Vulkan ® 1.2.148 - A Specification (with all registered extensions)

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY

To create an instance object, call:

// Provided by VK_VERSION_1_0
VkResult vkCreateInstance(
    const VkInstanceCreateInfo*                 pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkInstance*                                 pInstance);

pCreateInfo is a pointer to a VkInstanceCreateInfo structure 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

All required extensions for each extension in the VkInstanceCreateInfo::ppEnabledExtensionNames list must also be present in that list

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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is reserved for future use.
pApplicationInfo is NULL or a pointer to a VkApplicationInfo structure. 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. The layers are loaded in the order they are listed in this array, with the first array element being the closest to the application, and the last array element being the closest to the driver. 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, VkValidationFeaturesEXT, or VkValidationFlagsEXT
The sType value of each struct 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

// Provided by VK_VERSION_1_0
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.

// Provided by VK_EXT_validation_flags
typedef struct VkValidationFlagsEXT {
    VkStructureType                sType;
    const void*                    pNext;
    uint32_t                       disabledValidationCheckCount;
    const VkValidationCheckEXT*    pDisabledValidationChecks;
} VkValidationFlagsEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 valid VkValidationCheckEXT values
disabledValidationCheckCount must be greater than 0

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

// Provided by VK_EXT_validation_flags
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.

When creating a Vulkan instance for which you wish to enable or disable specific validation features, add a VkValidationFeaturesEXT structure to the pNext chain of the VkInstanceCreateInfo structure, specifying the features to be enabled or disabled.

// Provided by VK_EXT_validation_features
typedef struct VkValidationFeaturesEXT {
    VkStructureType                         sType;
    const void*                             pNext;
    uint32_t                                enabledValidationFeatureCount;
    const VkValidationFeatureEnableEXT*     pEnabledValidationFeatures;
    uint32_t                                disabledValidationFeatureCount;
    const VkValidationFeatureDisableEXT*    pDisabledValidationFeatures;
} VkValidationFeaturesEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
enabledValidationFeatureCount is the number of features to enable.
pEnabledValidationFeatures is a pointer to an array of VkValidationFeatureEnableEXT values specifying the validation features to be enabled.
disabledValidationFeatureCount is the number of features to disable.
pDisabledValidationFeatures is a pointer to an array of VkValidationFeatureDisableEXT values specifying the validation features to be disabled.

Valid Usage

If the pEnabledValidationFeatures array contains VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_RESERVE_BINDING_SLOT_EXT, then it must also contain VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_EXT
If the pEnabledValidationFeatures array contains VK_VALIDATION_FEATURE_ENABLE_DEBUG_PRINTF_EXT, then it must not contain VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_EXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_VALIDATION_FEATURES_EXT
If enabledValidationFeatureCount is not 0, pEnabledValidationFeatures must be a valid pointer to an array of enabledValidationFeatureCount valid VkValidationFeatureEnableEXT values
If disabledValidationFeatureCount is not 0, pDisabledValidationFeatures must be a valid pointer to an array of disabledValidationFeatureCount valid VkValidationFeatureDisableEXT values

Possible values of elements of the VkValidationFeaturesEXT::pEnabledValidationFeatures array, specifying validation features to be enabled, are:

// Provided by VK_EXT_validation_features
typedef enum VkValidationFeatureEnableEXT {
    VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_EXT = 0,
    VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_RESERVE_BINDING_SLOT_EXT = 1,
    VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT = 2,
    VK_VALIDATION_FEATURE_ENABLE_DEBUG_PRINTF_EXT = 3,
} VkValidationFeatureEnableEXT;

VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_EXT specifies that GPU-assisted validation is enabled. Activating this feature instruments shader programs to generate additional diagnostic data. This feature is disabled by default.
VK_VALIDATION_FEATURE_ENABLE_GPU_ASSISTED_RESERVE_BINDING_SLOT_EXT specifies that the validation layers reserve a descriptor set binding slot for their own use. The layer reports a value for VkPhysicalDeviceLimits::maxBoundDescriptorSets that is one less than the value reported by the device. If the device supports the binding of only one descriptor set, the validation layer does not perform GPU-assisted validation. This feature is disabled by default.
VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT specifies that Vulkan best-practices validation is enabled. Activating this feature enables the output of warnings related to common misuse of the API, but which are not explicitly prohibited by the specification. This feature is disabled by default.
VK_VALIDATION_FEATURE_ENABLE_DEBUG_PRINTF_EXT specifies that the layers will process debugPrintfEXT operations in shaders and send the resulting output to the debug callback. This feature is disabled by default.

Possible values of elements of the VkValidationFeaturesEXT::pDisabledValidationFeatures array, specifying validation features to be disabled, are:

// Provided by VK_EXT_validation_features
typedef enum VkValidationFeatureDisableEXT {
    VK_VALIDATION_FEATURE_DISABLE_ALL_EXT = 0,
    VK_VALIDATION_FEATURE_DISABLE_SHADERS_EXT = 1,
    VK_VALIDATION_FEATURE_DISABLE_THREAD_SAFETY_EXT = 2,
    VK_VALIDATION_FEATURE_DISABLE_API_PARAMETERS_EXT = 3,
    VK_VALIDATION_FEATURE_DISABLE_OBJECT_LIFETIMES_EXT = 4,
    VK_VALIDATION_FEATURE_DISABLE_CORE_CHECKS_EXT = 5,
    VK_VALIDATION_FEATURE_DISABLE_UNIQUE_HANDLES_EXT = 6,
} VkValidationFeatureDisableEXT;

VK_VALIDATION_FEATURE_DISABLE_ALL_EXT specifies that all validation checks are disabled.
VK_VALIDATION_FEATURE_DISABLE_SHADERS_EXT specifies that shader validation is disabled. This feature is enabled by default.
VK_VALIDATION_FEATURE_DISABLE_THREAD_SAFETY_EXT specifies that thread safety validation is disabled. This feature is enabled by default.
VK_VALIDATION_FEATURE_DISABLE_API_PARAMETERS_EXT specifies that stateless parameter validation is disabled. This feature is enabled by default.
VK_VALIDATION_FEATURE_DISABLE_OBJECT_LIFETIMES_EXT specifies that object lifetime validation is disabled. This feature is enabled by default.
VK_VALIDATION_FEATURE_DISABLE_CORE_CHECKS_EXT specifies that core validation checks are disabled. This feature is enabled by default. If this feature is disabled, the shader validation and GPU-assisted validation features are also disabled.
VK_VALIDATION_FEATURE_DISABLE_UNIQUE_HANDLES_EXT specifies that protection against duplicate non-dispatchable object handles is disabled. This feature is enabled by default.

Note

Disabling checks such as parameter validation and object lifetime validation prevents the reporting of error conditions that can cause other validation checks to behave incorrectly or crash. Some validation checks assume that their inputs are already valid and do not always revalidate them.

Note

The VK_EXT_validation_features extension subsumes all the functionality provided in the VK_EXT_validation_flags extension.

The VkApplicationInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 Version Numbers. 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.

As long as the instance supports at least Vulkan 1.1, an application can use different versions of Vulkan with an instance than it does with a device or physical device.

Note

The Khronos validation layers will treat apiVersion as the highest API version the application targets, and will validate API usage against the minimum of that version and the implementation version (instance or device, depending on context). If an application tries to use functionality from a greater version than this, a validation error will be triggered.

For example, if the instance supports Vulkan 1.1 and three physical devices support Vulkan 1.0, Vulkan 1.1, and Vulkan 1.2, respectively, and if the application sets apiVersion to 1.2, the application can use the following versions of Vulkan:

Vulkan 1.0 can be used with the instance and with all physical devices.
Vulkan 1.1 can be used with the instance and with the physical devices that support Vulkan 1.1 and Vulkan 1.2.
Vulkan 1.2 can be used with the physical device that supports Vulkan 1.2.

If we modify the above example so that the application sets apiVersion to 1.1, then the application must not use Vulkan 1.2 functionality on the physical device that supports Vulkan 1.2.

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

If apiVersion is not 0, then it must be greater or equal to VK_API_VERSION_1_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:

// Provided by VK_VERSION_1_0
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
Host access to all VkPhysicalDevice objects enumerated from 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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS
VK_INCOMPLETE

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:

// Provided by VK_VERSION_1_0
void vkGetPhysicalDeviceProperties(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties*                 pProperties);

physicalDevice is the handle to the physical device whose properties will be queried.
pProperties is a pointer to a VkPhysicalDeviceProperties structure in which properties are returned.

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:

// Provided by VK_VERSION_1_0
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 Version Numbers.
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 an array of VK_MAX_PHYSICAL_DEVICE_NAME_SIZE char containing a null-terminated UTF-8 string which is the name of the device.
pipelineCacheUUID is an array of VK_UUID_SIZE uint8_t values representing a universally unique identifier for the device.
limits is the VkPhysicalDeviceLimits structure specifying device-specific limits of the physical device. See Limits for details.
sparseProperties is the VkPhysicalDeviceSparseProperties structure specifying 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: Procedures and Conventions 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. Khronos vendor IDs are symbolically defined in the VkVendorId type.

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.

Khronos vendor IDs which may be returned in VkPhysicalDeviceProperties::vendorID are:

// Provided by VK_VERSION_1_0
typedef enum VkVendorId {
    VK_VENDOR_ID_VIV = 0x10001,
    VK_VENDOR_ID_VSI = 0x10002,
    VK_VENDOR_ID_KAZAN = 0x10003,
    VK_VENDOR_ID_CODEPLAY = 0x10004,
    VK_VENDOR_ID_MESA = 0x10005,
} VkVendorId;

Note

Khronos vendor IDs may be allocated by vendors at any time. Only the latest canonical versions of this Specification, of the corresponding vk.xml API Registry, and of the corresponding vulkan_core.h header file must contain all reserved Khronos vendor IDs.

Only Khronos vendor IDs are given symbolic names at present. PCI vendor IDs returned by the implementation can be looked up in the PCI-SIG database.

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

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_1
void vkGetPhysicalDeviceProperties2(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties2*                pProperties);

or the equivalent command

// Provided by VK_KHR_get_physical_device_properties2
void vkGetPhysicalDeviceProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceProperties2*                pProperties);

physicalDevice is the handle to the physical device whose properties will be queried.
pProperties is a pointer to a VkPhysicalDeviceProperties2 structure in which properties are returned.

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:

// Provided by VK_VERSION_1_1
typedef struct VkPhysicalDeviceProperties2 {
    VkStructureType               sType;
    void*                         pNext;
    VkPhysicalDeviceProperties    properties;
} VkPhysicalDeviceProperties2;

or the equivalent

// Provided by VK_KHR_get_physical_device_properties2
typedef VkPhysicalDeviceProperties2 VkPhysicalDeviceProperties2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
properties is a VkPhysicalDeviceProperties structure describing 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.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PROPERTIES_2
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 VkPhysicalDeviceBlendOperationAdvancedPropertiesEXT, VkPhysicalDeviceConservativeRasterizationPropertiesEXT, VkPhysicalDeviceCooperativeMatrixPropertiesNV, VkPhysicalDeviceCustomBorderColorPropertiesEXT, VkPhysicalDeviceDepthStencilResolveProperties, VkPhysicalDeviceDescriptorIndexingProperties, VkPhysicalDeviceDeviceGeneratedCommandsPropertiesNV, VkPhysicalDeviceDiscardRectanglePropertiesEXT, VkPhysicalDeviceDriverProperties, VkPhysicalDeviceExternalMemoryHostPropertiesEXT, VkPhysicalDeviceFloatControlsProperties, VkPhysicalDeviceFragmentDensityMap2PropertiesEXT, VkPhysicalDeviceFragmentDensityMapPropertiesEXT, VkPhysicalDeviceIDProperties, VkPhysicalDeviceInlineUniformBlockPropertiesEXT, VkPhysicalDeviceLineRasterizationPropertiesEXT, VkPhysicalDeviceMaintenance3Properties, VkPhysicalDeviceMeshShaderPropertiesNV, VkPhysicalDeviceMultiviewPerViewAttributesPropertiesNVX, VkPhysicalDeviceMultiviewProperties, VkPhysicalDevicePCIBusInfoPropertiesEXT, VkPhysicalDevicePerformanceQueryPropertiesKHR, VkPhysicalDevicePointClippingProperties, VkPhysicalDeviceProtectedMemoryProperties, VkPhysicalDevicePushDescriptorPropertiesKHR, VkPhysicalDeviceRayTracingPropertiesKHR, VkPhysicalDeviceRayTracingPropertiesNV, VkPhysicalDeviceRobustness2PropertiesEXT, VkPhysicalDeviceSampleLocationsPropertiesEXT, VkPhysicalDeviceSamplerFilterMinmaxProperties, VkPhysicalDeviceShaderCoreProperties2AMD, VkPhysicalDeviceShaderCorePropertiesAMD, VkPhysicalDeviceShaderSMBuiltinsPropertiesNV, VkPhysicalDeviceShadingRateImagePropertiesNV, VkPhysicalDeviceSubgroupProperties, VkPhysicalDeviceSubgroupSizeControlPropertiesEXT, VkPhysicalDeviceTexelBufferAlignmentPropertiesEXT, VkPhysicalDeviceTimelineSemaphoreProperties, VkPhysicalDeviceTransformFeedbackPropertiesEXT, VkPhysicalDeviceVertexAttributeDivisorPropertiesEXT, VkPhysicalDeviceVulkan11Properties, or VkPhysicalDeviceVulkan12Properties
The sType value of each struct in the pNext chain must be unique

To query the properties of the driver corresponding to Vulkan 1.1 functionality, add VkPhysicalDeviceVulkan11Properties to the pNext chain of the VkPhysicalDeviceProperties2 structure.

The VkPhysicalDeviceVulkan11Properties structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkPhysicalDeviceVulkan11Properties {
    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;
    uint32_t                   subgroupSize;
    VkShaderStageFlags         subgroupSupportedStages;
    VkSubgroupFeatureFlags     subgroupSupportedOperations;
    VkBool32                   subgroupQuadOperationsInAllStages;
    VkPointClippingBehavior    pointClippingBehavior;
    uint32_t                   maxMultiviewViewCount;
    uint32_t                   maxMultiviewInstanceIndex;
    VkBool32                   protectedNoFault;
    uint32_t                   maxPerSetDescriptors;
    VkDeviceSize               maxMemoryAllocationSize;
} VkPhysicalDeviceVulkan11Properties;

deviceUUID is an array of VK_UUID_SIZE uint8_t values representing a universally unique identifier for the device.
driverUUID is an array of VK_UUID_SIZE uint8_t values representing a universally unique identifier for the driver build in use by the device.
deviceLUID is an array of VK_LUID_SIZE uint8_t values representing a locally unique identifier for the device.
deviceNodeMask is a uint32_t 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.
subgroupSize is the default number of invocations in each subgroup. subgroupSize is at least 1 if any of the physical device’s queues support VK_QUEUE_GRAPHICS_BIT or VK_QUEUE_COMPUTE_BIT. subgroupSize is a power-of-two.
subgroupSupportedStages is a bitfield of VkShaderStageFlagBits describing the shader stages that subgroup operations are supported in. subgroupSupportedStages will have the VK_SHADER_STAGE_COMPUTE_BIT bit set if any of the physical device’s queues support VK_QUEUE_COMPUTE_BIT.
subgroupSupportedOperations is a bitmask of VkSubgroupFeatureFlagBits specifying the sets of subgroup operations supported on this device. subgroupSupportedOperations 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.
subgroupQuadOperationsInAllStages is a boolean specifying whether quad subgroup operations are available in all stages, or are restricted to fragment and compute stages.
pointClippingBehavior is a VkPointClippingBehavior value specifying the point clipping behavior supported by the implementation.
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.
protectedNoFault specifies the behavior of the implementation when protected memory access rules are broken. If protectedNoFault is VK_TRUE, breaking those rules will not result in process termination or device loss.
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.

The members of VkPhysicalDeviceVulkan11Properties must have the same values as the corresponding members of VkPhysicalDeviceIDProperties, VkPhysicalDeviceSubgroupProperties, VkPhysicalDevicePointClippingProperties, VkPhysicalDeviceMultiviewProperties, VkPhysicalDeviceProtectedMemoryProperties, and VkPhysicalDeviceMaintenance3Properties.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VULKAN_1_1_PROPERTIES

To query the properties of the driver corresponding to Vulkan 1.2 functionality, add VkPhysicalDeviceVulkan12Properties to the pNext chain of the VkPhysicalDeviceProperties2 structure.

The VkPhysicalDeviceVulkan12Properties structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkPhysicalDeviceVulkan12Properties {
    VkStructureType                      sType;
    void*                                pNext;
    VkDriverId                           driverID;
    char                                 driverName[VK_MAX_DRIVER_NAME_SIZE];
    char                                 driverInfo[VK_MAX_DRIVER_INFO_SIZE];
    VkConformanceVersion                 conformanceVersion;
    VkShaderFloatControlsIndependence    denormBehaviorIndependence;
    VkShaderFloatControlsIndependence    roundingModeIndependence;
    VkBool32                             shaderSignedZeroInfNanPreserveFloat16;
    VkBool32                             shaderSignedZeroInfNanPreserveFloat32;
    VkBool32                             shaderSignedZeroInfNanPreserveFloat64;
    VkBool32                             shaderDenormPreserveFloat16;
    VkBool32                             shaderDenormPreserveFloat32;
    VkBool32                             shaderDenormPreserveFloat64;
    VkBool32                             shaderDenormFlushToZeroFloat16;
    VkBool32                             shaderDenormFlushToZeroFloat32;
    VkBool32                             shaderDenormFlushToZeroFloat64;
    VkBool32                             shaderRoundingModeRTEFloat16;
    VkBool32                             shaderRoundingModeRTEFloat32;
    VkBool32                             shaderRoundingModeRTEFloat64;
    VkBool32                             shaderRoundingModeRTZFloat16;
    VkBool32                             shaderRoundingModeRTZFloat32;
    VkBool32                             shaderRoundingModeRTZFloat64;
    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;
    VkResolveModeFlags                   supportedDepthResolveModes;
    VkResolveModeFlags                   supportedStencilResolveModes;
    VkBool32                             independentResolveNone;
    VkBool32                             independentResolve;
    VkBool32                             filterMinmaxSingleComponentFormats;
    VkBool32                             filterMinmaxImageComponentMapping;
    uint64_t                             maxTimelineSemaphoreValueDifference;
    VkSampleCountFlags                   framebufferIntegerColorSampleCounts;
} VkPhysicalDeviceVulkan12Properties;

driverID is a unique identifier for the driver of the physical device.
driverName is an array of VK_MAX_DRIVER_NAME_SIZE char containing a null-terminated UTF-8 string which is the name of the driver.
driverInfo is an array of VK_MAX_DRIVER_INFO_SIZE char containing a null-terminated UTF-8 string with additional information about the driver.
conformanceVersion is the version of the Vulkan conformance test this driver is conformant against (see VkConformanceVersion).
denormBehaviorIndependence is a VkShaderFloatControlsIndependence value indicating whether, and how, denorm behavior can be set independently for different bit widths.
roundingModeIndependence is a VkShaderFloatControlsIndependence value indicating whether, and how, rounding modes can be set independently for different bit widths.
shaderSignedZeroInfNanPreserveFloat16 is a boolean value indicating whether sign of a zero, Nans and $\pm \infty$ can be preserved in 16-bit floating-point computations. It also indicates whether the SignedZeroInfNanPreserve execution mode can be used for 16-bit floating-point types.
shaderSignedZeroInfNanPreserveFloat32 is a boolean value indicating whether sign of a zero, Nans and $\pm \infty$ can be preserved in 32-bit floating-point computations. It also indicates whether the SignedZeroInfNanPreserve execution mode can be used for 32-bit floating-point types.
shaderSignedZeroInfNanPreserveFloat64 is a boolean value indicating whether sign of a zero, Nans and $\pm \infty$ can be preserved in 64-bit floating-point computations. It also indicates whether the SignedZeroInfNanPreserve execution mode can be used for 64-bit floating-point types.
shaderDenormPreserveFloat16 is a boolean value indicating whether denormals can be preserved in 16-bit floating-point computations. It also indicates whether the DenormPreserve execution mode can be used for 16-bit floating-point types.
shaderDenormPreserveFloat32 is a boolean value indicating whether denormals can be preserved in 32-bit floating-point computations. It also indicates whether the DenormPreserve execution mode can be used for 32-bit floating-point types.
shaderDenormPreserveFloat64 is a boolean value indicating whether denormals can be preserved in 64-bit floating-point computations. It also indicates whether the DenormPreserve execution mode can be used for 64-bit floating-point types.
shaderDenormFlushToZeroFloat16 is a boolean value indicating whether denormals can be flushed to zero in 16-bit floating-point computations. It also indicates whether the DenormFlushToZero execution mode can be used for 16-bit floating-point types.
shaderDenormFlushToZeroFloat32 is a boolean value indicating whether denormals can be flushed to zero in 32-bit floating-point computations. It also indicates whether the DenormFlushToZero execution mode can be used for 32-bit floating-point types.
shaderDenormFlushToZeroFloat64 is a boolean value indicating whether denormals can be flushed to zero in 64-bit floating-point computations. It also indicates whether the DenormFlushToZero execution mode can be used for 64-bit floating-point types.
shaderRoundingModeRTEFloat16 is a boolean value indicating whether an implementation supports the round-to-nearest-even rounding mode for 16-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTE execution mode can be used for 16-bit floating-point types.
shaderRoundingModeRTEFloat32 is a boolean value indicating whether an implementation supports the round-to-nearest-even rounding mode for 32-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTE execution mode can be used for 32-bit floating-point types.
shaderRoundingModeRTEFloat64 is a boolean value indicating whether an implementation supports the round-to-nearest-even rounding mode for 64-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTE execution mode can be used for 64-bit floating-point types.
shaderRoundingModeRTZFloat16 is a boolean value indicating whether an implementation supports the round-towards-zero rounding mode for 16-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTZ execution mode can be used for 16-bit floating-point types.
shaderRoundingModeRTZFloat32 is a boolean value indicating whether an implementation supports the round-towards-zero rounding mode for 32-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTZ execution mode can be used for 32-bit floating-point types.
shaderRoundingModeRTZFloat64 is a boolean value indicating whether an implementation supports the round-towards-zero rounding mode for 64-bit floating-point arithmetic and conversion instructions. It also indicates whether the RoundingModeRTZ execution mode can be used for 64-bit floating-point types.
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 bit set. Pool creation may fail when this limit is exceeded, or when the space this limit represents is unable to 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 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 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 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 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 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 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 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 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 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 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 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 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 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 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 bit set.
supportedDepthResolveModes is a bitmask of VkResolveModeFlagBits indicating the set of supported depth resolve modes. VK_RESOLVE_MODE_SAMPLE_ZERO_BIT must be included in the set but implementations may support additional modes.
supportedStencilResolveModes is a bitmask of VkResolveModeFlagBits indicating the set of supported stencil resolve modes. VK_RESOLVE_MODE_SAMPLE_ZERO_BIT must be included in the set but implementations may support additional modes. VK_RESOLVE_MODE_AVERAGE_BIT must not be included in the set.
independentResolveNone is VK_TRUE if the implementation supports setting the depth and stencil resolve modes to different values when one of those modes is VK_RESOLVE_MODE_NONE. Otherwise the implementation only supports setting both modes to the same value.
independentResolve is VK_TRUE if the implementation supports all combinations of the supported depth and stencil resolve modes, including setting either depth or stencil resolve mode to VK_RESOLVE_MODE_NONE. An implementation that supports independentResolve must also support independentResolveNone.
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.
maxTimelineSemaphoreValueDifference indicates the maximum difference allowed by the implementation between the current value of a timeline semaphore and any pending signal or wait operations.
framebufferIntegerColorSampleCounts is a bitmask of VkSampleCountFlagBits indicating the color sample counts that are supported for all framebuffer color attachments with integer formats.

The members of VkPhysicalDeviceVulkan12Properties must have the same values as the corresponding members of VkPhysicalDeviceDriverProperties, VkPhysicalDeviceFloatControlsProperties, VkPhysicalDeviceDescriptorIndexingProperties, VkPhysicalDeviceDepthStencilResolveProperties, VkPhysicalDeviceSamplerFilterMinmaxProperties, and VkPhysicalDeviceTimelineSemaphoreProperties.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_VULKAN_1_2_PROPERTIES

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

// Provided by VK_VERSION_1_1
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

// Provided by VK_KHR_external_memory_capabilities, VK_KHR_external_semaphore_capabilities, VK_KHR_external_fence_capabilities
typedef VkPhysicalDeviceIDProperties VkPhysicalDeviceIDPropertiesKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.

deviceUUID is an array of VK_UUID_SIZE uint8_t values representing a universally unique identifier for the device.
driverUUID is an array of VK_UUID_SIZE uint8_t values representing a universally unique identifier for the driver build in use by the device.
deviceLUID is an array of VK_LUID_SIZE uint8_t values representing a locally unique identifier for the device.
deviceNodeMask is a uint32_t 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:

External memory handle types compatibility
External semaphore handle types compatibility
External fence handle types compatibility

If deviceLUIDValid is VK_FALSE, the values 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.

Implementations should return deviceUUID values which are likely to be unique even in the presence of multiple Vulkan implementations (such as a GPU driver and a software renderer; two drivers for different GPUs; or the same Vulkan driver running on two logically different devices).

Khronos' conformance testing can not guarantee that deviceUUID values are actually unique, so implementers should make their own best efforts to ensure this. In particular, hard-coded deviceUUID values, especially all-0 bits, should never be used.

A combination of values unique to the vendor, the driver, and the hardware environment can be used to provide a deviceUUID which is unique to a high degree of certainty. Some possible inputs to such a computation are:

Information reported by vkGetPhysicalDeviceProperties
PCI device ID (if defined)
PCI bus ID, or similar system configuration information.
Driver binary checksums.

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 the properties of the driver corresponding to a physical device, add a VkPhysicalDeviceDriverProperties structure to the pNext chain of the VkPhysicalDeviceProperties2 structure. The VkPhysicalDeviceDriverProperties structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkPhysicalDeviceDriverProperties {
    VkStructureType         sType;
    void*                   pNext;
    VkDriverId              driverID;
    char                    driverName[VK_MAX_DRIVER_NAME_SIZE];
    char                    driverInfo[VK_MAX_DRIVER_INFO_SIZE];
    VkConformanceVersion    conformanceVersion;
} VkPhysicalDeviceDriverProperties;

or the equivalent

// Provided by VK_KHR_driver_properties
typedef VkPhysicalDeviceDriverProperties VkPhysicalDeviceDriverPropertiesKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.

driverID is a unique identifier for the driver of the physical device.
driverName is an array of VK_MAX_DRIVER_NAME_SIZE char containing a null-terminated UTF-8 string which is the name of the driver.
driverInfo is an array of VK_MAX_DRIVER_INFO_SIZE char containing a null-terminated UTF-8 string with additional information about the driver.
conformanceVersion is the version of the Vulkan conformance test this driver is conformant against (see VkConformanceVersion).

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

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_DRIVER_PROPERTIES

Khronos driver IDs which may be returned in VkPhysicalDeviceDriverProperties::driverID are:

// Provided by VK_VERSION_1_2
typedef enum VkDriverId {
    VK_DRIVER_ID_AMD_PROPRIETARY = 1,
    VK_DRIVER_ID_AMD_OPEN_SOURCE = 2,
    VK_DRIVER_ID_MESA_RADV = 3,
    VK_DRIVER_ID_NVIDIA_PROPRIETARY = 4,
    VK_DRIVER_ID_INTEL_PROPRIETARY_WINDOWS = 5,
    VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA = 6,
    VK_DRIVER_ID_IMAGINATION_PROPRIETARY = 7,
    VK_DRIVER_ID_QUALCOMM_PROPRIETARY = 8,
    VK_DRIVER_ID_ARM_PROPRIETARY = 9,
    VK_DRIVER_ID_GOOGLE_SWIFTSHADER = 10,
    VK_DRIVER_ID_GGP_PROPRIETARY = 11,
    VK_DRIVER_ID_BROADCOM_PROPRIETARY = 12,
    VK_DRIVER_ID_MESA_LLVMPIPE = 13,
    VK_DRIVER_ID_MOLTENVK = 14,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_AMD_PROPRIETARY_KHR = VK_DRIVER_ID_AMD_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_AMD_OPEN_SOURCE_KHR = VK_DRIVER_ID_AMD_OPEN_SOURCE,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_MESA_RADV_KHR = VK_DRIVER_ID_MESA_RADV,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_NVIDIA_PROPRIETARY_KHR = VK_DRIVER_ID_NVIDIA_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_INTEL_PROPRIETARY_WINDOWS_KHR = VK_DRIVER_ID_INTEL_PROPRIETARY_WINDOWS,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA_KHR = VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_IMAGINATION_PROPRIETARY_KHR = VK_DRIVER_ID_IMAGINATION_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_QUALCOMM_PROPRIETARY_KHR = VK_DRIVER_ID_QUALCOMM_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_ARM_PROPRIETARY_KHR = VK_DRIVER_ID_ARM_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_GOOGLE_SWIFTSHADER_KHR = VK_DRIVER_ID_GOOGLE_SWIFTSHADER,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_GGP_PROPRIETARY_KHR = VK_DRIVER_ID_GGP_PROPRIETARY,
  // Provided by VK_KHR_driver_properties
    VK_DRIVER_ID_BROADCOM_PROPRIETARY_KHR = VK_DRIVER_ID_BROADCOM_PROPRIETARY,
} VkDriverId;

or the equivalent

// Provided by VK_KHR_driver_properties
typedef VkDriverId VkDriverIdKHR;

Note

Khronos driver IDs may be allocated by vendors at any time. There may be multiple driver IDs for the same vendor, representing different drivers (for e.g. different platforms, proprietary or open source, etc.). Only the latest canonical versions of this Specification, of the corresponding vk.xml API Registry, and of the corresponding vulkan_core.h header file must contain all reserved Khronos driver IDs.

Only driver IDs registered with Khronos are given symbolic names. There may be unregistered driver IDs returned.

The conformance test suite version an implementation is compliant with is described with the VkConformanceVersion structure:

// Provided by VK_VERSION_1_2
typedef struct VkConformanceVersion {
    uint8_t    major;
    uint8_t    minor;
    uint8_t    subminor;
    uint8_t    patch;
} VkConformanceVersion;

or the equivalent

// Provided by VK_KHR_driver_properties
typedef VkConformanceVersion VkConformanceVersionKHR;

major is the major version number of the conformance test suite.
minor is the minor version number of the conformance test suite.
subminor is the subminor version number of the conformance test suite.
patch is the patch version number of the conformance test suite.

To query the PCI bus information of a physical device, add a VkPhysicalDevicePCIBusInfoPropertiesEXT structure to the pNext chain of the VkPhysicalDeviceProperties2 structure. The VkPhysicalDevicePCIBusInfoPropertiesEXT structure is defined as:

// Provided by VK_EXT_pci_bus_info
typedef struct VkPhysicalDevicePCIBusInfoPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           pciDomain;
    uint32_t           pciBus;
    uint32_t           pciDevice;
    uint32_t           pciFunction;
} VkPhysicalDevicePCIBusInfoPropertiesEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
pciDomain is the PCI bus domain.
pciBus is the PCI bus identifier.
pciDevice is the PCI device identifier.
pciFunction is the PCI device function identifier.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_PCI_BUS_INFO_PROPERTIES_EXT

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

// Provided by VK_VERSION_1_0
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. Implementations must support at least one queue family. 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:

// Provided by VK_VERSION_1_0
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. Each queue family must support at least one queue.
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.
(A_x, A_y, A_z) where A_x, A_y, and A_z 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 A_x, A_y, and A_z, respectively.
- width of a VkExtent3D parameter must be an integer multiple of A_x, 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 A_y, 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 A_z, 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:

// Provided by VK_VERSION_1_0
typedef enum VkQueueFlagBits {
    VK_QUEUE_GRAPHICS_BIT = 0x00000001,
    VK_QUEUE_COMPUTE_BIT = 0x00000002,
    VK_QUEUE_TRANSFER_BIT = 0x00000004,
    VK_QUEUE_SPARSE_BINDING_BIT = 0x00000008,
  // Provided by VK_VERSION_1_1
    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.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_1
void vkGetPhysicalDeviceQueueFamilyProperties2(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pQueueFamilyPropertyCount,
    VkQueueFamilyProperties2*                   pQueueFamilyProperties);

or the equivalent command

// Provided by VK_KHR_get_physical_device_properties2
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:

// Provided by VK_VERSION_1_1
typedef struct VkQueueFamilyProperties2 {
    VkStructureType            sType;
    void*                      pNext;
    VkQueueFamilyProperties    queueFamilyProperties;
} VkQueueFamilyProperties2;

or the equivalent

// Provided by VK_KHR_get_physical_device_properties2
typedef VkQueueFamilyProperties2 VkQueueFamilyProperties2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
queueFamilyProperties is a VkQueueFamilyProperties structure which is populated with the same values as in vkGetPhysicalDeviceQueueFamilyProperties.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_QUEUE_FAMILY_PROPERTIES_2
pNext must be NULL or a pointer to a valid instance of VkQueueFamilyCheckpointPropertiesNV
The sType value of each struct in the pNext chain must be unique

Additional queue family information can be queried by setting VkQueueFamilyProperties2::pNext to point to a VkQueueFamilyCheckpointPropertiesNV structure.

The VkQueueFamilyCheckpointPropertiesNV structure is defined as:

// Provided by VK_NV_device_diagnostic_checkpoints
typedef struct VkQueueFamilyCheckpointPropertiesNV {
    VkStructureType         sType;
    void*                   pNext;
    VkPipelineStageFlags    checkpointExecutionStageMask;
} VkQueueFamilyCheckpointPropertiesNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
checkpointExecutionStageMask is a mask indicating which pipeline stages the implementation can execute checkpoint markers in.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_QUEUE_FAMILY_CHECKPOINT_PROPERTIES_NV

To enumerate the performance query counters available on a queue family of a physical device, call:

// Provided by VK_KHR_performance_query
VkResult vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR(
    VkPhysicalDevice                            physicalDevice,
    uint32_t                                    queueFamilyIndex,
    uint32_t*                                   pCounterCount,
    VkPerformanceCounterKHR*                    pCounters,
    VkPerformanceCounterDescriptionKHR*         pCounterDescriptions);

physicalDevice is the handle to the physical device whose queue family performance query counter properties will be queried.
queueFamilyIndex is the index into the queue family of the physical device we want to get properties for.
pCounterCount is a pointer to an integer related to the number of counters available or queried, as described below.
pCounters is either NULL or a pointer to an array of VkPerformanceCounterKHR structures.
pCounterDescriptions is either NULL or a pointer to an array of VkPerformanceCounterDescriptionKHR structures.

If pCounters is NULL and pCounterDescriptions is NULL, then the number of counters available is returned in pCounterCount. Otherwise, pCounterCount must point to a variable set by the user to the number of elements in the pCounters, pCounterDescriptions, or both arrays and on return the variable is overwritten with the number of structures actually written out. If pCounterCount is less than the number of counters available, at most pCounterCount structures will be written and VK_INCOMPLETE will be returned instead of VK_SUCCESS.

Valid Usage (Implicit)

physicalDevice must be a valid VkPhysicalDevice handle
pCounterCount must be a valid pointer to a uint32_t value
If the value referenced by pCounterCount is not 0, and pCounters is not NULL, pCounters must be a valid pointer to an array of pCounterCount VkPerformanceCounterKHR structures
If the value referenced by pCounterCount is not 0, and pCounterDescriptions is not NULL, pCounterDescriptions must be a valid pointer to an array of pCounterCount VkPerformanceCounterDescriptionKHR structures

Return Codes

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INITIALIZATION_FAILED

The VkPerformanceCounterKHR structure is defined as:

// Provided by VK_KHR_performance_query
typedef struct VkPerformanceCounterKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkPerformanceCounterUnitKHR       unit;
    VkPerformanceCounterScopeKHR      scope;
    VkPerformanceCounterStorageKHR    storage;
    uint8_t                           uuid[VK_UUID_SIZE];
} VkPerformanceCounterKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
unit is a VkPerformanceCounterUnitKHR specifying the unit that the counter data will record.
scope is a VkPerformanceCounterScopeKHR specifying the scope that the counter belongs to.
storage is a VkPerformanceCounterStorageKHR specifying the storage type that the counter’s data uses.
uuid is an array of size VK_UUID_SIZE, containing 8-bit values that represent a universally unique identifier for the counter of the physical device.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_COUNTER_KHR
pNext must be NULL

Performance counters have an associated unit. This unit describes how to interpret the performance counter result.

