory(device, image, memory, 0);
    }

    // Get the external memory opaque FD handle
    vkGetMemoryWin32HandleNV(device, memory, &memoryHnd);

Version History

  • Revision 1, 2016-08-11 (James Jones)

    • Initial draft

VK_NV_fill_rectangle

Name String

VK_NV_fill_rectangle

Extension Type

Device extension

Registered Extension Number

154

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-05-22

Contributors
  • Jeff Bolz, NVIDIA

This extension adds a new VkPolygonMode enum where a triangle is rasterized by computing and filling its axis-aligned screen-space bounding box, disregarding the actual triangle edges. This can be useful for drawing a rectangle without being split into two triangles with an internal edge. It is also useful to minimize the number of primitives that need to be drawn, particularly for a user interface.

New Object Types

None.

New Enum Constants

New Enums

None.

New Structures

None.

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-05-22 (Jeff Bolz)

    • Internal revisions

VK_NV_fragment_coverage_to_color

Name String

VK_NV_fragment_coverage_to_color

Extension Type

Device extension

Registered Extension Number

150

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-05-21

Contributors
  • Jeff Bolz, NVIDIA

This extension allows the fragment coverage value, represented as an integer bitmask, to be substituted for a color output being written to a single-component color attachment with integer components (e.g. VK_FORMAT_R8_UINT). The functionality provided by this extension is different from simply writing the SampleMask fragment shader output, in that the coverage value written to the framebuffer is taken after stencil test and depth test, as well as after fragment operations such as alpha-to-coverage.

This functionality may be useful for deferred rendering algorithms, where the second pass needs to know which samples belong to which original fragments.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PIPELINE_COVERAGE_TO_COLOR_STATE_CREATE_INFO_NV

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-05-21 (Jeff Bolz)

    • Internal revisions

VK_NV_framebuffer_mixed_samples

Name String

VK_NV_framebuffer_mixed_samples

Extension Type

Device extension

Registered Extension Number

153

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-06-04

Contributors
  • Jeff Bolz, NVIDIA

This extension allows multisample rendering with a raster and depth/stencil sample count that is larger than the color sample count. Rasterization and the results of the depth and stencil tests together determine the portion of a pixel that is “covered”. It can be useful to evaluate coverage at a higher frequency than color samples are stored. This coverage is then “reduced” to a collection of covered color samples, each having an opacity value corresponding to the fraction of the color sample covered. The opacity can optionally be blended into individual color samples.

Rendering with fewer color samples than depth/stencil samples greatly reduces the amount of memory and bandwidth consumed by the color buffer. However, converting the coverage values into opacity introduces artifacts where triangles share edges and may not be suitable for normal triangle mesh rendering.

One expected use case for this functionality is Stencil-then-Cover path rendering (similar to the OpenGL GL_NV_path_rendering extension). The stencil step determines the coverage (in the stencil buffer) for an entire path at the higher sample frequency, and then the cover step draws the path into the lower frequency color buffer using the coverage information to antialias path edges. With this two-step process, internal edges are fully covered when antialiasing is applied and there is no corruption on these edges.

The key features of this extension are:

  • It allows render pass and framebuffer objects to be created where the number of samples in the depth/stencil attachment in a subpass is a multiple of the number of samples in the color attachments in the subpass.

  • A coverage reduction step is added to Fragment Operations which converts a set of covered raster/depth/stencil samples to a set of color samples that perform blending and color writes. The coverage reduction step also includes an optional coverage modulation step, multiplying color values by a fractional opacity corresponding to the number of associated raster/depth/stencil samples covered.

New Object Types

None.

New Enum Constants

None.

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2017-06-04 (Jeff Bolz)

    • Internal revisions

VK_NV_geometry_shader_passthrough

Name String

VK_NV_geometry_shader_passthrough

Extension Type

Device extension

Registered Extension Number

96

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2017-02-15

Interactions and External Dependencies
Contributors
  • Piers Daniell, NVIDIA

  • Jeff Bolz, NVIDIA

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_NV_geometry_shader_passthrough

Geometry shaders provide the ability for applications to process each primitive sent through the graphics pipeline using a programmable shader. However, one common use case treats them largely as a “passthrough”. In this use case, the bulk of the geometry shader code simply copies inputs from each vertex of the input primitive to corresponding outputs in the vertices of the output primitive. Such shaders might also compute values for additional built-in or user-defined per-primitive attributes (e.g., Layer) to be assigned to all the vertices of the output primitive.

This extension provides access to the PassthroughNV decoration under the GeometryShaderPassthroughNV capability. Adding this to a geometry shader input variable specifies that the values of this input are copied to the corresponding vertex of the output primitive.

When using GLSL source-based shading languages, the passthrough layout qualifier from GL_NV_geometry_shader_passthrough maps to the PassthroughNV decoration. To use the passthrough layout, in GLSL the GL_NV_geometry_shader_passthrough extension must be enabled. Behaviour is described in the GL_NV_geometry_shader_passthrough extension specification.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New Built-In Variables

None.

New Variable Decoration

New SPIR-V Capabilities

Issues

1) Should we require or allow a passthrough geometry shader to specify the output layout qualifiers for the output primitive type and maximum vertex count in the SPIR-V?

RESOLVED: Yes they should be required in the SPIR-V. Per GL_NV_geometry_shader_passthrough they are not permitted in the GLSL source shader, but SPIR-V is lower-level. It is straightforward for the GLSL compiler to infer them from the input primitive type and to explicitly emit them in the SPIR-V according to the following table.

Input Layout Implied Output Layout

points

layout(points, max_vertices=1)

lines

layout(line_strip, max_vertices=2)

triangles

layout(triangle_strip, max_vertices=3)

2) How does interface matching work with passthrough geometry shaders?

RESOLVED: This is described in Passthrough Interface Matching. In GL when using passthough geometry shaders in separable mode, all inputs must also be explicitly assigned location layout qualifiers. In Vulkan all SPIR-V shader inputs (except built-ins) must also have location decorations specified. Redeclarations of built-in varables that add the passthrough layout qualifier are exempted from the rule requiring location assignment because built-in variables do not have locations and are matched by BuiltIn decoration.

Sample Code

Consider the following simple geometry shader in unextended GLSL:

layout(triangles) in;
layout(triangle_strip) out;
layout(max_vertices=3) out;

in Inputs {
    vec2 texcoord;
    vec4 baseColor;
} v_in[];
out Outputs {
    vec2 texcoord;
    vec4 baseColor;
};

void main()
{
    int layer = compute_layer();
    for (int i = 0; i < 3; i++) {
        gl_Position = gl_in[i].gl_Position;
        texcoord = v_in[i].texcoord;
        baseColor = v_in[i].baseColor;
        gl_Layer = layer;
        EmitVertex();
    }
}

In this shader, the inputs gl_Position, Inputs.texcoord, and Inputs.baseColor are simply copied from the input vertex to the corresponding output vertex. The only “interesting” work done by the geometry shader is computing and emitting a gl_Layer value for the primitive.

The following geometry shader, using this extension, is equivalent:

#extension GL_NV_geometry_shader_passthrough : require

layout(triangles) in;
// No output primitive layout qualifiers required.

// Redeclare gl_PerVertex to pass through "gl_Position".
layout(passthrough) in gl_PerVertex {
    vec4 gl_Position;
} gl_in[];

// Declare "Inputs" with "passthrough" to automatically copy members.
layout(passthrough) in Inputs {
    vec2 texcoord;
    vec4 baseColor;
} v_in[];

// No output block declaration required.

void main()
{
    // The shader simply computes and writes gl_Layer.  We don't
    // loop over three vertices or call EmitVertex().
    gl_Layer = compute_layer();
}

Version History

  • Revision 1, 2017-02-15 (Daniel Koch)

    • Internal revisions

VK_NV_glsl_shader

Name String

VK_NV_glsl_shader

Extension Type

Device extension

Registered Extension Number

13

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2016-02-14

IP Status

No known IP claims.

Contributors
  • Piers Daniell, NVIDIA

This extension allows GLSL shaders written to the GL_KHR_vulkan_glsl extension specification to be used instead of SPIR-V. The implementation will automatically detect whether the shader is SPIR-V or GLSL, and compile it appropriately.

New Object Types

New Enum Constants

  • Extending VkResult:

    • VK_ERROR_INVALID_SHADER_NV

New Enums

New Structures

New Functions

Issues

Examples

Example 1

Passing in GLSL code

    char const vss[] =
        "#version 450 core\n"
        "layout(location = 0) in vec2 aVertex;\n"
        "layout(location = 1) in vec4 aColor;\n"
        "out vec4 vColor;\n"
        "void main()\n"
        "{\n"
        "    vColor = aColor;\n"
        "    gl_Position = vec4(aVertex, 0, 1);\n"
        "}\n"
    ;
    VkShaderModuleCreateInfo vertexShaderInfo = { VK_STRUCTURE_TYPE_SHADER_MODULE_CREATE_INFO };
    vertexShaderInfo.codeSize = sizeof vss;
    vertexShaderInfo.pCode = vss;
    VkShaderModule vertexShader;
    vkCreateShaderModule(device, &vertexShaderInfo, 0, &vertexShader);

Version History

  • Revision 1, 2016-02-14 (Piers Daniell)

    • Initial draft

VK_NV_sample_mask_override_coverage

Name String

VK_NV_sample_mask_override_coverage

Extension Type

Device extension

Registered Extension Number

95

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2016-12-08

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Daniel Koch, NVIDIA

  • Jeff Bolz, NVIDIA

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_NV_sample_mask_override_coverage

The extension provides access to the OverrideCoverageNV decoration under the SampleMaskOverrideCoverageNV capability. Adding this decoration to a variable with the SampleMask builtin decoration allows the shader to modify the coverage mask and affect which samples are used to process the fragment.

When using GLSL source-based shader languages, the override_coverage layout qualifier from GL_NV_sample_mask_override_coverage maps to the OverrideCoverageNV decoration. To use the override_coverage layout qualifier in GLSL the GL_NV_sample_mask_override_coverage extension must be enabled. Behavior is described in the GL_NV_sample_mask_override_coverage extension spec.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New Built-In Variables

None.

New Variable Decoration

New SPIR-V Capabilities

Issues

None.

Version History

  • Revision 1, 2016-12-08 (Piers Daniell)

    • Internal revisions

VK_NV_shader_subgroup_partitioned

Name String

VK_NV_shader_subgroup_partitioned

Extension Type

Device extension

Registered Extension Number

199

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.1

Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2018-03-17

Contributors
  • Jeff Bolz, NVIDIA

This extension enables support for a new class of subgroup operations via the GL_NV_shader_subgroup_partitioned GLSL extension and SPV_NV_shader_subgroup_partitioned SPIR-V extension. Support for these new operations is advertised via the VK_SUBGROUP_FEATURE_PARTITIONED_BIT_NV bit.

This extension requires Vulkan 1.1, for general subgroup support.

New Object Types

None.

New Enum Constants

New Enums

None.

New Structures

None.

New Functions

None.

Issues

None.

Version History

  • Revision 1, 2018-03-17 (Jeff Bolz)

    • Internal revisions

VK_NV_viewport_array2

Name String

VK_NV_viewport_array2

Extension Type

Device extension

Registered Extension Number

97

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Daniel Koch @dgkoch

Last Modified Date

2017-02-15

Interactions and External Dependencies
  • This extension requires the SPV_NV_viewport_array2 SPIR-V extension.

  • This extension requires the GL_NV_viewport_array2 extension for GLSL source languages.

  • This extension requires the geometryShader and multiViewport features.

  • This extension interacts with the tessellationShader feature.

Contributors
  • Piers Daniell, NVIDIA

  • Jeff Bolz, NVIDIA

This extension adds support for the following SPIR-V extension in Vulkan:

  • SPV_NV_viewport_array2

which allows a single primitive to be broadcast to multiple viewports and/or multiple layers. A new shader built-in output ViewportMaskNV is provided, which allows a single primitive to be output to multiple viewports simultaneously. Also, a new SPIR-V decoration is added to control whether the effective viewport index is added into the variable decorated with the Layer built-in decoration. These capabilities allow a single primitive to be output to multiple layers simultaneously.

This extension allows variables decorated with the Layer and ViewportIndex built-ins to be exported from vertex or tessellation shaders, using the ShaderViewportIndexLayerNV capability.

This extension adds a new ViewportMaskNV built-in decoration that is available for output variables in vertex, tessellation evaluation, and geometry shaders, and a new ViewportRelativeNV decoration that can be added on variables decorated with Layer when using the ShaderViewportMaskNV capability.

When using GLSL source-based shading languages, the gl_ViewportMask[] built-in output variable and viewport_relative layout qualifier from GL_NV_viewport_array2 map to the ViewportMaskNV and ViewportRelativeNV decorations, respectively. Behaviour is described in the GL_NV_viewport_array2 extension specificiation.

Note

The ShaderViewportIndexLayerNV capability is equivalent to the ShaderViewportIndexLayerEXT capability added by VK_EXT_shader_viewport_index_layer.

New Object Types

None.

New Enum Constants

None.

New Enums

None.

New Structures

None.

New Functions

None.

New or Modified Built-In Variables

New Variable Decoration

Issues

None yet!

Version History

  • Revision 1, 2017-02-15 (Daniel Koch)

    • Internal revisions

VK_NV_viewport_swizzle

Name String

VK_NV_viewport_swizzle

Extension Type

Device extension

Registered Extension Number

99

Revision

1

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Piers Daniell @pdaniell-nv

Last Modified Date

2016-12-22

Interactions and External Dependencies
  • This extension requires multiViewport and geometryShader features to be useful.

Contributors
  • Daniel Koch, NVIDIA

  • Jeff Bolz, NVIDIA

This extension provides a new per-viewport swizzle that can modify the position of primitives sent to each viewport. New viewport swizzle state is added for each viewport, and a new position vector is computed for each vertex by selecting from and optionally negating any of the four components of the original position vector.

This new viewport swizzle is useful for a number of algorithms, including single-pass cubemap rendering (broadcasting a primitive to multiple faces and reorienting the vertex position for each face) and voxel rasterization. The per-viewport component remapping and negation provided by the swizzle allows application code to re-orient three-dimensional geometry with a view along any of the X, Y, or Z axes. If a perspective projection and depth buffering is required, 1/W buffering should be used, as described in the single-pass cubemap rendering example in the “Issues” section below.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_SWIZZLE_STATE_CREATE_INFO_NV

New Functions

None.

Issues

1) Where does viewport swizzling occur in the pipeline?

RESOLVED: Despite being associated with the viewport, viewport swizzling must happen prior to the viewport transform. In particular, it needs to be performed before clipping and perspective division.

The viewport mask expansion (VK_NV_viewport_array2) and the viewport swizzle could potentially be performed before or after transform feedback, but feeding back several viewports worth of primitives with different swizzles doesn’t seem particularly useful. This specification applies the viewport mask and swizzle after transform feedback, and makes primitive queries only count each primitive once.

2) Any interesting examples of how this extension, VK_NV_viewport_array2, and VK_NV_geometry_shader_passthrough can be used together in practice?

RESOLVED: One interesting use case for this extension is for single-pass rendering to a cubemap. In this example, the application would attach a cubemap texture to a layered FBO where the six cube faces are treated as layers. Vertices are sent through the vertex shader without applying a projection matrix, where the gl_Position output is (x,y,z,1) and the center of the cubemap is at (0,0,0). With unextended Vulkan, one could have a conventional instanced geometry shader that looks something like the following:

layout(invocations = 6) in;     // separate invocation per face
layout(triangles) in;
layout(triangle_strip) out;
layout(max_vertices = 3) out;

in Inputs {
vec2 texcoord;
vec3 normal;
vec4 baseColor;
} v[];

    out Outputs {
    vec2 texcoord;
    vec3 normal;
    vec4 baseColor;
    };

    void main()
    {
    int face = gl_InvocationID;  // which face am I?

    // Project gl_Position for each vertex onto the cube map face.
    vec4 positions[3];
    for (int i = 0; i < 3; i++) {
        positions[i] = rotate(gl_in[i].gl_Position, face);
    }

    // If the primitive doesn't project onto this face, we're done.
    if (shouldCull(positions)) {
        return;
    }

    // Otherwise, emit a copy of the input primitive to the
    // appropriate face (using gl_Layer).
    for (int i = 0; i < 3; i++) {
        gl_Layer = face;
        gl_Position = positions[i];
        texcoord = v[i].texcoord;
        normal = v[i].normal;
        baseColor = v[i].baseColor;
        EmitVertex();
    }
}

With passthrough geometry shaders, this can be done using a much simpler shader:

layout(triangles) in;
layout(passthrough) in Inputs {
    vec2 texcoord;
    vec3 normal;
    vec4 baseColor;
}
layout(passthrough) in gl_PerVertex {
    vec4 gl_Position;
} gl_in[];
layout(viewport_relative) out int gl_Layer;

void main()
{
    // Figure out which faces the primitive projects onto and
    // generate a corresponding viewport mask.
    uint mask = 0;
    for (int i = 0; i < 6; i++) {
        if (!shouldCull(face)) {
        mask |= 1U << i;
        }
    }
    gl_ViewportMask = mask;
    gl_Layer = 0;
}

The application code is set up so that each of the six cube faces has a separate viewport (numbered 0 to 5). Each face also has a separate swizzle, programmed via the VkPipelineViewportSwizzleStateCreateInfoNV pipeline state. The viewport swizzle feature performs the coordinate transformation handled by the rotate() function in the original shader. The viewport_relative layout qualifier says that the viewport number (0 to 5) is added to the base gl_Layer value of 0 to determine which layer (cube face) the primitive should be sent to.

Note that the use of the passed through input normal in this example suggests that the fragment shader in this example would perform an operation like per-fragment lighting. The viewport swizzle would transform the position to be face-relative, but normal would remain in the original coordinate system. It seems likely that the fragment shader in either version of the example would want to perform lighting in the original coordinate system. It would likely do this by reconstructing the position of the fragment in the original coordinate system using gl_FragCoord, a constant or uniform holding the size of the cube face, and the input gl_ViewportIndex (or gl_Layer), which identifies the cube face. Since the value of normal is in the original coordinate system, it would not need to be modified as part of this coordinate transformation.

Note that while the rotate() operation in the regular geometry shader above could include an arbitrary post-rotation projection matrix, the viewport swizzle does not support arbitrary math. To get proper projection, 1/W buffering should be used. To do this:

1. Program the viewport swizzles to move the pre-projection W eye coordinate (typically 1.0) into the Z coordinate of the swizzle output and the eye coordinate component used for depth into the W coordinate. For example, the viewport corresponding to the +Z face might use a swizzle of (+X, -Y, +W, +Z). The Z normalized device coordinate computed after swizzling would then be z'/w' = 1/Zeye.

2. On NVIDIA implementations supporting floating-point depth buffers with values outside [0,1], prevent unwanted near plane clipping by enabling depthClampEnable. Ensure that the depth clamp doesn’t mess up depth testing by programming the depth range to very large values, such as minDepthBounds=-z, maxDepthBounds=+z, where z = 2127. It should be possible to use IEEE infinity encodings also (0xFF800000 for -INF, 0x7F800000 for +INF). Even when near/far clipping is disabled, primitives extending behind the eye will still be clipped because one or more vertices will have a negative W coordinate and fail X/Y clipping tests.

On other implementations, scale X, Y, and Z eye coordinates so that vertices on the near plane have a post-swizzle W coordinate of 1.0. For example, if the near plane is at Zeye = 1/256, scale X, Y, and Z by 256.

3. Adjust depth testing to reflect the fact that 1/W values are large near the eye and small away from the eye. Clear the depth buffer to zero (infinitely far away) and use a depth test of VK_COMPARE_OP_GREATER instead of VK_COMPARE_OP_LESS.

Version History

  • Revision 1, 2016-12-22 (Piers Daniell)

    • Internal revisions

VK_NV_win32_keyed_mutex

Name String

VK_NV_win32_keyed_mutex

Extension Type

Device extension

Registered Extension Number

59

Revision

1

Extension and Version Dependencies
Contact
  • Carsten Rohde @crohde

Last Modified Date

2016-08-19

IP Status

No known IP claims.

Contributors
  • James Jones, NVIDIA

  • Carsten Rohde, NVIDIA

Applications that wish to import Direct3D 11 memory objects into the Vulkan API may wish to use the native keyed mutex mechanism to synchronize access to the memory between Vulkan and Direct3D. This extension provides a way for an application to access the keyed mutex associated with an imported Vulkan memory object when submitting command buffers to a queue.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType:

    • VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_NV

New Enums

None.

New Functions

None.

Issues

None.

Examples

    //
    // Import a memory object from Direct3D 11, and synchronize
    // access to it in Vulkan using keyed mutex objects.
    //

    extern VkPhysicalDevice physicalDevice;
    extern VkDevice device;
    extern HANDLE sharedNtHandle;

    static const VkFormat format = VK_FORMAT_R8G8B8A8_UNORM;
    static const VkExternalMemoryHandleTypeFlagsNV handleType =
        VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_IMAGE_BIT_NV;

    VkPhysicalDeviceMemoryProperties memoryProperties;
    VkExternalImageFormatPropertiesNV properties;
    VkExternalMemoryImageCreateInfoNV externalMemoryImageCreateInfo;
    VkImageCreateInfo imageCreateInfo;
    VkImage image;
    VkMemoryRequirements imageMemoryRequirements;
    uint32_t numMemoryTypes;
    uint32_t memoryType;
    VkImportMemoryWin32HandleInfoNV importMemoryInfo;
    VkMemoryAllocateInfo memoryAllocateInfo;
    VkDeviceMemory mem;
    VkResult result;

    // Figure out how many memory types the device supports
    vkGetPhysicalDeviceMemoryProperties(physicalDevice,
                                        &memoryProperties);
    numMemoryTypes = memoryProperties.memoryTypeCount;

    // Check the external handle type capabilities for the chosen format
    // Importable 2D image support with at least 1 mip level, 1 array
    // layer, and VK_SAMPLE_COUNT_1_BIT using optimal tiling and supporting
    // texturing and color rendering is required.
    result = vkGetPhysicalDeviceExternalImageFormatPropertiesNV(
        physicalDevice,
        format,
        VK_IMAGE_TYPE_2D,
        VK_IMAGE_TILING_OPTIMAL,
        VK_IMAGE_USAGE_SAMPLED_BIT |
        VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT,
        0,
        handleType,
        &properties);

    if ((result != VK_SUCCESS) ||
        !(properties.externalMemoryFeatures &
          VK_EXTERNAL_MEMORY_FEATURE_IMPORTABLE_BIT_NV)) {
        abort();
    }

    // Set up the external memory image creation info
    memset(&externalMemoryImageCreateInfo,
           0, sizeof(externalMemoryImageCreateInfo));
    externalMemoryImageCreateInfo.sType =
        VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_NV;
    externalMemoryImageCreateInfo.handleTypes = handleType;
    // Set up the  core image creation info
    memset(&imageCreateInfo, 0, sizeof(imageCreateInfo));
    imageCreateInfo.sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO;
    imageCreateInfo.pNext = &externalMemoryImageCreateInfo;
    imageCreateInfo.format = format;
    imageCreateInfo.extent.width = 64;
    imageCreateInfo.extent.height = 64;
    imageCreateInfo.extent.depth = 1;
    imageCreateInfo.mipLevels = 1;
    imageCreateInfo.arrayLayers = 1;
    imageCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
    imageCreateInfo.tiling = VK_IMAGE_TILING_OPTIMAL;
    imageCreateInfo.usage = VK_IMAGE_USAGE_SAMPLED_BIT |
        VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT;
    imageCreateInfo.sharingMode = VK_SHARING_MODE_EXCLUSIVE;
    imageCreateInfo.initialLayout = VK_IMAGE_LAYOUT_UNDEFINED;

    vkCreateImage(device, &imageCreateInfo, NULL, &image);
    vkGetImageMemoryRequirements(device,
                                 image,
                                 &imageMemoryRequirements);

    // For simplicity, just pick the first compatible memory type.
    for (memoryType = 0; memoryType < numMemoryTypes; memoryType++) {
        if ((1 << memoryType) & imageMemoryRequirements.memoryTypeBits) {
            break;
        }
    }

    // At least one memory type must be supported given the prior external
    // handle capability check.
    assert(memoryType < numMemoryTypes);

    // Allocate the external memory object.
    memset(&exportMemoryAllocateInfo, 0, sizeof(exportMemoryAllocateInfo));
    exportMemoryAllocateInfo.sType =
        VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_NV;
    importMemoryInfo.handleTypes = handleType;
    importMemoryInfo.handle = sharedNtHandle;

    memset(&memoryAllocateInfo, 0, sizeof(memoryAllocateInfo));
    memoryAllocateInfo.sType = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO;
    memoryAllocateInfo.pNext = &exportMemoryAllocateInfo;
    memoryAllocateInfo.allocationSize = imageMemoryRequirements.size;
    memoryAllocateInfo.memoryTypeIndex = memoryType;

    vkAllocateMemory(device, &memoryAllocateInfo, NULL, &mem);

    vkBindImageMemory(device, image, mem, 0);

    ...

    const uint64_t acquireKey = 1;
    const uint32_t timeout = INFINITE;
    const uint64_t releaseKey = 2;

    VkWin32KeyedMutexAcquireReleaseInfoNV keyedMutex =
        { VK_STRUCTURE_TYPE_WIN32_KEYED_MUTEX_ACQUIRE_RELEASE_INFO_NV };
    keyedMutex.acquireCount = 1;
    keyedMutex.pAcquireSyncs = &mem;
    keyedMutex.pAcquireKeys = &acquireKey;
    keyedMutex.pAcquireTimeoutMilliseconds = &timeout;
    keyedMutex.releaseCount = 1;
    keyedMutex.pReleaseSyncs = &mem;
    keyedMutex.pReleaseKeys = &releaseKey;

    VkSubmitInfo submit_info = { VK_STRUCTURE_TYPE_SUBMIT_INFO, &keyedMutex };
    submit_info.commandBufferCount = 1;
    submit_info.pCommandBuffers = &cmd_buf;
    vkQueueSubmit(queue, 1, &submit_info, VK_NULL_HANDLE);

Version History

  • Revision 2, 2016-08-11 (James Jones)

    • Updated sample code based on the NV external memory extensions.