The performance counter unit types which may be returned in VkPerformanceCounterKHR::unit are:

// Provided by VK_KHR_performance_query
typedef enum VkPerformanceCounterUnitKHR {
    VK_PERFORMANCE_COUNTER_UNIT_GENERIC_KHR = 0,
    VK_PERFORMANCE_COUNTER_UNIT_PERCENTAGE_KHR = 1,
    VK_PERFORMANCE_COUNTER_UNIT_NANOSECONDS_KHR = 2,
    VK_PERFORMANCE_COUNTER_UNIT_BYTES_KHR = 3,
    VK_PERFORMANCE_COUNTER_UNIT_BYTES_PER_SECOND_KHR = 4,
    VK_PERFORMANCE_COUNTER_UNIT_KELVIN_KHR = 5,
    VK_PERFORMANCE_COUNTER_UNIT_WATTS_KHR = 6,
    VK_PERFORMANCE_COUNTER_UNIT_VOLTS_KHR = 7,
    VK_PERFORMANCE_COUNTER_UNIT_AMPS_KHR = 8,
    VK_PERFORMANCE_COUNTER_UNIT_HERTZ_KHR = 9,
    VK_PERFORMANCE_COUNTER_UNIT_CYCLES_KHR = 10,
} VkPerformanceCounterUnitKHR;

VK_PERFORMANCE_COUNTER_UNIT_GENERIC_KHR - the performance counter unit is a generic data point.
VK_PERFORMANCE_COUNTER_UNIT_PERCENTAGE_KHR - the performance counter unit is a percentage (%).
VK_PERFORMANCE_COUNTER_UNIT_NANOSECONDS_KHR - the performance counter unit is a value of nanoseconds (ns).
VK_PERFORMANCE_COUNTER_UNIT_BYTES_KHR - the performance counter unit is a value of bytes.
VK_PERFORMANCE_COUNTER_UNIT_BYTES_PER_SECOND_KHR - the performance counter unit is a value of bytes/s.
VK_PERFORMANCE_COUNTER_UNIT_KELVIN_KHR - the performance counter unit is a temperature reported in Kelvin.
VK_PERFORMANCE_COUNTER_UNIT_WATTS_KHR - the performance counter unit is a value of watts (W).
VK_PERFORMANCE_COUNTER_UNIT_VOLTS_KHR - the performance counter unit is a value of volts (V).
VK_PERFORMANCE_COUNTER_UNIT_AMPS_KHR - the performance counter unit is a value of amps (A).
VK_PERFORMANCE_COUNTER_UNIT_HERTZ_KHR - the performance counter unit is a value of hertz (Hz).
VK_PERFORMANCE_COUNTER_UNIT_CYCLES_KHR - the performance counter unit is a value of cycles.

Performance counters have an associated scope. This scope describes the granularity of a performance counter.

The performance counter scope types which may be returned in VkPerformanceCounterKHR::scope are:

// Provided by VK_KHR_performance_query
typedef enum VkPerformanceCounterScopeKHR {
    VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR = 0,
    VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR = 1,
    VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_KHR = 2,
    VK_QUERY_SCOPE_COMMAND_BUFFER_KHR = VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR,
    VK_QUERY_SCOPE_RENDER_PASS_KHR = VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR,
    VK_QUERY_SCOPE_COMMAND_KHR = VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_KHR,
} VkPerformanceCounterScopeKHR;

VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR - the performance counter scope is a single complete command buffer.
VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR - the performance counter scope is zero or more complete render passes. The performance query containing the performance counter must begin and end outside a render pass instance.
VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_KHR - the performance counter scope is zero or more commands.

Performance counters have an associated storage. This storage describes the payload of a counter result.

The performance counter storage types which may be returned in VkPerformanceCounterKHR::storage are:

// Provided by VK_KHR_performance_query
typedef enum VkPerformanceCounterStorageKHR {
    VK_PERFORMANCE_COUNTER_STORAGE_INT32_KHR = 0,
    VK_PERFORMANCE_COUNTER_STORAGE_INT64_KHR = 1,
    VK_PERFORMANCE_COUNTER_STORAGE_UINT32_KHR = 2,
    VK_PERFORMANCE_COUNTER_STORAGE_UINT64_KHR = 3,
    VK_PERFORMANCE_COUNTER_STORAGE_FLOAT32_KHR = 4,
    VK_PERFORMANCE_COUNTER_STORAGE_FLOAT64_KHR = 5,
} VkPerformanceCounterStorageKHR;

VK_PERFORMANCE_COUNTER_STORAGE_INT32_KHR - the performance counter storage is a 32-bit signed integer.
VK_PERFORMANCE_COUNTER_STORAGE_INT64_KHR - the performance counter storage is a 64-bit signed integer.
VK_PERFORMANCE_COUNTER_STORAGE_UINT32_KHR - the performance counter storage is a 32-bit unsigned integer.
VK_PERFORMANCE_COUNTER_STORAGE_UINT64_KHR - the performance counter storage is a 64-bit unsigned integer.
VK_PERFORMANCE_COUNTER_STORAGE_FLOAT32_KHR - the performance counter storage is a 32-bit floating-point.
VK_PERFORMANCE_COUNTER_STORAGE_FLOAT64_KHR - the performance counter storage is a 64-bit floating-point.

The VkPerformanceCounterDescriptionKHR structure is defined as:

// Provided by VK_KHR_performance_query
typedef struct VkPerformanceCounterDescriptionKHR {
    VkStructureType                            sType;
    const void*                                pNext;
    VkPerformanceCounterDescriptionFlagsKHR    flags;
    char                                       name[VK_MAX_DESCRIPTION_SIZE];
    char                                       category[VK_MAX_DESCRIPTION_SIZE];
    char                                       description[VK_MAX_DESCRIPTION_SIZE];
} VkPerformanceCounterDescriptionKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkPerformanceCounterDescriptionFlagBitsKHR indicating the usage behavior for the counter.
name is an array of size VK_MAX_DESCRIPTION_SIZE, containing a null-terminated UTF-8 string specifying the name of the counter.
category is an array of size VK_MAX_DESCRIPTION_SIZE, containing a null-terminated UTF-8 string specifying the category of the counter.
description is an array of size VK_MAX_DESCRIPTION_SIZE, containing a null-terminated UTF-8 string specifying the description of the counter.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_COUNTER_DESCRIPTION_KHR
pNext must be NULL

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

// Provided by VK_KHR_performance_query
typedef enum VkPerformanceCounterDescriptionFlagBitsKHR {
    VK_PERFORMANCE_COUNTER_DESCRIPTION_PERFORMANCE_IMPACTING_KHR = 0x00000001,
    VK_PERFORMANCE_COUNTER_DESCRIPTION_CONCURRENTLY_IMPACTED_KHR = 0x00000002,
} VkPerformanceCounterDescriptionFlagBitsKHR;

VK_PERFORMANCE_COUNTER_DESCRIPTION_PERFORMANCE_IMPACTING_KHR specifies that recording the counter may have a noticeable performance impact.
VK_PERFORMANCE_COUNTER_DESCRIPTION_CONCURRENTLY_IMPACTED_KHR specifies that concurrently recording the counter while other submitted command buffers are running may impact the accuracy of the recording.

// Provided by VK_KHR_performance_query
typedef VkFlags VkPerformanceCounterDescriptionFlagsKHR;

VkPerformanceCounterDescriptionFlagsKHR is a bitmask type for setting a mask of zero or more VkPerformanceCounterDescriptionFlagBitsKHR.

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. For two physical devices to be in the same device group, they must support identical extensions, features, and properties.

Note

Physical devices in the same device group must be so similar because there are no rules for how different features/properties would interact. They must return the same values for nearly every invariant vkGetPhysicalDevice* feature, property, capability, etc., but could potentially differ for certain queries based on things like having a different display connected, or different compositor, etc.. The specification does not attempt to enumerate which state is in each category, because such a list would quickly become out of date.

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

// Provided by VK_VERSION_1_1
VkResult vkEnumeratePhysicalDeviceGroups(
    VkInstance                                  instance,
    uint32_t*                                   pPhysicalDeviceGroupCount,
    VkPhysicalDeviceGroupProperties*            pPhysicalDeviceGroupProperties);

or the equivalent command

// Provided by VK_KHR_device_group_creation
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

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INITIALIZATION_FAILED

The VkPhysicalDeviceGroupProperties structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkPhysicalDeviceGroupProperties {
    VkStructureType     sType;
    void*               pNext;
    uint32_t            physicalDeviceCount;
    VkPhysicalDevice    physicalDevices[VK_MAX_DEVICE_GROUP_SIZE];
    VkBool32            subsetAllocation;
} VkPhysicalDeviceGroupProperties;

or the equivalent

// Provided by VK_KHR_device_group_creation
typedef VkPhysicalDeviceGroupProperties VkPhysicalDeviceGroupPropertiesKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
physicalDeviceCount is the number of physical devices in the group.
physicalDevices is an array of VK_MAX_DEVICE_GROUP_SIZE VkPhysicalDevice 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.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_GROUP_PROPERTIES
pNext must be NULL

4.2.1. Device Creation

Logical devices are represented by VkDevice handles:

// Provided by VK_VERSION_1_0
VK_DEFINE_HANDLE(VkDevice)

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

// Provided by VK_VERSION_1_0
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 is a pointer 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

All required extensions for each extension in the VkDeviceCreateInfo::ppEnabledExtensionNames list must also be present in that list

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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 containing 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
ppEnabledExtensionNames must not contain both VK_KHR_buffer_device_address and VK_EXT_buffer_device_address
If the pNext chain includes a VkPhysicalDeviceVulkan11Features structure, then it must not include a VkPhysicalDevice16BitStorageFeatures, VkPhysicalDeviceMultiviewFeatures, VkPhysicalDeviceVariablePointersFeatures, VkPhysicalDeviceProtectedMemoryFeatures, VkPhysicalDeviceSamplerYcbcrConversionFeatures, or VkPhysicalDeviceShaderDrawParametersFeatures structure
If the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then it must not include a VkPhysicalDevice8BitStorageFeatures, VkPhysicalDeviceShaderAtomicInt64Features, VkPhysicalDeviceShaderFloat16Int8Features, VkPhysicalDeviceDescriptorIndexingFeatures, VkPhysicalDeviceScalarBlockLayoutFeatures, VkPhysicalDeviceImagelessFramebufferFeatures, VkPhysicalDeviceUniformBufferStandardLayoutFeatures, VkPhysicalDeviceShaderSubgroupExtendedTypesFeatures, VkPhysicalDeviceSeparateDepthStencilLayoutsFeatures, VkPhysicalDeviceHostQueryResetFeatures, VkPhysicalDeviceTimelineSemaphoreFeatures, VkPhysicalDeviceBufferDeviceAddressFeatures, or VkPhysicalDeviceVulkanMemoryModelFeatures structure
If ppEnabledExtensions contains "VK_KHR_draw_indirect_count" and the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then VkPhysicalDeviceVulkan12Features::drawIndirectCount must be VK_TRUE
If ppEnabledExtensions contains "VK_KHR_sampler_mirror_clamp_to_edge" and the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then VkPhysicalDeviceVulkan12Features::samplerMirrorClampToEdge must be VK_TRUE
If ppEnabledExtensions contains "VK_EXT_descriptor_indexing" and the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then VkPhysicalDeviceVulkan12Features::descriptorIndexing must be VK_TRUE
If ppEnabledExtensions contains "VK_EXT_sampler_filter_minmax" and the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then VkPhysicalDeviceVulkan12Features::samplerFilterMinmax must be VK_TRUE
If ppEnabledExtensions contains "VK_EXT_shader_viewport_index_layer" and the pNext chain includes a VkPhysicalDeviceVulkan12Features structure, then VkPhysicalDeviceVulkan12Features::shaderOutputViewportIndex and VkPhysicalDeviceVulkan12Features::shaderOutputLayer must both be VK_TRUE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DEVICE_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 VkDeviceDiagnosticsConfigCreateInfoNV, VkDeviceGroupDeviceCreateInfo, VkDeviceMemoryOverallocationCreateInfoAMD, VkDevicePrivateDataCreateInfoEXT, VkPhysicalDevice16BitStorageFeatures, VkPhysicalDevice8BitStorageFeatures, VkPhysicalDeviceASTCDecodeFeaturesEXT, VkPhysicalDeviceBlendOperationAdvancedFeaturesEXT, VkPhysicalDeviceBufferDeviceAddressFeatures, VkPhysicalDeviceBufferDeviceAddressFeaturesEXT, VkPhysicalDeviceCoherentMemoryFeaturesAMD, VkPhysicalDeviceComputeShaderDerivativesFeaturesNV, VkPhysicalDeviceConditionalRenderingFeaturesEXT, VkPhysicalDeviceCooperativeMatrixFeaturesNV, VkPhysicalDeviceCornerSampledImageFeaturesNV, VkPhysicalDeviceCoverageReductionModeFeaturesNV, VkPhysicalDeviceCustomBorderColorFeaturesEXT, VkPhysicalDeviceDedicatedAllocationImageAliasingFeaturesNV, VkPhysicalDeviceDepthClipEnableFeaturesEXT, VkPhysicalDeviceDescriptorIndexingFeatures, VkPhysicalDeviceDeviceGeneratedCommandsFeaturesNV, VkPhysicalDeviceDiagnosticsConfigFeaturesNV, VkPhysicalDeviceExclusiveScissorFeaturesNV, VkPhysicalDeviceExtendedDynamicStateFeaturesEXT, VkPhysicalDeviceFeatures2, VkPhysicalDeviceFragmentDensityMap2FeaturesEXT, VkPhysicalDeviceFragmentDensityMapFeaturesEXT, VkPhysicalDeviceFragmentShaderBarycentricFeaturesNV, VkPhysicalDeviceFragmentShaderInterlockFeaturesEXT, VkPhysicalDeviceHostQueryResetFeatures, VkPhysicalDeviceImageRobustnessFeaturesEXT, VkPhysicalDeviceImagelessFramebufferFeatures, VkPhysicalDeviceIndexTypeUint8FeaturesEXT, VkPhysicalDeviceInlineUniformBlockFeaturesEXT, VkPhysicalDeviceLineRasterizationFeaturesEXT, VkPhysicalDeviceMemoryPriorityFeaturesEXT, VkPhysicalDeviceMeshShaderFeaturesNV, VkPhysicalDeviceMultiviewFeatures, VkPhysicalDevicePerformanceQueryFeaturesKHR, VkPhysicalDevicePipelineCreationCacheControlFeaturesEXT, VkPhysicalDevicePipelineExecutablePropertiesFeaturesKHR, VkPhysicalDevicePrivateDataFeaturesEXT, VkPhysicalDeviceProtectedMemoryFeatures, VkPhysicalDeviceRayTracingFeaturesKHR, VkPhysicalDeviceRepresentativeFragmentTestFeaturesNV, VkPhysicalDeviceRobustness2FeaturesEXT, VkPhysicalDeviceSamplerYcbcrConversionFeatures, VkPhysicalDeviceScalarBlockLayoutFeatures, VkPhysicalDeviceSeparateDepthStencilLayoutsFeatures, VkPhysicalDeviceShaderAtomicFloatFeaturesEXT, VkPhysicalDeviceShaderAtomicInt64Features, VkPhysicalDeviceShaderClockFeaturesKHR, VkPhysicalDeviceShaderDemoteToHelperInvocationFeaturesEXT, VkPhysicalDeviceShaderDrawParametersFeatures, VkPhysicalDeviceShaderFloat16Int8Features, VkPhysicalDeviceShaderImageFootprintFeaturesNV, VkPhysicalDeviceShaderIntegerFunctions2FeaturesINTEL, VkPhysicalDeviceShaderSMBuiltinsFeaturesNV, VkPhysicalDeviceShaderSubgroupExtendedTypesFeatures, VkPhysicalDeviceShadingRateImageFeaturesNV, VkPhysicalDeviceSubgroupSizeControlFeaturesEXT, VkPhysicalDeviceTexelBufferAlignmentFeaturesEXT, VkPhysicalDeviceTextureCompressionASTCHDRFeaturesEXT, VkPhysicalDeviceTimelineSemaphoreFeatures, VkPhysicalDeviceTransformFeedbackFeaturesEXT, VkPhysicalDeviceUniformBufferStandardLayoutFeatures, VkPhysicalDeviceVariablePointersFeatures, VkPhysicalDeviceVertexAttributeDivisorFeaturesEXT, VkPhysicalDeviceVulkan11Features, VkPhysicalDeviceVulkan12Features, VkPhysicalDeviceVulkanMemoryModelFeatures, or VkPhysicalDeviceYcbcrImageArraysFeaturesEXT
The sType value of each struct in the pNext chain must be unique, with the exception of structures of type VkDevicePrivateDataCreateInfoEXT
flags must be 0
pQueueCreateInfos must be a valid pointer to an array of queueCreateInfoCount valid VkDeviceQueueCreateInfo structures
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
If pEnabledFeatures is not NULL, pEnabledFeatures must be a valid pointer to a valid VkPhysicalDeviceFeatures structure
queueCreateInfoCount must be greater than 0

// Provided by VK_VERSION_1_0
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 adding a VkDeviceGroupDeviceCreateInfo structure to the pNext chain of VkDeviceCreateInfo. The VkDeviceGroupDeviceCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkDeviceGroupDeviceCreateInfo {
    VkStructureType            sType;
    const void*                pNext;
    uint32_t                   physicalDeviceCount;
    const VkPhysicalDevice*    pPhysicalDevices;
} VkDeviceGroupDeviceCreateInfo;

or the equivalent

// Provided by VK_KHR_device_group_creation
typedef VkDeviceGroupDeviceCreateInfo VkDeviceGroupDeviceCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
physicalDeviceCount is the number of elements in the pPhysicalDevices array.
pPhysicalDevices is a pointer to 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

To specify whether device memory allocation is allowed beyond the size reported by VkPhysicalDeviceMemoryProperties, add a VkDeviceMemoryOverallocationCreateInfoAMD structure to the pNext chain of the VkDeviceCreateInfo structure. If this structure is not specified, it is as if the VK_MEMORY_OVERALLOCATION_BEHAVIOR_DEFAULT_AMD value is used.

// Provided by VK_AMD_memory_overallocation_behavior
typedef struct VkDeviceMemoryOverallocationCreateInfoAMD {
    VkStructureType                      sType;
    const void*                          pNext;
    VkMemoryOverallocationBehaviorAMD    overallocationBehavior;
} VkDeviceMemoryOverallocationCreateInfoAMD;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
overallocationBehavior is the desired overallocation behavior.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DEVICE_MEMORY_OVERALLOCATION_CREATE_INFO_AMD
overallocationBehavior must be a valid VkMemoryOverallocationBehaviorAMD value

Possible values for VkDeviceMemoryOverallocationCreateInfoAMD::overallocationBehavior include:

// Provided by VK_AMD_memory_overallocation_behavior
typedef enum VkMemoryOverallocationBehaviorAMD {
    VK_MEMORY_OVERALLOCATION_BEHAVIOR_DEFAULT_AMD = 0,
    VK_MEMORY_OVERALLOCATION_BEHAVIOR_ALLOWED_AMD = 1,
    VK_MEMORY_OVERALLOCATION_BEHAVIOR_DISALLOWED_AMD = 2,
} VkMemoryOverallocationBehaviorAMD;

VK_MEMORY_OVERALLOCATION_BEHAVIOR_DEFAULT_AMD lets the implementation decide if overallocation is allowed.
VK_MEMORY_OVERALLOCATION_BEHAVIOR_ALLOWED_AMD specifies overallocation is allowed if platform permits.
VK_MEMORY_OVERALLOCATION_BEHAVIOR_DISALLOWED_AMD specifies the application is not allowed to allocate device memory beyond the heap sizes reported by VkPhysicalDeviceMemoryProperties. Allocations that are not explicitly made by the application within the scope of the Vulkan instance are not accounted for.

When using the Nsight^™ Aftermath SDK, to configure how device crash dumps are created, add a VkDeviceDiagnosticsConfigCreateInfoNV structure to the pNext chain of the VkDeviceCreateInfo structure.

// Provided by VK_NV_device_diagnostics_config
typedef struct VkDeviceDiagnosticsConfigCreateInfoNV {
    VkStructureType                     sType;
    const void*                         pNext;
    VkDeviceDiagnosticsConfigFlagsNV    flags;
} VkDeviceDiagnosticsConfigCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkDeviceDiagnosticsConfigFlagBitsNV specifying addtional parameters for configuring diagnostic tools.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DEVICE_DIAGNOSTICS_CONFIG_CREATE_INFO_NV
flags must be a valid combination of VkDeviceDiagnosticsConfigFlagBitsNV values

Bits which can be set in VkDeviceDiagnosticsConfigCreateInfoNV::flags include:

// Provided by VK_NV_device_diagnostics_config
typedef enum VkDeviceDiagnosticsConfigFlagBitsNV {
    VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_SHADER_DEBUG_INFO_BIT_NV = 0x00000001,
    VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_RESOURCE_TRACKING_BIT_NV = 0x00000002,
    VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_AUTOMATIC_CHECKPOINTS_BIT_NV = 0x00000004,
} VkDeviceDiagnosticsConfigFlagBitsNV;

VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_SHADER_DEBUG_INFO_BIT_NV enables the generation of debug information for shaders.
VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_RESOURCE_TRACKING_BIT_NV enables driver side tracking of resources (images, buffers, etc.) used to augment the device fault information.
VK_DEVICE_DIAGNOSTICS_CONFIG_ENABLE_AUTOMATIC_CHECKPOINTS_BIT_NV enables automatic insertion of diagnostic checkpoints for draw calls, dispatches, trace rays, and copies. The CPU call stack at the time of the command will be associated as the marker data for the automatically inserted checkpoints.

// Provided by VK_NV_device_diagnostics_config
typedef VkFlags VkDeviceDiagnosticsConfigFlagsNV;

VkDeviceDiagnosticsConfigFlagsNV is a bitmask type for setting a mask of zero or more VkDeviceDiagnosticsConfigFlagBitsNV.

To reserve private data storage slots, add a VkDevicePrivateDataCreateInfoEXT structure to the pNext chain of the VkDeviceCreateInfo structure. Reserving slots in this manner is not strictly necessary, but doing so may improve performance.

// Provided by VK_EXT_private_data
typedef struct VkDevicePrivateDataCreateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           privateDataSlotRequestCount;
} VkDevicePrivateDataCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
privateDataSlotRequestCount is the amount of slots to reserve.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DEVICE_PRIVATE_DATA_CREATE_INFO_EXT

4.2.2. Device Use

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

Creation of queues. See the Queues section below for further details.
Creation and tracking of various synchronization constructs. See Synchronization and Cache Control for further details.
Allocating, freeing, and managing memory. See Memory Allocation and Resource Creation for further details.
Creation and destruction of command buffers and command buffer pools. See Command Buffers for further details.
Creation, destruction, and management of graphics state. See Pipelines and Resource Descriptors, among others, for further details.

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, implementation errors.

Applications not adhering to valid usage may also result in device loss being reported, however this is not guaranteed. Even if device loss is reported, the system may be in an unrecoverable state, and further usage of the API is still considered invalid.

When this happens, certain commands will return VK_ERROR_DEVICE_LOST. 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.

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 runtime 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:

// Provided by VK_VERSION_1_0
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
Host access to all VkQueue objects received from 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_HANDLE(VkQueue)

The VkDeviceQueueCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 a pointer to 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
If the protected memory feature is not enabled, the VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT bit of flags must not be set

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
The sType value of each struct in the pNext chain must be unique
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:

// Provided by VK_VERSION_1_0
typedef enum VkDeviceQueueCreateFlagBits {
  // Provided by VK_VERSION_1_1
    VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT = 0x00000001,
} VkDeviceQueueCreateFlagBits;

VK_DEVICE_QUEUE_CREATE_PROTECTED_BIT specifies that the device queue is a protected-capable queue.

// Provided by VK_VERSION_1_0
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 adding a VkDeviceQueueGlobalPriorityCreateInfoEXT structure to the pNext chain of VkDeviceQueueCreateInfo.

The VkDeviceQueueGlobalPriorityCreateInfoEXT structure is defined as:

// Provided by VK_EXT_global_priority
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 a structure extending this 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:

// Provided by VK_EXT_global_priority
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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_1
void vkGetDeviceQueue2(
    VkDevice                                    device,
    const VkDeviceQueueInfo2*                   pQueueInfo,
    VkQueue*                                    pQueue);

device is the logical device that owns the queue.
pQueueInfo is a pointer to a 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:

// Provided by VK_VERSION_1_1
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 a structure extending this 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

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 Pipeline Barriers), 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:

Zero or more semaphores to wait on before execution of the rest of the batch.
- If present, these describe a semaphore wait operation.
Zero or more work items to execute.
- If present, these describe a queue operation matching the work described.
Zero or more semaphores to signal upon completion of the work items.
- If present, these describe a semaphore signal operation.

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:

// Provided by VK_VERSION_1_0
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. For state dependent commands (such as draws and dispatches), any state consumed by those commands must not be 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 either reverts back to the executable state, or if it was recorded with VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT, it moves to the invalid state. 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.

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 it does for any secondary recorded within it. After all executions of each command buffer complete, they each move to their appropriate completion state (either to the execution state or the invalid state, as specified above).

If a secondary moves to the invalid state or the initial state, then all primary buffers it is recorded in move to the invalid state. A primary moving to any other state does not affect the state of a secondary recorded in it.

Note

Resetting or freeing a primary command buffer removes the lifecycle linkage to all secondary command buffers that were recorded into 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkCommandPool)

To create a command pool, call:

// Provided by VK_VERSION_1_0
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 a VkCommandPoolCreateInfo structure specifying the state of the command pool object.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pCommandPool is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandPoolCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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

If the protected memory feature is not enabled, the VK_COMMAND_POOL_CREATE_PROTECTED_BIT bit of flags must not be set

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:

// Provided by VK_VERSION_1_0
typedef enum VkCommandPoolCreateFlagBits {
    VK_COMMAND_POOL_CREATE_TRANSIENT_BIT = 0x00000001,
    VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT = 0x00000002,
  // Provided by VK_VERSION_1_1
    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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkCommandPoolCreateFlags;

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

To trim a command pool, call:

// Provided by VK_VERSION_1_1
void vkTrimCommandPool(
    VkDevice                                    device,
    VkCommandPool                               commandPool,
    VkCommandPoolTrimFlags                      flags);

or the equivalent command

// Provided by VK_KHR_maintenance1
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” behavior.

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

// Provided by VK_VERSION_1_1
typedef VkFlags VkCommandPoolTrimFlags;

or the equivalent

// Provided by VK_KHR_maintenance1
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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkCommandPoolResetFlags;

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

To destroy a command pool, call:

// Provided by VK_VERSION_1_0
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.

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:

// Provided by VK_VERSION_1_0
VkResult vkAllocateCommandBuffers(
    VkDevice                                    device,
    const VkCommandBufferAllocateInfo*          pAllocateInfo,
    VkCommandBuffer*                            pCommandBuffers);

device is the logical device that owns the command pool.
pAllocateInfo is a pointer to a 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
pAllocateInfo->commandBufferCount must be greater than 0

Host Synchronization

Host access to pAllocateInfo->commandPool must be externally synchronized

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandBufferAllocateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
commandPool is the command pool from which the command buffers are allocated.
level is a 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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
VkResult vkResetCommandBuffer(
    VkCommandBuffer                             commandBuffer,
    VkCommandBufferResetFlags                   flags);

commandBuffer is the command buffer to reset. The command buffer can be in any state other than pending, and is moved into the initial state.
flags is a bitmask of VkCommandBufferResetFlagBits controlling the reset operation.

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)

commandBuffer must be a valid VkCommandBuffer handle
flags must be a valid combination of VkCommandBufferResetFlagBits values

Host Synchronization

Host access to commandBuffer must be externally synchronized

Return Codes

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkCommandBufferResetFlags;

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

To free command buffers, call:

// Provided by VK_VERSION_1_0
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 a pointer to 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:

// Provided by VK_VERSION_1_0
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 points to a VkCommandBufferBeginInfo structure defining 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
If commandBuffer is a primary command buffer, then pBeginInfo->flags must not set both the VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT and the VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT flags

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCommandBufferBeginInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkCommandBufferUsageFlagBits specifying usage behavior for the command buffer.
pInheritanceInfo is a pointer to a VkCommandBufferInheritanceInfo structure, 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)

sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO
pNext must be NULL or a pointer to a valid instance of VkDeviceGroupCommandBufferBeginInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkCommandBufferUsageFlagBits values

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

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 a structure extending this 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

If the inherited queries feature is not enabled, occlusionQueryEnable must be VK_FALSE
If the inherited queries feature is enabled, queryFlags must be a valid combination of VkQueryControlFlagBits values
If the inherited queries feature is not enabled, queryFlags must be 0
If the pipeline statistics queries feature is enabled, pipelineStatistics must be a valid combination of VkQueryPipelineStatisticFlagBits values
If the pipeline statistics queries feature is not enabled, pipelineStatistics must be 0

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_INHERITANCE_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 VkCommandBufferInheritanceConditionalRenderingInfoEXT or VkCommandBufferInheritanceRenderPassTransformInfoQCOM
The sType value of each struct in the pNext chain must be unique
Both of framebuffer, and renderPass that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

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.

If the pNext chain of VkCommandBufferInheritanceInfo includes a VkCommandBufferInheritanceConditionalRenderingInfoEXT structure, then that structure controls whether a command buffer can be executed while conditional rendering is active in the primary command buffer.

The VkCommandBufferInheritanceConditionalRenderingInfoEXT structure is defined as:

// Provided by VK_EXT_conditional_rendering
typedef struct VkCommandBufferInheritanceConditionalRenderingInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkBool32           conditionalRenderingEnable;
} VkCommandBufferInheritanceConditionalRenderingInfoEXT;

sType is the type of this structure
pNext is NULL or a pointer to a structure extending this structure
conditionalRenderingEnable specifies whether the command buffer can be executed while conditional rendering 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 active conditional rendering or not. If this is VK_FALSE, then the primary command buffer must not have conditional rendering active.

If this structure is not present, the behavior is as if conditionalRenderingEnable is VK_FALSE.

Valid Usage

If the inherited conditional rendering feature is not enabled, conditionalRenderingEnable must be VK_FALSE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_INHERITANCE_CONDITIONAL_RENDERING_INFO_EXT

To begin recording a secondary command buffer compatible with execution inside a render pass using render pass transform, add the VkCommandBufferInheritanceRenderPassTransformInfoQCOM to the pNext chain of VkCommandBufferInheritanceInfo structure passed to the vkBeginCommandBuffer command specifying the parameters for transformed rasterization.

The VkCommandBufferInheritanceRenderPassTransformInfoQCOM structure is defined as:

// Provided by VK_QCOM_render_pass_transform
typedef struct VkCommandBufferInheritanceRenderPassTransformInfoQCOM {
    VkStructureType                  sType;
    void*                            pNext;
    VkSurfaceTransformFlagBitsKHR    transform;
    VkRect2D                         renderArea;
} VkCommandBufferInheritanceRenderPassTransformInfoQCOM;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
transform is a VkSurfaceTransformFlagBitsKHR value describing the transform to be applied to the render pass.
renderArea is the render area that is affected by the command buffer.

When the secondary is recorded to execute within a render pass instance using vkCmdExecuteCommands, the render pass transform parameters of the secondary command buffer must be consistent with the render pass transform parameters specified for the render pass instance. In particular, the transform and renderArea for command buffer must be identical to the transform and renderArea of the render pass instance.

Valid Usage

transform must be VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR, VK_SURFACE_TRANSFORM_ROTATE_90_BIT_KHR, VK_SURFACE_TRANSFORM_ROTATE_180_BIT_KHR, or VK_SURFACE_TRANSFORM_ROTATE_270_BIT_KHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_COMMAND_BUFFER_INHERITANCE_RENDER_PASS_TRANSFORM_INFO_QCOM

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:

// Provided by VK_VERSION_1_0
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
Conditional rendering must not be active
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

VK_SUCCESS

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

Note

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

To submit command buffers to a queue, call:

// Provided by VK_VERSION_1_0
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.

vkQueueSubmit is a queue submission command, with each batch defined by an element of pSubmits. 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 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 wait operation referring to a binary semaphore defined by any element of the pWaitSemaphores member of any element of pSubmits executes on queue, there must be no other queues waiting on the same semaphore
All elements of the pWaitSemaphores member of all elements of pSubmits created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY must reference a semaphore signal operation that has been submitted for execution and any semaphore signal operations on which it depends (if any) must have also been 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
If any element of pSubmits->pCommandBuffers includes a Queue Family Transfer Acquire Operation, there must exist a previously submitted Queue Family Transfer Release Operation on a queue in the queue family identified by the acquire operation, with parameters matching the acquire operation as defined in the definition of such acquire operations, and which happens-before the acquire operation
If a command recorded into any element of pCommandBuffers was a vkCmdBeginQuery whose queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the profiling lock must have been held continuously on the VkDevice that queue was retrieved from, throughout recording of those command buffers
Any resource created with VK_SHARING_MODE_EXCLUSIVE that is read by an operation specified by pSubmits must not be owned by any queue family other than the one which queue belongs to, at the time it is executed

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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

Host Synchronization

Host access to queue must be externally synchronized
Host access to fence must be externally synchronized

Command Properties

Command Buffer Levels	Render Pass Scope	Supported Queue Types	Pipeline Type
-	-	Any	-

Any

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_DEVICE_LOST

The VkSubmitInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 VkSemaphore handles 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 VkCommandBuffer handles 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 VkSemaphore handles 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
If any element of pWaitSemaphores or pSignalSemaphores was created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, then the pNext chain must include a VkTimelineSemaphoreSubmitInfo structure
If the pNext chain of this structure includes a VkTimelineSemaphoreSubmitInfo structure and any element of pWaitSemaphores was created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, then its waitSemaphoreValueCount member must equal waitSemaphoreCount
If the pNext chain of this structure includes a VkTimelineSemaphoreSubmitInfo structure and any element of pSignalSemaphores was created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, then its signalSemaphoreValueCount member must equal signalSemaphoreCount
For each element of pSignalSemaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE the corresponding element of VkTimelineSemaphoreSubmitInfo::pSignalSemaphoreValues must have a value greater than the current value of the semaphore when the semaphore signal operation is executed
For each element of pWaitSemaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE the corresponding element of VkTimelineSemaphoreSubmitInfo::pWaitSemaphoreValues must have a value which does not differ from the current value of the semaphore or the value of any outstanding semaphore wait or signal operation on that semaphore by more than maxTimelineSemaphoreValueDifference
For each element of pSignalSemaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE the corresponding element of VkTimelineSemaphoreSubmitInfo::pSignalSemaphoreValues must have a value which does not differ from the current value of the semaphore or the value of any outstanding semaphore wait or signal operation on that semaphore by more than maxTimelineSemaphoreValueDifference
If the mesh shaders feature is not enabled, each element of pWaitDstStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, each element of pWaitDstStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the pNext chain of this structure does not include a VkProtectedSubmitInfo structure with protectedSubmit set to VK_TRUE, 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_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, VkPerformanceQuerySubmitInfoKHR, VkProtectedSubmitInfo, VkTimelineSemaphoreSubmitInfo, VkWin32KeyedMutexAcquireReleaseInfoKHR, or VkWin32KeyedMutexAcquireReleaseInfoNV
The sType value of each struct 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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

To specify the values to use when waiting for and signaling semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, add a VkTimelineSemaphoreSubmitInfo structure to the pNext chain of the VkSubmitInfo structure when using vkQueueSubmit or the VkBindSparseInfo structure when using vkQueueBindSparse. The VkTimelineSemaphoreSubmitInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkTimelineSemaphoreSubmitInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           waitSemaphoreValueCount;
    const uint64_t*    pWaitSemaphoreValues;
    uint32_t           signalSemaphoreValueCount;
    const uint64_t*    pSignalSemaphoreValues;
} VkTimelineSemaphoreSubmitInfo;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkTimelineSemaphoreSubmitInfo VkTimelineSemaphoreSubmitInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
waitSemaphoreValueCount is the number of semaphore wait values specified in pWaitSemaphoreValues.
pWaitSemaphoreValues is an array of length waitSemaphoreValueCount containing values for the corresponding semaphores in VkSubmitInfo::pWaitSemaphores to wait for.
signalSemaphoreValueCount is the number of semaphore signal values specified in pSignalSemaphoreValues.
pSignalSemaphoreValues is an array of length signalSemaphoreValueCount 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 was not created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, the implementation must ignore the value in the pWaitSemaphoreValues or pSignalSemaphoreValues entry.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_TIMELINE_SEMAPHORE_SUBMIT_INFO
If waitSemaphoreValueCount is not 0, and pWaitSemaphoreValues is not NULL, pWaitSemaphoreValues must be a valid pointer to an array of waitSemaphoreValueCount uint64_t values
If signalSemaphoreValueCount is not 0, and pSignalSemaphoreValues is not NULL, pSignalSemaphoreValues must be a valid pointer to an array of signalSemaphoreValueCount uint64_t values

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

// Provided by VK_KHR_external_semaphore_win32
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 a structure extending this structure.
waitSemaphoreValuesCount is the number of semaphore wait values specified in pWaitSemaphoreValues.
pWaitSemaphoreValues is a pointer to an array of waitSemaphoreValuesCount values for the corresponding semaphores in VkSubmitInfo::pWaitSemaphores to wait for.
signalSemaphoreValuesCount is the number of semaphore signal values specified in pSignalSemaphoreValues.
pSignalSemaphoreValues is a pointer to an array of signalSemaphoreValuesCount 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.

Note

As the introduction of the external semaphore handle type VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT predates that of timeline semaphores, support for importing semaphore payloads from external handles of that type into semaphores created (implicitly or explicitly) with a VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY is preserved for backwards compatibility. However, applications should prefer importing such handle types into semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE, and use the VkTimelineSemaphoreSubmitInfo structure instead of the VkD3D12FenceSubmitInfoKHR structure to specify the values to use when waiting for and signaling such semaphores.