    • Renamed from NVX to NV extension.

    • Added Overview and Description sections.

    • Updated sample code to use the NV external memory extensions.

  • Revision 1, 2016-06-14 (Carsten Rohde)

    • Initial draft.

VK_NVX_device_generated_commands

Name String

VK_NVX_device_generated_commands

Extension Type

Device extension

Registered Extension Number

87

Revision

3

Extension and Version Dependencies
  • Requires Vulkan 1.0

Contact
  • Christoph Kubisch @pixeljetstream

Last Modified Date

2017-07-25

Contributors
  • Pierre Boudier, NVIDIA

  • Christoph Kubisch, NVIDIA

  • Mathias Schott, NVIDIA

  • Jeff Bolz, NVIDIA

  • Eric Werness, NVIDIA

  • Detlef Roettger, NVIDIA

  • Daniel Koch, NVIDIA

  • Chris Hebert, NVIDIA

This extension allows the device to generate a number of critical commands for command buffers.

When rendering a large number of objects, the device can be leveraged to implement a number of critical functions, like updating matrices, or implementing occlusion culling, frustum culling, front to back sorting, etc. Implementing those on the device does not require any special extension, since an application is free to define its own data structure, and just process them using shaders.

However, if the application desires to quickly kick off the rendering of the final stream of objects, then unextended Vulkan forces the application to read back the processed stream and issue graphics command from the host. For very large scenes, the synchronization overhead, and cost to generate the command buffer can become the bottleneck. This extension allows an application to generate a device side stream of state changes and commands, and convert it efficiently into a command buffer without having to read it back on the host.

Furthermore, it allows incremental changes to such command buffers by manipulating only partial sections of a command stream — for example pipeline bindings. Unextended Vulkan requires re-creation of entire command buffers in such scenario, or updates synchronized on the host.

The intended usage for this extension is for the application to:

  • create its objects as in unextended Vulkan

  • create a VkObjectTableNVX, and register the various Vulkan objects that are needed to evaluate the input parameters.

  • create a VkIndirectCommandsLayoutNVX, which lists the VkIndirectCommandsTokenTypeNVX it wants to dynamically change as atomic command sequence. This step likely involves some internal device code compilation, since the intent is for the GPU to generate the command buffer in the pipeline.

  • fill the input buffers with the data for each of the inputs it needs. Each input is an array that will be filled with an index in the object table, instead of using CPU pointers.

  • set up a target secondary command buffer

  • reserve command buffer space via vkCmdReserveSpaceForCommandsNVX in a target command buffer at the position you want the generated commands to be executed.

  • call vkCmdProcessCommandsNVX to create the actual device commands for all sequences based on the array contents into a provided target command buffer.

  • execute the target command buffer like a regular secondary command buffer

For each draw/dispatch, the following can be specified:

  • a different pipeline state object

  • a number of descriptor sets, with dynamic offsets

  • a number of vertex buffer bindings, with an optional dynamic offset

  • a different index buffer, with an optional dynamic offset

Applications should register a small number of objects, and use dynamic offsets whenever possible.

While the GPU can be faster than a CPU to generate the commands, it may not happen asynchronously, therefore the primary use-case is generating “less” total work (occlusion culling, classification to use specialized shaders, etc.).

New Enum Constants

Extending VkStructureType:

  • VK_STRUCTURE_TYPE_OBJECT_TABLE_CREATE_INFO_NVX

  • VK_STRUCTURE_TYPE_INDIRECT_COMMANDS_LAYOUT_CREATE_INFO_NVX

  • VK_STRUCTURE_TYPE_CMD_PROCESS_COMMANDS_INFO_NVX

  • VK_STRUCTURE_TYPE_CMD_RESERVE_SPACE_FOR_COMMANDS_INFO_NVX

  • VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_LIMITS_NVX

  • VK_STRUCTURE_TYPE_DEVICE_GENERATED_COMMANDS_FEATURES_NVX

  • VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

Extending VkAccessFlagBits:

  • VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX

  • VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX

Issues

1) How to name this extension ?

RESOLVED: VK_NVX_device_generated_commands

As usual, one of the hardest issues ;)

Alternatives: VK_gpu_commands, VK_execute_commands, VK_device_commands, VK_device_execute_commands, VK_device_execute, VK_device_created_commands, VK_device_recorded_commands, VK_device_generated_commands

2) Should we use serial tokens or redundant sequence description?

Similarly to VkPipeline, signatures have the most likelihood to be cross-vendor adoptable. They also benefit from being processable in parallel.

3) How to name sequence description

ExecuteCommandSignature is a bit long. Maybe just ExecuteSignature, or actually more following Vulkan nomenclature: VkIndirectCommandsLayoutNVX.

4) Do we want to provide indirectCommands inputs with layout or at indirectCommands time?

Separate layout from data as Vulkan does. Provide full flexibilty for indirectCommands.

5) Should the input be provided as SoA or AoS?

It is desirable for the application to reuse the list of objects and render them with some kind of an override. This can be done by just selecting a different input for a push constant or a descriptor set, if they are defined as independent arrays. If the data was interleaved, this would not be as easily possible.

Allowing input divisors can also reduce the conservative command buffer allocation.

6) How do we know the size of the GPU command buffer generated by vkCmdProcessCommandsNVX ?

maxSequenceCount can give an upper estimate, even if the actual count is sourced from the gpu buffer at (buffer, countOffset). As such maxSequenceCount must always be set correctly.

Developers are encouraged to make well use the VkIndirectCommandsLayoutNVX’s pTokens[].divisor, as they allow less conservative storage costs. Especially pipeline changes on a per-draw basis can be costly memory wise.

7) How to deal with dynamic offsets in DescriptorSets?

Maybe additional token VK_EXECUTE_DESCRIPTOR_SET_OFFSET_COMMAND_NVX that works for a “single dynamic buffer” descriptor set and then use (32 bit tableEntry + 32bit offset)

added dynamicCount field, variable sized input

8) Should we allow updates to the object table, similar to DescriptorSet?

Desired yes, people may change “material” shaders and not want to recreate the entire register table. However the developer must ensure to not overwrite a registered objectIndex while it is still being used.

9) Should we allow dynamic state changes?

Seems a bit excessive for “per-draw” type of scenario, but GPU could partition work itself with viewport/scissor…​

10) How do we allow re-using already “filled” indirectCommands buffers?

just use a VkCommandBuffer for the output, and it can be reused easily.

11) How portable should such re-use be?

Same as secondary command buffer

12) Should sequenceOrdered be part of IndirectCommandsLayout or vkCmdProcessCommandsNVX?

Seems better for IndirectCommandsLayout, as that is when most heavy lifting in terms of internal device code generation is done.

13) Under which conditions is vkCmdProcessCommandsNVX legal?

Options:

a) on the host command buffer like a regular draw call

c) The targetCommandbuffer must be inside the “begin” state already at the moment of being passed. This very likely suggests a new VkCommandBufferUsageFlags VK_COMMAND_BUFFER_USAGE_DEVICE_GENERATED_BIT.

d) The targetCommandbuffer must reserve space via a new function.

used a) and d).

14) What if different pipelines have different DescriptorSetLayouts at a certain set unit that mismatches in token.dynamicCount?

Considered legal, as long as the maximum dynamic count of all used DescriptorSetLayouts is provided.

15) Should we add “strides” to input arrays, so that “Array of Structures” type setups can be supported more easily?

Maybe provide a usage flag for packed tokens stream (all inputs from same buffer, implicit stride).

No, given performance test was worse.

16) Should we allow re-using the target command buffer directly, without need to reset command buffer?

17) Is vkCmdProcessCommandsNVX copying the input data or referencing it ?

There are multiple implementations possible:

  • one could have some emulation code that parse the inputs, and generates an output command buffer, therefore copying the inputs.

  • one could just reference the inputs, and have the processing done in pipe at execution time.

If the data is mandated to be copied, then it puts a penalty on implementation that could process the inputs directly in pipe. If the data is “referenced”, then it allows both types of implementation

The inputs are “referenced”, and should not be modified after the call to vkCmdProcessCommandsNVX and until after the rendering of the target command buffer is finished.

18) Why is this NVX and not NV?

To allow early experimentation and feedback. We expect that a version with a refined design as multi-vendor variant will follow up.

19) Should we make the availability for each token type a device limit?

Only distinguish between graphics/compute for now, further splitting up may lead to too much fractioning.

20) When can the objectTable be modified?

Similar to the other inputs for vkCmdProcessCommandsNVX, only when all device access via vkCmdProcessCommandsNVX or execution of target command buffer has completed can an object at a given objectIndex be unregistered or re-registered again.

21) Which buffer usage flags are required for the buffers referenced by vkCmdProcessCommandsNVX

reuse existing VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT

22) In which pipeline stage do the device generated command expansion happen?

vkCmdProcessCommandsNVX is treated as if it occurs in a separate logical pipeline from either graphics or compute, and that pipeline only includes VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT, a new stage VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX, and VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT. This new stage has two corresponding new access types, VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX and VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX, used to synchronize reading the buffer inputs and writing the command buffer memory output. The output written in the target command buffer is considered to be consumed by the VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT pipeline stage.

Thus, to synchronize from writing the input buffers to executing vkCmdProcessCommandsNVX, use:

  • dstStageMask = VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

  • dstAccessMask = VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX

To synchronize from executing vkCmdProcessCommandsNVX to executing the generated commands, use

  • srcStageMask = VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX

  • srcAccessMask = VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX

  • dstStageMask = VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT

  • dstAccessMask = VK_ACCESS_INDIRECT_COMMAND_READ_BIT

When vkCmdProcessCommandsNVX is used with a targetCommandBuffer of NULL, the generated commands are immediately executed and there is implicit synchronization between generation and execution.

23) What if most token data is “static”, but we frequently want to render a subsection?

added “sequencesIndexBuffer”. This allows to easier sort and filter what should actually be processed.

Example Code

Open-Source samples illustrating the usage of the extension can be found at the following locations:

  // setup secondary command buffer
    vkBeginCommandBuffer(generatedCmdBuffer, &beginInfo);
    ... setup its state as usual

  // insert the reservation (there can only be one per command buffer)
  // where the generated calls should be filled into
    VkCmdReserveSpaceForCommandsInfoNVX reserveInfo = { VK_STRUCTURE_TYPE_CMD_RESERVE_SPACE_FOR_COMMANDS_INFO_NVX };
    reserveInfo.objectTable = objectTable;
    reserveInfo.indirectCommandsLayout = deviceGeneratedLayout;
    reserveInfo.maxSequencesCount = myCount;
    vkCmdReserveSpaceForCommandsNVX(generatedCmdBuffer, &reserveInfo);

    vkEndCommandBuffer(generatedCmdBuffer);

  // trigger the generation at some point in another primary command buffer
    VkCmdProcessCommandsInfoNVX processInfo = { VK_STRUCTURE_TYPE_CMD_PROCESS_COMMANDS_INFO_NVX };
    processInfo.objectTable = objectTable;
    processInfo.indirectCommandsLayout = deviceGeneratedLayout;
    processInfo.maxSequencesCount = myCount;
    // set the target of the generation (if null we would directly execute with mainCmd)
    processInfo.targetCommandBuffer = generatedCmdBuffer;
    // provide input data
    processInfo.indirectCommandsTokenCount = 3;
    processInfo.pIndirectCommandsTokens = myTokens;

  // If you modify the input buffer data referenced by VkCmdProcessCommandsInfoNVX,
  // ensure you have added the appropriate barriers prior generation process.
  // When regenerating the content of the same reserved space, ensure prior operations have completed

    VkMemoryBarrier memoryBarrier = { VK_STRUCTURE_TYPE_MEMORY_BARRIER };
    memoryBarrier.srcAccessMask = ...;
    memoryBarrier.dstAccessMask = VK_ACCESS_COMMAND_PROCESS_READ_BIT_NVX;

    vkCmdPipelineBarrier(mainCmd,
                         /*srcStageMask*/VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
                         /*dstStageMask*/VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX,
                         /*dependencyFlags*/0,
                         /*memoryBarrierCount*/1,
                         /*pMemoryBarriers*/&memoryBarrier,
                         ...);

    vkCmdProcessCommandsNVX(mainCmd, &processInfo);
    ...
  // execute the secondary command buffer and ensure the processing that modifies command-buffer content
  // has completed

    memoryBarrier.srcAccessMask = VK_ACCESS_COMMAND_PROCESS_WRITE_BIT_NVX;
    memoryBarrier.dstAccessMask = VK_ACCESS_INDIRECT_COMMAND_READ_BIT;

    vkCmdPipelineBarrier(mainCmd,
                         /*srcStageMask*/VK_PIPELINE_STAGE_COMMAND_PROCESS_BIT_NVX,
                         /*dstStageMask*/VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT,
                         /*dependencyFlags*/0,
                         /*memoryBarrierCount*/1,
                         /*pMemoryBarriers*/&memoryBarrier,
                         ...)
    vkCmdExecuteCommands(mainCmd, 1, &generatedCmdBuffer);

Version History

  • Revision 3, 2017-07-25 (Chris Hebert)

    • Correction to specification of dynamicCount for push_constant token in VkIndirectCommandsLayoutNVX. Stride was incorrectly computed as dynamicCount was not treated as byte size.

  • Revision 2, 2017-06-01 (Christoph Kubisch)

    • header compatibility break: add missing _TYPE to VkIndirectCommandsTokenTypeNVX and VkObjectEntryTypeNVX enums to follow Vulkan naming convention

    • behavior clarification: only allow a single work provoking token per sequence when creating a VkIndirectCommandsLayoutNVX

  • Revision 1, 2016-10-31 (Christoph Kubisch)

    • Initial draft

VK_NVX_multiview_per_view_attributes

Name String

VK_NVX_multiview_per_view_attributes

Extension Type

Device extension

Registered Extension Number

98

Revision

1

Extension and Version Dependencies
Contact
  • Jeff Bolz @jeffbolznv

Last Modified Date

2017-01-13

IP Status

No known IP claims.

Interactions and External Dependencies
Contributors
  • Jeff Bolz, NVIDIA

  • Daniel Koch, NVIDIA

This extension adds a new way to write shaders to be used with multiview subpasses, where the attributes for all views are written out by a single invocation of the vertex processing stages. Related SPIR-V and GLSL extensions SPV_NVX_multiview_per_view_attributes and GL_NVX_multiview_per_view_attributes introduce per-view position and viewport mask attributes arrays, and this extension defines how those per-view attribute arrays are interpreted by Vulkan. Pipelines using per-view attributes may only execute the vertex processing stages once for all views rather than once per-view, which reduces redundant shading work.

A subpass creation flag controls whether the subpass uses this extension. A subpass must either exclusively use this extension or not use it at all.

Some Vulkan implementations only support the position attribute varying between views in the X component. A subpass can declare via a second creation flag whether all pipelines compiled for this subpass will obey this restriction.

Shaders that use the new per-view outputs (e.g. gl_PositionPerViewNV) must also write the non-per-view output (gl_Position), and the values written must be such that gl_Position = gl_PositionPerViewNV[gl_ViewIndex] for all views in the subpass. Implementations are free to either use the per-view outputs or the non-per-view outputs, whichever would be more efficient.

If VK_NV_viewport_array2 is not also supported and enabled, the per-view viewport mask must not be used.

New Object Types

None.

New Enum Constants

  • Extending VkStructureType

    • VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MULTIVIEW_PER_VIEW_ATTRIBUTES_PROPERTIES_NVX

  • Extending VkSubpassDescriptionFlagBits

    • VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX

    • VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX

New Enums

None.

New Functions

None.

New Built-In Variables

New SPIR-V Capabilities

Issues

None.

Examples

#version 450 core

#extension GL_KHX_multiview : enable
#extension GL_NVX_multiview_per_view_attributes : enable

layout(location = 0) in vec4 position;
layout(set = 0, binding = 0) uniform Block { mat4 mvpPerView[2]; } buf;

void main()
{
    // Output both per-view positions and gl_Position as a function
    // of gl_ViewIndex
    gl_PositionPerViewNV[0] = buf.mvpPerView[0] * position;
    gl_PositionPerViewNV[1] = buf.mvpPerView[1] * position;
    gl_Position = buf.mvpPerView[gl_ViewIndex] * position;
}

Version History

  • Revision 1, 2017-01-13 (Jeff Bolz)

    • Internal revisions

Appendix E: API Boilerplate

This appendix defines Vulkan API features that are infrastructure required for a complete functional description of Vulkan, but do not logically belong elsewhere in the Specification.

Vulkan Header Files

Vulkan is defined as an API in the C99 language. Khronos provides a corresponding set of header files for applications using the API, which may be used in either C or C++ code. The interface descriptions in the specification are the same as the interfaces defined in these header files, and both are derived from the vk.xml XML API Registry, which is the canonical machine-readable description of the Vulkan API. The Registry, scripts used for processing it into various forms, and documentation of the registry schema are available as described at https://www.khronos.org/registry/vulkan/#apiregistry .

Language bindings for other languages can be defined using the information in the Specification and the Registry. Khronos does not provide any such bindings, but third-party developers have created some additional bindings.

Vulkan Combined API Header vulkan.h (Informative)

Applications normally will include the header vulkan.h. In turn, vulkan.h always includes the following headers:

  • vk_platform.h, defining platform-specific macros and headers.

  • vulkan_core.h, defining APIs for the Vulkan core and all registered extensions other than window system-specific extensions.

In addition, specific preprocessor macros defined at the time vulkan.h is included cause header files for the corresponding window system-specific extension interfaces to be included.

Vulkan Platform-Specific Header vk_platform.h (Informative)

Platform-specific macros and interfaces are defined in vk_platform.h. These macros are used to control platform-dependent behavior, and their exact definitions are under the control of specific platforms and Vulkan implementations.

Platform-Specific Calling Conventions

On many platforms the following macros are empty strings, causing platform- and compiler-specific default calling conventions to be used.

VKAPI_ATTR is a macro placed before the return type in Vulkan API function declarations. This macro controls calling conventions for C++11 and GCC/Clang-style compilers.

VKAPI_CALL is a macro placed after the return type in Vulkan API function declarations. This macro controls calling conventions for MSVC-style compilers.

VKAPI_PTR is a macro placed between the '(' and '*' in Vulkan API function pointer declarations. This macro also controls calling conventions, and typically has the same definition as VKAPI_ATTR or VKAPI_CALL, depending on the compiler.

With these macros, a Vulkan function declaration takes the form of:

VKAPI_ATTR <return_type> VKAPI_CALL <command_name>(<command_parameters>);

Additionaly, a Vulkan function pointer type declaration takes the form of:

typedef <return_type> (VKAPI_PTR *PFN_<command_name>)(<command_parameters>);
Platform-Specific Header Control

If the VK_NO_STDINT_H macro is defined by the application at compile time, extended integer types used by the Vulkan API, such as uint8_t, must also be defined by the application. Otherwise, the Vulkan headers will not compile. If VK_NO_STDINT_H is not defined, the system <stdint.h> is used to define these types. There is a fallback path when Microsoft Visual Studio version 2008 and earlier versions are detected at compile time.

Vulkan Core API Header vulkan_core.h

Applications that do not make use of window system-specific extensions may simply include vulkan_core.h instead of vulkan.h, although there is usually no reason to do so. In addition to the Vulkan API, vulkan_core.h also defines a small number of C preprocessor macros that are described below.

Vulkan Version Number Macros

API Version Numbers are packed into integers. These macros manipulate version numbers in useful ways.

VK_VERSION_MAJOR extracts the API major version number from a packed version number:

#define VK_VERSION_MAJOR(version) ((uint32_t)(version) >> 22)

VK_VERSION_MINOR extracts the API minor version number from a packed version number:

#define VK_VERSION_MINOR(version) (((uint32_t)(version) >> 12) & 0x3ff)

VK_VERSION_PATCH extracts the API patch version number from a packed version number:

#define VK_VERSION_PATCH(version) ((uint32_t)(version) & 0xfff)

VK_API_VERSION_1_0 returns the API version number for Vulkan 1.0. The patch version number in this macro will always be zero. The supported patch version for a physical device can be queried with vkGetPhysicalDeviceProperties.

// Vulkan 1.0 version number
#define VK_API_VERSION_1_0 VK_MAKE_VERSION(1, 0, 0)// Patch version should always be set to 0

VK_API_VERSION_1_1 returns the API version number for Vulkan 1.1. The patch version number in this macro will always be zero. The supported patch version for a physical device can be queried with vkGetPhysicalDeviceProperties.

// Vulkan 1.1 version number
#define VK_API_VERSION_1_1 VK_MAKE_VERSION(1, 1, 0)// Patch version should always be set to 0

VK_API_VERSION is now commented out of vulkan_core.h and cannot be used.

// DEPRECATED: This define has been removed. Specific version defines (e.g. VK_API_VERSION_1_0), or the VK_MAKE_VERSION macro, should be used instead.
//#define VK_API_VERSION VK_MAKE_VERSION(1, 0, 0) // Patch version should always be set to 0

VK_MAKE_VERSION constructs an API version number.

#define VK_MAKE_VERSION(major, minor, patch) \
    (((major) << 22) | ((minor) << 12) | (patch))
  • major is the major version number.

  • minor is the minor version number.

  • patch is the patch version number.

This macro can be used when constructing the VkApplicationInfo::apiVersion parameter passed to vkCreateInstance.

Vulkan Header File Version Number

VK_HEADER_VERSION is the version number of the vulkan_core.h header. This value is kept synchronized with the patch version of the released Specification.

// Version of this file
#define VK_HEADER_VERSION 75
Vulkan Handle Macros

VK_DEFINE_HANDLE defines a dispatchable handle type.

#define VK_DEFINE_HANDLE(object) typedef struct object##_T* object;
  • object is the name of the resulting C type.

The only dispatchable handle types are those related to device and instance management, such as VkDevice.

VK_DEFINE_NON_DISPATCHABLE_HANDLE defines a non-dispatchable handle type.

#if !defined(VK_DEFINE_NON_DISPATCHABLE_HANDLE)
#if defined(__LP64__) || defined(_WIN64) || (defined(__x86_64__) && !defined(__ILP32__) ) || defined(_M_X64) || defined(__ia64) || defined (_M_IA64) || defined(__aarch64__) || defined(__powerpc64__)
        #define VK_DEFINE_NON_DISPATCHABLE_HANDLE(object) typedef struct object##_T *object;
#else
        #define VK_DEFINE_NON_DISPATCHABLE_HANDLE(object) typedef uint64_t object;
#endif
#endif
  • object is the name of the resulting C type.

Most Vulkan handle types, such as VkBuffer, are non-dispatchable.

Note

The vulkan_core.h header allows the VK_DEFINE_NON_DISPATCHABLE_HANDLE definition to be overridden by the application. If VK_DEFINE_NON_DISPATCHABLE_HANDLE is already defined when vulkan_core.h is compiled, the default definition is skipped. This allows the application to define a binary-compatible custom handle which may provide more type-safety or other features needed by the application. Behavior is undefined if the application defines a non-binary-compatible handle and may result in memory corruption or application termination. Binary compatibility is platform dependent so the application must be careful if it overrides the default VK_DEFINE_NON_DISPATCHABLE_HANDLE definition.

VK_NULL_HANDLE is a reserved value representing a non-valid object handle. It may be passed to and returned from Vulkan commands only when specifically allowed.

#define VK_NULL_HANDLE 0

Window System-Specific Header Control (Informative)

To use a Vulkan extension supporting a platform-specific window system, header files for that window systems must be included at compile time, or platform-specific types must be forward-declared. The Vulkan header files cannot determine whether or not an external header is available at compile time, so platform-specific extensions are provided in separate headers from the core API and platform-independent extensions, allowing applications to decide which ones should be defined and how the external headers are included.

Extensions dependent on particular sets of platform headers, or that forward-declare platform-specific types, are declared in a header named for that platform. Before including these platform-specific Vulkan headers, applications must include both vulkan_core.h and any external native headers the platform extensions depend on.

As a convenience for applications that do not need the flexibility of separate platform-specific Vulkan headers, vulkan.h includes vulkan_core.h, and then conditionally includes platform-specific Vulkan headers and the external headers they depend on. Applications control which platform-specific headers are included by #defining macros before including vulkan.h.

The correspondence between platform-specific extensions, external headers they require, the platform-specific header which declares them, and the preprocessor macros which enable inclusion by vulkan.h are shown in the following table.

Table 82. Window System Extensions and Headers
Extension Name Window System Name Platform-specific Header Required External Headers Controlling vulkan.h Macro

VK_KHR_android_surface

Android

vulkan_android.h

None

VK_USE_PLATFORM_ANDROID_KHR

VK_KHR_mir_surface

Mir

vulkan_mir.h

<mir_toolkit/client_types.h>

VK_USE_PLATFORM_MIR_KHR

VK_KHR_wayland_surface

Wayland

vulkan_wayland.h

<wayland-client.h>

VK_USE_PLATFORM_WAYLAND_KHR

VK_KHR_win32_surface, VK_KHR_external_memory_win32, VK_KHR_win32_keyed_mutex, VK_KHR_external_semaphore_win32, VK_KHR_external_fence_win32, VK_NV_external_memory_win32, VK_NV_win32_keyed_mutex

Microsoft Windows

vulkan_win32.h

<windows.h>

VK_USE_PLATFORM_WIN32_KHR

VK_KHR_xcb_surface

X11 Xcb

vulkan_xcb.h

<xcb/xcb.h>

VK_USE_PLATFORM_XCB_KHR

VK_KHR_xlib_surface

X11 Xlib

vulkan_xlib.h

<X11/Xlib.h>

VK_USE_PLATFORM_XLIB_KHR

VK_EXT_acquire_xlib_display

X11 XRAndR

vulkan_xlib_xrandr.h

<X11/Xlib.h>, <X11/extensions/Xrandr.h>

VK_USE_PLATFORM_XLIB_XRANDR_EXT

VK_MVK_ios_surface

iOS

vulkan_ios.h

None

VK_USE_PLATFORM_IOS_MVK

VK_MVK_macos_surface

macOS

vulkan_macos.h

None

VK_USE_PLATFORM_MACOS_MVK

VK_NN_vi_surface

VI

vulkan_vi.h

None

VK_USE_PLATFORM_VI_NN

Note

This section describes the purpose of the headers independently of the specific underlying functionality of the window system extensions themselves. Each extension name will only link to a description of that extension when viewing a specification built with that extension included.