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:

// Provided by VK_KHR_win32_keyed_mutex
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 a pointer to 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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

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:

// Provided by VK_NV_win32_keyed_mutex
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 a pointer to 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 of non-ignored parameters 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:

// Provided by VK_VERSION_1_1
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 include 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

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:

// Provided by VK_VERSION_1_1
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

// Provided by VK_KHR_device_group
typedef VkDeviceGroupSubmitInfo VkDeviceGroupSubmitInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
waitSemaphoreCount is the number of elements in the pWaitSemaphoreDeviceIndices array.
pWaitSemaphoreDeviceIndices is a pointer to an array of waitSemaphoreCount 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 a pointer to an array of commandBufferCount 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 a pointer to an array of signalSemaphoreCount 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

If the pNext chain of VkSubmitInfo includes a VkPerformanceQuerySubmitInfoKHR structure, then the structure indicates which counter pass is active for the batch in that submit.

The VkPerformanceQuerySubmitInfoKHR structure is defined as:

// Provided by VK_KHR_performance_query
typedef struct VkPerformanceQuerySubmitInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           counterPassIndex;
} VkPerformanceQuerySubmitInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
counterPassIndex specifies which counter pass index is active.

If the VkSubmitInfo::pNext chain does not include this structure, the batch defaults to use counter pass index 0.

Valid Usage

counterPassIndex must be less than the number of counter passes required by any queries within the batch. The required number of counter passes for a performance query is obtained by calling vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_QUERY_SUBMIT_INFO_KHR

5.6. Queue Forward Progress

When using binary semaphores, 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 (or other queue operation), for every queued wait on a semaphore created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY there must be a prior signal of that semaphore that will not be consumed by a different wait on the semaphore.

When using timeline semaphores, wait-before-signal behavior is well-defined and applications can submit work via vkQueueSubmit which defines a timeline semaphore wait operation before submitting a corresponding semaphore signal operation. For each timeline semaphore wait operation defined by a call to vkQueueSubmit, the application must ensure that a corresponding semaphore signal operation is executed before forward progress can be made.

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. The event must be set before the vkCmdWaitEvents command is executed.

Note

Implementations may have some tolerance for waiting on events to be set, but this is defined outside of the scope of Vulkan.

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:

// Provided by VK_VERSION_1_0
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 a pointer to an array of commandBufferCount 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

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, 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 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 being called within a render pass instance that included VkRenderPassTransformBeginInfoQCOM in the pNext chain of VkRenderPassBeginInfo, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceRenderPassTransformInfoQCOM in the pNext chain of VkCommandBufferBeginInfo
If vkCmdExecuteCommands is being called within a render pass instance that included VkRenderPassTransformBeginInfoQCOM in the pNext chain of VkRenderPassBeginInfo, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceRenderPassTransformInfoQCOM::transform identical to VkRenderPassTransformBeginInfoQCOM::transform
If vkCmdExecuteCommands is being called within a render pass instance that included VkRenderPassTransformBeginInfoQCOM in the pNext chain of VkRenderPassBeginInfo, then each element of pCommandBuffers must have been recorded with VkCommandBufferInheritanceRenderPassTransformInfoQCOM::renderArea identical to VkRenderPassBeginInfo::renderArea
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
This command must not be recorded when transform feedback is active

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

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, exclusive scissor, and viewport state (excluding the count of each) 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:

// Provided by VK_VERSION_1_1
typedef struct VkDeviceGroupCommandBufferBeginInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceMask;
} VkDeviceGroupCommandBufferBeginInfo;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkDeviceGroupCommandBufferBeginInfo VkDeviceGroupCommandBufferBeginInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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:

// Provided by VK_VERSION_1_1
void vkCmdSetDeviceMask(
    VkCommandBuffer                             commandBuffer,
    uint32_t                                    deviceMask);

or the equivalent command

// Provided by VK_KHR_device_group
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, except commands beginning a render pass instance, commands transitioning to the next subpass in the render pass instance, and commands ending a render pass instance, which always execute on the set of physical devices whose bit indices are included in the deviceMask member of the VkDeviceGroupRenderPassBeginInfo structure passed to the command beginning the corresponding render pass instance.

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

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 A_S and B_S be the synchronization scopes of S.
Let A' be the intersection of sets A and A_S.
Let B' be the intersection of sets B and B_S.
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 B_S in the first dependency and A_S 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 A_S be the first synchronization scope of the first command in S.
Let B_S be the second synchronization scope of the last command in S.

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.

Three additional types of operation are used to control memory access. Availability operations cause the values generated by specified memory write accesses to become available to a memory domain 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. Memory domain operations cause writes that are available to a source memory domain to become available to a destination memory domain (an example of this is making writes available to the host domain available to the device domain). Visibility operations cause values available to a memory domain to become visible to specified memory accesses.

Availability, visibility, memory domains, and memory domain operations are formally defined in the Availability and Visibility section of the Memory Model chapter. Which API operations perform each of these operations is defined in Availability, Visibility, and Domain Operations.

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 a_S be the first access scope of the first command in S.
Let b_S be the second access scope of the last command in S.
Let a' be the intersection of sets a and a_S.
Let b' be the intersection of sets b and b_S.
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. The old layout must either be VK_IMAGE_LAYOUT_UNDEFINED, or match the current layout of the image subresource range. If the old layout matches the current layout of the image subresource range, the transition preserves the contents of that range. If the old layout is VK_IMAGE_LAYOUT_UNDEFINED, the contents of that range may be discarded.

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

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.

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 image memory barrier 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 any image layout transition.

If the VkSampleLocationsInfoEXT structure 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 present, then the contents of the given image subresource range becomes undefined as if oldLayout would equal VK_IMAGE_LAYOUT_UNDEFINED.

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 do not execute in a defined pipeline.

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 by using the synchronization scopes to create a dependency chain.

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.

The VkPipelineStageFlagBits enum is defined as:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_EXT_transform_feedback
    VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT = 0x01000000,
  // Provided by VK_EXT_conditional_rendering
    VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT = 0x00040000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR = 0x00200000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR = 0x02000000,
  // Provided by VK_NV_shading_rate_image
    VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV = 0x00400000,
  // Provided by VK_NV_mesh_shader
    VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV = 0x00080000,
  // Provided by VK_NV_mesh_shader
    VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV = 0x00100000,
  // Provided by VK_EXT_fragment_density_map
    VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT = 0x00800000,
  // Provided by VK_NV_device_generated_commands
    VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV = 0x00020000,
  // Provided by VK_NV_ray_tracing
    VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_NV = VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_NV = VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR,
} VkPipelineStageFlagBits;

VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT is equivalent to VK_PIPELINE_STAGE_ALL_COMMANDS_BIT with VkAccessFlags set to 0 when specified in the second synchronization scope, but specifies no stages in the first scope.
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 vkCmdExecuteGeneratedCommandsNV.
VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV specifies the task shader stage.
VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV specifies the mesh shader stage.
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 or depth/stencil format.
VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT specifies the execution of a compute shader.
VK_PIPELINE_STAGE_TRANSFER_BIT specifies the following commands:
- All copy commands, including vkCmdCopyQueryPoolResults
- vkCmdBlitImage
- vkCmdResolveImage
- All clear commands, with the exception of vkCmdClearAttachments
VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT is equivalent to VK_PIPELINE_STAGE_ALL_COMMANDS_BIT with VkAccessFlags set to 0 when specified in the first synchronization scope, but specifies no stages in the second scope.
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_RAY_TRACING_SHADER_BIT_KHR specifies the execution of the ray tracing shader stages.
VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR specifies the execution of acceleration structure commands.
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_DRAW_INDIRECT_BIT
- VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
- VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
- 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_CONDITIONAL_RENDERING_BIT_EXT
- VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
- VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
- VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
VK_PIPELINE_STAGE_ALL_COMMANDS_BIT specifies all commands supported on the queue it is used with.
VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT specifies the stage of the pipeline where the predicate of conditional rendering is consumed.
VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT specifies the stage of the pipeline where vertex attribute output values are written to the transform feedback buffers.
VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV specifies the stage of the pipeline where device-side preprocessing for generated commands via vkCmdPreprocessGeneratedCommandsNV is handled.
VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV specifies the stage of the pipeline where the shading rate image is read to determine the shading rate for portions of a rasterized primitive.
VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT specifies the stage of the pipeline where the fragment density map is read to generate the fragment areas.

// Provided by VK_VERSION_1_0
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_CONDITIONAL_RENDERING_BIT_EXT`	`VK_QUEUE_GRAPHICS_BIT` or `VK_QUEUE_COMPUTE_BIT`
`VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT`	`VK_QUEUE_GRAPHICS_BIT`
`VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV`	`VK_QUEUE_GRAPHICS_BIT` or `VK_QUEUE_COMPUTE_BIT`
`VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV`	`VK_QUEUE_GRAPHICS_BIT`
`VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV`	`VK_QUEUE_GRAPHICS_BIT`
`VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV`	`VK_QUEUE_GRAPHICS_BIT`
`VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR`	`VK_QUEUE_COMPUTE_BIT`
`VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR`	`VK_QUEUE_COMPUTE_BIT`
`VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT`	`VK_QUEUE_GRAPHICS_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 A_S 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 B_S 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.

Graphics pipelines are executable on queues supporting VK_QUEUE_GRAPHICS_BIT. Stages executed by graphics pipelines can only be specified in commands recorded for queues supporting VK_QUEUE_GRAPHICS_BIT.

The graphics primitive pipeline executes the following stages, with the logical ordering of the stages matching the order specified here:

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_TRANSFORM_FEEDBACK_BIT_EXT
VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
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

The graphics mesh pipeline executes the following stages, with the logical ordering of the stages matching the order specified here:

VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT
VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
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

For the compute pipeline, the following stages occur in this order:

VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT
VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT

For graphics pipeline commands executing in a render pass with a fragment density map attachment, the following pipeline stage where the fragment density map read happens has no particular order relative to the other stages, except that it is logically earlier than VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT:

VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT

The conditional rendering stage is formally part of both the graphics, and the compute pipeline. The pipeline stage where the predicate read happens has unspecified order relative to other stages of these pipelines:

VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT

For the transfer pipeline, the following stages occur in this order:

VK_PIPELINE_STAGE_TRANSFER_BIT

For host operations, only one pipeline stage occurs, so no order is guaranteed:

VK_PIPELINE_STAGE_HOST_BIT

For the command preprocessing pipeline, the following stages occur in this order:

VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV

For the ray tracing shader pipeline, only one pipeline stage occurs, so no order is guaranteed:

VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR

For ray tracing acceleration structure operations, only one pipeline stage occurs, so no order is guaranteed:

VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR

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.

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.

The VkAccessFlagBits enums is defined as:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_EXT_transform_feedback
    VK_ACCESS_TRANSFORM_FEEDBACK_WRITE_BIT_EXT = 0x02000000,
  // Provided by VK_EXT_transform_feedback
    VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_READ_BIT_EXT = 0x04000000,
  // Provided by VK_EXT_transform_feedback
    VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_WRITE_BIT_EXT = 0x08000000,
  // Provided by VK_EXT_conditional_rendering
    VK_ACCESS_CONDITIONAL_RENDERING_READ_BIT_EXT = 0x00100000,
  // Provided by VK_EXT_blend_operation_advanced
    VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT = 0x00080000,
  // Provided by VK_KHR_ray_tracing
    VK_ACCESS_ACCELERATION_STRUCTURE_READ_BIT_KHR = 0x00200000,
  // Provided by VK_KHR_ray_tracing
    VK_ACCESS_ACCELERATION_STRUCTURE_WRITE_BIT_KHR = 0x00400000,
  // Provided by VK_NV_shading_rate_image
    VK_ACCESS_SHADING_RATE_IMAGE_READ_BIT_NV = 0x00800000,
  // Provided by VK_EXT_fragment_density_map
    VK_ACCESS_FRAGMENT_DENSITY_MAP_READ_BIT_EXT = 0x01000000,
  // Provided by VK_NV_device_generated_commands
    VK_ACCESS_COMMAND_PREPROCESS_READ_BIT_NV = 0x00020000,
  // Provided by VK_NV_device_generated_commands
    VK_ACCESS_COMMAND_PREPROCESS_WRITE_BIT_NV = 0x00040000,
  // Provided by VK_NV_ray_tracing
    VK_ACCESS_ACCELERATION_STRUCTURE_READ_BIT_NV = VK_ACCESS_ACCELERATION_STRUCTURE_READ_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_ACCESS_ACCELERATION_STRUCTURE_WRITE_BIT_NV = VK_ACCESS_ACCELERATION_STRUCTURE_WRITE_BIT_KHR,
} VkAccessFlagBits;

VK_ACCESS_INDIRECT_COMMAND_READ_BIT specifies read access to indirect command data 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, physical storage buffer, shader binding table, uniform texel buffer, storage texel buffer, sampled image, or storage image.
VK_ACCESS_SHADER_WRITE_BIT specifies write access to a storage buffer, physical 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_WRITE_BIT specifies write access to a color, resolve, or depth/stencil 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 all read accesses. It is always valid in any access mask, and is treated as equivalent to setting all READ access flags that are valid where it is used.
VK_ACCESS_MEMORY_WRITE_BIT specifies all write accesses. It is always valid in any access mask, and is treated as equivalent to setting all WRITE access flags that are valid where it is used.
VK_ACCESS_CONDITIONAL_RENDERING_READ_BIT_EXT specifies read access to a predicate as part of conditional rendering.
VK_ACCESS_TRANSFORM_FEEDBACK_WRITE_BIT_EXT specifies write access to a transform feedback buffer made when transform feedback is active.
VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_READ_BIT_EXT specifies read access to a transform feedback counter buffer which is read when vkCmdBeginTransformFeedbackEXT executes.
VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_WRITE_BIT_EXT specifies write access to a transform feedback counter buffer which is written when vkCmdEndTransformFeedbackEXT executes.
VK_ACCESS_COMMAND_PREPROCESS_READ_BIT_NV specifies reads from VkBuffer inputs to vkCmdPreprocessGeneratedCommandsNV.
VK_ACCESS_COMMAND_PREPROCESS_WRITE_BIT_NV specifies writes to the VkBuffer preprocess outputs in vkCmdPreprocessGeneratedCommandsNV.
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_SHADING_RATE_IMAGE_READ_BIT_NV specifies read access to a shading rate image as part of a drawing command, as bound by vkCmdBindShadingRateImageNV.
VK_ACCESS_ACCELERATION_STRUCTURE_READ_BIT_KHR specifies read access to an acceleration structure as part of a trace or build command, or to an acceleration structure scratch buffer as part of a build command.
VK_ACCESS_ACCELERATION_STRUCTURE_WRITE_BIT_KHR specifies write access to an acceleration structure or acceleration structure scratch buffer as part of a build command.
VK_ACCESS_FRAGMENT_DENSITY_MAP_READ_BIT_EXT specifies read access to a fragment density map attachment during dynamic fragment density map operations

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_TASK_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR`, `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_READ_BIT`	`VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR`, `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_TASK_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV`, `VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR`, `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_COLOR_ATTACHMENT_READ_BIT`	`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`	Any
`VK_ACCESS_MEMORY_WRITE_BIT`	Any
`VK_ACCESS_COLOR_ATTACHMENT_READ_NONCOHERENT_BIT_EXT`	`VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT`
`VK_ACCESS_COMMAND_PREPROCESS_READ_BIT_NV`	`VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV`
`VK_ACCESS_COMMAND_PREPROCESS_WRITE_BIT_NV`	`VK_PIPELINE_STAGE_COMMAND_PREPROCESS_BIT_NV`
`VK_ACCESS_CONDITIONAL_RENDERING_READ_BIT_EXT`	`VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT`
`VK_ACCESS_SHADING_RATE_IMAGE_READ_BIT_NV`	`VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV`
`VK_ACCESS_TRANSFORM_FEEDBACK_WRITE_BIT_EXT`	`VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT`
`VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_WRITE_BIT_EXT`	`VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT`
`VK_ACCESS_TRANSFORM_FEEDBACK_COUNTER_READ_BIT_EXT`	`VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT`
`VK_ACCESS_ACCELERATION_STRUCTURE_READ_BIT_KHR`	`VK_PIPELINE_STAGE_RAY_TRACING_SHADER_BIT_KHR`, or `VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR`
`VK_ACCESS_ACCELERATION_STRUCTURE_WRITE_BIT_KHR`	`VK_PIPELINE_STAGE_ACCELERATION_STRUCTURE_BUILD_BIT_KHR`
`VK_ACCESS_FRAGMENT_DENSITY_MAP_READ_BIT_EXT`	`VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT`

// Provided by VK_VERSION_1_0
typedef VkFlags VkAccessFlags;

VkAccessFlags is a bitmask type for setting a mask of zero or more VkAccessFlagBits.

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 available to the host domain, where they can be further made available to the device domain via a domain operation. Similarly, vkInvalidateMappedMemoryRanges must be called to guarantee that writes which are available to the host domain 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 available to the host domain. Similarly, writes made available to the host domain are automatically made visible to the host.

Note

Queue submission commands automatically perform a domain operation from host to device for all writes performed before the command executes, 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.

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 subset of the entire framebuffer, and can either be:

A sample region, which is set of sample (x, y, layer, sample) coordinates that is a subset of the entire framebuffer, or
A fragment region, which is a set of fragment (x, y, layer) 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, the framebuffer region is a fragment region and it is a pixel granularity dependency. If N equals M, and if the VkSubpassDescription::flags does not specify the VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM flag, 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, the framebuffer region is a sample region and 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, either the VkSubpassDescription::flags VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM flag is required, or 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:

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.
The order in which VkSubmitInfo structures are specified in the pSubmits parameter of vkQueueSubmit, from lowest index to highest.
The order in which command buffers are specified in the pCommandBuffers member of VkSubmitInfo, from lowest index to highest.
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.

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.

Query commands, the order of primitives passing through the graphics pipeline and image layout transitions as part of an image memory barrier provide additional guarantees based on submission order.

Execution of pipeline stages within a given command also has a loose ordering, dependent only on a single command.

Signal operation order is a fundamental ordering in Vulkan, giving meaning to the order in which semaphore and fence signal operations occur when submitted to a single queue. The signal operation order for queue operations is determined as follows:

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.
The order in which VkSubmitInfo structures are specified in the pSubmits parameter of vkQueueSubmit, from lowest index to highest.
The fence signal operation defined by the fence parameter of a vkQueueSubmit or vkQueueBindSparse command is ordered after all semaphore signal operations defined by that command.

Semaphore signal operations defined by a single VkSubmitInfo or VkBindSparseInfo structure are unordered with respect to other semaphore signal operations defined within the same VkSubmitInfo or VkBindSparseInfo structure.

The vkSignalSemaphore command does not execute on a queue but instead performs the signal operation from the host. The semaphore signal operation defined by executing a vkSignalSemaphore command happens-after the vkSignalSemaphore command is invoked and happens-before the command returns.

Note

When signaling timeline semaphores, it is the responsibility of the application to ensure that they are ordered such that the semaphore value is strictly increasing. Because the first synchronization scope for a semaphore signal operation contains all semaphore signal operations which occur earlier in submission order, all semaphore signal operations contained in any given batch are guaranteed to happen-after all semaphore signal operations contained in any previous batches. However, no ordering guarantee is provided between the semaphore signal operations defined within a single batch. This, combined with the requirement that timeline semaphore values strictly increase, means that it is invalid to signal the same timeline semaphore twice within a single batch.

If an application wishes to ensure that some semaphore signal operation happens-after some other semaphore signal operation, it can submit a separate batch containing only semaphore signal operations, which will happen-after the semaphore signal operations in any earlier batches.

When signaling a semaphore from the host, the only ordering guarantee is that the signal operation happens-after when vkSignalSemaphore is called and happens-before it returns. Therefore, it is invalid to call vkSignalSemaphore while there are any outstanding signal operations on that semaphore from any queue submissions unless those queue submissions have some dependency which ensures that they happen-after the host signal operation. One example of this would be if the pending signal operation is, itself, waiting on the same semaphore at a lower value and the call to vkSignalSemaphore signals that lower value. Furthermore, if there are two or more processes or threads signaling the same timeline semaphore from the host, the application must ensure that the vkSignalSemaphore with the lower semaphore value returns before vkSignalSemaphore is called with the higher value.

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkFence)

To create a fence, call:

// Provided by VK_VERSION_1_0
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 a VkFenceCreateInfo structure containing information about how the fence is to be created.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pFence is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkFenceCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkFenceCreateFlagBits specifying the initial state and behavior of the fence.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_FENCE_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 VkExportFenceCreateInfo or VkExportFenceWin32HandleInfoKHR
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkFenceCreateFlagBits values

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
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 a VkExportFenceCreateInfo structure to the pNext chain of the VkFenceCreateInfo structure. The VkExportFenceCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkExportFenceCreateInfo {
    VkStructureType                   sType;
    const void*                       pNext;
    VkExternalFenceHandleTypeFlags    handleTypes;
} VkExportFenceCreateInfo;

or the equivalent

// Provided by VK_KHR_external_fence
typedef VkExportFenceCreateInfo VkExportFenceCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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

The bits in handleTypes must be supported and compatible, as reported by VkExternalFenceProperties

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_EXPORT_FENCE_CREATE_INFO
handleTypes must be a valid combination of VkExternalFenceHandleTypeFlagBits values

To specify additional attributes of NT handles exported from a fence, add a VkExportFenceWin32HandleInfoKHR structure to the pNext chain of the VkFenceCreateInfo structure. The VkExportFenceWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_fence_win32
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 a structure extending this 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 VkExportFenceCreateInfo is not present in the same pNext chain, this structure is ignored.

If VkExportFenceCreateInfo is present in the pNext chain of VkFenceCreateInfo with a Windows handleType, but either VkExportFenceWin32HandleInfoKHR is not present in the pNext chain, or if it is but 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

1: https://docs.microsoft.com/en-us/windows/win32/sync/synchronization-object-security-and-access-rights

Valid Usage

If VkExportFenceCreateInfo::handleTypes does not include VK_EXTERNAL_FENCE_HANDLE_TYPE_OPAQUE_WIN32_BIT, a VkExportFenceWin32HandleInfoKHR structure must not be included in the pNext chain of VkFenceCreateInfo

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:

// Provided by VK_KHR_external_fence_win32
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 a 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

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkFenceGetWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_fence_win32
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 a structure extending this 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:

// Provided by VK_KHR_external_fence_fd
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 a 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 an int value

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkFenceGetFdInfoKHR structure is defined as:

// Provided by VK_KHR_external_fence_fd
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 a structure extending this 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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS
VK_NOT_READY

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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

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. Fence signal operations that are defined by vkQueueSubmit or vkQueueBindSparse additionally include in the first synchronization scope any semaphore and fence signal operations that occur earlier in signal operation 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:

// Provided by VK_VERSION_1_0
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 until the condition is satisfied or the timeout has expired, whichever is sooner.

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 condition is satisfied before the timeout has expired, vkWaitForFences returns VK_SUCCESS. Otherwise, vkWaitForFences returns VK_TIMEOUT after the timeout has expired.

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

VK_SUCCESS
VK_TIMEOUT

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:

// Provided by VK_EXT_display_control
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 a VkDeviceEventInfoEXT structure describing the event of interest to the application.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pFence is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY

The VkDeviceEventInfoEXT structure is defined as:

// Provided by VK_EXT_display_control
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 a structure extending this 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:

// Provided by VK_EXT_display_control
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:

// Provided by VK_EXT_display_control
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 a VkDisplayEventInfoEXT structure describing the event of interest to the application.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pFence is a pointer 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
Both of device, and display must have been created, allocated, or retrieved from the same VkPhysicalDevice

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY

The VkDisplayEventInfoEXT structure is defined as:

// Provided by VK_EXT_display_control
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 a structure extending this 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:

// Provided by VK_EXT_display_control
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:

// Provided by VK_KHR_external_fence_win32
VkResult vkImportFenceWin32HandleKHR(
    VkDevice                                    device,
    const VkImportFenceWin32HandleInfoKHR*      pImportFenceWin32HandleInfo);

device is the logical device that created the fence.
pImportFenceWin32HandleInfo is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportFenceWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_fence_win32
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 a structure extending this 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 a 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:

// Provided by VK_KHR_external_fence_fd
VkResult vkImportFenceFdKHR(
    VkDevice                                    device,
    const VkImportFenceFdInfoKHR*               pImportFenceFdInfo);

device is the logical device that created the fence.
pImportFenceFdInfo is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportFenceFdInfoKHR structure is defined as:

// Provided by VK_KHR_external_fence_fd
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 a structure extending this 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

handleType must be a value included in the Handle Types Supported by VkImportFenceFdInfoKHR table
fd must obey any requirements listed for handleType in external fence handle types compatibility

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 does not 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)

sType must be VK_STRUCTURE_TYPE_IMPORT_FENCE_FD_INFO_KHR
pNext must be NULL
fence must be a valid VkFence handle
flags must be a valid combination of VkFenceImportFlagBits values
handleType must be a valid VkExternalFenceHandleTypeFlagBits value

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:

// Provided by VK_VERSION_1_1
typedef enum VkFenceImportFlagBits {
    VK_FENCE_IMPORT_TEMPORARY_BIT = 0x00000001,
  // Provided by VK_KHR_external_fence
    VK_FENCE_IMPORT_TEMPORARY_BIT_KHR = VK_FENCE_IMPORT_TEMPORARY_BIT,
} VkFenceImportFlagBits;

or the equivalent

// Provided by VK_KHR_external_fence
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.

// Provided by VK_VERSION_1_1
typedef VkFlags VkFenceImportFlags;

or the equivalent

// Provided by VK_KHR_external_fence
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 queue operations or between a queue operation and the host. Binary semaphores have two states - signaled and unsignaled. Timeline semaphores have a monotonically increasing 64-bit unsigned integer payload and are signaled with respect to a particular reference value. A semaphore can be signaled after execution of a queue operation is completed, and a queue operation can wait for a semaphore to become signaled before it begins execution. A timeline semaphore can additionally be signaled from the host with the vkSignalSemaphore command and waited on from the host with the vkWaitSemaphores command.

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.

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSemaphore)

To create a semaphore, call:

// Provided by VK_VERSION_1_0
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 a VkSemaphoreCreateInfo structure containing information about how the semaphore is to be created.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pSemaphore is a pointer to a handle in which the resulting semaphore object is returned.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkSemaphoreCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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, VkExportSemaphoreWin32HandleInfoKHR, or VkSemaphoreTypeCreateInfo
The sType value of each struct in the pNext chain must be unique
flags must be 0

// Provided by VK_VERSION_1_0
typedef VkFlags VkSemaphoreCreateFlags;

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

To create a semaphore of a specific type, add a VkSemaphoreTypeCreateInfo structure to the pNext chain of the VkSemaphoreCreateInfo structure. The VkSemaphoreTypeCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSemaphoreTypeCreateInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkSemaphoreType    semaphoreType;
    uint64_t           initialValue;
} VkSemaphoreTypeCreateInfo;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreTypeCreateInfo VkSemaphoreTypeCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
semaphoreType is a VkSemaphoreType value specifying the type of the semaphore.
initialValue is the initial payload value if semaphoreType is VK_SEMAPHORE_TYPE_TIMELINE.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SEMAPHORE_TYPE_CREATE_INFO
semaphoreType must be a valid VkSemaphoreType value

Valid Usage

If the timelineSemaphore feature is not enabled, semaphoreType must not equal VK_SEMAPHORE_TYPE_TIMELINE
If semaphoreType is VK_SEMAPHORE_TYPE_BINARY, initialValue must be zero

If no VkSemaphoreTypeCreateInfo structure is included in the pNext chain of VkSemaphoreCreateInfo, then the created semaphore will have a default VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY.

Possible values of VkSemaphoreTypeCreateInfo::semaphoreType, specifying the type of a semaphore, are:

// Provided by VK_VERSION_1_2
typedef enum VkSemaphoreType {
    VK_SEMAPHORE_TYPE_BINARY = 0,
    VK_SEMAPHORE_TYPE_TIMELINE = 1,
  // Provided by VK_KHR_timeline_semaphore
    VK_SEMAPHORE_TYPE_BINARY_KHR = VK_SEMAPHORE_TYPE_BINARY,
  // Provided by VK_KHR_timeline_semaphore
    VK_SEMAPHORE_TYPE_TIMELINE_KHR = VK_SEMAPHORE_TYPE_TIMELINE,
} VkSemaphoreType;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreType VkSemaphoreTypeKHR;

VK_SEMAPHORE_TYPE_BINARY specifies a binary semaphore type that has a boolean payload indicating whether the semaphore is currently signaled or unsignaled. When created, the semaphore is in the unsignaled state.
VK_SEMAPHORE_TYPE_TIMELINE specifies a timeline semaphore type that has a monotonically increasing 64-bit unsigned integer payload indicating whether the semaphore is signaled with respect to a particular reference value. When created, the semaphore payload has the value given by the initialValue field of VkSemaphoreTypeCreateInfo.

To create a semaphore whose payload can be exported to external handles, add a VkExportSemaphoreCreateInfo structure to the pNext chain of the VkSemaphoreCreateInfo structure. The VkExportSemaphoreCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkExportSemaphoreCreateInfo {
    VkStructureType                       sType;
    const void*                           pNext;
    VkExternalSemaphoreHandleTypeFlags    handleTypes;
} VkExportSemaphoreCreateInfo;

or the equivalent

// Provided by VK_KHR_external_semaphore
typedef VkExportSemaphoreCreateInfo VkExportSemaphoreCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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

The bits in handleTypes must be supported and compatible, as reported by VkExternalSemaphoreProperties

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_EXPORT_SEMAPHORE_CREATE_INFO
handleTypes must be a valid combination of VkExternalSemaphoreHandleTypeFlagBits values

To specify additional attributes of NT handles exported from a semaphore, add a VkExportSemaphoreWin32HandleInfoKHR structure to the pNext chain of the VkSemaphoreCreateInfo structure. The VkExportSemaphoreWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_semaphore_win32
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 a structure extending this 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 VkExportSemaphoreCreateInfo is not present in the same pNext chain, this structure is ignored.

If VkExportSemaphoreCreateInfo is present in the pNext chain of VkSemaphoreCreateInfo with a Windows handleType, but either VkExportSemaphoreWin32HandleInfoKHR is not present in the pNext chain, or if it is but 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 used depend on the handle type.

For handles of the following types:

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT

The implementation must ensure the access rights allow both signal and wait operations on the semaphore.

For handles of the following types:

VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE_BIT

The access rights must be:

GENERIC_ALL

1: https://docs.microsoft.com/en-us/windows/win32/sync/synchronization-object-security-and-access-rights

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 included 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:

// Provided by VK_KHR_external_semaphore_win32
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 a 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

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkSemaphoreGetWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_semaphore_win32
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 a structure extending this 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:

// Provided by VK_KHR_external_semaphore_fd
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 a 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

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 an int value

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkSemaphoreGetFdInfoKHR structure is defined as:

// Provided by VK_KHR_external_semaphore_fd
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 a structure extending this 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
If handleType refers to a handle type with copy payload transference semantics, semaphore must have been created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY
If handleType refers to a handle type with copy payload transference semantics, semaphore must have an associated semaphore signal operation that has been submitted for execution and any semaphore signal operations on which it depends (if any) must have also been submitted for execution

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:

// Provided by VK_VERSION_1_0
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.

In case of semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE the semaphore is considered signaled with respect to the counter value set to be signaled as specified in VkTimelineSemaphoreSubmitInfo or VkSemaphoreSignalInfo.

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. Semaphore signal operations that are defined by vkQueueSubmit or vkQueueBindSparse additionally include in the first synchronization scope any semaphore and fence signal operations that occur earlier in signal operation 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

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 wait operations.

Such semaphore wait operations set the semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_BINARY to the unsignaled state. In case of semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE a prior semaphore signal operation defines a memory dependency with a semaphore wait operation if the value the semaphore is signaled with is greater than or equal to the value the semaphore is waited with, thus the semaphore will continue to be considered signaled with respect to the counter value waited on as specified in VkTimelineSemaphoreSubmitInfo.

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 wait 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 timeline semaphores, fences or events, the act of waiting for a binary semaphore also unsignals that semaphore. Applications must ensure that between two such wait operations, the semaphore is signaled again, with execution dependencies used to ensure these occur in order. Binary 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 operation is submitted to a queue:

A binary semaphore must be signaled, or have an associated semaphore signal operation that is pending execution.
Any semaphore signal operations on which the pending binary semaphore signal operation depends must also be completed or pending execution.
There must be no other queue waiting on the same binary semaphore when the operation executes.

6.4.4. Host Operations on Semaphores

In addition to semaphore signal operations and semaphore wait operations submitted to device queues, timeline semaphores support the following host operations:

Query the current counter value of the semaphore using the vkGetSemaphoreCounterValue command.
Wait for a set of semaphores to reach particular counter values using the vkWaitSemaphores command.
Signal the semaphore with a particular counter value from the host using the vkSignalSemaphore command.

To query the current counter value of a semaphore created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE from the host, call:

// Provided by VK_VERSION_1_2
VkResult vkGetSemaphoreCounterValue(
    VkDevice                                    device,
    VkSemaphore                                 semaphore,
    uint64_t*                                   pValue);

or the equivalent command

// Provided by VK_KHR_timeline_semaphore
VkResult vkGetSemaphoreCounterValueKHR(
    VkDevice                                    device,
    VkSemaphore                                 semaphore,
    uint64_t*                                   pValue);

device is the logical device that owns the semaphore.
semaphore is the handle of the semaphore to query.
pValue is a pointer to a 64-bit integer value in which the current counter value of the semaphore is returned.

Note

If a queue submission command is pending execution, then the value returned by this command may immediately be out of date.

Valid Usage

semaphore must have been created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE

Valid Usage (Implicit)

device must be a valid VkDevice handle
semaphore must be a valid VkSemaphore handle
pValue must be a valid pointer to a uint64_t value
semaphore must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_DEVICE_LOST

To wait for a set of semaphores created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE to reach particular counter values on the host, call:

// Provided by VK_VERSION_1_2
VkResult vkWaitSemaphores(
    VkDevice                                    device,
    const VkSemaphoreWaitInfo*                  pWaitInfo,
    uint64_t                                    timeout);

or the equivalent command

// Provided by VK_KHR_timeline_semaphore
VkResult vkWaitSemaphoresKHR(
    VkDevice                                    device,
    const VkSemaphoreWaitInfo*                  pWaitInfo,
    uint64_t                                    timeout);

device is the logical device that owns the semaphore.
pWaitInfo is a pointer to a VkSemaphoreWaitInfo structure containing information about the wait condition.
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 vkWaitSemaphores is called, then vkWaitSemaphores returns immediately. If the condition is not satisfied at the time vkWaitSemaphores is called, then vkWaitSemaphores will block and wait until the condition is satisfied or the timeout has expired, whichever is sooner.

If timeout is zero, then vkWaitSemaphores does not wait, but simply returns information about the current state of the semaphore. VK_TIMEOUT will be returned in this case if the condition is not satisfied, even though no actual wait was performed.

If the condition is satisfied before the timeout has expired, vkWaitSemaphores returns VK_SUCCESS. Otherwise, vkWaitSemaphores returns VK_TIMEOUT after the timeout has expired.

If device loss occurs (see Lost Device) before the timeout has expired, vkWaitSemaphores must return in finite time with either VK_SUCCESS or VK_ERROR_DEVICE_LOST.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pWaitInfo must be a valid pointer to a valid VkSemaphoreWaitInfo structure

Return Codes

VK_SUCCESS
VK_TIMEOUT

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_DEVICE_LOST

The VkSemaphoreWaitInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSemaphoreWaitInfo {
    VkStructureType         sType;
    const void*             pNext;
    VkSemaphoreWaitFlags    flags;
    uint32_t                semaphoreCount;
    const VkSemaphore*      pSemaphores;
    const uint64_t*         pValues;
} VkSemaphoreWaitInfo;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreWaitInfo VkSemaphoreWaitInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkSemaphoreWaitFlagBits specifying additional parameters for the semaphore wait operation.
semaphoreCount is the number of semaphores to wait on.
pSemaphores is a pointer to an array of semaphoreCount semaphore handles to wait on.
pValues is a pointer to an array of semaphoreCount timeline semaphore values.

Valid Usage

All of the elements of pSemaphores must reference a semaphore that was created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SEMAPHORE_WAIT_INFO
pNext must be NULL
flags must be a valid combination of VkSemaphoreWaitFlagBits values
pSemaphores must be a valid pointer to an array of semaphoreCount valid VkSemaphore handles
pValues must be a valid pointer to an array of semaphoreCount uint64_t values
semaphoreCount must be greater than 0

Bits which can be set in VkSemaphoreWaitInfo::flags, specifying additional parameters of a semaphore wait operation, are:

// Provided by VK_VERSION_1_2
typedef enum VkSemaphoreWaitFlagBits {
    VK_SEMAPHORE_WAIT_ANY_BIT = 0x00000001,
  // Provided by VK_KHR_timeline_semaphore
    VK_SEMAPHORE_WAIT_ANY_BIT_KHR = VK_SEMAPHORE_WAIT_ANY_BIT,
} VkSemaphoreWaitFlagBits;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreWaitFlagBits VkSemaphoreWaitFlagBitsKHR;

VK_SEMAPHORE_WAIT_ANY_BIT specifies that the semaphore wait condition is that at least one of the semaphores in VkSemaphoreWaitInfo::pSemaphores has reached the value specified by the corresponding element of VkSemaphoreWaitInfo::pValues. If VK_SEMAPHORE_WAIT_ANY_BIT is not set, the semaphore wait condition is that all of the semaphores in VkSemaphoreWaitInfo::pSemaphores have reached the value specified by the corresponding element of VkSemaphoreWaitInfo::pValues.

// Provided by VK_VERSION_1_2
typedef VkFlags VkSemaphoreWaitFlags;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreWaitFlags VkSemaphoreWaitFlagsKHR;

VkSemaphoreWaitFlags is a bitmask type for setting a mask of zero or more VkSemaphoreWaitFlagBits.

To signal a semaphore created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE with a particular counter value, on the host, call:

// Provided by VK_VERSION_1_2
VkResult vkSignalSemaphore(
    VkDevice                                    device,
    const VkSemaphoreSignalInfo*                pSignalInfo);

or the equivalent command

// Provided by VK_KHR_timeline_semaphore
VkResult vkSignalSemaphoreKHR(
    VkDevice                                    device,
    const VkSemaphoreSignalInfo*                pSignalInfo);

device is the logical device that owns the semaphore.
pSignalInfo is a pointer to a VkSemaphoreSignalInfo structure containing information about the signal operation.

When vkSignalSemaphore is executed on the host, it defines and immediately executes a semaphore signal operation which sets the timeline semaphore to the given value.

The first synchronization scope is defined by the host execution model, but includes execution of vkSignalSemaphore on the host and anything that happened-before it.

The second synchronization scope is empty.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pSignalInfo must be a valid pointer to a valid VkSemaphoreSignalInfo structure

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkSemaphoreSignalInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSemaphoreSignalInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkSemaphore        semaphore;
    uint64_t           value;
} VkSemaphoreSignalInfo;

or the equivalent

// Provided by VK_KHR_timeline_semaphore
typedef VkSemaphoreSignalInfo VkSemaphoreSignalInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
semaphore is the handle of the semaphore to signal.
value is the value to signal.

Valid Usage

semaphore must have been created with a VkSemaphoreType of VK_SEMAPHORE_TYPE_TIMELINE
value must have a value greater than the current value of the semaphore
value must be less than the value of any pending semaphore signal operations
value must have a value which does not differ from the current value of the semaphore or the value of any outstanding semaphore wait or signal operation on semaphore by more than maxTimelineSemaphoreValueDifference

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SEMAPHORE_SIGNAL_INFO
pNext must be NULL
semaphore must be a valid VkSemaphore handle

6.4.5. 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 occurred 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.

In addition, when importing a semaphore payload that is not compatible with the payload type corresponding to the VkSemaphoreType the semaphore was created with, the implementation may fail the semaphore payload import operation with the error code VK_ERROR_INVALID_EXTERNAL_HANDLE.

Note

To import a semaphore payload from a Windows handle, call:

// Provided by VK_KHR_external_semaphore_win32
VkResult vkImportSemaphoreWin32HandleKHR(
    VkDevice                                    device,
    const VkImportSemaphoreWin32HandleInfoKHR*  pImportSemaphoreWin32HandleInfo);

device is the logical device that created the semaphore.
pImportSemaphoreWin32HandleInfo is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportSemaphoreWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_semaphore_win32
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 a structure extending this 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
If handleType is VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT or VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT, the VkSemaphoreCreateInfo::flags field must match that of the semaphore from which handle or name was exported
If handleType is VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_BIT or VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT_BIT, the VkSemaphoreTypeCreateInfo::semaphoreType field must match that of the semaphore from which handle or name was exported
If flags contains VK_SEMAPHORE_IMPORT_TEMPORARY_BIT, the VkSemaphoreTypeCreateInfo::semaphoreType field of the semaphore from which handle or name was exported must not be VK_SEMAPHORE_TYPE_TIMELINE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_WIN32_HANDLE_INFO_KHR
pNext must be NULL
semaphore must be a valid VkSemaphore handle
flags must be a valid combination of VkSemaphoreImportFlagBits values
If handleType is not 0, handleType must be a valid VkExternalSemaphoreHandleTypeFlagBits value

Host Synchronization

Host access to semaphore must be externally synchronized

To import a semaphore payload from a POSIX file descriptor, call:

// Provided by VK_KHR_external_semaphore_fd
VkResult vkImportSemaphoreFdKHR(
    VkDevice                                    device,
    const VkImportSemaphoreFdInfoKHR*           pImportSemaphoreFdInfo);

device is the logical device that created the semaphore.
pImportSemaphoreFdInfo is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkImportSemaphoreFdInfoKHR structure is defined as:

// Provided by VK_KHR_external_semaphore_fd
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 a structure extending this 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

handleType must be a value included in the Handle Types Supported by VkImportSemaphoreFdInfoKHR table
fd must obey any requirements listed for handleType in external semaphore handle types compatibility
If handleType is VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT, the VkSemaphoreCreateInfo::flags field must match that of the semaphore from which fd was exported
If handleType is VK_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD_BIT, the VkSemaphoreTypeCreateInfo::semaphoreType field must match that of the semaphore from which fd was exported
If flags contains VK_SEMAPHORE_IMPORT_TEMPORARY_BIT, the VkSemaphoreTypeCreateInfo::semaphoreType field of the semaphore from which fd was exported must not be VK_SEMAPHORE_TYPE_TIMELINE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMPORT_SEMAPHORE_FD_INFO_KHR
pNext must be NULL
semaphore must be a valid VkSemaphore handle
flags must be a valid combination of VkSemaphoreImportFlagBits values
handleType must be a valid VkExternalSemaphoreHandleTypeFlagBits value

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:

// Provided by VK_VERSION_1_1
typedef enum VkSemaphoreImportFlagBits {
    VK_SEMAPHORE_IMPORT_TEMPORARY_BIT = 0x00000001,
  // Provided by VK_KHR_external_semaphore
    VK_SEMAPHORE_IMPORT_TEMPORARY_BIT_KHR = VK_SEMAPHORE_IMPORT_TEMPORARY_BIT,
} VkSemaphoreImportFlagBits;

or the equivalent

// Provided by VK_KHR_external_semaphore
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.

// Provided by VK_VERSION_1_1
typedef VkFlags VkSemaphoreImportFlags;

or the equivalent

// Provided by VK_KHR_external_semaphore
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 or unsignal an event either on the host or on the device. A device can be made to 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkEvent)

To create an event, call:

// Provided by VK_VERSION_1_0
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 a VkEventCreateInfo structure containing information about how the event is to be created.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pEvent is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkEventCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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

// Provided by VK_VERSION_1_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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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

VK_EVENT_SET
VK_EVENT_RESET

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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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 the first synchronization scope.

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

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
If the conditional rendering feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in stageMask 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
stageMask must not include VK_PIPELINE_STAGE_HOST_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

Primary
Secondary

Outside

Graphics
Compute

To set the state of an event to unsignaled from a device, call:

// Provided by VK_VERSION_1_0
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 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

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
If the conditional rendering feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, stageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in stageMask 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
stageMask must not include VK_PIPELINE_STAGE_HOST_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

Primary
Secondary

Outside

Graphics
Compute

To wait for one or more events to enter the signaled state on a device, call:

// Provided by VK_VERSION_1_0
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 a pointer to 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 occurring 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.

Unlike vkCmdPipelineBarrier, a queue family ownership transfer cannot be performed using vkCmdWaitEvents.

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

If the geometry shaders feature is not enabled, srcStageMask 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 conditional rendering feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in srcStageMask 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

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, dstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT
If the conditional rendering feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in 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

The srcAccessMask member of each element of pMemoryBarriers must only include access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
The dstAccessMask member of each element of pMemoryBarriers must only include access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types
For any element of pBufferMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its srcQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its srcAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
For any element of pBufferMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its dstQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its dstAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types
For any element of pImageMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its srcQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its srcAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
For any element of pImageMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its dstQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its dstAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types
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 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
The srcQueueFamilyIndex and dstQueueFamilyIndex members of any element of pBufferMemoryBarriers or pImageMemoryBarriers must be equal
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

Primary
Secondary

Both

Graphics
Compute

6.6. Pipeline Barriers

To record a pipeline barrier, call:

// Provided by VK_VERSION_1_0
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.

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 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 conditional rendering feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in srcStageMask 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

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, dstStageMask must not contain VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT
If the conditional rendering feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
Any pipeline stage included in 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

The srcAccessMask member of each element of pMemoryBarriers must only include access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
The dstAccessMask member of each element of pMemoryBarriers must only include access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types
For any element of pBufferMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its srcQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its srcAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
For any element of pBufferMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its dstQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its dstAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types
For any element of pImageMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its srcQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its srcAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in srcStageMask, as specified in the table of supported access types
For any element of pImageMemoryBarriers, if its srcQueueFamilyIndex and dstQueueFamilyIndex members are equal, or if its dstQueueFamilyIndex is the queue family index that was used to create the command pool that commandBuffer was allocated from, then its dstAccessMask member must only contain access flags that are supported by one or more of the pipeline stages in dstStageMask, as specified in the table of supported access types

If vkCmdPipelineBarrier is called within a render pass instance, the render pass must have been created with at least one VkSubpassDependency instance in VkRenderPassCreateInfo::pDependencies that expresses a dependency from the current subpass to itself, with synchronization scopes and access scopes that are all supersets of the scopes defined in this command
If vkCmdPipelineBarrier is called within a render pass instance, it must not include any buffer memory barriers
If vkCmdPipelineBarrier is called within a render pass instance, the image member of any image memory barrier included in this command must be an attachment used in the current subpass both as an input attachment, and as either a color or depth/stencil attachment
If vkCmdPipelineBarrier is called within a render pass instance, the oldLayout and newLayout members of any image memory barrier included in this command must be equal
If vkCmdPipelineBarrier is called within a render pass instance, the srcQueueFamilyIndex and dstQueueFamilyIndex members of any image memory barrier included in this command must be equal
If vkCmdPipelineBarrier is called outside of a render pass instance, VK_DEPENDENCY_VIEW_LOCAL_BIT must not be included in the dependency flags

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

Primary
Secondary

Both

Transfer
Graphics
Compute

Bits which can be set in vkCmdPipelineBarrier::dependencyFlags, specifying how execution and memory dependencies are formed, are:

// Provided by VK_VERSION_1_0
typedef enum VkDependencyFlagBits {
    VK_DEPENDENCY_BY_REGION_BIT = 0x00000001,
  // Provided by VK_VERSION_1_1
    VK_DEPENDENCY_DEVICE_GROUP_BIT = 0x00000004,
  // Provided by VK_VERSION_1_1
    VK_DEPENDENCY_VIEW_LOCAL_BIT = 0x00000002,
  // Provided by VK_KHR_multiview
    VK_DEPENDENCY_VIEW_LOCAL_BIT_KHR = VK_DEPENDENCY_VIEW_LOCAL_BIT,
  // Provided by VK_KHR_device_group
    VK_DEPENDENCY_DEVICE_GROUP_BIT_KHR = VK_DEPENDENCY_DEVICE_GROUP_BIT,
} VkDependencyFlagBits;

VK_DEPENDENCY_BY_REGION_BIT specifies that dependencies will be framebuffer-local.
VK_DEPENDENCY_VIEW_LOCAL_BIT specifies that a subpass has more than one view.
VK_DEPENDENCY_DEVICE_GROUP_BIT specifies that dependencies are non-device-local dependency.

// Provided by VK_VERSION_1_0
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 subpass dependency with srcSubpass and dstSubpass both 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. If any of the stages in srcStages are framebuffer-space stages, dstStages must only contain framebuffer-space stages. This means that pseudo-stages like VK_PIPELINE_STAGE_ALL_COMMANDS_BIT which include the execution of both framebuffer-space stages and non-framebuffer-space stages must not be used.

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.

Each of the synchronization scopes and access scopes of a vkCmdPipelineBarrier command inside a render pass instance must be a subset of the scopes of one of the self-dependencies for the current subpass.

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 must not include buffer memory barriers. Image memory barriers must only specify image subresources that are used as attachments within the subpass, and must not define an image layout transition or queue family ownership transfer.

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:

// Provided by VK_VERSION_1_0
typedef struct VkMemoryBarrier {
    VkStructureType    sType;
    const void*        pNext;
    VkAccessFlags      srcAccessMask;
    VkAccessFlags      dstAccessMask;
} VkMemoryBarrier;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
srcAccessMask is a bitmask of VkAccessFlagBits specifying a source access mask.
dstAccessMask is a bitmask of VkAccessFlagBits specifying a destination access mask.

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 a 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:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object
If srcQueueFamilyIndex is not equal to dstQueueFamilyIndex, at least one must not be 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_CONCURRENT, srcQueueFamilyIndex and dstQueueFamilyIndex are not equal, and one of srcQueueFamilyIndex and dstQueueFamilyIndex is a special queue family values reserved for external memory transfers, the other must be VK_QUEUE_FAMILY_IGNORED
If buffer was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex and dstQueueFamilyIndex are not equal, srcQueueFamilyIndex and dstQueueFamilyIndex must both be valid queue families, or one of the special queue family values reserved for external memory transfers, as described in Queue Family Ownership Transfer
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

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BUFFER_MEMORY_BARRIER
pNext must be NULL
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:

// Provided by VK_VERSION_1_0
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;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
srcAccessMask is a bitmask of VkAccessFlagBits specifying a source access mask.
dstAccessMask is a bitmask of VkAccessFlagBits specifying a destination access mask.
oldLayout is the old layout in an image layout transition.
newLayout is the new layout in an image layout transition.
srcQueueFamilyIndex is the source queue family for a queue family ownership transfer.
dstQueueFamilyIndex is the destination queue family for a queue family ownership transfer.
image is a handle to the image affected by this barrier.
subresourceRange describes the image subresource range within image that is affected by this barrier.

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.

oldLayout and newLayout define an image layout transition for the specified image subresource range.

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

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 is non-sparse then it must be bound completely and contiguously to a single VkDeviceMemory object
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, oldLayout must be VK_IMAGE_LAYOUT_UNDEFINED or the current layout of the image subresources affected by the barrier
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, newLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and 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 srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL_KHR then image must have been created with at least one of VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, VK_IMAGE_USAGE_SAMPLED_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT set
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and oldLayout or newLayout is VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL_KHR then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and oldLayout or newLayout is VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL_KHR then image must have been created with at least one of VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, VK_IMAGE_USAGE_SAMPLED_BIT, or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT set
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and oldLayout or newLayout is VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL_KHR then image must have been created with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT set
If srcQueueFamilyIndex and dstQueueFamilyIndex define a queue family ownership transfer or oldLayout and newLayout define a image layout transition, and oldLayout or newLayout is VK_IMAGE_LAYOUT_SHADING_RATE_OPTIMAL_NV then image must have been created with VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV set
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 image has a depth/stencil format with both depth and stencil and the separateDepthStencilLayouts feature is enabled, then the aspectMask member of subresourceRange must include either or both VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT
If image has a depth/stencil format with both depth and stencil and the separateDepthStencilLayouts feature is not enabled, then the aspectMask member of subresourceRange must include both VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT
If srcQueueFamilyIndex is not equal to dstQueueFamilyIndex, at least one must not be a special queue family reserved for external memory ownership transfers, as described in Queue Family Ownership Transfer
If image was created with a sharing mode of VK_SHARING_MODE_CONCURRENT, srcQueueFamilyIndex and dstQueueFamilyIndex are not equal, and one of srcQueueFamilyIndex and dstQueueFamilyIndex is a special queue family values reserved for external memory transfers, the other must be VK_QUEUE_FAMILY_IGNORED
If image was created with a sharing mode of VK_SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex and dstQueueFamilyIndex are not equal, srcQueueFamilyIndex and dstQueueFamilyIndex must both be valid queue families, or one of the special queue family values reserved for external memory transfers, as described in Queue Family Ownership Transfer
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

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
The sType value of each struct in the pNext chain must be unique
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:

Release exclusive ownership from the source queue family
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) using vkCmdPipelineBarrier, 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. dstAccessMask 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, and happens-before operations specified in the second synchronization scope of the calling command.

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) using vkCmdPipelineBarrier, on a queue from the destination queue family. The buffer range or image subresource range specified in an acquire operation must match exactly that of a previous release operation. 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. srcAccessMask 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-after operations in the first synchronization scope of the calling command, and 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 operation's memory barrier must be equal to values of oldLayout and newLayout in the acquire operation's 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 a 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.

Note

Because events cannot be used directly for inter-queue synchronization, and because vkCmdSetEvent does not have the queue family index or memory barrier parameters needed by a release operation, the release and acquire operations of a queue family ownership transfer can only be performed using vkCmdPipelineBarrier.

6.8. Wait Idle Operations

To wait on the host for the completion of outstanding queue operations for a given queue, call:

// Provided by VK_VERSION_1_0
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

Host Synchronization

Host access to queue must be externally synchronized

Command Properties

Any

Return Codes

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

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 a queue submission command, 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 available to the host memory domain.

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).

6.11. Calibrated timestamps

In order to be able to correlate the time a particular operation took place at on timelines of different time domains (e.g. a device operation vs a host operation), Vulkan allows querying calibrated timestamps from multiple time domains.

To query calibrated timestamps from a set of time domains, call:

// Provided by VK_EXT_calibrated_timestamps
VkResult vkGetCalibratedTimestampsEXT(
    VkDevice                                    device,
    uint32_t                                    timestampCount,
    const VkCalibratedTimestampInfoEXT*         pTimestampInfos,
    uint64_t*                                   pTimestamps,
    uint64_t*                                   pMaxDeviation);

device is the logical device used to perform the query.
timestampCount is the number of timestamps to query.
pTimestampInfos is a pointer to an array of timestampCount VkCalibratedTimestampInfoEXT structures, describing the time domains the calibrated timestamps should be captured from.
pTimestamps is a pointer to an array of timestampCount 64-bit unsigned integer values in which the requested calibrated timestamp values are returned.
pMaxDeviation is a pointer to a 64-bit unsigned integer value in which the strictly positive maximum deviation, in nanoseconds, of the calibrated timestamp values is returned.

Note

The maximum deviation may vary between calls to vkGetCalibratedTimestampsEXT even for the same set of time domains due to implementation and platform specific reasons. It is the application’s responsibility to assess whether the returned maximum deviation makes the timestamp values suitable for any particular purpose and can choose to re-issue the timestamp calibration call pursuing a lower devation value.

Calibrated timestamp values can be extrapolated to estimate future coinciding timestamp values, however, depending on the nature of the time domains and other properties of the platform extrapolating values over a sufficiently long period of time may no longer be accurate enough to fit any particular purpose so applications are expected to re-calibrate the timestamps on a regular basis.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pTimestampInfos must be a valid pointer to an array of timestampCount valid VkCalibratedTimestampInfoEXT structures
pTimestamps must be a valid pointer to an array of timestampCount uint64_t values
pMaxDeviation must be a valid pointer to a uint64_t value
timestampCount must be greater than 0

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkCalibratedTimestampInfoEXT structure is defined as:

// Provided by VK_EXT_calibrated_timestamps
typedef struct VkCalibratedTimestampInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkTimeDomainEXT    timeDomain;
} VkCalibratedTimestampInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
timeDomain is a VkTimeDomainEXT value specifying the time domain from which the calibrated timestamp value should be returned.

Valid Usage

timeDomain must be one of the VkTimeDomainEXT values returned by vkGetPhysicalDeviceCalibrateableTimeDomainsEXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_CALIBRATED_TIMESTAMP_INFO_EXT
pNext must be NULL
timeDomain must be a valid VkTimeDomainEXT value

The set of supported time domains consists of:

// Provided by VK_EXT_calibrated_timestamps
typedef enum VkTimeDomainEXT {
    VK_TIME_DOMAIN_DEVICE_EXT = 0,
    VK_TIME_DOMAIN_CLOCK_MONOTONIC_EXT = 1,
    VK_TIME_DOMAIN_CLOCK_MONOTONIC_RAW_EXT = 2,
    VK_TIME_DOMAIN_QUERY_PERFORMANCE_COUNTER_EXT = 3,
} VkTimeDomainEXT;

VK_TIME_DOMAIN_DEVICE_EXT specifies the device time domain. Timestamp values in this time domain use the same units and are comparable with device timestamp values captured using vkCmdWriteTimestamp and are defined to be incrementing according to the timestampPeriod of the device.
VK_TIME_DOMAIN_CLOCK_MONOTONIC_EXT specifies the CLOCK_MONOTONIC time domain available on POSIX platforms. Timestamp values in this time domain are in units of nanoseconds and are comparable with platform timestamp values captured using the POSIX clock_gettime API as computed by this example:

struct timespec tv;
clock_gettime(CLOCK_MONOTONIC, &tv);
return tv.tv_nsec + tv.tv_sec*1000000000ull;

VK_TIME_DOMAIN_CLOCK_MONOTONIC_RAW_EXT specifies the CLOCK_MONOTONIC_RAW time domain available on POSIX platforms. Timestamp values in this time domain are in units of nanoseconds and are comparable with platform timestamp values captured using the POSIX clock_gettime API as computed by this example:

struct timespec tv;
clock_gettime(CLOCK_MONOTONIC_RAW, &tv);
return tv.tv_nsec + tv.tv_sec*1000000000ull;

VK_TIME_DOMAIN_QUERY_PERFORMANCE_COUNTER_EXT specifies the performance counter (QPC) time domain available on Windows. Timestamp values in this time domain are in the same units as those provided by the Windows QueryPerformanceCounter API and are comparable with platform timestamp values captured using that API as computed by this example:

LARGE_INTEGER counter;
QueryPerformanceCounter(&counter);
return counter.QuadPart;

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:

// Provided by VK_VERSION_1_0
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, perform shader resolve operations to 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, depth/stencil resolve, or input attachment for that subpass (as determined by the pColorAttachments, pDepthStencilAttachment, pResolveAttachments, VkSubpassDescriptionDepthStencilResolve::pDepthStencilResolveAttachment, 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:

// Provided by VK_VERSION_1_0
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 a VkRenderPassCreateInfo structure describing the parameters of the render pass.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pRenderPass is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkRenderPassCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkRenderPassCreateFlagBits
attachmentCount is the number of attachments used by this render pass.
pAttachments is a pointer to an array of attachmentCount VkAttachmentDescription structures describing the attachments used by the render pass.
subpassCount is the number of subpasses to create.
pSubpasses is a pointer to an array of subpassCount VkSubpassDescription structures describing each subpass.
dependencyCount is the number of memory dependencies between pairs of subpasses.
pDependencies is a pointer to an array of dependencyCount VkSubpassDependency structures describing dependencies between pairs of subpasses.

Note

Care should be taken to avoid a data race here; if any subpasses access attachments with overlapping memory locations, and one of those accesses is a write, a subpass dependency needs to be included between them.

Valid Usage

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
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 stencilLoadOp 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 a VkRenderPassInputAttachmentAspectCreateInfo structure, the subpass member of each element of its pAspectReferences member must be less than subpassCount
If the pNext chain includes a VkRenderPassInputAttachmentAspectCreateInfo structure, 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 a VkRenderPassInputAttachmentAspectCreateInfo structure, for any element of the pInputAttachments member of any element of pSubpasses where the attachment member is not VK_ATTACHMENT_UNUSED, the aspectMask member of the corresponding element of VkRenderPassInputAttachmentAspectCreateInfo::pAspectReferences must only include aspects that are present in images of the format specified by the element of pAttachments at attachment
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, and its subpassCount member is not zero, that member must be equal to the value of subpassCount
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, if its dependencyCount member is not zero, it must be equal to dependencyCount
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, for each non-zero element of pViewOffsets, the srcSubpass and dstSubpass members of pDependencies at the same index must not be equal
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, for any element of pDependencies with a dependencyFlags member that does not include VK_DEPENDENCY_VIEW_LOCAL_BIT, the corresponding element of the pViewOffsets member of that VkRenderPassMultiviewCreateInfo instance must be 0
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, elements of its pViewMasks member must either all be 0, or all not be 0
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, and each element of its pViewMasks member is 0, the dependencyFlags member of each element of pDependencies must not include VK_DEPENDENCY_VIEW_LOCAL_BIT
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, and each element of its pViewMasks member is 0, correlatedViewMaskCount must be 0
If the pNext chain includes a VkRenderPassMultiviewCreateInfo structure, each element of its pViewMask member must not have a bit set at an index greater than or equal to VkPhysicalDeviceLimits::maxFramebufferLayers
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 destination subpass
The srcSubpass member of each element of pDependencies must be less than subpassCount
The dstSubpass member of each element of pDependencies must be less than subpassCount

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 VkRenderPassFragmentDensityMapCreateInfoEXT, VkRenderPassInputAttachmentAspectCreateInfo, or VkRenderPassMultiviewCreateInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkRenderPassCreateFlagBits values
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

Bits which can be set in VkRenderPassCreateInfo::flags describing additional properties of the render pass are:

// Provided by VK_VERSION_1_0
typedef enum VkRenderPassCreateFlagBits {
  // Provided by VK_QCOM_render_pass_transform
    VK_RENDER_PASS_CREATE_TRANSFORM_BIT_QCOM = 0x00000002,
} VkRenderPassCreateFlagBits;

VK_RENDER_PASS_CREATE_TRANSFORM_BIT_QCOM specifies that the created renderpass is compatible with render pass transform.

// Provided by VK_VERSION_1_0
typedef VkFlags VkRenderPassCreateFlags;

VkRenderPassCreateFlags is a bitmask type for setting a mask of zero or more VkRenderPassCreateFlagBits.

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:

// Provided by VK_VERSION_1_1
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

// Provided by VK_KHR_multiview
typedef VkRenderPassMultiviewCreateInfo VkRenderPassMultiviewCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
subpassCount is zero or the number of subpasses in the render pass.
pViewMasks is a pointer to an array of subpassCount 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 is a pointer 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 the number of correlation masks.
pCorrelationMasks is a pointer to an array of correlationMaskCount 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

If the VkRenderPassCreateInfo::pNext chain includes a VkRenderPassFragmentDensityMapCreateInfoEXT structure, then that structure includes a fragment density map attachment for the render pass.

The VkRenderPassFragmentDensityMapCreateInfoEXT structure is defined as:

// Provided by VK_EXT_fragment_density_map
typedef struct VkRenderPassFragmentDensityMapCreateInfoEXT {
    VkStructureType          sType;
    const void*              pNext;
    VkAttachmentReference    fragmentDensityMapAttachment;
} VkRenderPassFragmentDensityMapCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
fragmentDensityMapAttachment is the fragment density map to use for the render pass.

The fragment density map is read at an implementation-dependent time with the following constraints determined by the attachment’s image view flags:

VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT specifies that the fragment density map will be read by the device during VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DEFERRED_BIT_EXT specifies that the fragment density map will be read by the host during vkEndCommandBuffer of the primary command buffer that the render pass is recorded into
Otherwise the fragment density map will be read by the host during vkCmdBeginRenderPass

The fragment density map may additionally be read by the device during VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT for any mode.

If this structure is not present, it is as if fragmentDensityMapAttachment was given as VK_ATTACHMENT_UNUSED.

Valid Usage

If fragmentDensityMapAttachment is not VK_ATTACHMENT_UNUSED, fragmentDensityMapAttachment must be less than VkRenderPassCreateInfo::attachmentCount
If fragmentDensityMapAttachment is not VK_ATTACHMENT_UNUSED, fragmentDensityMapAttachment must not be an element of VkSubpassDescription::pInputAttachments, VkSubpassDescription::pColorAttachments, VkSubpassDescription::pResolveAttachments, VkSubpassDescription::pDepthStencilAttachment, or VkSubpassDescription::pPreserveAttachments for any subpass
If fragmentDensityMapAttachment is not VK_ATTACHMENT_UNUSED, layout must be equal to VK_IMAGE_LAYOUT_FRAGMENT_DENSITY_MAP_OPTIMAL_EXT, or VK_IMAGE_LAYOUT_GENERAL
If fragmentDensityMapAttachment is not VK_ATTACHMENT_UNUSED, fragmentDensityMapAttachment must reference an attachment with a loadOp equal to VK_ATTACHMENT_LOAD_OP_LOAD or VK_ATTACHMENT_LOAD_OP_DONT_CARE
If fragmentDensityMapAttachment is not VK_ATTACHMENT_UNUSED, fragmentDensityMapAttachment must reference an attachment with a storeOp equal to VK_ATTACHMENT_STORE_OP_DONT_CARE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RENDER_PASS_FRAGMENT_DENSITY_MAP_CREATE_INFO_EXT
fragmentDensityMapAttachment must be a valid VkAttachmentReference structure

The VkAttachmentDescription structure is defined as:

// Provided by VK_VERSION_1_0
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.

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.

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.

Valid Usage

finalLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED
If format is a color format, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
If format is a depth/stencil format, initialLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL
If format is a color format, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
If format is a depth/stencil format, finalLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL
If the separateDepthStencilLayouts feature is not enabled, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If the separateDepthStencilLayouts feature is not enabled, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a color format, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a color format, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes both depth and stencil aspects, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes both depth and stencil aspects, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the depth aspect, initialLayout must not be VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the depth aspect, finalLayout must not be VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the stencil aspect, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the stencil aspect, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL

Valid Usage (Implicit)

flags must be a valid combination of VkAttachmentDescriptionFlagBits values
format must be a valid VkFormat value
samples must be a valid VkSampleCountFlagBits value
loadOp must be a valid VkAttachmentLoadOp value
storeOp must be a valid VkAttachmentStoreOp value
stencilLoadOp must be a valid VkAttachmentLoadOp value
stencilStoreOp must be a valid VkAttachmentStoreOp value
initialLayout must be a valid VkImageLayout value
finalLayout must be a valid VkImageLayout value

Bits which can be set in VkAttachmentDescription::flags describing additional properties of the attachment are:

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
typedef enum VkAttachmentStoreOp {
    VK_ATTACHMENT_STORE_OP_STORE = 0,
    VK_ATTACHMENT_STORE_OP_DONT_CARE = 1,
  // Provided by VK_QCOM_render_pass_store_ops
    VK_ATTACHMENT_STORE_OP_NONE_QCOM = 1000301000,
} 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.
VK_ATTACHMENT_STORE_OP_NONE_QCOM specifies that the contents within the render area were not written during rendering, and may not be written to memory. If the attachment was written to during the renderpass, the contents of the attachment will be undefined inside the render area.

Note

VK_ATTACHMENT_STORE_OP_DONT_CARE can cause contents generated during previous render passes to be discarded before reaching memory, even if no write to the attachment occurs during the current render pass.

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:

// Provided by VK_VERSION_1_1
typedef struct VkRenderPassInputAttachmentAspectCreateInfo {
    VkStructureType                            sType;
    const void*                                pNext;
    uint32_t                                   aspectReferenceCount;
    const VkInputAttachmentAspectReference*    pAspectReferences;
} VkRenderPassInputAttachmentAspectCreateInfo;

or the equivalent

// Provided by VK_KHR_maintenance2
typedef VkRenderPassInputAttachmentAspectCreateInfo VkRenderPassInputAttachmentAspectCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
aspectReferenceCount is the number of elements in the pAspectReferences array.
pAspectReferences is a pointer to an array of aspectReferenceCount VkInputAttachmentAspectReference structures containing a mask describing which aspect(s) can be accessed for a given input attachment within a given subpass.

An application can access any aspect of an input attachment that does not have a specified aspect mask in the pAspectReferences array. Otherwise, an application must not access aspect(s) of an input attachment other than those in its specified aspect mask.

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]

// Provided by VK_VERSION_1_1
typedef struct VkInputAttachmentAspectReference {
    uint32_t              subpass;
    uint32_t              inputAttachmentIndex;
    VkImageAspectFlags    aspectMask;
} VkInputAttachmentAspectReference;

or the equivalent

// Provided by VK_KHR_maintenance2
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
aspectMask must not include VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT for any index i

Valid Usage (Implicit)

aspectMask must be a valid combination of VkImageAspectFlagBits values
aspectMask must not be 0

The VkSubpassDescription structure is defined as:

// Provided by VK_VERSION_1_0
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 the pipeline type supported for this subpass.
inputAttachmentCount is the number of input attachments.
pInputAttachments is a pointer to an array of VkAttachmentReference structures defining the input attachments for this subpass and their layouts.
colorAttachmentCount is the number of color attachments.
pColorAttachments is a pointer to an array of VkAttachmentReference structures defining the color attachments for this subpass and their layouts.
pResolveAttachments is an optional array of colorAttachmentCount VkAttachmentReference structures defining the resolve attachments for this subpass and their layouts.
pDepthStencilAttachment is a pointer to a VkAttachmentReference structure specifying the depth/stencil attachment for this subpass and its layout.
preserveAttachmentCount is the number of preserved attachments.
pPreserveAttachments is a pointer to an array of preserveAttachmentCount render pass attachment indices identifying attachments that are not used by this subpass, but whose contents must be preserved throughout the subpass.

Each element of the pInputAttachments array corresponds to an input attachment index in a fragment shader, i.e. if a shader declares an image variable decorated with a InputAttachmentIndex value of X, then it uses the attachment provided in pInputAttachments[X]. Input attachments must also be bound to the pipeline in a descriptor set. If the attachment member of any element of pInputAttachments is VK_ATTACHMENT_UNUSED, the application must not read from the corresponding input attachment index. Fragment shaders can use subpass input variables to access the contents of an input attachment at the fragment’s (x, y, layer) framebuffer coordinates. Input attachments must not be used by any subpasses within a renderpass that enables render pass transform.

Each element of the pColorAttachments array corresponds to an output location in the shader, i.e. if the shader declares an output variable decorated with a Location value of X, then it uses the attachment provided in pColorAttachments[X]. If the attachment member of any element of pColorAttachments is VK_ATTACHMENT_UNUSED, writes to the corresponding location by a fragment are discarded.

If flags does not include VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM, and 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 within the render area to the same pixel location in the corresponding resolve attachments, unless the resolve attachment index is VK_ATTACHMENT_UNUSED.

Similarly, if flags does not include VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM, and VkSubpassDescriptionDepthStencilResolve::pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED, it corresponds to the depth/stencil attachment in pDepthStencilAttachment, and multisample resolve operations for depth and stencil are defined by VkSubpassDescriptionDepthStencilResolve::depthResolveMode and VkSubpassDescriptionDepthStencilResolve::stencilResolveMode, respectively. At the end of each subpass, multisample resolve operations read the subpass’s depth/stencil attachment, and resolve the samples for each pixel to the same pixel location in the corresponding resolve attachment. If VkSubpassDescriptionDepthStencilResolve::depthResolveMode is VK_RESOLVE_MODE_NONE, then the depth component of the resolve attachment is not written to and its contents are preserved. Similarly, if VkSubpassDescriptionDepthStencilResolve::stencilResolveMode is VK_RESOLVE_MODE_NONE, then the stencil component of the resolve attachment is not written to and its contents are preserved. VkSubpassDescriptionDepthStencilResolve::depthResolveMode is ignored if the VkFormat of the pDepthStencilResolveAttachment does not have a depth component. Similarly, VkSubpassDescriptionDepthStencilResolve::stencilResolveMode is ignored if the VkFormat of the pDepthStencilResolveAttachment does not have a stencil component.

If the image subresource range referenced by the depth/stencil attachment is created with VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT, then the multisample resolve operation uses the sample locations state specified in the sampleLocationsInfo member of the element of the VkRenderPassSampleLocationsBeginInfoEXT::pPostSubpassSampleLocations for the subpass.

If pDepthStencilAttachment is NULL, or if its attachment index is VK_ATTACHMENT_UNUSED, it indicates that no depth/stencil attachment will be used in 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.

In addition, 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:

VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM is set.
The attachment is used as a color or depth/stencil in the subpass.

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 is not VK_ATTACHMENT_UNUSED, the corresponding color attachment must not be VK_ATTACHMENT_UNUSED
If pResolveAttachments is not NULL, for each resolve attachment that is not VK_ATTACHMENT_UNUSED, the corresponding color attachment must not have a sample count of VK_SAMPLE_COUNT_1_BIT
If pResolveAttachments is not NULL, each resolve attachment that is not VK_ATTACHMENT_UNUSED must have a sample count of VK_SAMPLE_COUNT_1_BIT
If pResolveAttachments is not NULL, each resolve attachment that is not VK_ATTACHMENT_UNUSED 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 pInputAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain at least VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT or VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
All attachments in pColorAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
All attachments in pResolveAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
If pDepthStencilAttachment is not NULL and the attachment is not VK_ATTACHMENT_UNUSED then it must have a image format whose potential format features contain VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
If the VK_AMD_mixed_attachment_samples extension is enabled, and 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 neither the VK_AMD_mixed_attachment_samples nor the VK_NV_framebuffer_mixed_samples extensions are enabled, and if pDepthStencilAttachment is not VK_ATTACHMENT_UNUSED and any attachments in pColorAttachments are not VK_ATTACHMENT_UNUSED, they must have the same sample count
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 by more than one VkAttachmentReference member, 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
If flags includes VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM, and if pResolveAttachments is not NULL, then each resolve attachment must be VK_ATTACHMENT_UNUSED
If flags includes VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM, and if pDepthStencilResolveAttachmentKHR is not NULL, then the depth/stencil resolve attachment must be VK_ATTACHMENT_UNUSED
If flags includes VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM, then the subpass must be the last subpass in a subpass dependency chain
If flags includes VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM, then the sample count of the input attachments must equal rasterizationSamples
If flags includes VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM, and if sampleShadingEnable is enabled (explicitly or implicitly) then minSampleShading must equal 0.0
If the render pass is created with VK_RENDER_PASS_CREATE_TRANSFORM_BIT_QCOM each of the elements of pInputAttachments must be VK_ATTACHMENT_UNUSED

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:

// Provided by VK_VERSION_1_0
typedef enum VkSubpassDescriptionFlagBits {
  // Provided by VK_NVX_multiview_per_view_attributes
    VK_SUBPASS_DESCRIPTION_PER_VIEW_ATTRIBUTES_BIT_NVX = 0x00000001,
  // Provided by VK_NVX_multiview_per_view_attributes
    VK_SUBPASS_DESCRIPTION_PER_VIEW_POSITION_X_ONLY_BIT_NVX = 0x00000002,
  // Provided by VK_QCOM_render_pass_shader_resolve
    VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM = 0x00000004,
  // Provided by VK_QCOM_render_pass_shader_resolve
    VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM = 0x00000008,
} 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 *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.
VK_SUBPASS_DESCRIPTION_FRAGMENT_REGION_BIT_QCOM specifies that the framebuffer region is the fragment region, that is, the minimum region dependencies are by pixel rather than by sample, such that any fragment shader invocation can access any sample associated with that fragment shader invocation.
VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM specifies that the subpass performs shader resolve operations.