Appendix F: Invariance

The Vulkan specification is not pixel exact. It therefore does not guarantee an exact match between images produced by different Vulkan implementations. However, the specification does specify exact matches, in some cases, for images produced by the same implementation. The purpose of this appendix is to identify and provide justification for those cases that require exact matches.

Repeatability

The obvious and most fundamental case is repeated issuance of a series of Vulkan commands. For any given Vulkan and framebuffer state vector, and for any Vulkan command, the resulting Vulkan and framebuffer state must be identical whenever the command is executed on that initial Vulkan and framebuffer state. This repeatability requirement does not apply when using shaders containing side effects (image and buffer variable stores and atomic operations), because these memory operations are not guaranteed to be processed in a defined order.

The repeatability requirement does not apply for rendering done using a graphics pipeline that uses VK_RASTERIZATION_ORDER_RELAXED_AMD.

One purpose of repeatability is avoidance of visual artifacts when a double-buffered scene is redrawn. If rendering is not repeatable, swapping between two buffers rendered with the same command sequence may result in visible changes in the image. Such false motion is distracting to the viewer. Another reason for repeatability is testability.

Repeatability, while important, is a weak requirement. Given only repeatability as a requirement, two scenes rendered with one (small) polygon changed in position might differ at every pixel. Such a difference, while within the law of repeatability, is certainly not within its spirit. Additional invariance rules are desirable to ensure useful operation.

Multi-pass Algorithms

Invariance is necessary for a whole set of useful multi-pass algorithms. Such algorithms render multiple times, each time with a different Vulkan mode vector, to eventually produce a result in the framebuffer. Examples of these algorithms include:

  • “Erasing” a primitive from the framebuffer by redrawing it, either in a different color or using the XOR logical operation.

  • Using stencil operations to compute capping planes.

Invariance Rules

For a given Vulkan device:

Rule 1 For any given Vulkan and framebuffer state vector, and for any given Vulkan command, the resulting Vulkan and framebuffer state must be identical each time the command is executed on that initial Vulkan and framebuffer state.

Rule 2 Changes to the following state values have no side effects (the use of any other state value is not affected by the change):

Required:

  • Color and depth/stencil attachment contents

  • Scissor parameters (other than enable)

  • Write masks (color, depth, stencil)

  • Clear values (color, depth, stencil)

Strongly suggested:

  • Stencil parameters (other than enable)

  • Depth test parameters (other than enable)

  • Blend parameters (other than enable)

  • Logical operation parameters (other than enable)

Corollary 1 Fragment generation is invariant with respect to the state values listed in Rule 2.

Rule 3 The arithmetic of each per-fragment operation is invariant except with respect to parameters that directly control it.

Corollary 2 Images rendered into different color attachments of the same framebuffer, either simultaneously or separately using the same command sequence, are pixel identical.

Rule 4 Identical pipelines will produce the same result when run multiple times with the same input. The wording “Identical pipelines” means VkPipeline objects that have been created with identical SPIR-V binaries and identical state, which are then used by commands executed using the same Vulkan state vector. Invariance is relaxed for shaders with side effects, such as performing stores or atomics.

Rule 5 All fragment shaders that either conditionally or unconditionally assign FragCoord.z to FragDepth are depth-invariant with respect to each other, for those fragments where the assignment to FragDepth actually is done.

If a sequence of Vulkan commands specifies primitives to be rendered with shaders containing side effects (image and buffer variable stores and atomic operations), invariance rules are relaxed. In particular, rule 1, corollary 2, and rule 4 do not apply in the presence of shader side effects.

The following weaker versions of rules 1 and 4 apply to Vulkan commands involving shader side effects:

Rule 6 For any given Vulkan and framebuffer state vector, and for any given Vulkan command, the contents of any framebuffer state not directly or indirectly affected by results of shader image or buffer variable stores or atomic operations must be identical each time the command is executed on that initial Vulkan and framebuffer state.

Rule 7 Identical pipelines will produce the same result when run multiple times with the same input as long as:

  • shader invocations do not use image atomic operations;

  • no framebuffer memory is written to more than once by image stores, unless all such stores write the same value; and

  • no shader invocation, or other operation performed to process the sequence of commands, reads memory written to by an image store.

Note

The OpenGL spec has the following invariance rule: Consider a primitive p' obtained by translating a primitive p through an offset (x, y) in window coordinates, where x and y are integers. As long as neither p' nor p is clipped, it must be the case that each fragment f' produced from p' is identical to a corresponding fragment f from p except that the center of f' is offset by (x, y) from the center of f.

This rule does not apply to Vulkan and is an intentional difference from OpenGL.

When any sequence of Vulkan commands triggers shader invocations that perform image stores or atomic operations, and subsequent Vulkan commands read the memory written by those shader invocations, these operations must be explicitly synchronized.

Tessellation Invariance

When using a pipeline containing tessellation evaluation shaders, the fixed-function tessellation primitive generator consumes the input patch specified by an application and emits a new set of primitives. The following invariance rules are intended to provide repeatability guarantees. Additionally, they are intended to allow an application with a carefully crafted tessellation evaluation shader to ensure that the sets of triangles generated for two adjacent patches have identical vertices along shared patch edges, avoiding “cracks” caused by minor differences in the positions of vertices along shared edges.

Rule 1 When processing two patches with identical outer and inner tessellation levels, the tessellation primitive generator will emit an identical set of point, line, or triangle primitives as long as the pipeline used to process the patch primitives has tessellation evaluation shaders specifying the same tessellation mode, spacing, vertex order, and point mode decorations. Two sets of primitives are considered identical if and only if they contain the same number and type of primitives and the generated tessellation coordinates for the vertex numbered m of the primitive numbered n are identical for all values of m and n.

Rule 2 The set of vertices generated along the outer edge of the subdivided primitive in triangle and quad tessellation, and the tessellation coordinates of each, depends only on the corresponding outer tessellation level and the spacing decorations in the tessellation shaders of the pipeline.

Rule 3 The set of vertices generated when subdividing any outer primitive edge is always symmetric. For triangle tessellation, if the subdivision generates a vertex with tessellation coordinates of the form (0, x, 1-x), (x, 0, 1-x), or (x, 1-x, 0), it will also generate a vertex with coordinates of exactly (0, 1-x, x), (1-x, 0, x), or (1-x, x, 0), respectively. For quad tessellation, if the subdivision generates a vertex with coordinates of (x, 0) or (0, x), it will also generate a vertex with coordinates of exactly (1-x, 0) or (0, 1-x), respectively. For isoline tessellation, if it generates vertices at (0, x) and (1, x) where x is not zero, it will also generate vertices at exactly (0, 1-x) and (1, 1-x), respectively.

Rule 4 The set of vertices generated when subdividing outer edges in triangular and quad tessellation must be independent of the specific edge subdivided, given identical outer tessellation levels and spacing. For example, if vertices at (x, 1 - x, 0) and (1-x, x, 0) are generated when subdividing the w = 0 edge in triangular tessellation, vertices must be generated at (x, 0, 1-x) and (1-x, 0, x) when subdividing an otherwise identical v = 0 edge. For quad tessellation, if vertices at (x, 0) and (1-x, 0) are generated when subdividing the v = 0 edge, vertices must be generated at (0, x) and (0, 1-x) when subdividing an otherwise identical u = 0 edge.

Rule 5 When processing two patches that are identical in all respects enumerated in rule 1 except for vertex order, the set of triangles generated for triangle and quad tessellation must be identical except for vertex and triangle order. For each triangle n1 produced by processing the first patch, there must be a triangle n2 produced when processing the second patch each of whose vertices has the same tessellation coordinates as one of the vertices in n1.

Rule 6 When processing two patches that are identical in all respects enumerated in rule 1 other than matching outer tessellation levels and/or vertex order, the set of interior triangles generated for triangle and quad tessellation must be identical in all respects except for vertex and triangle order. For each interior triangle n1 produced by processing the first patch, there must be a triangle n2 produced when processing the second patch each of whose vertices has the same tessellation coordinates as one of the vertices in n1. A triangle produced by the tessellator is considered an interior triangle if none of its vertices lie on an outer edge of the subdivided primitive.

Rule 7 For quad and triangle tessellation, the set of triangles connecting an inner and outer edge depends only on the inner and outer tessellation levels corresponding to that edge and the spacing decorations.

Rule 8 The value of all defined components of TessCoord will be in the range [0, 1]. Additionally, for any defined component x of TessCoord, the results of computing 1.0-x in a tessellation evaluation shader will be exact. If any floating-point values in the range [0, 1] fail to satisfy this property, such values must not be used as tessellation coordinate components.

Glossary

The terms defined in this section are used consistently throughout this Specification and may be used with or without capitalization.

Accessible (Descriptor Binding)

A descriptor binding is accessible to a shader stage if that stage is included in the stageFlags of the descriptor binding. Descriptors using that binding can only be used by stages in which they are accessible.

Acquire Operation (Resource)

An operation that acquires ownership of an image subresource or buffer range.

Adjacent Vertex

A vertex in an adjacency primitive topology that is not part of a given primitive, but is accessible in geometry shaders.

Advanced Blend Operation

Blending performed using one of the blend operation enums introduced by the VK_EXT_blend_operation_advanced extension. See Advanced Blending Operations.

Aliased Range (Memory)

A range of a device memory allocation that is bound to multiple resources simultaneously.

Allocation Scope

An association of a host memory allocation to a parent object or command, where the allocation’s lifetime ends before or at the same time as the parent object is freed or destroyed, or during the parent command.

API Order

A set of ordering rules that govern how primitives in draw commands affect the framebuffer.

Aspect (Image)

An image may contain multiple kinds, or aspects, of data for each pixel, where each aspect is used in a particular way by the pipeline and may be stored differently or separately from other aspects. For example, the color components of an image format make up the color aspect of the image, and may be used as a framebuffer color attachment. Some operations, like depth testing, operate only on specific aspects of an image. Others operations, like image/buffer copies, only operate on one aspect at a time.

Attachment (Render Pass)

A zero-based integer index name used in render pass creation to refer to a framebuffer attachment that is accessed by one or more subpasses. The index also refers to an attachment description which includes information about the properties of the image view that will later be attached.

Availability Operation

An operation that causes the values generated by specified memory write accesses to become available for future access.

Available

A state of values written to memory that allows them to be made visible.

Back-Facing

See Facingness.

Batch

A single structure submitted to a queue as part of a queue submission command, describing a set of queue operations to execute.

Backwards Compatibility

A given version of the API is backwards compatible with an earlier version if an application, relying only on valid behavior and functionality defined by the earlier specification, is able to correctly run against each version without any modification. This assumes no active attempt by that application to not run when it detects a different version.

Full Compatibility

A given version of the API is fully compatible with another version if an application, relying only on valid behavior and functionality defined by either of those specifications, is able to correctly run against each version without any modification. This assumes no active attempt by that application to not run when it detects a different version.

Binding (Memory)

An association established between a range of a resource object and a range of a memory object. These associations determine the memory locations affected by operations performed on elements of a resource object. Memory bindings are established using the vkBindBufferMemory command for non-sparse buffer objects, using the vkBindImageMemory command for non-sparse image objects, and using the vkQueueBindSparse command for sparse resources.

Blend Constant

Four floating point (RGBA) values used as an input to blending.

Blending

Arithmetic operations between a fragment color value and a value in a color attachment that produce a final color value to be written to the attachment.

Buffer

A resource that represents a linear array of data in device memory. Represented by a VkBuffer object.

Buffer View

An object that represents a range of a specific buffer, and state that controls how the contents are interpreted. Represented by a VkBufferView object.

Built-In Variable

A variable decorated in a shader, where the decoration makes the variable take values provided by the execution environment or values that are generated by fixed-function pipeline stages.

Built-In Interface Block

A block defined in a shader that contains only variables decorated with built-in decorations, and is used to match against other shader stages.

Clip Coordinates

The homogeneous coordinate space that vertex positions (Position decoration) are written in by vertex processing stages.

Clip Distance

A built-in output from vertex processing stages that defines a clip half-space against which the primitive is clipped.

Clip Volume

The intersection of the view volume with all clip half-spaces.

Color Attachment

A subpass attachment point, or image view, that is the target of fragment color outputs and blending.

Color Fragment

A unique color value within a pixel of a multisampled color image. The fragment mask will contain indices to the color fragment.

Color Renderable Format

A VkFormat where VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT is set in the optimalTilingFeatures or linearTilingFeatures field of VkFormatProperties::optimalTilingFeatures returned by vkGetPhysicalDeviceFormatProperties, depending on the tiling used.

Color Sample Mask

A bitfield associated with a fragment, with one bit for each sample in the color attachment(s). Samples are considered to be covered based on the result of the Coverage Reduction stage. Uncovered samples do not write to color attachments.

Combined Image Sampler

A descriptor type that includes both a sampled image and a sampler.

Command Buffer

An object that records commands to be submitted to a queue. Represented by a VkCommandBuffer object.

Command Pool

An object that command buffer memory is allocated from, and that owns that memory. Command pools aid multithreaded performance by enabling different threads to use different allocators, without internal synchronization on each use. Represented by a VkCommandPool object.

Compatible Allocator

When allocators are compatible, allocations from each allocator can be freed by the other allocator.