Note

Shader resolve operations allow for custom resolve operations, but overdrawing pixels may have a performance and/or power cost. Furthermore, since the content of any depth stencil attachment or color attachment is undefined at the begining of a shader resolve subpass, any depth testing, stencil testing, or blending operation which sources these undefined values also has undefined result value.

// Provided by VK_VERSION_1_0
typedef VkFlags VkSubpassDescriptionFlags;

VkSubpassDescriptionFlags is a bitmask type for setting a mask of zero or more VkSubpassDescriptionFlagBits.

The VkAttachmentReference structure is defined as:

// Provided by VK_VERSION_1_0
typedef struct VkAttachmentReference {
    uint32_t         attachment;
    VkImageLayout    layout;
} VkAttachmentReference;

attachment is either an integer value identifying an attachment at the corresponding index in VkRenderPassCreateInfo::pAttachments, or VK_ATTACHMENT_UNUSED to signify that this attachment is not used.
layout is a VkImageLayout value specifying the layout the attachment uses during the subpass.

Valid Usage

If attachment is not VK_ATTACHMENT_UNUSED, layout must not be VK_IMAGE_LAYOUT_UNDEFINED, VK_IMAGE_LAYOUT_PREINITIALIZED, VK_IMAGE_LAYOUT_PRESENT_SRC_KHR, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL_KHR, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL_KHR, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL_KHR, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL_KHR

Valid Usage (Implicit)

layout must be a valid VkImageLayout value

The VkSubpassDependency structure is defined as:

// Provided by VK_VERSION_1_0
typedef struct VkSubpassDependency {
    uint32_t                srcSubpass;
    uint32_t                dstSubpass;
    VkPipelineStageFlags    srcStageMask;
    VkPipelineStageFlags    dstStageMask;
    VkAccessFlags           srcAccessMask;
    VkAccessFlags           dstAccessMask;
    VkDependencyFlags       dependencyFlags;
} VkSubpassDependency;

srcSubpass is the subpass index of the first subpass in the dependency, or VK_SUBPASS_EXTERNAL.
dstSubpass is the subpass index of the second subpass in the dependency, or VK_SUBPASS_EXTERNAL.
srcStageMask is a bitmask of VkPipelineStageFlagBits specifying the source stage mask.
dstStageMask is a bitmask of VkPipelineStageFlagBits specifying the destination stage mask
srcAccessMask is a bitmask of VkAccessFlagBits specifying a source access mask.
dstAccessMask is a bitmask of VkAccessFlagBits specifying a destination access mask.
dependencyFlags is a bitmask of VkDependencyFlagBits.

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 and dstAccessMask parameters) submitted as a part of a vkCmdPipelineBarrier (with matching srcStageMask and 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 a 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 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 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 srcSubpass equals dstSubpass, and srcStageMask and dstStageMask both include a framebuffer-space stage, then dependencyFlags must include VK_DEPENDENCY_BY_REGION_BIT
If dependencyFlags includes VK_DEPENDENCY_VIEW_LOCAL_BIT, srcSubpass must not be equal to VK_SUBPASS_EXTERNAL
If dependencyFlags includes VK_DEPENDENCY_VIEW_LOCAL_BIT, dstSubpass must not be equal to VK_SUBPASS_EXTERNAL
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
If the mesh shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the mesh shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV

Valid Usage (Implicit)

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
srcAccessMask must be a valid combination of VkAccessFlagBits values
dstAccessMask must be a valid combination of VkAccessFlagBits values
dependencyFlags must be a valid combination of VkDependencyFlagBits values

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.

If any two subpasses of a render pass activate transform feedback to the same bound transform feedback buffers, a subpass dependency must be included (either directly or via some intermediate subpasses) between them.

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 are 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 implicit subpass dependency only exists if there exists an automatic layout transition away from initialLayout. 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 implicit subpass dependency only exists if there exists an automatic layout transition into finalLayout. 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, and both subpasses use the attachment in a read-only layout, 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.

A more extensible version of render pass creation is also defined below.

To create a render pass, call:

// Provided by VK_VERSION_1_2
VkResult vkCreateRenderPass2(
    VkDevice                                    device,
    const VkRenderPassCreateInfo2*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkRenderPass*                               pRenderPass);

or the equivalent command

// Provided by VK_KHR_create_renderpass2
VkResult vkCreateRenderPass2KHR(
    VkDevice                                    device,
    const VkRenderPassCreateInfo2*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkRenderPass*                               pRenderPass);

device is the logical device that creates the render pass.
pCreateInfo is a pointer to a VkRenderPassCreateInfo2 structure describing the parameters of the render pass.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pRenderPass is a pointer to a VkRenderPass handle in which the resulting render pass object is returned.

This command is functionally identical to vkCreateRenderPass, but includes extensible sub-structures that include sType and pNext parameters, allowing them to be more easily extended.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pCreateInfo must be a valid pointer to a valid VkRenderPassCreateInfo2 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkRenderPassCreateInfo2 structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkRenderPassCreateInfo2 {
    VkStructureType                    sType;
    const void*                        pNext;
    VkRenderPassCreateFlags            flags;
    uint32_t                           attachmentCount;
    const VkAttachmentDescription2*    pAttachments;
    uint32_t                           subpassCount;
    const VkSubpassDescription2*       pSubpasses;
    uint32_t                           dependencyCount;
    const VkSubpassDependency2*        pDependencies;
    uint32_t                           correlatedViewMaskCount;
    const uint32_t*                    pCorrelatedViewMasks;
} VkRenderPassCreateInfo2;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkRenderPassCreateInfo2 VkRenderPassCreateInfo2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is reserved for future use.
attachmentCount is the number of attachments used by this render pass.
pAttachments is a pointer to an array of attachmentCount VkAttachmentDescription2 structures describing the attachments used by the render pass.
subpassCount is the number of subpasses to create.
pSubpasses is a pointer to an array of subpassCount VkSubpassDescription2 structures describing each subpass.
dependencyCount is the number of dependencies between pairs of subpasses.
pDependencies is a pointer to an array of dependencyCount VkSubpassDependency structures describing dependencies between pairs of subpasses.
correlatedViewMaskCount is the number of correlation masks.
pCorrelatedViewMasks is a pointer to an array of view masks indicating sets of views that may be more efficient to render concurrently.

Parameters defined by this structure with the same name as those in VkRenderPassCreateInfo have the identical effect to those parameters; the child structures are variants of those used in VkRenderPassCreateInfo which add sType and pNext parameters, allowing them to be extended.

If the VkSubpassDescription2::viewMask member of any element of pSubpasses is not zero, multiview functionality is considered to be enabled for this render pass.

correlatedViewMaskCount and pCorrelatedViewMasks have the same effect as VkRenderPassMultiviewCreateInfo::correlationMaskCount and VkRenderPassMultiviewCreateInfo::pCorrelationMasks, respectively.

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 given 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 VkAttachmentDescription2 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 given element of pSubpasses is not VK_ATTACHMENT_UNUSED, it must be less than attachmentCount
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, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, or 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_SHADER_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL
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 destination subpass
The set of bits included in any element of pCorrelatedViewMasks must not overlap with the set of bits included in any other element of pCorrelatedViewMasks
If the VkSubpassDescription2::viewMask member of all elements of pSubpasses is 0, correlatedViewMaskCount must be 0
The VkSubpassDescription2::viewMask member of all elements of pSubpasses must either all be 0, or all not be 0
If the VkSubpassDescription2::viewMask member of all elements of pSubpasses is 0, the dependencyFlags member of any element of pDependencies must not include VK_DEPENDENCY_VIEW_LOCAL_BIT
For any element of pDependencies where its srcSubpass member equals its dstSubpass member, if the viewMask member of the corresponding element of pSubpasses includes more than one bit, its dependencyFlags member must include VK_DEPENDENCY_VIEW_LOCAL_BIT
The viewMask member must not have a bit set at an index greater than or equal to VkPhysicalDeviceLimits::maxFramebufferLayers
If the attachment member of any element of the pInputAttachments member of any element of pSubpasses is not VK_ATTACHMENT_UNUSED, the aspectMask member of that element of pInputAttachments must only include aspects that are present in images of the format specified by the element of pAttachments specified by attachment
The srcSubpass member of each element of pDependencies must be less than subpassCount
The dstSubpass member of each element of pDependencies must be less than subpassCount

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO_2
pNext must be NULL or a pointer to a valid instance of VkRenderPassFragmentDensityMapCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkRenderPassCreateFlagBits values
If attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkAttachmentDescription2 structures
pSubpasses must be a valid pointer to an array of subpassCount valid VkSubpassDescription2 structures
If dependencyCount is not 0, pDependencies must be a valid pointer to an array of dependencyCount valid VkSubpassDependency2 structures
If correlatedViewMaskCount is not 0, pCorrelatedViewMasks must be a valid pointer to an array of correlatedViewMaskCount uint32_t values
subpassCount must be greater than 0

The VkAttachmentDescription2 structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkAttachmentDescription2 {
    VkStructureType                 sType;
    const void*                     pNext;
    VkAttachmentDescriptionFlags    flags;
    VkFormat                        format;
    VkSampleCountFlagBits           samples;
    VkAttachmentLoadOp              loadOp;
    VkAttachmentStoreOp             storeOp;
    VkAttachmentLoadOp              stencilLoadOp;
    VkAttachmentStoreOp             stencilStoreOp;
    VkImageLayout                   initialLayout;
    VkImageLayout                   finalLayout;
} VkAttachmentDescription2;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkAttachmentDescription2 VkAttachmentDescription2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkAttachmentDescriptionFlagBits specifying additional properties of the attachment.
format is a VkFormat value specifying the format of the image 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.

Parameters defined by this structure with the same name as those in VkAttachmentDescription have the identical effect to those parameters.

If the separateDepthStencilLayouts feature is enabled, and format is a depth/stencil format, initialLayout and finalLayout can be set to a layout that only specifies the layout of the depth aspect.

If format is a depth/stencil format, and initialLayout only specifies the initial layout of the depth aspect of the attachment, the initial layout of the stencil aspect is specified by the stencilInitialLayout member of a VkAttachmentDescriptionStencilLayout structure included in the pNext chain. Otherwise, initialLayout describes the initial layout for all relevant image aspects.

If format is a depth/stencil format, and finalLayout only specifies the final layout of the depth aspect of the attachment, the final layout of the stencil aspect is specified by the stencilFinalLayout member of a VkAttachmentDescriptionStencilLayout structure included in the pNext chain. Otherwise, finalLayout describes the final layout for all relevant image aspects.

Valid Usage

finalLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED
If format is a color format, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
If format is a depth/stencil format, initialLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL
If format is a color format, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
If format is a depth/stencil format, finalLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL
If the separateDepthStencilLayouts feature is not enabled, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If the separateDepthStencilLayouts feature is not enabled, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a color format, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a color format, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes both depth and stencil aspects, and initialLayout is VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, the pNext chain must include a VkAttachmentDescriptionStencilLayout structure
If format is a depth/stencil format which includes both depth and stencil aspects, and finalLayout is VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, the pNext chain must include a VkAttachmentDescriptionStencilLayout structure
If format is a depth/stencil format which includes only the depth aspect, initialLayout must not be VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the depth aspect, finalLayout must not be VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the stencil aspect, initialLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL
If format is a depth/stencil format which includes only the stencil aspect, finalLayout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ATTACHMENT_DESCRIPTION_2
flags must be a valid combination of VkAttachmentDescriptionFlagBits values
format must be a valid VkFormat value
samples must be a valid VkSampleCountFlagBits value
loadOp must be a valid VkAttachmentLoadOp value
storeOp must be a valid VkAttachmentStoreOp value
stencilLoadOp must be a valid VkAttachmentLoadOp value
stencilStoreOp must be a valid VkAttachmentStoreOp value
initialLayout must be a valid VkImageLayout value
finalLayout must be a valid VkImageLayout value

The VkAttachmentDescriptionStencilLayout structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkAttachmentDescriptionStencilLayout {
    VkStructureType    sType;
    void*              pNext;
    VkImageLayout      stencilInitialLayout;
    VkImageLayout      stencilFinalLayout;
} VkAttachmentDescriptionStencilLayout;

or the equivalent

// Provided by VK_KHR_separate_depth_stencil_layouts
typedef VkAttachmentDescriptionStencilLayout VkAttachmentDescriptionStencilLayoutKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
stencilInitialLayout is the layout the stencil aspect of the attachment image subresource will be in when a render pass instance begins.
stencilFinalLayout is the layout the stencil aspect of the attachment image subresource will be transitioned to when a render pass instance ends.

Valid Usage

stencilInitialLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
stencilFinalLayout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
stencilFinalLayout must not be VK_IMAGE_LAYOUT_UNDEFINED or VK_IMAGE_LAYOUT_PREINITIALIZED

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ATTACHMENT_DESCRIPTION_STENCIL_LAYOUT
stencilInitialLayout must be a valid VkImageLayout value
stencilFinalLayout must be a valid VkImageLayout value

The VkSubpassDescription2 structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSubpassDescription2 {
    VkStructureType                  sType;
    const void*                      pNext;
    VkSubpassDescriptionFlags        flags;
    VkPipelineBindPoint              pipelineBindPoint;
    uint32_t                         viewMask;
    uint32_t                         inputAttachmentCount;
    const VkAttachmentReference2*    pInputAttachments;
    uint32_t                         colorAttachmentCount;
    const VkAttachmentReference2*    pColorAttachments;
    const VkAttachmentReference2*    pResolveAttachments;
    const VkAttachmentReference2*    pDepthStencilAttachment;
    uint32_t                         preserveAttachmentCount;
    const uint32_t*                  pPreserveAttachments;
} VkSubpassDescription2;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkSubpassDescription2 VkSubpassDescription2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkSubpassDescriptionFlagBits specifying usage of the subpass.
pipelineBindPoint is a VkPipelineBindPoint value specifying the pipeline type supported for this subpass.
viewMask is a bitfield of view indices describing which views rendering is broadcast to in this subpass, when multiview is enabled.
inputAttachmentCount is the number of input attachments.
pInputAttachments is a pointer to an array of VkAttachmentReference2 structures defining the input attachments for this subpass and their layouts.
colorAttachmentCount is the number of color attachments.
pColorAttachments is a pointer to an array of VkAttachmentReference2 structures defining the color attachments for this subpass and their layouts.
pResolveAttachments is an optional array of colorAttachmentCount VkAttachmentReference2 structures defining the resolve attachments for this subpass and their layouts.
pDepthStencilAttachment is a pointer to a VkAttachmentReference2 structure specifying the depth/stencil attachment for this subpass and its layout.
preserveAttachmentCount is the number of preserved attachments.
pPreserveAttachments is a pointer to an array of preserveAttachmentCount render pass attachment indices identifying attachments that are not used by this subpass, but whose contents must be preserved throughout the subpass.

Parameters defined by this structure with the same name as those in VkSubpassDescription have the identical effect to those parameters.

viewMask has the same effect for the described subpass as VkRenderPassMultiviewCreateInfo::pViewMasks has on each corresponding subpass.

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, for each resolve attachment that is not VK_ATTACHMENT_UNUSED, the corresponding color attachment must not have a sample count of VK_SAMPLE_COUNT_1_BIT
If pResolveAttachments is not NULL, each resolve attachment that is not VK_ATTACHMENT_UNUSED must have a sample count of VK_SAMPLE_COUNT_1_BIT
Any given 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 pInputAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain at least VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT or VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
All attachments in pColorAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
All attachments in pResolveAttachments that are not VK_ATTACHMENT_UNUSED must have image formats whose potential format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
If pDepthStencilAttachment is not NULL and the attachment is not VK_ATTACHMENT_UNUSED then it must have a image format whose potential format features contain VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
If the VK_AMD_mixed_attachment_samples extension is enabled, 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 neither the VK_AMD_mixed_attachment_samples nor the VK_NV_framebuffer_mixed_samples extensions are enabled, and if pDepthStencilAttachment is not VK_ATTACHMENT_UNUSED and any attachments in pColorAttachments are not VK_ATTACHMENT_UNUSED, they must have the same sample count
The attachment member of any element of pPreserveAttachments must not be VK_ATTACHMENT_UNUSED
Any given element of pPreserveAttachments must not also be an element of any other member of the subpass description
If any attachment is used by more than one VkAttachmentReference member, 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
If the attachment member of any element of pInputAttachments is not VK_ATTACHMENT_UNUSED, then the aspectMask member must be a valid combination of VkImageAspectFlagBits
If the attachment member of any element of pInputAttachments is not VK_ATTACHMENT_UNUSED, then the aspectMask member must not be 0
If the attachment member of any element of pInputAttachments is not VK_ATTACHMENT_UNUSED, then the aspectMask member must not include VK_IMAGE_ASPECT_METADATA_BIT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SUBPASS_DESCRIPTION_2
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 VkAttachmentReference2 structures
If colorAttachmentCount is not 0, pColorAttachments must be a valid pointer to an array of colorAttachmentCount valid VkAttachmentReference2 structures
If colorAttachmentCount is not 0, and pResolveAttachments is not NULL, pResolveAttachments must be a valid pointer to an array of colorAttachmentCount valid VkAttachmentReference2 structures
If pDepthStencilAttachment is not NULL, pDepthStencilAttachment must be a valid pointer to a valid VkAttachmentReference2 structure
If preserveAttachmentCount is not 0, pPreserveAttachments must be a valid pointer to an array of preserveAttachmentCount uint32_t values

If the pNext list of VkSubpassDescription2 includes a VkSubpassDescriptionDepthStencilResolve structure, then that structure describes multisample resolve operations for the depth/stencil attachment in a subpass.

The VkSubpassDescriptionDepthStencilResolve structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSubpassDescriptionDepthStencilResolve {
    VkStructureType                  sType;
    const void*                      pNext;
    VkResolveModeFlagBits            depthResolveMode;
    VkResolveModeFlagBits            stencilResolveMode;
    const VkAttachmentReference2*    pDepthStencilResolveAttachment;
} VkSubpassDescriptionDepthStencilResolve;

or the equivalent

// Provided by VK_KHR_depth_stencil_resolve
typedef VkSubpassDescriptionDepthStencilResolve VkSubpassDescriptionDepthStencilResolveKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
depthResolveMode is a bitmask of VkResolveModeFlagBits describing the depth resolve mode.
stencilResolveMode is a bitmask of VkResolveModeFlagBits describing the stencil resolve mode.
pDepthStencilResolveAttachment is an optional VkAttachmentReference structure defining the depth/stencil resolve attachment for this subpass and its layout.

Valid Usage

If pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED, pDepthStencilAttachment must not have the value VK_ATTACHMENT_UNUSED
If pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED, depthResolveMode and stencilResolveMode must not both be VK_RESOLVE_MODE_NONE
If pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED, pDepthStencilAttachment must not have a sample count of VK_SAMPLE_COUNT_1_BIT
If pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED, pDepthStencilResolveAttachment must have a sample count of VK_SAMPLE_COUNT_1_BIT
If pDepthStencilResolveAttachment is not NULL and does not have the value VK_ATTACHMENT_UNUSED then it must have a format whose features contain VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
If the VkFormat of pDepthStencilResolveAttachment has a depth component, then the VkFormat of pDepthStencilAttachment must have a depth component with the same number of bits and numerical type
If the VkFormat of pDepthStencilResolveAttachment has a stencil component, then the VkFormat of pDepthStencilAttachment must have a stencil component with the same number of bits and numerical type
The value of depthResolveMode must be one of the bits set in VkPhysicalDeviceDepthStencilResolveProperties::supportedDepthResolveModes or VK_RESOLVE_MODE_NONE
The value of stencilResolveMode must be one of the bits set in VkPhysicalDeviceDepthStencilResolveProperties::supportedStencilResolveModes or VK_RESOLVE_MODE_NONE
If the VkFormat of pDepthStencilResolveAttachment has both depth and stencil components, VkPhysicalDeviceDepthStencilResolveProperties::independentResolve is VK_FALSE, and VkPhysicalDeviceDepthStencilResolveProperties::independentResolveNone is VK_FALSE, then the values of depthResolveMode and stencilResolveMode must be identical
If the VkFormat of pDepthStencilResolveAttachment has both depth and stencil components, VkPhysicalDeviceDepthStencilResolveProperties::independentResolve is VK_FALSE and VkPhysicalDeviceDepthStencilResolveProperties::independentResolveNone is VK_TRUE, then the values of depthResolveMode and stencilResolveMode must be identical or one of them must be VK_RESOLVE_MODE_NONE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SUBPASS_DESCRIPTION_DEPTH_STENCIL_RESOLVE
depthResolveMode must be a valid VkResolveModeFlagBits value
stencilResolveMode must be a valid VkResolveModeFlagBits value
If pDepthStencilResolveAttachment is not NULL, pDepthStencilResolveAttachment must be a valid pointer to a valid VkAttachmentReference2 structure

Possible values of VkSubpassDescriptionDepthStencilResolve::depthResolveMode and stencilResolveMode, specifying the depth and stencil resolve modes, are:

// Provided by VK_VERSION_1_2
typedef enum VkResolveModeFlagBits {
    VK_RESOLVE_MODE_NONE = 0,
    VK_RESOLVE_MODE_SAMPLE_ZERO_BIT = 0x00000001,
    VK_RESOLVE_MODE_AVERAGE_BIT = 0x00000002,
    VK_RESOLVE_MODE_MIN_BIT = 0x00000004,
    VK_RESOLVE_MODE_MAX_BIT = 0x00000008,
  // Provided by VK_KHR_depth_stencil_resolve
    VK_RESOLVE_MODE_NONE_KHR = VK_RESOLVE_MODE_NONE,
  // Provided by VK_KHR_depth_stencil_resolve
    VK_RESOLVE_MODE_SAMPLE_ZERO_BIT_KHR = VK_RESOLVE_MODE_SAMPLE_ZERO_BIT,
  // Provided by VK_KHR_depth_stencil_resolve
    VK_RESOLVE_MODE_AVERAGE_BIT_KHR = VK_RESOLVE_MODE_AVERAGE_BIT,
  // Provided by VK_KHR_depth_stencil_resolve
    VK_RESOLVE_MODE_MIN_BIT_KHR = VK_RESOLVE_MODE_MIN_BIT,
  // Provided by VK_KHR_depth_stencil_resolve
    VK_RESOLVE_MODE_MAX_BIT_KHR = VK_RESOLVE_MODE_MAX_BIT,
} VkResolveModeFlagBits;

or the equivalent

// Provided by VK_KHR_depth_stencil_resolve
typedef VkResolveModeFlagBits VkResolveModeFlagBitsKHR;

VK_RESOLVE_MODE_NONE indicates that no resolve operation is done.
VK_RESOLVE_MODE_SAMPLE_ZERO_BIT indicates that result of the resolve operation is equal to the value of sample 0.
VK_RESOLVE_MODE_AVERAGE_BIT indicates that result of the resolve operation is the average of the sample values.
VK_RESOLVE_MODE_MIN_BIT indicates that result of the resolve operation is the minimum of the sample values.
VK_RESOLVE_MODE_MAX_BIT indicates that result of the resolve operation is the maximum of the sample values.

// Provided by VK_VERSION_1_2
typedef VkFlags VkResolveModeFlags;

or the equivalent

// Provided by VK_KHR_depth_stencil_resolve
typedef VkResolveModeFlags VkResolveModeFlagsKHR;

VkResolveModeFlags is a bitmask type for setting a mask of zero or more VkResolveModeFlagBits.

The VkAttachmentReference2 structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkAttachmentReference2 {
    VkStructureType       sType;
    const void*           pNext;
    uint32_t              attachment;
    VkImageLayout         layout;
    VkImageAspectFlags    aspectMask;
} VkAttachmentReference2;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkAttachmentReference2 VkAttachmentReference2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
attachment is either an integer value identifying an attachment at the corresponding index in VkRenderPassCreateInfo::pAttachments, or VK_ATTACHMENT_UNUSED to signify that this attachment is not used.
layout is a VkImageLayout value specifying the layout the attachment uses during the subpass.
aspectMask is a mask of which aspect(s) can be accessed within the specified subpass as an input attachment.

Parameters defined by this structure with the same name as those in VkAttachmentReference have the identical effect to those parameters.

aspectMask is ignored when this structure is used to describe anything other than an input attachment reference.

If the separateDepthStencilLayouts feature is enabled, and attachment has a depth/stencil format, layout can be set to a layout that only specifies the layout of the depth aspect.

If layout only specifies the layout of the depth aspect of the attachment, the layout of the stencil aspect is specified by the stencilLayout member of a VkAttachmentReferenceStencilLayout structure included in the pNext chain. Otherwise, layout describes the layout for all relevant image aspects.

Valid Usage

If attachment is not VK_ATTACHMENT_UNUSED, layout must not be VK_IMAGE_LAYOUT_UNDEFINED, VK_IMAGE_LAYOUT_PREINITIALIZED, or VK_IMAGE_LAYOUT_PRESENT_SRC_KHR
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask does not include VK_IMAGE_ASPECT_STENCIL_BIT or VK_IMAGE_ASPECT_DEPTH_BIT, layout must not be VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask does not include VK_IMAGE_ASPECT_COLOR_BIT, layout must not be VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL
If the separateDepthStencilLayouts feature is not enabled, and attachment is not VK_ATTACHMENT_UNUSED, layout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL,
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask includes VK_IMAGE_ASPECT_COLOR_BIT, layout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL,
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask includes both VK_IMAGE_ASPECT_DEPTH_BIT and VK_IMAGE_ASPECT_STENCIL_BIT, and layout is VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, the pNext chain must include a VkAttachmentReferenceStencilLayout structure
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask includes only VK_IMAGE_ASPECT_DEPTH_BIT then layout must not be VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL
If attachment is not VK_ATTACHMENT_UNUSED, and aspectMask includes only VK_IMAGE_ASPECT_STENCIL_BIT then layout must not be VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ATTACHMENT_REFERENCE_2
layout must be a valid VkImageLayout value

The VkAttachmentReferenceStencilLayout structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkAttachmentReferenceStencilLayout {
    VkStructureType    sType;
    void*              pNext;
    VkImageLayout      stencilLayout;
} VkAttachmentReferenceStencilLayout;

or the equivalent

// Provided by VK_KHR_separate_depth_stencil_layouts
typedef VkAttachmentReferenceStencilLayout VkAttachmentReferenceStencilLayoutKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
stencilLayout is a VkImageLayout value specifying the layout the stencil aspect of the attachment uses during the subpass.

Valid Usage

stencilLayout must not be VK_IMAGE_LAYOUT_UNDEFINED, VK_IMAGE_LAYOUT_PREINITIALIZED, VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_PRESENT_SRC_KHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ATTACHMENT_REFERENCE_STENCIL_LAYOUT
stencilLayout must be a valid VkImageLayout value

The VkSubpassDependency2 structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSubpassDependency2 {
    VkStructureType         sType;
    const void*             pNext;
    uint32_t                srcSubpass;
    uint32_t                dstSubpass;
    VkPipelineStageFlags    srcStageMask;
    VkPipelineStageFlags    dstStageMask;
    VkAccessFlags           srcAccessMask;
    VkAccessFlags           dstAccessMask;
    VkDependencyFlags       dependencyFlags;
    int32_t                 viewOffset;
} VkSubpassDependency2;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkSubpassDependency2 VkSubpassDependency2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
srcSubpass is the subpass index of the first subpass in the dependency, or VK_SUBPASS_EXTERNAL.
dstSubpass is the subpass index of the second subpass in the dependency, or VK_SUBPASS_EXTERNAL.
srcStageMask is a bitmask of VkPipelineStageFlagBits specifying the source stage mask.
dstStageMask is a bitmask of VkPipelineStageFlagBits specifying the destination stage mask
srcAccessMask is a bitmask of VkAccessFlagBits specifying a source access mask.
dstAccessMask is a bitmask of VkAccessFlagBits specifying a destination access mask.
dependencyFlags is a bitmask of VkDependencyFlagBits.
viewOffset controls which views in the source subpass the views in the destination subpass depend on.

Parameters defined by this structure with the same name as those in VkSubpassDependency have the identical effect to those parameters.

viewOffset has the same effect for the described subpass dependency as VkRenderPassMultiviewCreateInfo::pViewOffsets has on each corresponding subpass dependency.

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
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 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, srcSubpass must not be equal to VK_SUBPASS_EXTERNAL
If dependencyFlags includes VK_DEPENDENCY_VIEW_LOCAL_BIT, dstSubpass must not be equal to VK_SUBPASS_EXTERNAL
If srcSubpass equals dstSubpass, and srcStageMask and dstStageMask both include a framebuffer-space stage, then dependencyFlags must include VK_DEPENDENCY_BY_REGION_BIT
If viewOffset is not equal to 0, srcSubpass must not be equal to dstSubpass
If dependencyFlags does not include VK_DEPENDENCY_VIEW_LOCAL_BIT, viewOffset must be 0
If the mesh shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, srcStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the mesh shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV
If the task shaders feature is not enabled, dstStageMask must not contain VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SUBPASS_DEPENDENCY_2
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
srcAccessMask must be a valid combination of VkAccessFlagBits values
dstAccessMask must be a valid combination of VkAccessFlagBits values
dependencyFlags must be a valid combination of VkDependencyFlagBits values

To destroy a render pass, call:

// Provided by VK_VERSION_1_0
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

As an additional special case, if two render passes have a single subpass, the resolve attachment reference and depth/stencil resolve mode compatibility requirements are ignored.

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkFramebuffer)

To create a framebuffer, call:

// Provided by VK_VERSION_1_0
VkResult vkCreateFramebuffer(
    VkDevice                                    device,
    const VkFramebufferCreateInfo*              pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkFramebuffer*                              pFramebuffer);

device is the logical device that creates the framebuffer.
pCreateInfo is a pointer to a VkFramebufferCreateInfo structure describing additional information about framebuffer creation.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pFramebuffer is a pointer to a VkFramebuffer handle in which the resulting framebuffer object is returned.

Valid Usage

If pCreateInfo->flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, and attachmentCount is not 0, each element of pCreateInfo->pAttachments must have been created on device

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkFramebufferCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkFramebufferCreateFlagBits
renderPass is a render pass defining what render passes the framebuffer will be compatible with. See Render Pass Compatibility for details.
attachmentCount is the number of attachments.
pAttachments is a pointer to an array of VkImageView handles, each of which will be used as the corresponding attachment in a render pass instance. If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, this parameter is ignored.
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, either because it has no attachment references or because all of them are VK_ATTACHMENT_UNUSED. This kind of subpass can 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 the subpass must have the same value for VkPipelineMultisampleStateCreateInfo::rasterizationSamples.

Valid Usage

attachmentCount must be equal to the attachment count specified in renderPass
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, and attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkImageView handles
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, 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
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_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
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments that is used as a depth/stencil resolve attachment by renderPass must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_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 that is used as a fragment density map attachment by renderPass must not have been created with a flags value including VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT
If renderPass has a fragment density map attachment and non-subsample image feature is not enabled, each element of pAttachments must have been created with a flags value including VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT unless that element is the fragment density map attachment
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments must have been created with a VkFormat value that matches the VkFormat specified by the corresponding VkAttachmentDescription in renderPass
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments must have been created with a samples value that matches the samples value specified by the corresponding VkAttachmentDescription in renderPass
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments must have dimensions at least as large as the corresponding framebuffer dimension except for any element that is referenced by fragmentDensityMapAttachment
If renderPass was specified with non-zero view masks, each element of pAttachments that is not referenced by fragmentDensityMapAttachment must have a layerCount greater than the index of the most significant bit set in any of those view masks
If renderPass was specified with non-zero view masks, each element of pAttachments that is referenced by fragmentDensityMapAttachment must have a layerCount equal to 1 or greater than the index of the most significant bit set in any of those view masks
If renderPass was not specified with non-zero view masks, each element of pAttachments that is referenced by fragmentDensityMapAttachment must have a layerCount equal to 1
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, an element of pAttachments that is referenced by fragmentDensityMapAttachment must have a width at least as large as $⌈ \frac{w i d t h}{m a x F r a g m e n t D e n s i t y T e x e l S i z e _{w i d t h}} ⌉$
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, an element of pAttachments that is referenced by fragmentDensityMapAttachment must have a height at least as large as $⌈ \frac{h e i g h t}{m a x F r a g m e n t D e n s i t y T e x e l S i z e _{h e i g h t}} ⌉$
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments must only specify a single mip level
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, 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
If renderPass was specified with non-zero view masks, layers must be 1
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, 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
If the imageless framebuffer feature is not enabled, flags must not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the pNext chain must include a VkFramebufferAttachmentsCreateInfo structure
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the attachmentImageInfoCount member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be equal to either zero or attachmentCount
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the width member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be greater than or equal to width, except for any element that is referenced by VkRenderPassFragmentDensityMapCreateInfoEXT::fragmentDensityMapAttachment in renderPass
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the height member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be greater than or equal to height, except for any element that is referenced by VkRenderPassFragmentDensityMapCreateInfoEXT::fragmentDensityMapAttachment in renderPass
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the width member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that is referenced by VkRenderPassFragmentDensityMapCreateInfoEXT::fragmentDensityMapAttachment in renderPass must be greater than or equal to $⌈ \frac{w i d t h}{m a x F r a g m e n t D e n s i t y T e x e l S i z e _{w i d t h}} ⌉$
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the height member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that is referenced by VkRenderPassFragmentDensityMapCreateInfoEXT::fragmentDensityMapAttachment in renderPass must be greater than or equal to $⌈ \frac{h e i g h t}{m a x F r a g m e n t D e n s i t y T e x e l S i z e _{h e i g h t}} ⌉$
If multiview is enabled for renderPass, and flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the layerCount member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be greater than the maximum bit index set in the view mask in the subpasses in which it is used in renderPass
If multiview is not enabled for renderPass, and flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the layerCount member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be greater than or equal to layers
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the usage member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that refers to an attachment used as a color attachment or resolve attachment by renderPass must include VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the usage member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that refers to an attachment used as a depth/stencil attachment by renderPass must include VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the usage member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that refers to an attachment used as a depth/stencil resolve attachment by renderPass must include VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the usage member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain that refers to an attachment used as an input attachment by renderPass must include VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT
If flags includes VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, at least one element of the pViewFormats member of any element of the pAttachmentImageInfos member of a VkFramebufferAttachmentsCreateInfo structure included in the pNext chain must be equal to the corresponding value of VkAttachmentDescription::format used to create renderPass
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of pAttachments must have been created with VkImageViewCreateInfo::viewType not equal to VK_IMAGE_VIEW_TYPE_3D

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO
pNext must be NULL or a pointer to a valid instance of VkFramebufferAttachmentsCreateInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkFramebufferCreateFlagBits values
renderPass must be a valid VkRenderPass handle
Both of renderPass, and the elements of pAttachments that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

The VkFramebufferAttachmentsCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkFramebufferAttachmentsCreateInfo {
    VkStructureType                            sType;
    const void*                                pNext;
    uint32_t                                   attachmentImageInfoCount;
    const VkFramebufferAttachmentImageInfo*    pAttachmentImageInfos;
} VkFramebufferAttachmentsCreateInfo;

or the equivalent

// Provided by VK_KHR_imageless_framebuffer
typedef VkFramebufferAttachmentsCreateInfo VkFramebufferAttachmentsCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
attachmentImageInfoCount is the number of attachments being described.
pAttachmentImageInfos is a pointer to an array of VkFramebufferAttachmentImageInfo instances, each of which describes a number of parameters of the corresponding attachment in a render pass instance.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_FRAMEBUFFER_ATTACHMENTS_CREATE_INFO
If attachmentImageInfoCount is not 0, pAttachmentImageInfos must be a valid pointer to an array of attachmentImageInfoCount valid VkFramebufferAttachmentImageInfo structures

The VkFramebufferAttachmentImageInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkFramebufferAttachmentImageInfo {
    VkStructureType       sType;
    const void*           pNext;
    VkImageCreateFlags    flags;
    VkImageUsageFlags     usage;
    uint32_t              width;
    uint32_t              height;
    uint32_t              layerCount;
    uint32_t              viewFormatCount;
    const VkFormat*       pViewFormats;
} VkFramebufferAttachmentImageInfo;

or the equivalent

// Provided by VK_KHR_imageless_framebuffer
typedef VkFramebufferAttachmentImageInfo VkFramebufferAttachmentImageInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is a bitmask of VkImageCreateFlagBits, matching the value of VkImageCreateInfo::flags used to create an image that will be used with this framebuffer.
usage is a bitmask of VkImageUsageFlagBits, matching the value of VkImageCreateInfo::usage used to create an image used with this framebuffer.
width is the width of the image view used for rendering.
height is the height of the image view used for rendering.
viewFormatCount is the number of entries in the pViewFormats array, matching the value of VkImageFormatListCreateInfo::viewFormatCount used to create an image used with this framebuffer.
pViewFormats is an array which lists of all formats which can be used when creating views of the image, matching the value of VkImageFormatListCreateInfo::pViewFormats used to create an image used with this framebuffer.