Compatible Image Formats

When formats are compatible, images created with one of the formats can have image views created from it using any of the compatible formats. Also see Size-Compatible Image Formats.

Compatible Queues

Queues within a queue family. Compatible queues have identical properties.

Component (Format)

A distinct part of a format. Depth, stencil, and color channels (e.g. R, G, B, A), are all separate components.

Compressed Texel Block

An element of an image having a block-compressed format, comprising a rectangular block of texel values that are encoded as a single value in memory. Compressed texel blocks of a particular block-compressed format have a corresponding width, height, and depth that define the dimensions of these elements in units of texels, and a size in bytes of the encoding in memory.

Coverage

A bitfield associated with a fragment, where each bit is associated to a rasterization sample. Samples are initially considered to be covered based on the result of rasterization, and then coverage can subsequently be turned on or off by other fragment operations or the fragment shader. Uncovered samples do not write to framebuffer attachments.

Cull Distance

A built-in output from vertex processing stages that defines a cull half-space where the primitive is rejected if all vertices have a negative value for the same cull distance.

Cull Volume

The intersection of the view volume with all cull half-spaces.

Decoration (SPIR-V)

Auxiliary information such as built-in variables, stream numbers, invariance, interpolation type, relaxed precision, etc., added to variables or structure-type members through decorations.

Depth/Stencil Attachment

A subpass attachment point, or image view, that is the target of depth and/or stencil test operations and writes.

Depth/Stencil Format

A VkFormat that includes depth and/or stencil components.

Depth/Stencil Image (or ImageView)

A VkImage (or VkImageView) with a depth/stencil format.

Derivative Group

A set of fragment shader invocations that cooperate to compute derivatives, including implicit derivatives for sampled image operations.

Descriptor

Information about a resource or resource view written into a descriptor set that is used to access the resource or view from a shader.

Descriptor Binding

An entry in a descriptor set layout corresponding to zero or more descriptors of a single descriptor type in a set. Defined by a VkDescriptorSetLayoutBinding structure.

Descriptor Pool

An object that descriptor sets are allocated from, and that owns the storage of those descriptor sets. Descriptor pools aid multithreaded performance by enabling different threads to use different allocators, without internal synchronization on each use. Represented by a VkDescriptorPool object.

Descriptor Set

An object that resource descriptors are written into via the API, and that can be bound to a command buffer such that the descriptors contained within it can be accessed from shaders. Represented by a VkDescriptorSet object.

Descriptor Set Layout

An object that defines the set of resources (types and counts) and their relative arrangement (in the binding namespace) within a descriptor set. Used when allocating descriptor sets and when creating pipeline layouts. Represented by a VkDescriptorSetLayout object.

Device

The processor(s) and execution environment that perform tasks requested by the application via the Vulkan API.

Device Group

A set of physical devices that support accessing each other’s memory and recording a single command buffer that can be executed on all the physical devices.

Device Index

A zero-based integer that identifies one physical device from a logical device. A device index is valid if it is less than the number of physical devices in the logical device.

Device Mask

A bitmask where each bit represents one device index. A device mask value is valid if every bit that is set in the mask is at a bit position that is less than the number of physical devices in the logical device.

Device Memory

Memory accessible to the device. Represented by a VkDeviceMemory object.

Device-Level Command

Any command that is dispatched from a logical device, or from a child object of a logical device.

Device-Level Functionality

All device-level commands and objects, and their structures, enumerated types, and enumerants.

Device-Level Object

Logical device objects and their child objects. For example, VkDevice, VkQueue, and VkCommandBuffer objects are device-level objects.

Device-Local Memory

Memory that is connected to the device, and may be more performant for device access than host-local memory.

Direct Drawing Commands

Drawing commands that take all their parameters as direct arguments to the command (and not sourced via structures in buffer memory as the indirect drawing commands). Includes vkCmdDraw, and vkCmdDrawIndexed.

Disjoint

Disjoint planes are image planes to which memory is bound independently.
A disjoint image consists of multiple disjoint planes, and is created with the VK_IMAGE_CREATE_DISJOINT_BIT bit set.

Dispatchable Handle

A handle of a pointer handle type which may be used by layers as part of intercepting API commands. The first argument to each Vulkan command is a dispatchable handle type.

Dispatching Commands

Commands that provoke work using a compute pipeline. Includes vkCmdDispatch and vkCmdDispatchIndirect.

Drawing Commands

Commands that provoke work using a graphics pipeline. Includes vkCmdDraw, vkCmdDrawIndexed, vkCmdDrawIndirect, vkCmdDrawIndirectCountAMD, vkCmdDrawIndexedIndirectCountAMD, and vkCmdDrawIndexedIndirect.

Duration (Command)

The duration of a Vulkan command refers to the interval between calling the command and its return to the caller.

Dynamic Storage Buffer

A storage buffer whose offset is specified each time the storage buffer is bound to a command buffer via a descriptor set.

Dynamic Uniform Buffer

A uniform buffer whose offset is specified each time the uniform buffer is bound to a command buffer via a descriptor set.

Dynamically Uniform

See Dynamically Uniform in section 2.2 “Terms” of the Khronos SPIR-V Specification.

Element Size

The size (in bytes) used to store one element of an uncompressed format or the size (in bytes) used to store one block of a block-compressed format.

Explicitly-Enabled Layer

A layer enabled by the application by adding it to the enabled layer list in vkCreateInstance or vkCreateDevice.

Event

A synchronization primitive that is signaled when execution of previous commands complete through a specified set of pipeline stages. Events can be waited on by the device and polled by the host. Represented by a VkEvent object.

Executable State (Command Buffer)

A command buffer that has ended recording commands and can be executed. See also Initial State and Recording State.

Execution Dependency

A dependency that guarantees that certain pipeline stages’ work for a first set of commands has completed execution before certain pipeline stages’ work for a second set of commands begins execution. This is accomplished via pipeline barriers, subpass dependencies, events, or implicit ordering operations.

Execution Dependency Chain

A sequence of execution dependencies that transitively act as a single execution dependency.

Explicit chroma reconstruction

An implementation of sampler Y’CBCR conversion which reconstructs reduced-resolution chroma samples to luma resolution and then separately performs texture sample interpolation. This is distinct from an implicit implementation, which incorporates chroma sample reconstruction into texture sample interpolation.

Extension Scope

The set of objects and commands that can be affected by an extension. Extensions are either device scope or instance scope.

External Handle

A resource handle which has meaning outside of a specific Vulkan device or its parent instance. External handles may be used to share resources between multiple Vulkan devices in different instances, or between Vulkan and other APIs. Some external handle types correspond to platform-defined handles, in which case the resource may outlive any particular Vulkan device or instance and may be transferred between processes, or otherwise manipulated via functionality defined by the platform for that handle type.

External synchronization

A type of synchronization required of the application, where parameters defined to be externally synchronized must not be used simultaneously in multiple threads.

Facingness (Polygon)

A classification of a polygon as either front-facing or back-facing, depending on the orientation (winding order) of its vertices.

Facingness (Fragment)

A fragment is either front-facing or back-facing, depending on the primitive it was generated from. If the primitive was a polygon (regardless of polygon mode), the fragment inherits the facingness of the polygon. All other fragments are front-facing.

Fence

A synchronization primitive that is signaled when a set of batches or sparse binding operations complete execution on a queue. Fences can be waited on by the host. Represented by a VkFence object.

Flat Shading

A property of a vertex attribute that causes the value from a single vertex (the provoking vertex) to be used for all vertices in a primitive, and for interpolation of that attribute to return that single value unaltered.

Fragment Input Attachment Interface

Variables with UniformConstant storage class and a decoration of InputAttachmentIndex that are statically used by a fragment shader’s entry point, which receive values from input attachments.

Fragment Mask

A lookup table that associates color samples with color fragment values.

Fragment Output Interface

A fragment shader entry point’s variables with Output storage class, which output to color and/or depth/stencil attachments.

Framebuffer

A collection of image views and a set of dimensions that, in conjunction with a render pass, define the inputs and outputs used by drawing commands. Represented by a VkFramebuffer object.

Framebuffer Attachment

One of the image views used in a framebuffer.

Framebuffer Coordinates

A coordinate system in which adjacent pixels’ coordinates differ by 1 in x and/or y, with (0,0) in the upper left corner and pixel centers at half-integers.

Framebuffer-Space

Operating with respect to framebuffer coordinates.

Framebuffer-Local

A framebuffer-local dependency guarantees that only for a single framebuffer region, the first set of operations happens-before the second set of operations.

Framebuffer-Global

A framebuffer-global dependency guarantees that for all framebuffer regions, the first set of operations happens-before the second set of operations.

Framebuffer Region

A framebuffer region is a set of sample (x, y, layer, sample) coordinates that is a subset of the entire framebuffer.

Front-Facing

See Facingness.

Global Workgroup

A collection of local workgroups dispatched by a single dispatch command.

Handle

An opaque integer or pointer value used to refer to a Vulkan object. Each object type has a unique handle type.

Happen-after

A transitive, irreflexive and antisymmetric ordering relation between operations. An execution dependency with a source of A and a destination of B enforces that B happens-after A. The inverse relation of happens-before.

Happen-before

A transitive, irreflexive and antisymmetric ordering relation between operations. An execution dependency with a source of A and a destination of B enforces that A happens-before B. The inverse relation of happens-after.

Helper Invocation

A fragment shader invocation that is created solely for the purposes of evaluating derivatives for use in non-helper fragment shader invocations, and which does not have side effects.

Host

The processor(s) and execution environment that the application runs on, and that the Vulkan API is exposed on.

Host Mapped Foreign Memory

Memory owned by a foreign device that is mapped for host access.

Host Memory

Memory not accessible to the device, used to store implementation data structures.

Host-Accessible Subresource

A buffer, or a linear image subresource in either the VK_IMAGE_LAYOUT_PREINITIALIZED or VK_IMAGE_LAYOUT_GENERAL layout. Host-accessible subresources have a well-defined addressing scheme which can be used by the host.

Host-Local Memory

Memory that is not local to the device, and may be less performant for device access than device-local memory.

Host-Visible Memory

Device memory that can be mapped on the host and can be read and written by the host.

Identically Defined Objects

Objects of the same type where all arguments to their creation or allocation functions, with the exception of pAllocator, are

  1. Vulkan handles which refer to the same object or

  2. identical scalar or enumeration values or

  3. Host pointers which point to an array of values or structures which also satisfy these three constraints.

Image

A resource that represents a multi-dimensional formatted interpretation of device memory. Represented by a VkImage object.

Image Subresource

A specific mipmap level and layer of an image.

Image Subresource Range

A set of image subresources that are contiguous mipmap levels and layers.

Image View

An object that represents an image subresource range of a specific image, and state that controls how the contents are interpreted. Represented by a VkImageView object.

Immutable Sampler

A sampler descriptor provided at descriptor set layout creation time, and that is used for that binding in all descriptor sets allocated from the layout, and cannot be changed.

Implicit chroma reconstruction

An implementation of sampler Y’CBCR conversion which reconstructs the reduced-resolution chroma samples directly at the sample point, as part of the normal texture sampling operation. This is distinct from an explicit chroma reconstruction implementation, which reconstructs the reduced-resolution chroma samples to the resolution of the luma samples, then filters the result as part of texture sample interpolation.

Implicitly-Enabled Layer

A layer enabled by a loader-defined mechanism outside the Vulkan API, rather than explicitly by the application during instance or device creation.

Index Buffer

A buffer bound via vkCmdBindIndexBuffer which is the source of index values used to fetch vertex attributes for a vkCmdDrawIndexed or vkCmdDrawIndexedIndirect command.

Indexed Drawing Commands

Drawing commands which use an index buffer as the source of index values used to fetch vertex attributes for a drawing command. Includes vkCmdDrawIndexed, vkCmdDrawIndexedIndirectCountAMD, and vkCmdDrawIndexedIndirect.

Indirect Commands

Drawing or dispatching commands that source some of their parameters from structures in buffer memory. Includes vkCmdDrawIndirect, vkCmdDrawIndexedIndirect, vkCmdDrawIndirectCountAMD, vkCmdDrawIndexedIndirectCountAMD, and vkCmdDispatchIndirect.

Indirect Commands Layout

A definition of a sequence of commands, that are generated on the device via vkCmdProcessCommandsNVX. Each sequence is comprised of multiple VkIndirectCommandsTokenTypeNVX, which represent a subset of traditional command buffer commands. Represented as VkIndirectCommandsLayoutNVX.

Indirect Drawing Commands

Drawing commands that source some of their parameters from structures in buffer memory. Includes vkCmdDrawIndirect, vkCmdDrawIndirectCountAMD, vkCmdDrawIndexedIndirectCountAMD, and vkCmdDrawIndexedIndirect.

Initial State (Command Buffer)

A command buffer that has not begun recording commands. See also Recorded State and Executable State.

Input Attachment

A descriptor type that represents an image view, and supports unfiltered read-only access in a shader, only at the fragment’s location in the view.

Instance

The top-level Vulkan object, which represents the application’s connection to the implementation. Represented by a VkInstance object.

Instance-Level Command

Any command that is dispatched from an instance, or from a child object of an instance, except for physical devices and their children.

Instance-Level Functionality

All instance-level commands and objects, and their structures, enumerated types, and enumerants.

Instance-Level Object

High-level Vulkan objects, which are not physical devices, nor children of physical devices. For example, VkInstance is an instance-level object.

Instance (Memory)

In a logical device representing more than one physical device, some device memory allocations have the requested amount of memory allocated multiple times, once for each physical device in a device mask. Each such replicated allocation is an instance of the device memory.

Instance (Resource)

In a logical device representing more than one physical device, buffer and image resources exist on all physical devices but can be bound to memory differently on each. Each such replicated resource is an instance of the resource.

Internal Synchronization

A type of synchronization required of the implementation, where parameters not defined to be externally synchronized may require internal mutexing to avoid multithreaded race conditions.

Invocation (Shader)

A single execution of an entry point in a SPIR-V module. For example, a single vertex’s execution of a vertex shader or a single fragment’s execution of a fragment shader.

Invocation Group

A set of shader invocations that are executed in parallel and that must execute the same control flow path in order for control flow to be considered dynamically uniform.

Linear Resource

A resource is linear if it is a VkBuffer, or a VkImage created with VK_IMAGE_TILING_LINEAR. A resource is non-linear if it is a VkImage created with VK_IMAGE_TILING_OPTIMAL.

Local Workgroup

A collection of compute shader invocations invoked by a single dispatch command, which share data via WorkgroupLocal variables and can synchronize with each other.

Logical Device

An object that represents the application’s interface to the physical device. The logical device is the parent of most Vulkan objects. Represented by a VkDevice object.

Logical Operation

Bitwise operations between a fragment color value and a value in a color attachment, that produce a final color value to be written to the attachment.

Lost Device

A state that a logical device may be in as a result of unrecoverable implementation errors, or other exceptional conditions.

Mappable

See Host-Visible Memory.

Memory Dependency

A memory dependency is an execution dependency which includes availability and visibility operations such that:

  • The first set of operations happens-before the availability operation

  • The availability operation happens-before the visibility operation

  • The visibility operation happens-before the second set of operations

Memory Heap

A region of memory from which device memory allocations can be made.

Memory Type

An index used to select a set of memory properties (e.g. mappable, cached) for a device memory allocation.

Mip Tail Region

The set of mipmap levels of a sparse residency texture that are too small to fill a sparse block, and that must all be bound to memory collectively and opaquely.

Multi-planar

A multi-planar format (or “planar format”) is an image format consisting of more than one plane, identifiable with a _2PLANE or _3PLANE component to the format name and listed in Formats requiring sampler Y’CBCR conversion for VK_IMAGE_ASPECT_COLOR_BIT image views. A multi-planar image (or “planar image”) is an image of a multi-planar format.

Non-Dispatchable Handle

A handle of an integer handle type. Handle values may not be unique, even for two objects of the same type.

Non-Indexed Drawing Commands

Drawing commands for which the vertex attributes are sourced in linear order from the vertex input attributes for a drawing command (i.e. they do not use an index buffer). Includes vkCmdDraw, vkCmdDrawIndirectCountAMD, and vkCmdDrawIndirect.

Normalized

A value that is interpreted as being in the range [0,1] as a result of being implicitly divided by some other value.

Normalized Device Coordinates

A coordinate space after perspective division is applied to clip coordinates, and before the viewport transformation converts to framebuffer coordinates.

Object Table

A binding table for various resources (VkPipeline, VkBuffer, VkDescriptorSet), so that they can be referenced in device-generated command processing. Represented as VkObjectTableNVX. Entries are registered or unregistered via uint32_t indices.

Overlapped Range (Aliased Range)

The aliased range of a device memory allocation that intersects a given image subresource of an image or range of a buffer.

Ownership (Resource)

If an entity (e.g. a queue family) has ownership of a resource, access to that resource is well-defined for access by that entity.

Packed Format

A format whose components are stored as a single data element in memory, with their relative locations defined within that element.

Passthrough Geometry Shader

A geometry shader which uses the PassthroughNV decoration on a variable in its input interface. Output primitives in a passthrough geometry shader always have the same topology as the input primitive and are not produced by emitting vertices.

Payload

Importable or exportable reference to the internal data of an object in Vulkan.

Peer Memory

An instance of memory corresponding to a different physical device than the physical device performing the memory access, in a logical device that represents multiple physical devices.

Physical Device

An object that represents a single device in the system. Represented by a VkPhysicalDevice object.

Physical-Device-Level Command

Any command that is dispatched from a physical device.

Physical-Device-Level Functionality

All physical-device-level commands and objects, and their structures, enumerated types, and enumerants.

Physical-Device-Level Object

Physical device objects. For example, VkPhysicalDevice is a physical-device-level object.

Pipeline

An object that controls how graphics or compute work is executed on the device. A pipeline includes one or more shaders, as well as state controlling any non-programmable stages of the pipeline. Represented by a VkPipeline object.

Pipeline Barrier

An execution and/or memory dependency recorded as an explicit command in a command buffer, that forms a dependency between the previous and subsequent commands.

Pipeline Cache

An object that can be used to collect and retrieve information from pipelines as they are created, and can be populated with previously retrieved information in order to accelerate pipeline creation. Represented by a VkPipelineCache object.

Pipeline Layout

An object that defines the set of resources (via a collection of descriptor set layouts) and push constants used by pipelines that are created using the layout. Used when creating a pipeline and when binding descriptor sets and setting push constant values. Represented by a VkPipelineLayout object.

Pipeline Stage

A logically independent execution unit that performs some of the operations defined by an action command.

pNext Chain

A set of structures chained together through their pNext members.

Planar

See multi-planar.

Plane

An image plane is part of the representation of an image, containing a subset of the color channels required to represent the texels in the image and with a contiguous mapping of coordinates to bound memory. Most images consist only of a single plane, but some formats spread the channels across multiple image planes. The host-accessible properties of each image plane are accessed in a linear layout using vkGetImageSubresourceLayout. If a multi-planar image is created with the VK_IMAGE_CREATE_DISJOINT_BIT bit set, the image is described as disjoint, and its planes are therefore are bound to memory independently.

Point Sampling (Rasterization)

A rule that determines whether a fragment sample location is covered by a polygon primitive by testing whether the sample location is in the interior of the polygon in framebuffer-space, or on the boundary of the polygon according to the tie-breaking rules.

Presentable image

A VkImage object obtained from a VkSwapchainKHR used to present to a VkSurfaceKHR object.

Preserve Attachment

One of a list of attachments in a subpass description that is not read or written by the subpass, but that is read or written on earlier and later subpasses and whose contents must be preserved through this subpass.

Primary Command Buffer

A command buffer that can execute secondary command buffers, and can be submitted directly to a queue.

Primitive Topology

State that controls how vertices are assembled into primitives, e.g. as lists of triangles, strips of lines, etc..

Promoted

An extension whose interfaces are later made available as part of a core version of the API, with the author ID suffixes removed, is said to be promoted to that core version. Minor differences, such as making the availability of specific features from the extension supported only if a corresponding feature bit is enabled, may still exist.

Protected Buffer

A buffer to which protected device memory can be bound.

Protected-capable Device Queue

A device queue to which protected command buffers can be submitted.

Protected Command Buffer

A command buffer which can be submitted to a protected-capable device queue.

Protected Device Memory

Device memory which can be visible to the device but must not be visible to the host.

Protected Image

An image to which protected device memory can be bound.

Provoking Vertex

The vertex in a primitive from which flat shaded attribute values are taken. This is generally the “first” vertex in the primitive, and depends on the primitive topology.

Push Constants

A small bank of values writable via the API and accessible in shaders. Push constants allow the application to set values used in shaders without creating buffers or modifying and binding descriptor sets for each update.

Push Constant Interface

The set of variables with PushConstant storage class that are statically used by a shader entry point, and which receive values from push constant commands.

Push Descriptors

Descriptors that are written directly into a command buffer rather than into a descriptor set. Push descriptors allow the application to set descriptors used in shaders without allocating or modifying descriptor sets for each update.

Descriptor Update Template

An object that specifies a mapping from descriptor update information in host memory to elements in a descriptor set, which helps enable more efficient descriptor set updates.

Query Pool

An object that contains a number of query entries and their associated state and results. Represented by a VkQueryPool object.

Queue

An object that executes command buffers and sparse binding operations on a device. Represented by a VkQueue object.

Queue Family

A set of queues that have common properties and support the same functionality, as advertised in VkQueueFamilyProperties.

Queue Operation

A unit of work to be executed by a specific queue on a device, submitted via a queue submission command. Each queue submission command details the specific queue operations that occur as a result of calling that command. Queue operations typically include work that is specific to each command, and synchronization tasks.

Queue Submission

Zero or more batches and an optional fence to be signaled, passed to a command for execution on a queue. See the Devices and Queues chapter for more information.

Recording State (Command Buffer)

A command buffer that is ready to record commands. See also Initial State and Executable State.

Release Operation (Resource)

An operation that releases ownership of an image subresource or buffer range.

Render Pass

An object that represents a set of framebuffer attachments and phases of rendering using those attachments. Represented by a VkRenderPass object.

Render Pass Instance

A use of a render pass in a command buffer.

Required Extensions

Extensions that must be enabled alongside extensions dependent on them (see Extension Dependencies).

Reset (Command Buffer)

Resetting a command buffer discards any previously recorded commands and puts a command buffer in the initial state.

Residency Code

An integer value returned by sparse image instructions, indicating whether any sparse unbound texels were accessed.

Resolve Attachment

A subpass attachment point, or image view, that is the target of a multisample resolve operation from the corresponding color attachment at the end of the subpass.

Retired Swapchain

A swapchain that has been used as the oldSwapchain parameter to vkCreateSwapchainKHR. Images cannot be acquired from a retired swapchain, however images that were acquired (but not presented) before the swapchain was retired can be presented.

Sampled Image

A descriptor type that represents an image view, and supports filtered (sampled) and unfiltered read-only access in a shader.

Sampler

An object that contains state that controls how sampled image data is sampled (or filtered) when accessed in a shader. Also a descriptor type describing the object. Represented by a VkSampler object.

Secondary Command Buffer

A command buffer that can be executed by a primary command buffer, and must not be submitted directly to a queue.

Self-Dependency

A subpass dependency from a subpass to itself, i.e. with srcSubpass equal to dstSubpass. A self-dependency is not automatically performed during a render pass instance, rather a subset of it can be performed via vkCmdPipelineBarrier during the subpass.

Semaphore

A synchronization primitive that supports signal and wait operations, and can be used to synchronize operations within a queue or across queues. Represented by a VkSemaphore object.

Shader

Instructions selected (via an entry point) from a shader module, which are executed in a shader stage.

Shader Code

A stream of instructions used to describe the operation of a shader.

Shader Module

A collection of shader code, potentially including several functions and entry points, that is used to create shaders in pipelines. Represented by a VkShaderModule object.

Shader Stage

A stage of the graphics or compute pipeline that executes shader code.

Shared presentable image

A presentable image created from a swapchain with VkPresentModeKHR set to either VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR or VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR.

Side Effect

A store to memory or atomic operation on memory from a shader invocation.

Single-plane format

A format that is not multi-planar.

Size-Compatible Image Formats

When a compressed image format and an uncompressed image format are size-compatible, it means that the element size of the uncompressed format must equal the element size (compressed texel block size) of the compressed format.

Sparse Block

An element of a sparse resource that can be independently bound to memory. Sparse blocks of a particular sparse resource have a corresponding size in bytes that they use in the bound memory.

Sparse Image Block

A sparse block in a sparse partially-resident image. In addition to the sparse block size in bytes, sparse image blocks have a corresponding width, height, and depth that define the dimensions of these elements in units of texels or compressed texel blocks, the latter being used in case of sparse images having a block-compressed format.

Sparse Unbound Texel

A texel read from a region of a sparse texture that does not have memory bound to it.

Static Use

An object in a shader is statically used by a shader entry point if any function in the entry point’s call tree contains an instruction using the object. Static use is used to constrain the set of descriptors used by a shader entry point.

Storage Buffer

A descriptor type that represents a buffer, and supports reads, writes, and atomics in a shader.

Storage Image

A descriptor type that represents an image view, and supports unfiltered loads, stores, and atomics in a shader.

Storage Texel Buffer

A descriptor type that represents a buffer view, and supports unfiltered, formatted reads, writes, and atomics in a shader.

Subgroup

A set of shader invocations that can synchronize and share data with each other efficiently. In compute shaders, the local workgroup is a superset of the subgroup.

Subgroup Mask

A bitmask for all invocations in the current subgroup with one bit per invocation, starting with the least significant bit in the first vector component, continuing to the last bit (less than SubgroupSize) in the last required vector component.

Subpass

A phase of rendering within a render pass, that reads and writes a subset of the attachments.

Subpass Dependency

An execution and/or memory dependency between two subpasses described as part of render pass creation, and automatically performed between subpasses in a render pass instance. A subpass dependency limits the overlap of execution of the pair of subpasses, and can provide guarantees of memory coherence between accesses in the subpasses.

Subpass Description

Lists of attachment indices for input attachments, color attachments, depth/stencil attachment, resolve attachments, and preserve attachments used by the subpass in a render pass.

Subset (Self-Dependency)

A subset of a self-dependency is a pipeline barrier performed during the subpass of the self-dependency, and whose stage masks and access masks each contain a subset of the bits set in the identically named mask in the self-dependency.