Images that can be used with the framebuffer when beginning a render pass, as specified by VkRenderPassAttachmentBeginInfo, must be created with parameters that are identical to those specified here.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_FRAMEBUFFER_ATTACHMENT_IMAGE_INFO
pNext must be NULL
flags must be a valid combination of VkImageCreateFlagBits values
usage must be a valid combination of VkImageUsageFlagBits values
usage must not be 0
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 VkFramebufferCreateInfo::flags to specify options for framebuffers are:

// Provided by VK_VERSION_1_0
typedef enum VkFramebufferCreateFlagBits {
  // Provided by VK_VERSION_1_2
    VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT = 0x00000001,
  // Provided by VK_KHR_imageless_framebuffer
    VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT_KHR = VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT,
} VkFramebufferCreateFlagBits;

VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT specifies that image views are not specified, and only attachment compatibility information will be provided via a VkFramebufferAttachmentImageInfo structure.

// Provided by VK_VERSION_1_0
typedef VkFlags VkFramebufferCreateFlags;

VkFramebufferCreateFlags is a bitmask type for setting a mask of zero or more VkFramebufferCreateFlagBits.

To destroy a framebuffer, call:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 specifying 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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
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_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL then the corresponding attachment image view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If any of the stencilInitialLayout or stencilFinalLayout member of the VkAttachmentDescriptionStencilLayout structures or the stencilLayout member of the VkAttachmentReferenceStencilLayout structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL then the corresponding attachment image view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_SAMPLED_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_TRANSFER_SRC_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_TRANSFER_DST_BIT
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
For any attachment in framebuffer that is used by renderPass and is bound to memory locations that are also bound to another attachment used by renderPass, and if at least one of those uses causes either attachment to be written to, both attachments must have had the VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT set
Each element of the pAttachments of framebuffer that is referenced by any element of the pInputAttachments of any element of pSubpasses of renderPass must have image view format features containing at least VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT or VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pColorAttachments of any element of pSubpasses of renderPass must have image view format features containing VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pResolveAttachments of any element of pSubpasses of renderPass must have image view format features containing VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pDepthStencilAttachment of any element of pSubpasses of renderPass must have image view format features containing VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

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

Primary

Outside

Graphics

Alternatively to begin a render pass, call:

// Provided by VK_VERSION_1_2
void vkCmdBeginRenderPass2(
    VkCommandBuffer                             commandBuffer,
    const VkRenderPassBeginInfo*                pRenderPassBegin,
    const VkSubpassBeginInfo*                   pSubpassBeginInfo);

or the equivalent command

// Provided by VK_KHR_create_renderpass2
void vkCmdBeginRenderPass2KHR(
    VkCommandBuffer                             commandBuffer,
    const VkRenderPassBeginInfo*                pRenderPassBegin,
    const VkSubpassBeginInfo*                   pSubpassBeginInfo);

commandBuffer is the command buffer in which to record the command.
pRenderPassBegin is a pointer to a VkRenderPassBeginInfo structure specifying the render pass to begin an instance of, and the framebuffer the instance uses.
pSubpassBeginInfo is a pointer to a VkSubpassBeginInfo structure containing information about the subpass which is about to begin rendering.

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

Both the framebuffer and renderPass members of pRenderPassBegin must have been created on the same VkDevice that commandBuffer was allocated on
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
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_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL then the corresponding attachment image view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If any of the stencilInitialLayout or stencilFinalLayout member of the VkAttachmentDescriptionStencilLayout structures or the stencilLayout member of the VkAttachmentReferenceStencilLayout structures specified when creating the render pass specified in the renderPass member of pRenderPassBegin is VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL, or VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL then the corresponding attachment image view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_SAMPLED_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_TRANSFER_SRC_BIT
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 view of the framebuffer specified in the framebuffer member of pRenderPassBegin must have been created with a usage value including VK_IMAGE_USAGE_TRANSFER_DST_BIT
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
For any attachment in framebuffer that is used by renderPass and is bound to memory locations that are also bound to another attachment used by renderPass, and if at least one of those uses causes either attachment to be written to, both attachments must have had the VK_ATTACHMENT_DESCRIPTION_MAY_ALIAS_BIT set
Each element of the pAttachments of framebuffer that is referenced by any element of the pInputAttachments of any element of pSubpasses of renderPass must have image view format features contain at least VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT or VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pColorAttachments of any element of pSubpasses of renderPass must have image view format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pResolveAttachments of any element of pSubpasses of renderPass must have image view format features contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
Each element of the pAttachments of framebuffer that is referenced by any element of the pDepthStencilAttachment of any element of pSubpasses of renderPass must have image view format features contain VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pRenderPassBegin must be a valid pointer to a valid VkRenderPassBeginInfo structure
pSubpassBeginInfo must be a valid pointer to a valid VkSubpassBeginInfo structure
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

Primary

Outside

Graphics

The VkRenderPassBeginInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 a pointer to an array of clearValueCount 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. The render area, after any transform specified by VkRenderPassTransformBeginInfoQCOM::transform is applied, must be contained within the framebuffer dimensions.

If render pass transform is enabled, then renderArea must equal the framebuffer pre-transformed dimensions. After renderArea has been transformed by VkRenderPassTransformBeginInfoQCOM::transform, the resulting render area must be equal to 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
renderPass must be compatible with the renderPass member of the VkFramebufferCreateInfo structure specified when creating framebuffer
If the pNext chain does not contain VkDeviceGroupRenderPassBeginInfo or its deviceRenderAreaCount member is equal to 0, renderArea.offset.x must be greater than or equal to 0
If the pNext chain does not contain VkDeviceGroupRenderPassBeginInfo or its deviceRenderAreaCount member is equal to 0, renderArea.offset.y must be greater than or equal to 0
If the pNext chain does not contain VkDeviceGroupRenderPassBeginInfo or its deviceRenderAreaCount member is equal to 0, renderArea.offset.x + renderArea.offset.width must be less than or equal to VkFramebufferCreateInfo::width the framebuffer was created with
If the pNext chain does not contain VkDeviceGroupRenderPassBeginInfo or its deviceRenderAreaCount member is equal to 0, renderArea.offset.y + renderArea.offset.height must be less than or equal to VkFramebufferCreateInfo::height the framebuffer was created with
If the pNext chain contains VkDeviceGroupRenderPassBeginInfo, the offset.x member of each element of pDeviceRenderAreas must be greater than or equal to 0
If the pNext chain contains VkDeviceGroupRenderPassBeginInfo, the offset.y member of each element of pDeviceRenderAreas must be greater than or equal to 0
If the pNext chain contains VkDeviceGroupRenderPassBeginInfo, offset.x + offset.width of each element of pDeviceRenderAreas must be less than or equal to VkFramebufferCreateInfo::width the framebuffer was created with
If the pNext chain contains VkDeviceGroupRenderPassBeginInfo, offset.y + offset.height of each element of pDeviceRenderAreas must be less than or equal to VkFramebufferCreateInfo::height the framebuffer was created with
If framebuffer was created with a VkFramebufferCreateInfo::flags value that did not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, and the pNext chain includes a VkRenderPassAttachmentBeginInfo structure, its attachmentCount must be zero
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, the attachmentCount of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be equal to the value of VkFramebufferAttachmentsCreateInfo::attachmentImageInfoCount used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT_KHR, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must have been created on the same VkDevice as framebuffer and renderPass
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageCreateInfo::flags equal to the flags member of the corresponding element of VkFramebufferAttachmentsCreateInfoKHR::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageCreateInfo::usage equal to the usage member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView with a width equal to the width member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView with a height equal to the height member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageViewCreateInfo::subresourceRange.layerCount equal to the layerCount member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageFormatListCreateInfo::viewFormatCount equal to the viewFormatCount member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a set of elements in VkImageFormatListCreateInfo::pViewFormats equal to the set of elements in the pViewFormats member of the corresponding element of VkFramebufferAttachmentsCreateInfo::pAttachments used to create framebuffer
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageViewCreateInfo::format equal to the corresponding value of VkAttachmentDescription::format in renderPass
If framebuffer was created with a VkFramebufferCreateInfo::flags value that included VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each element of the pAttachments member of a VkRenderPassAttachmentBeginInfo structure included in the pNext chain must be a VkImageView of an image created with a value of VkImageCreateInfo::samples equal to the corresponding value of VkAttachmentDescription::samples in renderPass
If the pNext chain includes VkRenderPassTransformBeginInfoQCOM, renderArea.offset must equal (0,0)
If the pNext chain includes VkRenderPassTransformBeginInfoQCOM, renderArea.extent transformed by VkRenderPassTransformBeginInfoQCOM::transform must equal the framebuffer dimensions

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, VkRenderPassAttachmentBeginInfo, VkRenderPassSampleLocationsBeginInfoEXT, or VkRenderPassTransformBeginInfoQCOM
The sType value of each struct in the pNext chain must be unique
renderPass must be a valid VkRenderPass handle
framebuffer must be a valid VkFramebuffer handle
If clearValueCount is not 0, pClearValues must be a valid pointer to an array of clearValueCount VkClearValue unions
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 adding a VkRenderPassSampleLocationsBeginInfoEXT structure to the pNext chain of VkRenderPassBeginInfo.

The VkRenderPassSampleLocationsBeginInfoEXT structure is defined as:

// Provided by VK_EXT_sample_locations
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 a structure extending this structure.
attachmentInitialSampleLocationsCount is the number of elements in the pAttachmentInitialSampleLocations array.
pAttachmentInitialSampleLocations is a pointer to 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 a pointer to 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:

// Provided by VK_EXT_sample_locations
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

attachmentIndex must be less than the attachmentCount specified in VkRenderPassCreateInfo the render pass specified by VkRenderPassBeginInfo::renderPass was created with

Valid Usage (Implicit)

sampleLocationsInfo must be a valid VkSampleLocationsInfoEXT structure

The VkSubpassSampleLocationsEXT structure is defined as:

// Provided by VK_EXT_sample_locations
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

subpassIndex must be less than the subpassCount specified in VkRenderPassCreateInfo the render pass specified by VkRenderPassBeginInfo::renderPass was created with

Valid Usage (Implicit)

sampleLocationsInfo must be a valid VkSampleLocationsInfoEXT structure

To begin a renderpass instance with render pass transform enabled, add the VkRenderPassTransformBeginInfoQCOM to the pNext chain of VkRenderPassBeginInfo structure passed to the vkCmdBeginRenderPass command specifying the renderpass transform.

The VkRenderPassTransformBeginInfoQCOM structure is defined as:

// Provided by VK_QCOM_render_pass_transform
typedef struct VkRenderPassTransformBeginInfoQCOM {
    VkStructureType                  sType;
    void*                            pNext;
    VkSurfaceTransformFlagBitsKHR    transform;
} VkRenderPassTransformBeginInfoQCOM;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
transform is a VkSurfaceTransformFlagBitsKHR value describing the transform to be applied to rasterization.

Valid Usage

transform must be VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR, VK_SURFACE_TRANSFORM_ROTATE_90_BIT_KHR, VK_SURFACE_TRANSFORM_ROTATE_180_BIT_KHR, or VK_SURFACE_TRANSFORM_ROTATE_270_BIT_KHR
The renderpass must have been created with VkRenderPassCreateInfo::flags containing VK_RENDER_PASS_CREATE_TRANSFORM_BIT_QCOM

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RENDER_PASS_TRANSFORM_BEGIN_INFO_QCOM

The VkSubpassBeginInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSubpassBeginInfo {
    VkStructureType      sType;
    const void*          pNext;
    VkSubpassContents    contents;
} VkSubpassBeginInfo;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkSubpassBeginInfo VkSubpassBeginInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
contents is a VkSubpassContents value specifying how the commands in the next subpass will be provided.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SUBPASS_BEGIN_INFO
pNext must be NULL
contents must be a valid VkSubpassContents value

Possible values of vkCmdBeginRenderPass::contents, specifying how the commands in the first subpass will be provided, are:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_1
typedef struct VkDeviceGroupRenderPassBeginInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceMask;
    uint32_t           deviceRenderAreaCount;
    const VkRect2D*    pDeviceRenderAreas;
} VkDeviceGroupRenderPassBeginInfo;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkDeviceGroupRenderPassBeginInfo VkDeviceGroupRenderPassBeginInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
deviceMask is the device mask for the render pass instance.
deviceRenderAreaCount is the number of elements in the pDeviceRenderAreas array.
pDeviceRenderAreas is a pointer to an array of VkRect2D structures 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. In addition, commands transitioning to the next subpass in the render pass instance and commands ending the render pass instance, and, accordingly render pass attachment load, store, and resolve operations and subpass dependencies corresponding to the render pass instance, are executed on the physical devices included in the device mask provided here.

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

The VkRenderPassAttachmentBeginInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkRenderPassAttachmentBeginInfo {
    VkStructureType       sType;
    const void*           pNext;
    uint32_t              attachmentCount;
    const VkImageView*    pAttachments;
} VkRenderPassAttachmentBeginInfo;

or the equivalent

// Provided by VK_KHR_imageless_framebuffer
typedef VkRenderPassAttachmentBeginInfo VkRenderPassAttachmentBeginInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
attachmentCount is the number of attachments.
pAttachments is a pointer to an array of VkImageView handles, each of which will be used as the corresponding attachment in the render pass instance.

Valid Usage

Each element of pAttachments must only specify a single mip level
Each element of pAttachments must have been created with the identity swizzle
Each element of pAttachments must have been created with VkImageViewCreateInfo::viewType not equal to VK_IMAGE_VIEW_TYPE_3D

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RENDER_PASS_ATTACHMENT_BEGIN_INFO
If attachmentCount is not 0, pAttachments must be a valid pointer to an array of attachmentCount valid VkImageView handles

To query the render area granularity, call:

// Provided by VK_VERSION_1_0
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 is a pointer 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:

// Provided by VK_VERSION_1_0
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. This applies to resolve operations for both color and depth/stencil attachments. 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
This command must not be recorded when transform feedback is active

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

Primary

Inside

Graphics

To transition to the next subpass in the render pass instance after recording the commands for a subpass, call:

// Provided by VK_VERSION_1_2
void vkCmdNextSubpass2(
    VkCommandBuffer                             commandBuffer,
    const VkSubpassBeginInfo*                   pSubpassBeginInfo,
    const VkSubpassEndInfo*                     pSubpassEndInfo);

or the equivalent command

// Provided by VK_KHR_create_renderpass2
void vkCmdNextSubpass2KHR(
    VkCommandBuffer                             commandBuffer,
    const VkSubpassBeginInfo*                   pSubpassBeginInfo,
    const VkSubpassEndInfo*                     pSubpassEndInfo);

commandBuffer is the command buffer in which to record the command.
pSubpassBeginInfo is a pointer to a VkSubpassBeginInfo structure containing information about the subpass which is about to begin rendering.
pSubpassEndInfo is a pointer to a VkSubpassEndInfo structure containing information about how the previous subpass will be ended.

vkCmdNextSubpass2 is semantically identical to vkCmdNextSubpass, except that it is extensible, and that contents is provided as part of an extensible structure instead of as a flat parameter.

Valid Usage

The current subpass index must be less than the number of subpasses in the render pass minus one
This command must not be recorded when transform feedback is active

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pSubpassBeginInfo must be a valid pointer to a valid VkSubpassBeginInfo structure
pSubpassEndInfo must be a valid pointer to a valid VkSubpassEndInfo structure
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

Primary

Inside

Graphics

To record a command to end a render pass instance after recording the commands for the last subpass, call:

// Provided by VK_VERSION_1_0
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
This command must not be recorded when transform feedback is active

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

Primary

Inside

Graphics

To record a command to end a render pass instance after recording the commands for the last subpass, call:

// Provided by VK_VERSION_1_2
void vkCmdEndRenderPass2(
    VkCommandBuffer                             commandBuffer,
    const VkSubpassEndInfo*                     pSubpassEndInfo);

or the equivalent command

// Provided by VK_KHR_create_renderpass2
void vkCmdEndRenderPass2KHR(
    VkCommandBuffer                             commandBuffer,
    const VkSubpassEndInfo*                     pSubpassEndInfo);

commandBuffer is the command buffer in which to end the current render pass instance.
pSubpassEndInfo is a pointer to a VkSubpassEndInfo structure containing information about how the previous subpass will be ended.

vkCmdEndRenderPass2 is semantically identical to vkCmdEndRenderPass, except that it is extensible.

Valid Usage

The current subpass index must be equal to the number of subpasses in the render pass minus one
This command must not be recorded when transform feedback is active

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pSubpassEndInfo must be a valid pointer to a valid VkSubpassEndInfo structure
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

Primary

Inside

Graphics

The VkSubpassEndInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSubpassEndInfo {
    VkStructureType    sType;
    const void*        pNext;
} VkSubpassEndInfo;

or the equivalent

// Provided by VK_KHR_create_renderpass2
typedef VkSubpassEndInfo VkSubpassEndInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SUBPASS_END_INFO
pNext must be NULL

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkShaderModule)

To create a shader module, call:

// Provided by VK_VERSION_1_0
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 a VkShaderModuleCreateInfo structure.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pShaderModule is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_SHADER_NV

The VkShaderModuleCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is reserved for future use.
codeSize is the size, in bytes, of the code pointed to by pCode.
pCode is a pointer 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 is a pointer 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 is a pointer 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 is a pointer 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 in the SPIR-V Environment appendix, one of the corresponding requirements must be satisfied

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
The sType value of each struct in the pNext chain must be unique
flags must be 0
pCode must be a valid pointer to an array of $\frac{codeSize}{4}$ uint32_t values

// Provided by VK_VERSION_1_0
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 structure to the pNext chain of the VkShaderModuleCreateInfo structure, specifying the cache object to use.

The VkShaderModuleValidationCacheCreateInfoEXT struct is defined as:

// Provided by VK_EXT_validation_cache
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 a structure extending this 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:

// Provided by VK_VERSION_1_0
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 execution 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 execution order of invocations of the same shader type is 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 execution 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.

The Memory Model appendix defines the terminology and rules for how to correctly communicate between shader invocations, such as when a write is Visible-To a read, and what constitutes a Data Race.

Applications must not cause a data race.

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. Task Shaders

Task shaders operate in conjunction with the mesh shaders to produce a collection of primitives that will be processed by subsequent stages of the graphics pipeline. Its primary purpose is to create a variable amount of subsequent mesh shader invocations.

Task shaders are invoked via the execution of the programmable mesh shading pipeline.

The task shader has no fixed-function inputs other than variables identifying the specific workgroup and invocation. The only fixed output of the task shader is a task count, identifying the number of mesh shader workgroups to create. The task shader can write additional outputs to task memory, which can be read by all of the mesh shader workgroups it created.

8.5.1. Task Shader Execution

Task workloads are formed from groups of work items called workgroups and processed by the task shader in the current graphics pipeline. A workgroup is a collection of shader invocations that execute the same shader, potentially in parallel. Task 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.

8.6. Mesh Shaders

Mesh shaders operate in workgroups to produce a collection of primitives that will be processed by subsequent stages of the graphics pipeline. Each workgroup emits zero or more output primitives and the group of vertices and their associated data required for each output primitive.

Mesh shaders are invoked via the execution of the programmable mesh shading pipeline.

The only inputs available to the mesh shader are variables identifying the specific workgroup and invocation and, if applicable, any outputs written to task memory by the task shader that spawned the mesh shader’s workgroup. The mesh shader can operate without a task shader as well.

The invocations of the mesh shader workgroup write an output mesh, comprising a set of primitives with per-primitive attributes, a set of vertices with per-vertex attributes, and an array of indices identifying the mesh vertices that belong to each primitive. The primitives of this mesh are then processed by subsequent graphics pipeline stages, where the outputs of the mesh shader form an interface with the fragment shader.

8.6.1. Mesh Shader Execution

Mesh workloads are formed from groups of work items called workgroups and processed by the mesh shader in the current graphics pipeline. A workgroup is a collection of shader invocations that execute the same shader, potentially in parallel. Mesh 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.

The global workgroups may be generated explcitly via the API, or implicitly through the task shader’s work creation mechanism.

8.7. 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 using primitive shading must include a vertex shader, and the vertex shader stage is always the first shader stage in the graphics pipeline.

8.7.1. 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).

8.8. 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.

8.8.1. 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.

8.9. 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.

8.9.1. 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.

8.10. 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.

8.10.1. 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. A shader can request that the geometry shader runs multiple instances. A geometry shader is invoked at least once for each instance. If the subpass includes multiple views in its view mask, the shader may be invoked separately for each view.

8.11. 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.

8.11.1. Fragment Shader Execution

Fragment shaders are invoked for each fragment generated by rasterization, or as helper invocations.

For fragment shaders invoked by fragments, the following rules apply:

A fragment shader must not be executed if a fragment operation that executes before fragment shading discards the fragment.
A fragment shader may not be executed if:
- An implementation determines that another fragment shader, invoked by a subsequent primitive in primitive order, overwrites all results computed by the shader (including writes to storage resources).
- Any other fragment operation discards the fragment, and the shader does not write to any storage resources.
Otherwise, at least one fragment shader must be executed.
- If sample shading is enabled and multiple invocations per fragment are required, additional invocations must be executed as specified.
- If a shading rate image is used and multiple invocations per fragment are required, additional invocations must be executed as specified.
- Each covered sample must be included in at least one fragment shader invocation.

Note

Multiple fragment shader invocations may be executed for the same fragment for any number of implementation dependent reasons. When there is more than one fragment shader invocation per fragment, the association of samples to invocations is implementation-dependent. Stores and atomics performed by these additional invocations have the normal effect.

For example, if the subpass includes multiple views in its view mask, a fragment shader may be invoked separately for each view.

Similarly, if the render pass has a fragment density map attachment, more than one fragment shader invocation may be invoked for each covered sample. Such additional invocations are only produced if VkPhysicalDeviceFragmentDensityMapPropertiesEXT::fragmentDensityInvocations is VK_TRUE. Implementations may generate these additional fragment shader invocations in order to make transitions between fragment areas with different fragment densities more smooth.

Note

Relative ordering of execution of different fragment shader invocations is explicitly not defined.

8.11.2. 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, additional per-fragment tests are performed prior to 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.

If early fragment tests are enabled, any depth value computed by the fragment shader has no effect.

8.11.3. Fragment Shader Interlock

In normal operation, it is possible for more than one fragment shader invocation to be executed simultaneously for the same pixel if there are overlapping primitives. If the fragmentShaderSampleInterlock, fragmentShaderPixelInterlock, or fragmentShaderShadingRateInterlock features are enabled, it is possible to define a critical section within the fragment shader that is guaranteed to not run simultaneously with another fragment shader invocation for the same sample(s) or pixel(s). It is also possible to control the relative ordering of execution of these critical sections across different fragment shader invovations.

If the FragmentShaderSampleInterlockEXT, FragmentShaderPixelInterlockEXT, or FragmentShaderShadingRateInterlockEXT capabilities are declared in the fragment shader, the OpBeginInvocationInterlockEXT and OpEndInvocationInterlockEXT instructions must be used to delimit a critical section of fragment shader code.

To ensure each invocation of the critical section is executed in primitive order, declare one of the PixelInterlockOrderedEXT, SampleInterlockOrderedEXT, or ShadingRateInterlockOrderedEXT execution modes. If the order of execution of each invocation of the critical section does not matter, declare one of the PixelInterlockUnorderedEXT, SampleInterlockUnorderedEXT, or ShadingRateInterlockUnorderedEXT execution modes.

The PixelInterlockOrderedEXT and PixelInterlockUnorderedEXT execution modes provide mutual exclusion in the critical section for any pair of fragments corresponding to the same pixel, or pixels if the fragment covers more than one pixel. With sample shading enabled, these execution modes are treated like SampleInterlockOrderedEXT or SampleInterlockUnorderedEXT respectively.

The SampleInterlockOrderedEXT and SampleInterlockUnorderedEXT execution modes only provide mutual exclusion for pairs of fragments that both cover at least one common sample in the same pixel; these are recommended for performance if shaders use per-sample data structures. If these execution modes are used in single-sample mode they are treated like PixelInterlockOrderedEXT or PixelInterlockUnorderedEXT respectively.

The ShadingRateInterlockOrderedEXT and ShadingRateInterlockUnorderedEXT execution modes provide mutual exclusion for pairs of fragments that both have at least one common sample in the same pixel, even if none of the common samples are covered by both fragments. With sample shading enabled, these execution modes are treated like SampleInterlockOrderedEXT or SampleInterlockUnorderedEXT respectively.

8.12. 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.

8.13. 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 fragment and a single value may be assigned to each sample within the fragment.

If a fragment shader input is decorated with Centroid, a single value may be assigned to that variable for all samples in the fragment, but that value must be interpolated to a location that lies in both the fragment and in the primitive being rendered, including any of the fragment’s samples covered by the primitive. Because the location at which the variable is interpolated may be different in neighboring fragments, and derivatives may be computed by computing differences between neighboring fragments, 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 fragment, and that value must be sampled at the location of the individual sample. When rasterizationSamples is VK_SAMPLE_COUNT_1_BIT, the fragment 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.

When the fragmentShaderBarycentric feature is enabled, inputs can be also decorated with the PerVertexNV interpolation decoration, including fragment shader inputs that are signed or unsigned integers, integer vectors, or any double-precision floating-point type. Inputs decorated with PerVertexNV can only be accessed using an extra array dimension, where the extra index identifies one of the vertices of the primitive that produced the fragment.

8.14. Ray Generation Shaders

A ray generation shader is similar to a compute shader. Its main purpose is to execute ray tracing queries using OpTraceRayKHR instructions and process the results.

8.14.1. Ray Generation Shader Execution

One ray generation shader is executed per ray tracing dispatch. Its location in the shader binding table (see Shader Binding Table for details) is passed directly into vkCmdTraceRaysKHR using the raygenShaderBindingTableBuffer and raygenShaderBindingOffset parameters.

8.15. Intersection Shaders

Intersection shaders enable the implementation of arbitrary, application defined geometric primitives. An intersection shader for a primitive is executed whenever its axis-aligned bounding box is hit by a ray.

Like other ray tracing shader domains, an intersection shader operates on a single ray at a time. It also operates on a single primitive at a time. It is therefore the purpose of an intersection shader to compute the ray-primitive intersections and report them. To report an intersection, the shader calls the OpReportIntersectionKHR instruction.

An intersection shader communicates with any-hit and closest shaders by generating attribute values that they can read. Intersection shaders cannot read or modify the ray payload.

8.15.1. Intersection Shader Execution

The order in which intersections are found along a ray, and therefore the order in which intersection shaders are executed, is unspecified.

The intersection shader of the closest AABB which intersects the ray is guaranteed to be executed at some point during traversal, unless the ray is forcibly terminated.

8.16. Any-Hit Shaders

The any-hit shader is executed after the intersection shader reports an intersection that lies within the current [t_min,t_max] of the ray. The main use of any-hit shaders is to programmatically decide whether or not an intersection will be accepted. The intersection will be accepted unless the shader calls the OpIgnoreIntersectionKHR instruction. Any-hit shaders have read-only access to the attributes generated by the corresponding intersection shader, and can read or modify the ray payload.

8.16.1. Any-Hit Shader Execution

The order in which intersections are found along a ray, and therefore the order in which any-hit shaders are executed, is unspecified.

The any-hit shader of the closest hit is guaranteed to be executed at some point during traversal, unless the ray is forcibly terminated.

8.17. Closest Hit Shaders

Closest hit shaders have read-only access to the attributes generated by the corresponding intersection shader, and can read or modify the ray payload. They also have access to a number of system-generated values. Closest hit shaders can call OpTraceRayKHR to recursively trace rays.

8.17.1. Closest Hit Shader Execution

Exactly one closest hit shader is executed when traversal is finished and an intersection has been found and accepted.

8.18. Miss Shaders

Miss shaders can access the ray payload and can trace new rays through the OpTraceRayKHR instruction, but cannot access attributes since they are not associated with an intersection.

8.18.1. Miss Shader Execution

A miss shader is executed instead of a closest hit shader if no intersection was found during traversal.

8.19. Callable Shaders

Callable shaders can access a callable payload that works similarly to ray payloads to do subroutine work.

8.19.1. Callable Shader Execution

A callable shader is executed by calling OpExecuteCallableKHR from an allowed shader stage.

8.20. 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 containing 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.

8.21. Scope

A scope describes a set of shader invocations, where each such set is a scope instance. Each invocation belongs to one or more scope instances, but belongs to no more than one scope instance for each scope.

The operations available between invocations in a given scope instance vary, with smaller scopes generally able to perform more operations, and with greater efficiency.

8.21.1. Cross Device

All invocations executed in a Vulkan instance fall into a single cross device scope instance.

Whilst the CrossDevice scope is defined in SPIR-V, it is disallowed in Vulkan. API synchronization commands can be used to communicate between devices.

8.21.2. Device

All invocations executed on a single device form a device scope instance.

If the vulkanMemoryModel and vulkanMemoryModelDeviceScope features are enabled, this scope is represented in SPIR-V by the Device Scope, which can be used as a Memory Scope for barrier and atomic operations.

If both the shaderDeviceClock and vulkanMemoryModelDeviceScope features are enabled, using the Device Scope with the OpReadClockKHR instruction will read from a clock that is consistent across invocations in the same device scope instance.

There is no method to synchronize the execution of these invocations within SPIR-V, and this can only be done with API synchronization primitives.

Invocations executing on different devices in a device group operate in separate device scope instances.

8.21.3. Queue Family

Invocations executed by queues in a given queue family form a queue family scope instance.

This scope is identified in SPIR-V as the QueueFamily Scope if the vulkanMemoryModel feature is enabled, or if not, the Device Scope, which can be used as a Memory Scope for barrier and atomic operations.

If the shaderDeviceClock feature is enabled, but the vulkanMemoryModelDeviceScope feature is not enabled, using the Device Scope with the OpReadClockKHR instruction will read from a clock that is consistent across invocations in the same queue family scope instance.

There is no method to synchronize the execution of these invocations within SPIR-V, and this can only be done with API synchronization primitives.

Each invocation in a queue family scope instance must be in the same device scope instance.

8.21.4. Command

Any shader invocations executed as the result of a single command such as vkCmdDispatch or vkCmdDraw form a command scope instance. For indirect drawing commands with drawCount greater than one, invocations from separate draws are in separate command scope instances. For ray tracing 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 trace rays command.

There is no specific Scope for communication across invocations in a command scope instance. As this has a clear boundary at the API level, coordination here can be performed in the API, rather than in SPIR-V.

Each invocation in a command scope instance must be in the same queue-family scope instance.

For shaders without defined workgroups, this set of invocations forms an invocation group as defined in the SPIR-V specification.

8.21.5. Primitive

Any fragment shader invocations executed as the result of rasterization of a single primitive form a primitive scope instance.

There is no specific Scope for communication across invocations in a primitive scope instance.

Any generated helper invocations are included in this scope instance.

Each invocation in a primitive scope instance must be in the same command scope instance.

Any input variables decorated with Flat are uniform within a primitive scope instance.

8.21.6. Shader Call

Any shader-call-related invocations that are executed in one or more ray tracing execution models form a shader call scope instance.

The ShaderCallKHR Scope can be used as Memory Scope for barrier and atomic operations.

Each invocation in a shader call scope instance must be in the same queue family scope instance.

8.21.7. Workgroup

A local workgroup is a set of invocations that can synchronize and share data with each other using memory in the Workgroup storage class.

The Workgroup Scope can be used as both an Execution Scope and Memory Scope for barrier and atomic operations.

Each invocation in a local workgroup must be in the same command scope instance.

Only task, mesh, and compute shaders have defined workgroups - other shader types cannot use workgroup functionality. For shaders that have defined workgroups, this set of invocations forms an invocation group as defined in the SPIR-V specification.

8.21.8. Subgroup

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.

The Subgroup Scope can be used as both an Execution Scope and Memory Scope for barrier and atomic operations. Other subgroup features allow the use of group operations with subgroup scope.

If the shaderSubgroupClock feature is enabled, using the Subgroup Scope with the OpReadClockKHR instruction will read from a clock that is consistent across invocations in the same subgroup.

For shaders that have defined workgroups, each invocation in a subgroup must be in the same local workgroup.

In other shader stages, each invocation in a subgroup must be in the same device scope instance.

Only shader stages that support subgroup operations have defined subgroups.

8.21.9. Quad

A quad scope instance is formed of four shader invocations.

In a fragment shader, each invocation in a quad scope instance is formed of invocations in neighboring framebuffer locations (x_i, y_i), where:

i is the index of the invocation within the scope instance.
w and h are the number of pixels the fragment covers in the x and y axes.
w and h are identical for all participating invocations.
(x₀) = (x₁ - w) = (x₂) = (x₃ - w)
(y₀) = (y₁) = (y₂ - h) = (y₃ - h)
Each invocation has the same layer and sample indices.

In a compute shader, if the DerivativeGroupQuadsNV execution mode is specified, each invocation in a quad scope instance is formed of invocations with adjacent local invocation IDs (x_i, y_i), where:

i is the index of the invocation within the quad scope instance.
(x₀) = (x₁ - 1) = (x₂) = (x₃ - 1)
(y₀) = (y₁) = (y₂ - 1) = (y₃ - 1)
x₀ and y₀ are integer multiples of 2.
Each invocation has the same z coordinate.

In a compute shader, if the DerivativeGroupLinearNV execution mode is specified, each invocation in a quad scope instance is formed of invocations with adjacent local invocation indices (l_i), where:

i is the index of the invocation within the quad scope instance.
(l₀) = (l₁ - 1) = (l₂ - 2) = (l₃ - 3)
l₀ is an integer multiple of 4.

In all shaders, each invocation in a quad scope instance is formed of invocations in adjacent subgroup invocation indices (s_i), where:

i is the index of the invocation within the quad scope instance.
(s₀) = (s₁ - 1) = (s₂ - 2) = (s₃ - 3)
s₀ is an integer multiple of 4.

Each invocation in a quad scope instance must be in the same subgroup.

Fragment and compute shaders have defined quad scope instances. If the quadOperationsInAllStages limit is supported, any shader stages that support subgroup operations also have defined quad scope instances.

8.21.10. Fragment Interlock

A fragment interlock scope instance is formed of fragment shader invocations based on their framebuffer locations (x,y,layer,sample), executed by commands inside a single subpass.

The specific set of invocations included varies based on the execution mode as follows:

If the SampleInterlockOrderedEXT or SampleInterlockUnorderedEXT execution modes are used, only invocations with identical framebuffer locations (x,y,layer,sample) are included.
If the PixelInterlockOrderedEXT or PixelInterlockUnorderedEXT execution modes are used, fragments with different sample ids are also included.
If the ShadingRateInterlockOrderedEXT or ShadingRateInterlockUnorderedEXT execution modes are used, fragments from neighbouring framebuffer locations are also included, as determined by the shading rate.

Only fragment shaders with one of the above execution modes have defined fragment interlock scope instances.

There is no specific Scope value for communication across invocations in a fragment interlock scope instance. However, this is implicitly used as a memory scope by OpBeginInvocationInterlockEXT and OpEndInvocationInterlockEXT.

Each invocation in a fragment interlock scope instance must be in the same queue family scope instance.

8.21.11. Invocation

The smallest scope is a single invocation; this is represented by the Invocation Scope in SPIR-V.

Fragment shader invocations must be in a primitive scope instance.

Invocations in fragment shaders that have a defined fragment interlock scope must be in a fragment interlock scope instance.

Invocations in shaders that have defined workgroups must be in a local workgroup.

Invocations in shaders that have a defined subgroup scope must be in a subgroup.

Invocations in shaders that have a defined quad scope must be in a quad scope instance.

All invocations in all stages must be in a command scope instance.

8.22. Group Operations

Group operations are executed by multiple invocations within a scope instance; with each invocation involved in calculating the result. This provides a mechanism for efficient communication between invocations in a particular scope instance.

Group operations all take a Scope defining the desired scope instance to operate within. Only the Subgroup scope can be used for these operations; the subgroupSupportedOperations limit defines which types of operation can be used.

8.22.1. Basic Group Operations

Basic group operations include the use of OpGroupNonUniformElect, OpControlBarrier, OpMemoryBarrier, and atomic operations.

OpGroupNonUniformElect can be used to choose a single invocation to perform a task for the whole group. Only the invocation with the lowest id in the group will return true.