Texel Coordinate System

One of three coordinate systems (normalized, unnormalized, integer) that define how texel coordinates are interpreted in an image or a specific mipmap level of an image.

Uniform Texel Buffer

A descriptor type that represents a buffer view, and supports unfiltered, formatted, read-only access in a shader.

Uniform Buffer

A descriptor type that represents a buffer, and supports read-only access in a shader.

Unnormalized

A value that is interpreted according to its conventional interpretation, and is not normalized.

Unprotected Buffer

A buffer to which unprotected device memory can be bound.

Unprotected Command Buffer

A command buffer which can be submitted to an unprotected device queue or a protected-capable device queue.

Unprotected Device Memory

Device memory which can be visible to the device and can be visible to the host.

Unprotected Image

An image to which unprotected device memory can be bound.

User-Defined Variable Interface

A shader entry point’s variables with Input or Output storage class that are not built-in variables.

Vertex Input Attribute

A graphics pipeline resource that produces input values for the vertex shader by reading data from a vertex input binding and converting it to the attribute’s format.

Validation Cache

An object that can be used to collect and retrieve validation results from the validation layers, and can be populated with previously retrieved results in order to accelerate the validation process. Represented by a VkValidationCacheEXT object.

Vertex Input Binding

A graphics pipeline resource that is bound to a buffer and includes state that affects addressing calculations within that buffer.

Vertex Input Interface

A vertex shader entry point’s variables with Input storage class, which receive values from vertex input attributes.

Vertex Processing Stages

A set of shader stages that comprises the vertex shader, tessellation control shader, tessellation evaluation shader, and geometry shader stages.

View Mask

When multiview is enabled, a view mask is a property of a subpass controlling which views the rendering commands are broadcast to.

View Volume

A subspace in homogeneous coordinates, corresponding to post-projection x and y values between -1 and +1, and z values between 0 and +1.

Viewport Transformation

A transformation from normalized device coordinates to framebuffer coordinates, based on a viewport rectangle and depth range.

Visibility Operation

An operation that causes available values to become visible to specified memory accesses.

Visible

A state of values written to memory that allows them to be accessed by a set of operations.

Common Abbreviations

Abbreviations and acronyms are sometimes used in the Specification and the API where they are considered clear and commonplace, and are defined here:

Src

Source

Dst

Destination

Min

Minimum

Max

Maximum

Rect

Rectangle

Info

Information

LOD

Level of Detail

ID

Identifier

UUID

Universally Unique Identifier

Op

Operation

R

Red color component

G

Green color component

B

Blue color component

A

Alpha color component

Prefixes

Prefixes are used in the API to denote specific semantic meaning of Vulkan names, or as a label to avoid name clashes, and are explained here:

VK/Vk/vk

Vulkan namespace
All types, commands, enumerants and defines in this specification are prefixed with these two characters.

PFN/pfn

Function Pointer
Denotes that a type is a function pointer, or that a variable is of a pointer type.

p

Pointer
Variable is a pointer.

vkCmd

Commands that record commands in command buffers
These API commands do not result in immediate processing on the device. Instead, they record the requested action in a command buffer for execution when the command buffer is submitted to a queue.

s

Structure
Used to denote the VK_STRUCTURE_TYPE* member of each structure in sType

Appendix G: Credits (Informative)

Vulkan 1.1 is the result of contributions from many people and companies participating in the Khronos Vulkan Working Group, as well as input from the Vulkan Advisory Panel.

Members of the Working Group, including the company that they represented at the time of their most recent contribution, are listed in the following sections. Some specific contributions made by individuals are listed together with their name.

Working Group Contributors to Vulkan 1.1 and 1.0

  • Adam Jackson, Red Hat

  • Alexander Galazin, Arm

  • Alex Bourd, Qualcomm Technologies, Inc.

  • Alon Or-bach, Samsung Electronics (WSI technical sub-group chair)

  • Andrew Garrard, Samsung Electronics (format wrangler)

  • Andrew Woloszyn, Google

  • Antoine Labour, Google

  • Bill Licea-Kane, Qualcomm Technologies, Inc.

  • Cass Everitt, Oculus VR

  • Chad Versace, Google

  • Christophe Riccio, Unity Technologies

  • Dan Baker, Oxide Games

  • Dan Ginsburg, Valve Software

  • Daniel Johnston, Intel

  • Daniel Koch, NVIDIA (Shader Interfaces; Features, Limits, and Formats)

  • Daniel Rakos, AMD

  • David Airlie, Red Hat

  • David Miller, Miller & Mattson (Vulkan reference card)

  • David Neto, Google

  • Dominik Witczak, AMD

  • Graeme Leese, Broadcom

  • Graham Sellers, AMD

  • Ian Romanick, Intel

  • James Jones, NVIDIA

  • Jan-Harald Fredriksen, Arm

  • Jan Hermes, Continental Corporation

  • Jason Ekstrand, Intel

  • Jeff Bolz, NVIDIA (extensive contributions, exhaustive review and rewrites for technical correctness)

  • Jeff Juliano, NVIDIA

  • Jesse Barker, Unity Technologies

  • Jesse Hall, Google

  • Johannes van Waveren, Oculus VR

  • John Kessenich, Google (SPIR-V and GLSL for Vulkan spec author)

  • John McDonald, Valve Software

  • Jonas Gustavsson, Samsung Electronics

  • Jon Ashburn, LunarG

  • Jon Leech, Independent (XML toolchain, normative language, release wrangler)

  • Jungwoo Kim, Samsung Electronics

  • Kathleen Mattson, Miller & Mattson (Vulkan reference card)

  • Kenneth Benzie, Codeplay Software Ltd.

  • Kerch Holt, NVIDIA (SPIR-V technical sub-group chair)

  • Kristian Kristensen, Intel

  • Mark Lobodzinski, LunarG

  • Mathias Heyer, NVIDIA

  • Mathias Schott, NVIDIA

  • Maurice Ribble, Qualcomm Technologies, Inc.

  • Michael Worcester, Imagination Technologies

  • Mika Isojarvi, Google

  • Mitch Singer, AMD

  • Neil Henning, Codeplay Software Ltd.

  • Neil Trevett, NVIDIA

  • Norbert Nopper, Independent

  • Pierre Boudier, NVIDIA

  • Pierre-Loup Griffais, Valve Software

  • Piers Daniell, NVIDIA (dynamic state, copy commands, memory types)

  • Pyry Haulos, Google (Vulkan conformance test subcommittee chair)

  • Ray Smith, Arm

  • Robert Simpson, Qualcomm Technologies, Inc.

  • Rolando Caloca Olivares, Epic Games

  • Sean Harmer, KDAB Group

  • Shannon Woods, Google

  • Slawomir Cygan, Intel

  • Slawomir Grajewski, Intel

  • Stuart Smith, Imagination Technologies

  • Timothy Lottes, AMD

  • Tobias Hector, Imagination Technologies (validity language and toolchain)

  • Tom Olson, Arm (working group chair)

  • Tony Barbour, LunarG

  • Yanjun Zhang, VeriSilicon

Working Group Contributors to Vulkan 1.1

  • Aaron Greig, Codeplay Software Ltd.

  • Aaron Hagan, AMD

  • Alan Ward, Google

  • Alejandro Piñeiro, Igalia

  • Andres Gomez, Igalia

  • Baldur Karlsson, Independent

  • Barthold Lichtenbelt, NVIDIA

  • Bas Nieuwenhuizen, Google

  • Bill Hollings, Brenwill

  • Colin Riley, AMD

  • Cort Stratton, Google

  • Courtney Goeltzenleuchter, Google

  • Dae Kim, Imagination Technologies

  • Daniel Stone, Collabora

  • David Pinedo, LunarG

  • Dejan Mircevski, Google

  • Dzmitry Malyshau, Mozilla

  • Erika Johnson, LunarG

  • Greg Fischer, LunarG

  • Hans-Kristian Arntzen, Arm

  • Iago Toral, Igalia

  • Ian Elliott, Google

  • Jeff Leger, Qualcomm Technologies, Inc.

  • Jeff Vigil, Samsung Electronics

  • Jens Owen, Google

  • Joe Davis, Samsung Electronics

  • John Zulauf, LunarG

  • Jordan Justen, Intel

  • Jörg Wagner, Arm

  • Kalle Raita, Google

  • Karen Ghavam, LunarG

  • Karl Schultz, LunarG

  • Kenneth Russell, Google

  • Kevin O’Neil, AMD

  • Lauri Ilola, Nokia

  • Lenny Komow, LunarG

  • Lionel Landwerlin, Intel

  • Maciej Jesionowski, AMD

  • Mais Alnasser, AMD

  • Marcin Rogucki, Mobica

  • Mark Callow, Independent

  • Mark Kilgard, NVIDIA

  • Markus Tavenrath, NVIDIA

  • Mark Young, LunarG

  • Matthäus Chajdas, AMD

  • Matt Netsch, Qualcomm Technologies, Inc.

  • Michael O’Hara, AMD

  • Michael Wong, Codeplay Software Ltd.

  • Mike Schuchardt, LunarG

  • Mike Weiblen, LunarG

  • Nicolai Hähnle, AMD

  • Nuno Subtil, NVIDIA

  • Patrick Cozzi, Independent

  • Petros Bantolas, Imagination Technologies

  • Ralph Potter, Codeplay Software Ltd.

  • Rob Barris, NVIDIA

  • Ruihao Zhang, Qualcomm Technologies, Inc.

  • Sorel Bosan, AMD

  • Stephen Huang, Mediatek

  • Tilmann Scheller, Samsung Electronics

  • Tomasz Bednarz, Independent

  • Victor Eruhimov, ???

  • Wolfgang Engel, ???

Working Group Contributors to Vulkan 1.0

  • Adam Śmigielski, Mobica

  • Allen Hux, Intel

  • Andrew Cox, Samsung Electronics

  • Andrew Poole, Samsung Electronics

  • Andrew Rafter, Samsung Electronics

  • Andrew Richards, Codeplay Software Ltd.

  • Aras Pranckevičius, Unity Technologies

  • Ashwin Kolhe, NVIDIA

  • Ben Bowman, Imagination Technologies

  • Benj Lipchak

  • Bill Hollings, The Brenwill Workshop

  • Brent E. Insko, Intel

  • Brian Ellis, Qualcomm Technologies, Inc.

  • Cemil Azizoglu, Canonical

  • Chang-Hyo Yu, Samsung Electronics

  • Chia-I Wu, LunarG

  • Chris Frascati, Qualcomm Technologies, Inc.

  • Cody Northrop, LunarG

  • Courtney Goeltzenleuchter, LunarG

  • Damien Leone, NVIDIA

  • David Mao, AMD

  • David Yu, Pixar

  • Frank (LingJun) Chen, Qualcomm Technologies, Inc.

  • Fred Liao, Mediatek

  • Gabe Dagani, Freescale

  • Graham Connor, Imagination Technologies

  • Hwanyong Lee, Kyungpook National University

  • Ian Elliott, LunarG

  • James Hughes, Oculus VR

  • Jeff Vigil, Qualcomm Technologies, Inc.

  • Jens Owen, LunarG

  • Jeremy Hayes, LunarG

  • Jonathan Hamilton, Imagination Technologies

  • Krzysztof Iwanicki, Samsung Electronics

  • Larry Seiler, Intel

  • Lutz Latta, Lucasfilm

  • Maria Rovatsou, Codeplay Software Ltd.

  • Mark Callow

  • Mateusz Przybylski, Intel

  • Maxim Lukyanov, Samsung Electronics

  • Michael Lentine, Google

  • Michal Pietrasiuk, Intel

  • Mike Stroyan, LunarG

  • Minyoung Son, Samsung Electronics

  • Mythri Venugopal, Samsung Electronics

  • Naveen Leekha, Google

  • Nick Penwarden, Epic Games

  • Niklas Smedberg, Unity Technologies

  • Pat Brown, NVIDIA

  • Patrick Doane, Blizzard Entertainment

  • Peter Lohrmann, Valve Software

  • Piotr Bialecki, Intel

  • Prabindh Sundareson, Samsung Electronics

  • Rob Stepinski, Transgaming

  • Roy Ju, Mediatek

  • Rufus Hamede, Imagination Technologies

  • Sean Ellis, Arm

  • Stefanus Du Toit, Google

  • Steve Hill, Broadcom

  • Steve Viggers, Core Avionics & Industrial Inc.

  • Tim Foley, Intel

  • Timo Suoranta, AMD

  • Tobin Ehlis, LunarG

  • Tomasz Kubale, Intel

  • Wayne Lister, Imagination Technologies

Other Credits

In addition to the Working Group, the Vulkan Advisory Panel members provided important real-world usage information and advice that helped guide design decisions.

Administrative support to the Working Group for Vulkan 1.1 was provided by Khronos staff including Angela Cheng, Ann Thorsnes, Emily Stearns, Liz Maitral, and Dominic Agoro-Ombaka; and by Alex Crabb of Caster Communications.

Administrative support for Vulkan 1.0 was provided by Andrew Riegel, Elizabeth Riegel, Glenn Fredericks, Kathleen Mattson and Michelle Clark of Gold Standard Group.

Technical support was provided by James Riordon, webmaster of Khronos.org and OpenGL.org.