The Memory Model appendix defines the operation of barriers and atomics.

8.22.2. Vote Group Operations

The vote group operations allow invocations within a group to compare values across a group. The types of votes enabled are:

Do all active group invocations agree that an expression is true?
Do any active group invocations evaluate an expression to true?
Do all active group 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 group and choose potentially faster code-paths in these cases.

8.22.3. Arithmetic Group Operations

The arithmetic group operations allow invocations to perform scans and reductions across a group. The operators supported are add, mul, min, max, and, or, xor.

For reductions, every invocation in a group will obtain the cumulative result of these operators applied to all values in the group. For exclusive scans, each invocation in a group will obtain the cumulative result of these operators applied to all values in invocations with a lower index in the group. Inclusive scans are identical to exclusive scans, except the cumulative result includes the operator applied to the value in the current invocation.

The order in which these operators are applied is implementation-dependent.

8.22.4. Ballot Group Operations

The ballot group operations allow invocations to perform more complex votes across the group. The ballot functionality allows all invocations within a group 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 group.

8.22.5. Shuffle Group Operations

The shuffle group operations allow invocations to read values from other invocations within a group.

8.22.6. Shuffle Relative Group Operations

The shuffle relative group operations allow invocations to read values from other invocations within the group 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 group.

8.22.7. Clustered Group Operations

The clustered group operations allow invocations to perform an operation among partitions of a group, such that the operation is only performed within the group invocations within a partition. The partitions for clustered group 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.

8.23. Quad Group Operations

Quad group operations (OpGroupNonUniformQuad*) are a specialized type of group operations that only operate on quad scope instances. Whilst these instructions do include a Scope parameter, this scope is always overridden; only the quad scope instance is included in its execution scope.

Fragment shaders that statically execute quad group operations must launch sufficient invocations to ensure their correct operation; additional helper invocations are launched for framebuffer locations not covered by rasterized fragments if necessary.

The index used to select participating invocations is i, as described for a quad scope instance, defined as the quad index in the SPIR-V specification.

For OpGroupNonUniformQuadBroadcast this value is equal to Index. For OpGroupNonUniformQuadSwap, it is equal to the implicit Index used by each participating invocation.

8.24. Derivative Operations

Derivative operations calculate the partial derivative for an expression P as a function of an invocation’s x and y coordinates.

Derivative operations operate on a set of invocations known as a derivative group as defined in the SPIR-V specification. A derivative group is equivalent to the local workgroup for a compute shader invocation, or the primitive scope instance for a fragment shader invocation.

Derivatives are calculated assuming that P is piecewise linear and continuous within the derivative group. All dynamic instances of explicit derivative instructions (OpDPdx*, OpDPdy*, and OpFwidth*) must be executed in control flow that is uniform within a derivative group. For other derivative operations, results are undefined if a dynamic instance is executed in control flow is not uniform within the derivative group.

Fragment shaders that statically execute derivative operations must launch sufficient invocations to ensure their correct operation; additional helper invocations are launched for framebuffer locations not covered by rasterized fragments if necessary.

Note

In a compute shader, it is the application’s responsibility to ensure that sufficient invocations are launched.

Derivative operations calculate their results as the difference between the result of P across invocations in the quad. For fine derivative operations (OpDPdxFine and OpDPdyFine), the values of DPdx(P_i) are calculated as

: DPdx(P₀) = DPdx(P₁) = P₁ - P₀
: DPdx(P₂) = DPdx(P₃) = P₃ - P₂

and the values of DPdy(P_i) are calculated as

: DPdy(P₀) = DPdy(P₂) = P₂ - P₀
: DPdy(P₁) = DPdy(P₃) = P₃ - P₁

where i is the index of each invocation as described in Quad.

Coarse derivative operations (OpDPdxCoarse and OpDPdyCoarse), calculate their results in roughly the same manner, but may only calculate two values instead of four (one for each of DPdx and DPdy), reusing the same result no matter the originating invocation. If an implementation does this, it should use the fine derivative calculations described for P₀.

Note

Derivative values are calculated between fragments rather than pixels. If the fragment shader invocations involved in the calculation covers multiple pixels, these operations cover a wider area, resulting in larger derivative values. This in turn will result in a coarser level of detail being selected for image sampling operations using derivatives.

Applications may want to account for this when using multi-pixel fragments; if pixel derivatives are desired, applications should use explicit derivative operations and divide the results by the size of the fragment in each dimension as follows:

: DPdx(P_n)' = DPdx(P_n) / w
: DPdy(P_n)' = DPdy(P_n) / h

where w and h are the size of the fragments in the quad, and DPdx(P_n)' and DPdy(P_n)' are the pixel derivatives.

The results for OpDPdx and OpDPdy may be calculated as either fine or coarse derivatives, with implementations favouring the most efficient approach. Implementations must choose coarse or fine consistently between the two.

Executing OpFwidthFine, OpFwidthCoarse, or OpFwidth is equivalent to executing the corresponding OpDPdx* and OpDPdy* instructions, taking the absolute value of the results, and summing them.

Executing a OpImage*Sample*ImplicitLod instruction is equivalent to executing OpDPdx(Coordinate) and OpDPdy(Coordinate), and passing the results as the Grad operands dx and dy.

Note

It is expected that using the ImplicitLod variants of sampling functions will be substantially more efficient than using the ExplicitLod variants with explicitly generated derivatives.

8.25. Helper Invocations

When performing derivative or quad group operations in a fragment shader, additional invocations may be spawned in order to ensure correct results. These additional invocations are known as helper invocations and can be identified by a non-zero value in the HelperInvocation built-in. 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.

Helper invocations may become inactive at any time for any reason, with one exception. If a helper invocation would be active if it were not a helper invocation, it must be active for derivative and quad group operations.

Helper invocations may become permanently inactive if all invocations in a quad scope instance become helper invocations.

8.26. Cooperative Matrices

A cooperative matrix type is a SPIR-V type where the storage for and computations performed on the matrix are spread across the invocations in a scope instance. These types give the implementation freedom in how to optimize matrix multiplies.

SPIR-V defines the types and instructions, but does not specify rules about what sizes/combinations are valid, and it is expected that different implementations may support different sizes.

To enumerate the supported cooperative matrix types and operations, call:

// Provided by VK_NV_cooperative_matrix
VkResult vkGetPhysicalDeviceCooperativeMatrixPropertiesNV(
    VkPhysicalDevice                            physicalDevice,
    uint32_t*                                   pPropertyCount,
    VkCooperativeMatrixPropertiesNV*            pProperties);

physicalDevice is the physical device.
pPropertyCount is a pointer to an integer related to the number of cooperative matrix properties available or queried.
pProperties is either NULL or a pointer to an array of VkCooperativeMatrixPropertiesNV structures.

If pProperties is NULL, then the number of cooperative matrix 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 cooperative matrix properties available, at most pPropertyCount structures will be written. If pPropertyCount is smaller than the number of cooperative matrix properties available, VK_INCOMPLETE will be returned instead of VK_SUCCESS, to indicate that not all the available cooperative matrix properties 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 VkCooperativeMatrixPropertiesNV structures

Return Codes

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

Each VkCooperativeMatrixPropertiesNV structure describes a single supported combination of types for a matrix multiply/add operation (OpCooperativeMatrixMulAddNV). The multiply can be described in terms of the following variables and types (in SPIR-V pseudocode):

    %A is of type OpTypeCooperativeMatrixNV %AType %scope %MSize %KSize
    %B is of type OpTypeCooperativeMatrixNV %BType %scope %KSize %NSize
    %C is of type OpTypeCooperativeMatrixNV %CType %scope %MSize %NSize
    %D is of type OpTypeCooperativeMatrixNV %DType %scope %MSize %NSize

    %D = %A * %B + %C // using OpCooperativeMatrixMulAddNV

A matrix multiply with these dimensions is known as an MxNxK matrix multiply.

The VkCooperativeMatrixPropertiesNV structure is defined as:

// Provided by VK_NV_cooperative_matrix
typedef struct VkCooperativeMatrixPropertiesNV {
    VkStructureType      sType;
    void*                pNext;
    uint32_t             MSize;
    uint32_t             NSize;
    uint32_t             KSize;
    VkComponentTypeNV    AType;
    VkComponentTypeNV    BType;
    VkComponentTypeNV    CType;
    VkComponentTypeNV    DType;
    VkScopeNV            scope;
} VkCooperativeMatrixPropertiesNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
MSize is the number of rows in matrices A, C, and D.
KSize is the number of columns in matrix A and rows in matrix B.
NSize is the number of columns in matrices B, C, D.
AType is the component type of matrix A, of type VkComponentTypeNV.
BType is the component type of matrix B, of type VkComponentTypeNV.
CType is the component type of matrix C, of type VkComponentTypeNV.
DType is the component type of matrix D, of type VkComponentTypeNV.
scope is the scope of all the matrix types, of type VkScopeNV.

If some types are preferred over other types (e.g. for performance), they should appear earlier in the list enumerated by vkGetPhysicalDeviceCooperativeMatrixPropertiesNV.

At least one entry in the list must have power of two values for all of MSize, KSize, and NSize.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_COOPERATIVE_MATRIX_PROPERTIES_NV
pNext must be NULL
AType must be a valid VkComponentTypeNV value
BType must be a valid VkComponentTypeNV value
CType must be a valid VkComponentTypeNV value
DType must be a valid VkComponentTypeNV value
scope must be a valid VkScopeNV value

Possible values for VkScopeNV include:

// Provided by VK_NV_cooperative_matrix
typedef enum VkScopeNV {
    VK_SCOPE_DEVICE_NV = 1,
    VK_SCOPE_WORKGROUP_NV = 2,
    VK_SCOPE_SUBGROUP_NV = 3,
    VK_SCOPE_QUEUE_FAMILY_NV = 5,
} VkScopeNV;

VK_SCOPE_DEVICE_NV corresponds to SPIR-V Device scope.
VK_SCOPE_WORKGROUP_NV corresponds to SPIR-V Workgroup scope.
VK_SCOPE_SUBGROUP_NV corresponds to SPIR-V Subgroup scope.
VK_SCOPE_QUEUE_FAMILY_NV corresponds to SPIR-V QueueFamily scope.

All enum values match the corresponding SPIR-V value.

Possible values for VkComponentTypeNV include:

// Provided by VK_NV_cooperative_matrix
typedef enum VkComponentTypeNV {
    VK_COMPONENT_TYPE_FLOAT16_NV = 0,
    VK_COMPONENT_TYPE_FLOAT32_NV = 1,
    VK_COMPONENT_TYPE_FLOAT64_NV = 2,
    VK_COMPONENT_TYPE_SINT8_NV = 3,
    VK_COMPONENT_TYPE_SINT16_NV = 4,
    VK_COMPONENT_TYPE_SINT32_NV = 5,
    VK_COMPONENT_TYPE_SINT64_NV = 6,
    VK_COMPONENT_TYPE_UINT8_NV = 7,
    VK_COMPONENT_TYPE_UINT16_NV = 8,
    VK_COMPONENT_TYPE_UINT32_NV = 9,
    VK_COMPONENT_TYPE_UINT64_NV = 10,
} VkComponentTypeNV;

VK_COMPONENT_TYPE_FLOAT16_NV corresponds to SPIR-V OpTypeFloat 16.
VK_COMPONENT_TYPE_FLOAT32_NV corresponds to SPIR-V OpTypeFloat 32.
VK_COMPONENT_TYPE_FLOAT64_NV corresponds to SPIR-V OpTypeFloat 64.
VK_COMPONENT_TYPE_SINT8_NV corresponds to SPIR-V OpTypeInt 8 1.
VK_COMPONENT_TYPE_SINT16_NV corresponds to SPIR-V OpTypeInt 16 1.
VK_COMPONENT_TYPE_SINT32_NV corresponds to SPIR-V OpTypeInt 32 1.
VK_COMPONENT_TYPE_SINT64_NV corresponds to SPIR-V OpTypeInt 64 1.
VK_COMPONENT_TYPE_UINT8_NV corresponds to SPIR-V OpTypeInt 8 0.
VK_COMPONENT_TYPE_UINT16_NV corresponds to SPIR-V OpTypeInt 16 0.
VK_COMPONENT_TYPE_UINT32_NV corresponds to SPIR-V OpTypeInt 32 0.
VK_COMPONENT_TYPE_UINT64_NV corresponds to SPIR-V OpTypeInt 64 0.

8.27. 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:

// Provided by VK_EXT_validation_cache
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkValidationCacheEXT)

To create validation cache objects, call:

// Provided by VK_EXT_validation_cache
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 containing 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 by adding this object to the VkShaderModuleCreateInfo structure’s pNext chain. If a VkShaderModuleValidationCacheCreateInfoEXT object is included in 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY

The VkValidationCacheCreateInfoEXT structure is defined as:

// Provided by VK_EXT_validation_cache
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 a structure extending this 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

// Provided by VK_EXT_validation_cache
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:

// Provided by VK_EXT_validation_cache
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 a pointer to 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

Data can be retrieved from a validation cache object using the command:

// Provided by VK_EXT_validation_cache
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

VK_SUCCESS
VK_INCOMPLETE

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:

// Provided by VK_EXT_validation_cache
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:

// Provided by VK_EXT_validation_cache
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, a ray tracing pipeline, or a compute pipeline.

The graphics pipeline can be operated in two modes, as either primitive shading or mesh shading pipeline.

Primitive Shading

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.

Mesh Shading

When using the mesh shading pipeline input primitives are not assembled implicitly, but explicitly through the (Mesh Shader). The work on the mesh pipeline is initiated by the application drawing a set of mesh tasks.

If an optional (Task Shader) is active, each task triggers the execution of a task shader workgroup that will generate a new set of tasks upon completion. Each of these spawned tasks, or each of the original dispatched tasks if no task shader is present, triggers the execution of a mesh shader workgroup that produces an output mesh with a variable-sized number of primitives assembled from vertices stored in the output mesh.

Common

The final resulting primitives are clipped to a clip volume in preparation for the next stage, Rasterization. The rasterizer produces a series of fragments associated with a region of the framebuffer, from a two-dimensional description of a point, line segment, or triangle. These fragments are processed by fragment operations to determine whether generated values will be written to the framebuffer. fragment shading determines the values to be written to the framebuffer attachments. Framebuffer operations then 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.

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, ray tracing, and graphics pipelines are each represented by VkPipeline handles:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 a pointer to an array of VkComputePipelineCreateInfo structures.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.

pPipelines is a pointer to an array of VkPipeline handles 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
If pipelineCache was created with VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT, host access to pipelineCache must be externally synchronized

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

VK_SUCCESS
VK_PIPELINE_COMPILE_REQUIRED_EXT

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_SHADER_NV

The VkComputePipelineCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkPipelineCreateFlagBits specifying how the pipeline will be generated.
stage is a VkPipelineShaderStageCreateInfo structure 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.

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
flags must not include VK_PIPELINE_CREATE_LIBRARY_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV
If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT or VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_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 VkPipelineCompilerControlCreateInfoAMD or VkPipelineCreationFeedbackCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

The VkPipelineShaderStageCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkPipelineShaderStageCreateFlagBits specifying how the pipeline shader stage will be generated.
stage is a VkShaderStageFlagBits value specifying a single pipeline stage.
module is a VkShaderModule object containing 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 a VkSpecializationInfo structure, as described in Specialization Constants, or 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
If the mesh shader feature is not enabled, stage must not be VK_SHADER_STAGE_MESH_BIT_NV
If the task shader feature is not enabled, stage must not be VK_SHADER_STAGE_TASK_BIT_NV
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 a vertex processing stage, 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 a vertex processing stage, 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
If stage is VK_SHADER_STAGE_MESH_BIT_NV, the identified entry point must have an OpExecutionMode instruction that specifies a maximum output vertex count, OutputVertices, that is greater than 0 and less than or equal to VkPhysicalDeviceMeshShaderPropertiesNV::maxMeshOutputVertices
If stage is VK_SHADER_STAGE_MESH_BIT_NV, the identified entry point must have an OpExecutionMode instruction that specifies a maximum output primitive count, OutputPrimitivesNV, that is greater than 0 and less than or equal to VkPhysicalDeviceMeshShaderPropertiesNV::maxMeshOutputPrimitives
If flags has the VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flag set, the subgroupSizeControl feature must be enabled
If flags has the VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT flag set, the computeFullSubgroups feature must be enabled
If a VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain, flags must not have the VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flag set
If a VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain, the subgroupSizeControl feature must be enabled, and stage must be a valid bit specified in requiredSubgroupSizeStages
If a VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain and stage is VK_SHADER_STAGE_COMPUTE_BIT, the local workgroup size of the shader must be less than or equal to the product of VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT::requiredSubgroupSize and maxComputeWorkgroupSubgroups
If a VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain, and flags has the VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT flag set, the local workgroup size in the X dimension of the pipeline must be a multiple of VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT::requiredSubgroupSize
If flags has both the VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT and VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flags set, the local workgroup size in the X dimension of the pipeline must be a multiple of maxSubgroupSize
If flags has the VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT flag set and flags does not have the VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flag set and no VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain, the local workgroup size in the X dimension of the pipeline must be a multiple of subgroupSize

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO
pNext must be NULL or a pointer to a valid instance of VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkPipelineShaderStageCreateFlagBits values
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

// Provided by VK_VERSION_1_0
typedef VkFlags VkPipelineShaderStageCreateFlags;

VkPipelineShaderStageCreateFlags is a bitmask type for setting a mask of zero or more VkPipelineShaderStageCreateFlagBits.

Possible values of the flags member of VkPipelineShaderStageCreateInfo specifying how a pipeline shader stage is created, are:

// Provided by VK_VERSION_1_0
typedef enum VkPipelineShaderStageCreateFlagBits {
  // Provided by VK_EXT_subgroup_size_control
    VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT = 0x00000001,
  // Provided by VK_EXT_subgroup_size_control
    VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT = 0x00000002,
} VkPipelineShaderStageCreateFlagBits;

VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT specifies that the SubgroupSize may vary in the shader stage.
VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT specifies that the subgroup sizes must be launched with all invocations active in the compute stage.

Note

If VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT and VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT are specified and minSubgroupSize does not equal maxSubgroupSize and no required subgroup size is specified, then the only way to guarantee that the 'X' dimension of the local workgroup size is a multiple of SubgroupSize is to make it a multiple of maxSubgroupSize. Under these conditions, you are guaranteed full subgroups but not any particular subgroup size.

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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_RAYGEN_BIT_KHR = 0x00000100,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_ANY_HIT_BIT_KHR = 0x00000200,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_CLOSEST_HIT_BIT_KHR = 0x00000400,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_MISS_BIT_KHR = 0x00000800,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_INTERSECTION_BIT_KHR = 0x00001000,
  // Provided by VK_KHR_ray_tracing
    VK_SHADER_STAGE_CALLABLE_BIT_KHR = 0x00002000,
  // Provided by VK_NV_mesh_shader
    VK_SHADER_STAGE_TASK_BIT_NV = 0x00000040,
  // Provided by VK_NV_mesh_shader
    VK_SHADER_STAGE_MESH_BIT_NV = 0x00000080,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_RAYGEN_BIT_NV = VK_SHADER_STAGE_RAYGEN_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_ANY_HIT_BIT_NV = VK_SHADER_STAGE_ANY_HIT_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_CLOSEST_HIT_BIT_NV = VK_SHADER_STAGE_CLOSEST_HIT_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_MISS_BIT_NV = VK_SHADER_STAGE_MISS_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_INTERSECTION_BIT_NV = VK_SHADER_STAGE_INTERSECTION_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_SHADER_STAGE_CALLABLE_BIT_NV = VK_SHADER_STAGE_CALLABLE_BIT_KHR,
} 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.
VK_SHADER_STAGE_TASK_BIT_NV specifies the task stage.
VK_SHADER_STAGE_MESH_BIT_NV specifies the mesh stage.
VK_SHADER_STAGE_RAYGEN_BIT_KHR specifies the ray generation stage.
VK_SHADER_STAGE_ANY_HIT_BIT_KHR specifies the any-hit stage.
VK_SHADER_STAGE_CLOSEST_HIT_BIT_KHR specifies the closest hit stage.
VK_SHADER_STAGE_MISS_BIT_KHR specifies the miss stage.
VK_SHADER_STAGE_INTERSECTION_BIT_KHR specifies the intersection stage.
VK_SHADER_STAGE_CALLABLE_BIT_KHR specifies the callable stage.

Note

VK_SHADER_STAGE_ALL_GRAPHICS only includes the original five graphics stages included in Vulkan 1.0, and not any stages added by extensions. Thus, it may not have the desired effect in all cases.

// Provided by VK_VERSION_1_0
typedef VkFlags VkShaderStageFlags;

VkShaderStageFlags is a bitmask type for setting a mask of zero or more VkShaderStageFlagBits.

The VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is defined as:

// Provided by VK_EXT_subgroup_size_control
typedef struct VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           requiredSubgroupSize;
} VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
requiredSubgroupSize is an unsigned integer value that specifies the required subgroup size for the newly created pipeline shader stage.

Valid Usage

requiredSubgroupSize must be a power-of-two integer
requiredSubgroupSize must be greater or equal to minSubgroupSize
requiredSubgroupSize must be less than or equal to maxSubgroupSize

If a VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure is included in the pNext chain of VkPipelineShaderStageCreateInfo, it specifies that the pipeline shader stage being compiled has a required subgroup size.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_REQUIRED_SUBGROUP_SIZE_CREATE_INFO_EXT

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:

// Provided by VK_VERSION_1_0
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 a pointer to an array of VkGraphicsPipelineCreateInfo structures.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pPipelines is a pointer to an array of VkPipeline handles 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
If pipelineCache was created with VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT, host access to pipelineCache must be externally synchronized

Note

An implicit cache may be provided by the implementation or a layer. For this reason, it is still valid to set VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT on flags for any element of pCreateInfos while passing VK_NULL_HANDLE for pipelineCache.

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

VK_SUCCESS
VK_PIPELINE_COMPILE_REQUIRED_EXT

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_SHADER_NV

The VkGraphicsPipelineCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 a pointer to an array of stageCount VkPipelineShaderStageCreateInfo structures describing the set of the shader stages to be included in the graphics pipeline.
pVertexInputState is a pointer to a VkPipelineVertexInputStateCreateInfo structure. It is ignored if the pipeline includes a mesh shader stage.
pInputAssemblyState is a pointer to a VkPipelineInputAssemblyStateCreateInfo structure which determines input assembly behavior, as described in Drawing Commands. It is ignored if the pipeline includes a mesh shader stage.
pTessellationState is a pointer to a 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 a VkPipelineViewportStateCreateInfo structure, and is ignored if the pipeline has rasterization disabled.
pRasterizationState is a pointer to a VkPipelineRasterizationStateCreateInfo structure.
pMultisampleState is a pointer to a VkPipelineMultisampleStateCreateInfo structure, and is ignored if the pipeline has rasterization disabled.
pDepthStencilState is a pointer to a 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 a 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 a VkPipelineDynamicStateCreateInfo structure, 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.

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 geometric shader stages provided in pStages must be either from the mesh shading pipeline (stage is VK_SHADER_STAGE_TASK_BIT_NV or VK_SHADER_STAGE_MESH_BIT_NV) or from the primitive shading pipeline (stage is VK_SHADER_STAGE_VERTEX_BIT, VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT, VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT, or VK_SHADER_STAGE_GEOMETRY_BIT)
The stage member of one element of pStages must be either VK_SHADER_STAGE_VERTEX_BIT or VK_SHADER_STAGE_MESH_BIT_NV
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 attached image’s format features does not contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT
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 valid 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 included in 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 included in 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 included in the pNext chain of pMultisampleState is VK_TRUE, sampleLocationsInfo.sampleLocationsPerPixel must equal rasterizationSamples
If the sampleLocationsEnable member of a VkPipelineSampleLocationsStateCreateInfoEXT structure included in 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 neither the VK_AMD_mixed_attachment_samples nor the VK_NV_framebuffer_mixed_samples extensions are enabled, and if subpass uses color and/or depth/stencil attachments, then the rasterizationSamples member of pMultisampleState must be the same as the sample count for those subpass attachments
If the VK_AMD_mixed_attachment_samples extension is enabled, and 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 the VK_NV_framebuffer_mixed_samples extension is enabled, and 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 the VK_NV_framebuffer_mixed_samples extension is enabled, and 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 the VK_NV_coverage_reduction_mode extension is enabled, the coverage reduction mode specified by VkPipelineCoverageReductionStateCreateInfoNV::coverageReductionMode, the rasterizationSamples member of pMultisampleState and the sample counts for the color and depth/stencil attachments (if the subpass has them) must be a valid combination returned by vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV
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 an aspectMask at renderPass creation time, its shader code must only read from the aspects that were specified for that input attachment
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, included in 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
If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_EXCLUSIVE_SCISSOR_NV, and if pViewportState->pNext chain includes a VkPipelineViewportExclusiveScissorStateCreateInfoNV structure, and if its exclusiveScissorCount member is not 0, then its pExclusiveScissors member must be a valid pointer to an array of exclusiveScissorCount VkRect2D structures
If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_VIEWPORT_SHADING_RATE_PALETTE_NV, and if pViewportState->pNext chain includes a VkPipelineViewportShadingRateImageStateCreateInfoNV structure, then its pShadingRatePalettes member must be a valid pointer to an array of viewportCount valid VkShadingRatePaletteNV structures
If no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT, and if pNext chain includes a VkPipelineDiscardRectangleStateCreateInfoEXT structure, and if its discardRectangleCount member is not 0, then its pDiscardRectangles member must be a valid pointer to an array of discardRectangleCount VkRect2D structures
If pStages includes a vertex shader stage, pVertexInputState must be a valid pointer to a valid VkPipelineVertexInputStateCreateInfo structure
If pStages includes a vertex shader stage, pInputAssemblyState must be a valid pointer to a valid VkPipelineInputAssemblyStateCreateInfo structure
The Xfb execution mode can be specified by only one shader stage in pStages
If any shader stage in pStages specifies Xfb execution mode it must be the last vertex processing stage
If a VkPipelineRasterizationStateStreamCreateInfoEXT::rasterizationStream value other than zero is specified, all variables in the output interface of the entry point being compiled decorated with Position, PointSize, ClipDistance, or CullDistance must all be decorated with identical Stream values that match the rasterizationStream
If VkPipelineRasterizationStateStreamCreateInfoEXT::rasterizationStream is zero, or not specified, all variables in the output interface of the entry point being compiled decorated with Position, PointSize, ClipDistance, or CullDistance must all be decorated with a Stream value of zero, or must not specify the Stream decoration
If the last vertex processing stage is a geometry shader, and that geometry shader uses the GeometryStreams capability, then VkPhysicalDeviceTransformFeedbackFeaturesEXT::geometryStreams feature must be enabled
If there are any mesh shader stages in the pipeline there must not be any shader stage in the pipeline with a Xfb execution mode
If the lineRasterizationMode member of a VkPipelineRasterizationLineStateCreateInfoEXT structure included in the pNext chain of pRasterizationState is VK_LINE_RASTERIZATION_MODE_BRESENHAM_EXT or VK_LINE_RASTERIZATION_MODE_RECTANGULAR_SMOOTH_EXT and if rasterization is enabled, then the alphaToCoverageEnable, alphaToOneEnable, and sampleShadingEnable members of pMultisampleState must all be VK_FALSE
If the stippledLineEnable member of VkPipelineRasterizationLineStateCreateInfoEXT is VK_TRUE and no element of the pDynamicStates member of pDynamicState is VK_DYNAMIC_STATE_LINE_STIPPLE_EXT, then the lineStippleFactor member of VkPipelineRasterizationLineStateCreateInfoEXT must be in the range [1,256]
flags must not include VK_PIPELINE_CREATE_LIBRARY_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR
If the extendedDynamicState feature is not enabled, there must be no element of the pDynamicStates member of pDynamicState set to VK_DYNAMIC_STATE_CULL_MODE_EXT, VK_DYNAMIC_STATE_FRONT_FACE_EXT, VK_DYNAMIC_STATE_PRIMITIVE_TOPOLOGY_EXT, VK_DYNAMIC_STATE_VIEWPORT_WITH_COUNT_EXT, VK_DYNAMIC_STATE_SCISSOR_WITH_COUNT_EXT, VK_DYNAMIC_STATE_VERTEX_INPUT_BINDING_STRIDE_EXT, VK_DYNAMIC_STATE_DEPTH_TEST_ENABLE_EXT, VK_DYNAMIC_STATE_DEPTH_WRITE_ENABLE_EXT, VK_DYNAMIC_STATE_DEPTH_COMPARE_OP_EXT, VK_DYNAMIC_STATE_DEPTH_BOUNDS_TEST_ENABLE_EXT, VK_DYNAMIC_STATE_STENCIL_TEST_ENABLE_EXT, or VK_DYNAMIC_STATE_STENCIL_OP_EXT
If VK_DYNAMIC_STATE_VIEWPORT_WITH_COUNT_EXT is included in the pDynamicStates array then viewportCount must be zero
If VK_DYNAMIC_STATE_SCISSOR_WITH_COUNT_EXT is included in the pDynamicStates array then scissorCount must be zero
If flags includes VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV, then the VkPhysicalDeviceDeviceGeneratedCommandsFeaturesNV::deviceGeneratedCommands feature must be enabled
If flags includes VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV, then all stages must not specify Xfb execution mode
If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT or VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_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 VkGraphicsPipelineShaderGroupsCreateInfoNV, VkPipelineCompilerControlCreateInfoAMD, VkPipelineCreationFeedbackCreateInfoEXT, VkPipelineDiscardRectangleStateCreateInfoEXT, or VkPipelineRepresentativeFragmentTestStateCreateInfoNV
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkPipelineCreateFlagBits values
pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures
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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

Possible values of the flags member of VkGraphicsPipelineCreateInfo, VkRayTracingPipelineCreateInfoKHR, VkRayTracingPipelineCreateInfoNV, and VkComputePipelineCreateInfo, specifying how a pipeline is created, are:

// Provided by VK_VERSION_1_0
typedef enum VkPipelineCreateFlagBits {
    VK_PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT = 0x00000001,
    VK_PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT = 0x00000002,
    VK_PIPELINE_CREATE_DERIVATIVE_BIT = 0x00000004,
  // Provided by VK_VERSION_1_1
    VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT = 0x00000008,
  // Provided by VK_VERSION_1_1
    VK_PIPELINE_CREATE_DISPATCH_BASE_BIT = 0x00000010,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR = 0x00004000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR = 0x00008000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR = 0x00010000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR = 0x00020000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR = 0x00001000,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR = 0x00002000,
  // Provided by VK_NV_ray_tracing
    VK_PIPELINE_CREATE_DEFER_COMPILE_BIT_NV = 0x00000020,
  // Provided by VK_KHR_pipeline_executable_properties
    VK_PIPELINE_CREATE_CAPTURE_STATISTICS_BIT_KHR = 0x00000040,
  // Provided by VK_KHR_pipeline_executable_properties
    VK_PIPELINE_CREATE_CAPTURE_INTERNAL_REPRESENTATIONS_BIT_KHR = 0x00000080,
  // Provided by VK_NV_device_generated_commands
    VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV = 0x00040000,
  // Provided by VK_KHR_pipeline_library
    VK_PIPELINE_CREATE_LIBRARY_BIT_KHR = 0x00000800,
  // Provided by VK_EXT_pipeline_creation_cache_control
    VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT = 0x00000100,
  // Provided by VK_EXT_pipeline_creation_cache_control
    VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT = 0x00000200,
    VK_PIPELINE_CREATE_DISPATCH_BASE = VK_PIPELINE_CREATE_DISPATCH_BASE_BIT,
  // Provided by VK_KHR_device_group
    VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT_KHR = VK_PIPELINE_CREATE_VIEW_INDEX_FROM_DEVICE_INDEX_BIT,
  // Provided by VK_KHR_device_group
    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 pipeline creation call.
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.
VK_PIPELINE_CREATE_DEFER_COMPILE_BIT_NV specifies that a pipeline is created with all shaders in the deferred state. Before using the pipeline the application must call vkCompileDeferredNV exactly once on each shader in the pipeline before using the pipeline.
VK_PIPELINE_CREATE_CAPTURE_STATISTICS_BIT_KHR specifies that the shader compiler should capture statistics for the executables produced by the compile process which can later be retrieved by calling vkGetPipelineExecutableStatisticsKHR. Enabling this flag must not affect the final compiled pipeline but may disable pipeline caching or otherwise affect pipeline creation time.
VK_PIPELINE_CREATE_CAPTURE_INTERNAL_REPRESENTATIONS_BIT_KHR specifies that the shader compiler should capture the internal representations of executables produced by the compile process which can later be retrieved by calling vkGetPipelineExecutableInternalRepresentationsKHR. Enabling this flag must not affect the final compiled pipeline but may disable pipeline caching or otherwise affect pipeline creation time.
VK_PIPELINE_CREATE_LIBRARY_BIT_KHR specifies that the pipeline cannot be used directly, and instead defines a pipeline library that can be combined with other pipelines using the VkPipelineLibraryCreateInfoKHR structure. This is available in raytracing pipelines.
VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR specifies that an any hit shader will always be present when an any hit shader would be executed.
VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR specifies that a closest hit shader will always be present when a closest hit shader would be executed.
VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR specifies that a miss shader will always be present when a miss shader would be executed.
VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR specifies that an intersection shader will always be present when an intersection shader would be executed.
VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR specifies that triangle primitives will be skipped during traversal using OpTraceKHR.
VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR specifies that AABB primitives will be skipped during traversal using OpTraceKHR.
VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV specifies that the pipeline can be used in combination with Device-Generated Commands.
VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT specifies that pipeline creation will fail if a compile is required for creation of a valid VkPipeline object; VK_PIPELINE_COMPILE_REQUIRED_EXT will be returned by pipeline creation, and the VkPipeline will be set to VK_NULL_HANDLE.
When creating multiple pipelines, VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT specifies that control will be returned to the application on failure of the corresponding pipeline rather than continuing to create additional pipelines.

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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkPipelineCreateFlags;

VkPipelineCreateFlags is a bitmask type for setting a mask of zero or more VkPipelineCreateFlagBits.

The VkPipelineDynamicStateCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is reserved for future use.
dynamicStateCount is the number of elements in the pDynamicStates array.
pDynamicStates is a pointer to 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
If dynamicStateCount is not 0, pDynamicStates must be a valid pointer to an array of dynamicStateCount valid VkDynamicState values

// Provided by VK_VERSION_1_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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_NV_clip_space_w_scaling
    VK_DYNAMIC_STATE_VIEWPORT_W_SCALING_NV = 1000087000,
  // Provided by VK_EXT_discard_rectangles
    VK_DYNAMIC_STATE_DISCARD_RECTANGLE_EXT = 1000099000,
  // Provided by VK_EXT_sample_locations
    VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_EXT = 1000143000,
  // Provided by VK_NV_shading_rate_image
    VK_DYNAMIC_STATE_VIEWPORT_SHADING_RATE_PALETTE_NV = 1000164004,
  // Provided by VK_NV_shading_rate_image
    VK_DYNAMIC_STATE_VIEWPORT_COARSE_SAMPLE_ORDER_NV = 1000164006,
  // Provided by VK_NV_scissor_exclusive
    VK_DYNAMIC_STATE_EXCLUSIVE_SCISSOR_NV = 1000205001,
  // Provided by VK_EXT_line_rasterization
    VK_DYNAMIC_STATE_LINE_STIPPLE_EXT = 1000259000,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_CULL_MODE_EXT = 1000267000,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_FRONT_FACE_EXT = 1000267001,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_PRIMITIVE_TOPOLOGY_EXT = 1000267002,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_VIEWPORT_WITH_COUNT_EXT = 1000267003,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_SCISSOR_WITH_COUNT_EXT = 1000267004,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_VERTEX_INPUT_BINDING_STRIDE_EXT = 1000267005,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_DEPTH_TEST_ENABLE_EXT = 1000267006,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_DEPTH_WRITE_ENABLE_EXT = 1000267007,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_DEPTH_COMPARE_OP_EXT = 1000267008,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_DEPTH_BOUNDS_TEST_ENABLE_EXT = 1000267009,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_STENCIL_TEST_ENABLE_EXT = 1000267010,
  // Provided by VK_EXT_extended_dynamic_state
    VK_DYNAMIC_STATE_STENCIL_OP_EXT = 1000267011,
} 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 VkPipelineViewportWScalingStateCreateInfoNV member viewportScalingEnable set to VK_TRUE
VK_DYNAMIC_STATE_DISCARD_RECTANGLE_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.
VK_DYNAMIC_STATE_EXCLUSIVE_SCISSOR_NV specifies that the pExclusiveScissors state in VkPipelineViewportExclusiveScissorStateCreateInfoNV will be ignored and must be set dynamically with vkCmdSetExclusiveScissorNV before any draw commands. The number of exclusive scissor rectangles used by a pipeline is still specified by the exclusiveScissorCount member of VkPipelineViewportExclusiveScissorStateCreateInfoNV.
VK_DYNAMIC_STATE_VIEWPORT_SHADING_RATE_PALETTE_NV specifies that the pShadingRatePalettes state in VkPipelineViewportShadingRateImageStateCreateInfoNV will be ignored and must be set dynamically with vkCmdSetViewportShadingRatePaletteNV before any draw commands.
VK_DYNAMIC_STATE_VIEWPORT_COARSE_SAMPLE_ORDER_NV specifies that the coarse sample order state in VkPipelineViewportCoarseSampleOrderStateCreateInfoNV will be ignored and must be set dynamically with vkCmdSetCoarseSampleOrderNV before any draw commands.
VK_DYNAMIC_STATE_LINE_STIPPLE_EXT specifies that the lineStippleFactor and lineStipplePattern state in VkPipelineRasterizationLineStateCreateInfoEXT will be ignored and must be set dynamically with vkCmdSetLineStippleEXT before any draws are performed with a pipeline state with VkPipelineRasterizationLineStateCreateInfoEXT member stippledLineEnable set to VK_TRUE.
VK_DYNAMIC_STATE_CULL_MODE_EXT specifies that the cullMode state in VkPipelineRasterizationStateCreateInfo will be ignored and must be set dynamically with vkCmdSetCullModeEXT before any draw commands.
VK_DYNAMIC_STATE_FRONT_FACE_EXT specifies that the frontFace state in VkPipelineRasterizationStateCreateInfo will be ignored and must be set dynamically with vkCmdSetFrontFaceEXT before any draw commands.
VK_DYNAMIC_STATE_PRIMITIVE_TOPOLOGY_EXT specifies that the topology state in VkPipelineInputAssemblyStateCreateInfo only specifies the topology class, and the specific topology order and adjacency must be set dynamically with vkCmdSetPrimitiveTopologyEXT before any draw commands.
VK_DYNAMIC_STATE_VIEWPORT_WITH_COUNT_EXT specifies that the viewportCount and pViewports state in VkPipelineViewportStateCreateInfo will be ignored and must be set dynamically with vkCmdSetViewportWithCountEXT before any draw call.
VK_DYNAMIC_STATE_SCISSOR_WITH_COUNT_EXT specifies that the scissorCount and pScissors state in VkPipelineViewportStateCreateInfo will be ignored and must be set dynamically with vkCmdSetScissorWithCountEXT before any draw call.
VK_DYNAMIC_STATE_VERTEX_INPUT_BINDING_STRIDE_EXT specifies that the stride state in VkVertexInputBindingDescription will be ignored and must be set dynamically with vkCmdBindVertexBuffers2EXT before any draw call.
VK_DYNAMIC_STATE_DEPTH_TEST_ENABLE_EXT specifies that the depthTestEnable state in VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthTestEnableEXT before any draw call.
VK_DYNAMIC_STATE_DEPTH_WRITE_ENABLE_EXT specifies that the depthWriteEnable state in VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthWriteEnableEXT before any draw call.
VK_DYNAMIC_STATE_DEPTH_COMPARE_OP_EXT specifies that the depthCompareOp state in VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthCompareOpEXT before any draw call.
VK_DYNAMIC_STATE_DEPTH_BOUNDS_TEST_ENABLE_EXT specifies that the depthBoundsTestEnable state in VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetDepthBoundsTestEnableEXT before any draw call.
VK_DYNAMIC_STATE_STENCIL_TEST_ENABLE_EXT specifies that the stencilTestEnable state in VkPipelineDepthStencilStateCreateInfo will be ignored and must be set dynamically with vkCmdSetStencilTestEnableEXT before any draw call.
VK_DYNAMIC_STATE_STENCIL_OP_EXT specifies that the failOp, passOp, depthFailOp, and compareOp states in VkPipelineDepthStencilStateCreateInfo for both front and back will be ignored and must be set dynamically with vkCmdSetStencilOpEXT before any draws are performed with a pipeline state with VkPipelineDepthStencilStateCreateInfo member stencilTestEnable set to VK_TRUE

9.2.1. Valid Combinations of Stages for Graphics Pipelines

The geometric primitive processing can either be handled on a per primitive basis by the vertex, tessellation, and geometry shader stages, or on a per mesh basis using task and mesh shader stages. If the pipeline includes a mesh shader stage, it uses the mesh pipeline, otherwise it uses the primitive pipeline.

If a task shader is omitted, the task shading stage is skipped.

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, fragment color outputs have undefined values, and the fragment depth value is 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:

Depth/stencil-only rendering in a subpass with no color attachments
- Active Pipeline Shader Stages
  - Vertex Shader
- Required: Fixed-Function Pipeline Stages
Color-only rendering in a subpass with no depth/stencil attachment
- Active Pipeline Shader Stages
  - Vertex Shader
  - Fragment Shader
- Required: Fixed-Function Pipeline Stages
Rendering pipeline with tessellation and geometry shaders
- Active Pipeline Shader Stages
  - Vertex Shader
  - Tessellation Control Shader
  - Tessellation Evaluation Shader
  - Geometry Shader
  - Fragment Shader
- Required: Fixed-Function Pipeline Stages
Rendering pipeline with task and mesh shaders
- Active Pipeline Shader Stages
  - Task Shader
  - Mesh Shader
  - Fragment Shader
- Required: Fixed-Function Pipeline Stages

9.2.2. Graphics Pipeline Shader Groups

Graphics pipelines can contain multiple shader groups that can be bound individually. Each shader group behaves as if it was a pipeline using the shader group’s state. When the pipeline is bound by regular means, it behaves as if the state of group 0 is active, use vkCmdBindPipelineShaderGroupNV to bind an invidual shader group.

The primary purpose of shader groups is allowing the device to bind different pipeline state using Device-Generated Commands.

The VkGraphicsPipelineShaderGroupsCreateInfoNV structure is defined as:

// Provided by VK_NV_device_generated_commands
typedef struct VkGraphicsPipelineShaderGroupsCreateInfoNV {
    VkStructureType                             sType;
    const void*                                 pNext;
    uint32_t                                    groupCount;
    const VkGraphicsShaderGroupCreateInfoNV*    pGroups;
    uint32_t                                    pipelineCount;
    const VkPipeline*                           pPipelines;
} VkGraphicsPipelineShaderGroupsCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
groupCount is the number of elements in the pGroups array.
pGroups is an array of VkGraphicsShaderGroupCreateInfoNV values specifying which state of the original VkGraphicsPipelineCreateInfo each shader group overrides.
pipelineCount is the number of elements in the pPipelines array.
pPipelines is an array of graphics VkPipeline, which are referenced within the created pipeline, including all their shader groups.

When referencing shader groups by index, groups defined in the referenced pipelines are treated as if they were defined as additional entries in pGroups. They are appended in the order they appear in the pPipelines array and in the pGroups array when those pipelines were defined.

The application must maintain the lifetime of all such referenced pipelines based on the pipelines that make use of them.

Valid Usage

groupCount must be at least 1 and as maximum VkPhysicalDeviceDeviceGeneratedCommandsPropertiesNV::maxGraphicsShaderGroupCount
The sum of groupCount including those groups added from referenced pPipelines must also be as maximum VkPhysicalDeviceDeviceGeneratedCommandsPropertiesNV::maxGraphicsShaderGroupCount
The state of the first element of pGroups must match its equivalent within the parent’s VkGraphicsPipelineCreateInfo
Each element of pGroups must in combination with the rest of the pipeline state yield a valid state configuration
All elements of pGroups must use the same shader stage combinations unless any mesh shader stage is used, then either combination of task and mesh or just mesh shader is valid
Mesh and regular primitive shading stages cannot be mixed across pGroups
Each element of the pPipelines member of libraries must have been created with identical state to the pipeline currently created except the state that can be overriden by VkGraphicsShaderGroupCreateInfoNV
The VkPhysicalDeviceDeviceGeneratedCommandsFeaturesNV::deviceGeneratedCommands feature must be enabled

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_SHADER_GROUPS_CREATE_INFO_NV
pGroups must be a valid pointer to an array of groupCount valid VkGraphicsShaderGroupCreateInfoNV structures
If pipelineCount is not 0, pPipelines must be a valid pointer to an array of pipelineCount valid VkPipeline handles
groupCount must be greater than 0

The VkGraphicsShaderGroupCreateInfoNV structure provides the state overrides for each shader group. Each shader group behaves like a pipeline that was created from its state as well as the remaining parent’s state. It is defined as:

// Provided by VK_NV_device_generated_commands
typedef struct VkGraphicsShaderGroupCreateInfoNV {
    VkStructureType                                 sType;
    const void*                                     pNext;
    uint32_t                                        stageCount;
    const VkPipelineShaderStageCreateInfo*          pStages;
    const VkPipelineVertexInputStateCreateInfo*     pVertexInputState;
    const VkPipelineTessellationStateCreateInfo*    pTessellationState;
} VkGraphicsShaderGroupCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
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 this shader group.
pVertexInputState is a pointer to an instance of the VkPipelineVertexInputStateCreateInfo structure.
pTessellationState is a pointer to an instance of the VkPipelineTessellationStateCreateInfo structure, and is ignored if the shader group does not include a tessellation control shader stage and tessellation evaluation shader stage.

Valid Usage

For stageCount, the same restrictions as in VkGraphicsPipelineCreateInfo::stageCount apply
For pStages, the same restrictions as in VkGraphicsPipelineCreateInfo::pStages apply
For pVertexInputState, the same restrictions as in VkGraphicsPipelineCreateInfo::pVertexInputState apply
For pTessellationState, the same restrictions as in VkGraphicsPipelineCreateInfo::pTessellationState apply

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GRAPHICS_SHADER_GROUP_CREATE_INFO_NV
pNext must be NULL
pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures
stageCount must be greater than 0

9.3. Pipeline destruction

To destroy a graphics or compute pipeline, call:

// Provided by VK_VERSION_1_0
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, VkRayTracingPipelineCreateInfoKHR, VkRayTracingPipelineCreateInfoNV, or VkComputePipelineCreateInfo structures into the vkCreateGraphicsPipelines, vkCreateRayTracingPipelinesKHR, vkCreateRayTracingPipelinesNV, 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 despite valid arguments (for example, due to out of memory errors), the VkResult code returned by vkCreate*Pipelines will indicate why. The implementation will attempt to create all pipelines, and only return VK_NULL_HANDLE values for those that actually failed.

If creation fails for a pipeline that had VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT set, pipelines at an index in the pPipelines array greater than or equal to that of the failing pipeline must be set to VK_NULL_HANDLE.

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPipelineCache)

To create pipeline cache objects, call:

// Provided by VK_VERSION_1_0
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 containing 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 vkCreateRayTracingPipelinesKHR, vkCreateRayTracingPipelinesNV, 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.

If flags of pCreateInfo includes VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT, all commands that modify the returned pipeline cache object must be externally synchronized.

Note

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineCacheCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkPipelineCacheCreateFlagBits specifying the behavior of the pipeline cache.
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
If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_CACHE_CREATE_INFO
pNext must be NULL
flags must be a valid combination of VkPipelineCacheCreateFlagBits values
If initialDataSize is not 0, pInitialData must be a valid pointer to an array of initialDataSize bytes

// Provided by VK_VERSION_1_0
typedef VkFlags VkPipelineCacheCreateFlags;

VkPipelineCacheCreateFlags is a bitmask type for setting a mask of zero or more VkPipelineCacheCreateFlagBits.

Possible values of the flags member of VkPipelineCacheCreateInfo, specifying the behavior of the pipeline cache, are:

// Provided by VK_VERSION_1_0
typedef enum VkPipelineCacheCreateFlagBits {
  // Provided by VK_EXT_pipeline_creation_cache_control
    VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT = 0x00000001,
} VkPipelineCacheCreateFlagBits;

VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT specifies that all commands that modify the created VkPipelineCache will be externally synchronized. When set, the implementation may skip any unnecessary processing needed to support simultaneous modification from multiple threads where allowed.

Pipeline cache objects can be merged using the command:

// Provided by VK_VERSION_1_0
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 a pointer to 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

Data can be retrieved from a pipeline cache object using the command:

// Provided by VK_VERSION_1_0
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 size_t 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

VK_SUCCESS
VK_INCOMPLETE

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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 VkPipelineShaderStageCreateInfo structure contains a pSpecializationInfo member, which can be NULL to indicate no specialization constants, or point to a VkSpecializationInfo structure.

The VkSpecializationInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 structures which map 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 is a pointer to a VkSpecializationMapEntry structure.

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

Valid Usage (Implicit)

If mapEntryCount is not 0, pMapEntries must be a valid pointer to an array of mapEntryCount valid VkSpecializationMapEntry structures
If dataSize is not 0, pData must be a valid pointer to an array of dataSize bytes

The VkSpecializationMapEntry structure is defined as:

// Provided by VK_VERSION_1_0
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 Libraries

A pipeline library is a special pipeline that was created using the VK_PIPELINE_CREATE_LIBRARY_BIT_KHR and cannot be bound, instead it defines a set of pipeline state which can be linked into other pipelines. For ray tracing pipelines this includes shaders and shader groups. The application must maintain the lifetime of pipeline libraries based on the pipelines that link with it. A pipeline library is considered in-use, as long as one of the linking pipelines is in-use.

This linkage is achieved by using the following structure within the appropriate creation mechanisms:

The VkPipelineLibraryCreateInfoKHR structure is defined as:

// Provided by VK_KHR_pipeline_library
typedef struct VkPipelineLibraryCreateInfoKHR {
    VkStructureType      sType;
    const void*          pNext;
    uint32_t             libraryCount;
    const VkPipeline*    pLibraries;
} VkPipelineLibraryCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
libraryCount is the number of pipeline libraries in pLibraries.
pLibraries is an array of pipeline libraries to use when creating a pipeline.

Valid Usage

Each element of pLibraries must have been created with VK_PIPELINE_CREATE_LIBRARY_BIT_KHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_LIBRARY_CREATE_INFO_KHR
pNext must be NULL
If libraryCount is not 0, pLibraries must be a valid pointer to an array of libraryCount valid VkPipeline handles

9.9. Pipeline Binding

Once a pipeline has been created, it can be bound to the command buffer using the command:

// Provided by VK_VERSION_1_0
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. The pipeline bound to VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR controls the behavior of vkCmdTraceRaysKHR. 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 uses 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
This command must not be recorded when transform feedback is active
If pipelineBindPoint is VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR, the VkCommandPool that commandBuffer was allocated from must support compute operations
If pipelineBindPoint is VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR, the pipeline must be a ray tracing pipeline
The pipeline must not have been created with VK_PIPELINE_CREATE_LIBRARY_BIT_KHR set

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

Primary
Secondary

Both

Graphics
Compute

Possible values of vkCmdBindPipeline::pipelineBindPoint, specifying the bind point of a pipeline object, are:

// Provided by VK_VERSION_1_0
typedef enum VkPipelineBindPoint {
    VK_PIPELINE_BIND_POINT_GRAPHICS = 0,
    VK_PIPELINE_BIND_POINT_COMPUTE = 1,
  // Provided by VK_KHR_ray_tracing
    VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR = 1000165000,
  // Provided by VK_NV_ray_tracing
    VK_PIPELINE_BIND_POINT_RAY_TRACING_NV = VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR,
} VkPipelineBindPoint;

VK_PIPELINE_BIND_POINT_COMPUTE specifies binding as a compute pipeline.
VK_PIPELINE_BIND_POINT_GRAPHICS specifies binding as a graphics pipeline.
VK_PIPELINE_BIND_POINT_RAY_TRACING_KHR specifies binding as a ray tracing pipeline.

For pipelines that were created with the support of multiple shader groups (see Graphics Pipeline Shader Groups), the regular vkCmdBindPipeline command will bind Shader Group 0. To explicitly bind a shader group use:

// Provided by VK_NV_device_generated_commands
void vkCmdBindPipelineShaderGroupNV(
    VkCommandBuffer                             commandBuffer,
    VkPipelineBindPoint                         pipelineBindPoint,
    VkPipeline                                  pipeline,
    uint32_t                                    groupIndex);

commandBuffer is the command buffer that the pipeline will be bound to.
pipelineBindPoint is a VkPipelineBindPoint value specifying to which bind point the pipeline is bound.
pipeline is the pipeline to be bound.
groupIndex is the shader group to be bound.

Valid Usage

groupIndex must be 0 or less than the effective VkGraphicsPipelineShaderGroupsCreateInfoNV::groupCount including the referenced pipelines
The pipelineBindPoint must be VK_PIPELINE_BIND_POINT_GRAPHICS
The same restrictions as vkCmdBindPipeline apply as if the bound pipeline was created only with the Shader Group from the groupIndex information
The VkPhysicalDeviceDeviceGeneratedCommandsFeaturesNV::deviceGeneratedCommands feature must be enabled

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

Primary
Secondary

Both

Graphics
Compute

9.10. 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. Before any draw or dispatch call with this pipeline there must not have been any call to any of the corresponding dynamic state setting commands after this pipeline was bound
If the state is specified as dynamic in the new pipeline object, then that command buffer state is not disturbed. Before any draw or dispatch call with this pipeline there must have been at least one call to each of the corresponding dynamic state setting commands since the command buffer recording was begun, or the last bound pipeline object with that state specified as static, whichever was the latter

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.

9.11. Pipeline Shader Information

When a pipeline is created, its state and shaders are compiled into zero or more device-specific executables, which are used when executing commands against that pipeline. To query the properties of these executables, call:

// Provided by VK_KHR_pipeline_executable_properties
VkResult vkGetPipelineExecutablePropertiesKHR(
    VkDevice                                    device,
    const VkPipelineInfoKHR*                    pPipelineInfo,
    uint32_t*                                   pExecutableCount,
    VkPipelineExecutablePropertiesKHR*          pProperties);

device is the device that created the pipeline.
pPipelineInfo describes the pipeline being queried.
pExecutableCount is a pointer to an integer related to the number of pipeline executables available or queried, as described below.
pProperties is either NULL or a pointer to an array of VkPipelineExecutablePropertiesKHR structures.

If pProperties is NULL, then the number of executables associated with the pipeline is returned in pExecutableCount. Otherwise, pExecutableCount 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 pExecutableCount is less than the number of executables associated with the pipeline, at most pExecutableCount structures will be written and vkGetPipelineExecutablePropertiesKHR will return VK_INCOMPLETE.

Valid Usage

pipelineExecutableInfo must be enabled
pipeline member of pPipelineInfo must have been created with device

Valid Usage (Implicit)

device must be a valid VkDevice handle
pPipelineInfo must be a valid pointer to a valid VkPipelineInfoKHR structure
pExecutableCount must be a valid pointer to a uint32_t value
If the value referenced by pExecutableCount is not 0, and pProperties is not NULL, pProperties must be a valid pointer to an array of pExecutableCount VkPipelineExecutablePropertiesKHR structures

Return Codes

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineInfoKHR structure is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef struct VkPipelineInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkPipeline         pipeline;
} VkPipelineInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
pipeline is a VkPipeline handle.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_INFO_KHR
pNext must be NULL
pipeline must be a valid VkPipeline handle

The VkPipelineExecutablePropertiesKHR structure is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef struct VkPipelineExecutablePropertiesKHR {
    VkStructureType       sType;
    void*                 pNext;
    VkShaderStageFlags    stages;
    char                  name[VK_MAX_DESCRIPTION_SIZE];
    char                  description[VK_MAX_DESCRIPTION_SIZE];
    uint32_t              subgroupSize;
} VkPipelineExecutablePropertiesKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
stages is a bitmask of VkShaderStageFlagBits indicating which shader stages (if any) were principally used as inputs to compile this pipeline executable.
name is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a short human readable name for this executable.
description is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a human readable description for this executable.
subgroupSize is the subgroup size with which this executable is dispatched.

The stages field may be zero or it may contain one or more bits describing the stages principally used to compile this pipeline. Not all implementations have a 1:1 mapping between shader stages and pipeline executables and some implementations may reduce a given shader stage to fixed function hardware programming such that no executable is available. No guarantees are provided about the mapping between shader stages and pipeline executables and stages should be considered a best effort hint. Because the application cannot rely on the stages field to provide an exact description, name and description provide a human readable name and description which more accurately describes the given pipeline executable.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_EXECUTABLE_PROPERTIES_KHR
pNext must be NULL

Each pipeline executable may have a set of statistics associated with it that are generated by the pipeline compilation process. These statistics may include things such as instruction counts, amount of spilling (if any), maximum number of simultaneous threads, or anything else which may aid developers in evaluating the expected performance of a shader. To query the compile-time statistics associated with a pipeline executable, call:

// Provided by VK_KHR_pipeline_executable_properties
VkResult vkGetPipelineExecutableStatisticsKHR(
    VkDevice                                    device,
    const VkPipelineExecutableInfoKHR*          pExecutableInfo,
    uint32_t*                                   pStatisticCount,
    VkPipelineExecutableStatisticKHR*           pStatistics);

device is the device that created the pipeline.
pExecutableInfo describes the pipeline executable being queried.
pStatisticCount is a pointer to an integer related to the number of statistics available or queried, as described below.
pStatistics is either NULL or a pointer to an array of VkPipelineExecutableStatisticKHR structures.

If pStatistics is NULL, then the number of statistics associated with the pipeline executable is returned in pStatisticCount. Otherwise, pStatisticCount must point to a variable set by the user to the number of elements in the pStatistics array, and on return the variable is overwritten with the number of structures actually written to pStatistics. If pStatisticCount is less than the number of statistics associated with the pipeline executable, at most pStatisticCount structures will be written and vkGetPipelineExecutableStatisticsKHR will return VK_INCOMPLETE.

Valid Usage

pipelineExecutableInfo must be enabled
pipeline member of pExecutableInfo must have been created with device
pipeline member of pExecutableInfo must have been created with VK_PIPELINE_CREATE_CAPTURE_STATISTICS_BIT_KHR set in the flags field of VkGraphicsPipelineCreateInfo or VkComputePipelineCreateInfo

Valid Usage (Implicit)

device must be a valid VkDevice handle
pExecutableInfo must be a valid pointer to a valid VkPipelineExecutableInfoKHR structure
pStatisticCount must be a valid pointer to a uint32_t value
If the value referenced by pStatisticCount is not 0, and pStatistics is not NULL, pStatistics must be a valid pointer to an array of pStatisticCount VkPipelineExecutableStatisticKHR structures

Return Codes

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineExecutableInfoKHR structure is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef struct VkPipelineExecutableInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    VkPipeline         pipeline;
    uint32_t           executableIndex;
} VkPipelineExecutableInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
pipeline is the pipeline to query.
executableIndex is the index of the executable to query in the array of executable properties returned by vkGetPipelineExecutablePropertiesKHR.

Valid Usage

executableIndex must be less than the number of executables associated with pipeline as returned in the pExecutableCount parameter of vkGetPipelineExecutablePropertiesKHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_EXECUTABLE_INFO_KHR
pNext must be NULL
pipeline must be a valid VkPipeline handle

The VkPipelineExecutableStatisticKHR structure is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef struct VkPipelineExecutableStatisticKHR {
    VkStructureType                           sType;
    void*                                     pNext;
    char                                      name[VK_MAX_DESCRIPTION_SIZE];
    char                                      description[VK_MAX_DESCRIPTION_SIZE];
    VkPipelineExecutableStatisticFormatKHR    format;
    VkPipelineExecutableStatisticValueKHR     value;
} VkPipelineExecutableStatisticKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
name is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a short human readable name for this statistic.
description is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a human readable description for this statistic.
format is a VkPipelineExecutableStatisticFormatKHR value specifying the format of the data found in value.
value is the value of this statistic.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_EXECUTABLE_STATISTIC_KHR
pNext must be NULL

The VkPipelineExecutableStatisticFormatKHR enum is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef enum VkPipelineExecutableStatisticFormatKHR {
    VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_BOOL32_KHR = 0,
    VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_INT64_KHR = 1,
    VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_UINT64_KHR = 2,
    VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_FLOAT64_KHR = 3,
} VkPipelineExecutableStatisticFormatKHR;

VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_BOOL32_KHR specifies that the statistic is returned as a 32-bit boolean value which must be either VK_TRUE or VK_FALSE and should be read from the b32 field of VkPipelineExecutableStatisticValueKHR.
VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_INT64_KHR specifies that the statistic is returned as a signed 64-bit integer and should be read from the i64 field of VkPipelineExecutableStatisticValueKHR.
VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_UINT64_KHR specifies that the statistic is returned as an unsigned 64-bit integer and should be read from the u64 field of VkPipelineExecutableStatisticValueKHR.
VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_FLOAT64_KHR specifies that the statistic is returned as a 64-bit floating-point value and should be read from the f64 field of VkPipelineExecutableStatisticValueKHR.

The VkPipelineExecutableStatisticValueKHR union is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef union VkPipelineExecutableStatisticValueKHR {
    VkBool32    b32;
    int64_t     i64;
    uint64_t    u64;
    double      f64;
} VkPipelineExecutableStatisticValueKHR;

b32 is the 32-bit boolean value if the VkPipelineExecutableStatisticFormatKHR is VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_BOOL32_KHR.
i64 is the signed 64-bit integer value if the VkPipelineExecutableStatisticFormatKHR is VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_INT64_KHR.
u64 is the unsigned 64-bit integer value if the VkPipelineExecutableStatisticFormatKHR is VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_UINT64_KHR.
f64 is the 64-bit floating-point value if the VkPipelineExecutableStatisticFormatKHR is VK_PIPELINE_EXECUTABLE_STATISTIC_FORMAT_FLOAT64_KHR.

Each pipeline executable may have one or more text or binary internal representations associated with it which are generated as part of the compile process. These may include the final shader assembly, a binary form of the compiled shader, or the shader compiler’s internal representation at any number of intermediate compile steps. To query the internal representations associated with a pipeline executable, call:

// Provided by VK_KHR_pipeline_executable_properties
VkResult vkGetPipelineExecutableInternalRepresentationsKHR(
    VkDevice                                    device,
    const VkPipelineExecutableInfoKHR*          pExecutableInfo,
    uint32_t*                                   pInternalRepresentationCount,
    VkPipelineExecutableInternalRepresentationKHR* pInternalRepresentations);

device is the device that created the pipeline.
pExecutableInfo describes the pipeline executable being queried.
pInternalRepresentationCount is a pointer to an integer related to the number of internal representations available or queried, as described below.
pInternalRepresentations is either NULL or a pointer to an array of VkPipelineExecutableInternalRepresentationKHR structures.

If pInternalRepresentations is NULL, then the number of internal representations associated with the pipeline executable is returned in pInternalRepresentationCount. Otherwise, pInternalRepresentationCount must point to a variable set by the user to the number of elements in the pInternalRepresentations array, and on return the variable is overwritten with the number of structures actually written to pInternalRepresentations. If pInternalRepresentationCount is less than the number of internal representations associated with the pipeline executable, at most pInternalRepresentationCount structures will be written and vkGetPipelineExecutableInternalRepresentationsKHR will return VK_INCOMPLETE.

While the details of the internal representations remain implementation dependent, the implementation should order the internal representations in the order in which they occur in the compile pipeline with the final shader assembly (if any) last.

Valid Usage

pipelineExecutableInfo must be enabled
pipeline member of pExecutableInfo must have been created with device
pipeline member of pExecutableInfo must have been created with VK_PIPELINE_CREATE_CAPTURE_INTERNAL_REPRESENTATIONS_BIT_KHR set in the flags field of VkGraphicsPipelineCreateInfo or VkComputePipelineCreateInfo

Valid Usage (Implicit)

device must be a valid VkDevice handle
pExecutableInfo must be a valid pointer to a valid VkPipelineExecutableInfoKHR structure
pInternalRepresentationCount must be a valid pointer to a uint32_t value
If the value referenced by pInternalRepresentationCount is not 0, and pInternalRepresentations is not NULL, pInternalRepresentations must be a valid pointer to an array of pInternalRepresentationCount VkPipelineExecutableInternalRepresentationKHR structures

Return Codes

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineExecutableInternalRepresentationKHR structure is defined as:

// Provided by VK_KHR_pipeline_executable_properties
typedef struct VkPipelineExecutableInternalRepresentationKHR {
    VkStructureType    sType;
    void*              pNext;
    char               name[VK_MAX_DESCRIPTION_SIZE];
    char               description[VK_MAX_DESCRIPTION_SIZE];
    VkBool32           isText;
    size_t             dataSize;
    void*              pData;
} VkPipelineExecutableInternalRepresentationKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
name is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a short human readable name for this internal representation.
description is an array of VK_MAX_DESCRIPTION_SIZE char containing a null-terminated UTF-8 string which is a human readable description for this internal representation.
isText specifies whether the returned data is text or opaque data. If isText is VK_TRUE then the data returned in pData is text and is guaranteed to be a null-terminated UTF-8 string.
dataSize is an integer related to the size, in bytes, of the internal representation data, as described below.
pData is either NULL or a pointer to an block of data into which the implementation will write the textual form of the internal representation.

If pData is NULL, then the size, in bytes, of the internal representation data is returned in dataSize. Otherwise, dataSize must be the size of the buffer, in bytes, pointed to by pData and on return dataSize is overwritten with the number of bytes of data actually written to pData including any trailing null character. If dataSize is less than the size, in bytes, of the internal representation data, at most dataSize bytes of data will be written to pData and vkGetPipelineExecutableInternalRepresentationsKHR will return VK_INCOMPLETE. If isText is VK_TRUE and pData is not NULL and dataSize is not zero, the last byte written to pData will be a null character.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_EXECUTABLE_INTERNAL_REPRESENTATION_KHR
pNext must be NULL

Information about a particular shader that has been compiled as part of a pipeline object can be extracted by calling:

// Provided by VK_AMD_shader_info
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 infoType VK_SHADER_INFO_TYPE_STATISTICS_AMD, a VkShaderStatisticsInfoAMD structure 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 infoType VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD, pInfo is a pointer 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 infoType 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

VK_SUCCESS
VK_INCOMPLETE

VK_ERROR_FEATURE_NOT_PRESENT
VK_ERROR_OUT_OF_HOST_MEMORY

Possible values of vkGetShaderInfoAMD::infoType, specifying the information being queried from a shader, are:

// Provided by VK_AMD_shader_info
typedef enum VkShaderInfoTypeAMD {
    VK_SHADER_INFO_TYPE_STATISTICS_AMD = 0,
    VK_SHADER_INFO_TYPE_BINARY_AMD = 1,
    VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD = 2,
} VkShaderInfoTypeAMD;

VK_SHADER_INFO_TYPE_STATISTICS_AMD specifies that device resources used by a shader will be queried.
VK_SHADER_INFO_TYPE_BINARY_AMD specifies that implementation-specific information will be queried.
VK_SHADER_INFO_TYPE_DISASSEMBLY_AMD specifies that human-readable dissassembly of a shader.

The VkShaderStatisticsInfoAMD structure is defined as:

// Provided by VK_AMD_shader_info
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 a VkShaderResourceUsageAMD structure 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:

// Provided by VK_AMD_shader_info
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.

9.12. Pipeline Compiler Control

The compilation of a pipeline can be tuned by adding a VkPipelineCompilerControlCreateInfoAMD structure to the pNext chain of VkGraphicsPipelineCreateInfo or VkComputePipelineCreateInfo.

// Provided by VK_AMD_pipeline_compiler_control
typedef struct VkPipelineCompilerControlCreateInfoAMD {
    VkStructureType                      sType;
    const void*                          pNext;
    VkPipelineCompilerControlFlagsAMD    compilerControlFlags;
} VkPipelineCompilerControlCreateInfoAMD;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
compilerControlFlags is a bitmask of VkPipelineCompilerControlFlagBitsAMD affecting how the pipeline will be compiled.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_COMPILER_CONTROL_CREATE_INFO_AMD
compilerControlFlags must be 0

There are currently no available flags for this extension; flags will be added by future versions of this extension.

// Provided by VK_AMD_pipeline_compiler_control
typedef enum VkPipelineCompilerControlFlagBitsAMD {
} VkPipelineCompilerControlFlagBitsAMD;

// Provided by VK_AMD_pipeline_compiler_control
typedef VkFlags VkPipelineCompilerControlFlagsAMD;

VkPipelineCompilerControlFlagsAMD is a bitmask type for setting a mask of zero or more VkPipelineCompilerControlFlagBitsAMD.

9.13. Ray Tracing Pipeline

Ray tracing pipelines consist of multiple shader stages, fixed-function traversal stages, and a pipeline layout.

To create ray tracing pipelines, call:

// Provided by VK_NV_ray_tracing
VkResult vkCreateRayTracingPipelinesNV(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    uint32_t                                    createInfoCount,
    const VkRayTracingPipelineCreateInfoNV*     pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkPipeline*                                 pPipelines);

device is the logical device that creates the ray tracing 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 a pointer to an array of VkRayTracingPipelineCreateInfoNV structures.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pPipelines is a pointer to an array in which the resulting ray tracing pipeline objects are returned.

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
If pipelineCache was created with VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT, host access to pipelineCache must be externally synchronized

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 VkRayTracingPipelineCreateInfoNV 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

VK_SUCCESS
VK_PIPELINE_COMPILE_REQUIRED_EXT

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_SHADER_NV

To create ray tracing pipelines, call:

// Provided by VK_KHR_ray_tracing
VkResult vkCreateRayTracingPipelinesKHR(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    uint32_t                                    createInfoCount,
    const VkRayTracingPipelineCreateInfoKHR*    pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkPipeline*                                 pPipelines);

device is the logical device that creates the ray tracing 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 a pointer to an array of VkRayTracingPipelineCreateInfoKHR structures.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pPipelines is a pointer to an array in which the resulting ray tracing pipeline objects are returned.

The VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS error is returned if the implementation is unable to re-use the shader group handles provided in VkRayTracingShaderGroupCreateInfoKHR::pShaderGroupCaptureReplayHandle when VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingShaderGroupHandleCaptureReplay is enabled.

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
If pipelineCache was created with VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT_EXT, host access to pipelineCache must be externally synchronized
The rayTracing feature must be enabled

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 VkRayTracingPipelineCreateInfoKHR 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

VK_SUCCESS
VK_OPERATION_DEFERRED_KHR
VK_OPERATION_NOT_DEFERRED_KHR
VK_PIPELINE_COMPILE_REQUIRED_EXT

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS

The VkRayTracingPipelineCreateInfoNV structure is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkRayTracingPipelineCreateInfoNV {
    VkStructureType                               sType;
    const void*                                   pNext;
    VkPipelineCreateFlags                         flags;
    uint32_t                                      stageCount;
    const VkPipelineShaderStageCreateInfo*        pStages;
    uint32_t                                      groupCount;
    const VkRayTracingShaderGroupCreateInfoNV*    pGroups;
    uint32_t                                      maxRecursionDepth;
    VkPipelineLayout                              layout;
    VkPipeline                                    basePipelineHandle;
    int32_t                                       basePipelineIndex;
} VkRayTracingPipelineCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 ray tracing pipeline.
groupCount is the number of entries in the pGroups array.
pGroups is an array of size groupCount structures of type VkRayTracingShaderGroupCreateInfoNV describing the set of the shader stages to be included in each shader group in the ray tracing pipeline.
maxRecursionDepth is the maximum recursion depth of shaders executed by this pipeline.
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.

Valid Usage

If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid handle to a ray tracing 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 at least one element of pStages must be VK_SHADER_STAGE_RAYGEN_BIT_KHR
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
layout must be consistent with all shaders specified in pStages
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
flags must not include VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV
If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT or VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT
flags must not include VK_PIPELINE_CREATE_LIBRARY_BIT_KHR
maxRecursionDepth must be less than or equal to VkPhysicalDeviceRayTracingPropertiesNV::maxRecursionDepth
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR
flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR
flags must not include both VK_PIPELINE_CREATE_DEFER_COMPILE_BIT_NV and VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT at the same time

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RAY_TRACING_PIPELINE_CREATE_INFO_NV
pNext must be NULL or a pointer to a valid instance of VkPipelineCreationFeedbackCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkPipelineCreateFlagBits values
pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures
pGroups must be a valid pointer to an array of groupCount valid VkRayTracingShaderGroupCreateInfoNV structures
layout must be a valid VkPipelineLayout handle
stageCount must be greater than 0
groupCount must be greater than 0
Both of basePipelineHandle, and layout that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

The VkRayTracingPipelineCreateInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkRayTracingPipelineCreateInfoKHR {
    VkStructureType                                      sType;
    const void*                                          pNext;
    VkPipelineCreateFlags                                flags;
    uint32_t                                             stageCount;
    const VkPipelineShaderStageCreateInfo*               pStages;
    uint32_t                                             groupCount;
    const VkRayTracingShaderGroupCreateInfoKHR*          pGroups;
    uint32_t                                             maxRecursionDepth;
    VkPipelineLibraryCreateInfoKHR                       libraries;
    const VkRayTracingPipelineInterfaceCreateInfoKHR*    pLibraryInterface;
    VkPipelineLayout                                     layout;
    VkPipeline                                           basePipelineHandle;
    int32_t                                              basePipelineIndex;
} VkRayTracingPipelineCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 a pointer to an array of stageCount VkPipelineShaderStageCreateInfo structures describing the set of the shader stages to be included in the ray tracing pipeline.
groupCount is the number of entries in the pGroups array.
pGroups is a pointer to an array of groupCount VkRayTracingShaderGroupCreateInfoKHR structures describing the set of the shader stages to be included in each shader group in the ray tracing pipeline.
maxRecursionDepth is the maximum recursion depth of shaders executed by this pipeline.
libraries is a VkPipelineLibraryCreateInfoKHR structure defining pipeline libraries to include.
pLibraryInterface is a pointer to a VkRayTracingPipelineInterfaceCreateInfoKHR structure defining additional information when using pipeline libraries.
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.

When VK_PIPELINE_CREATE_LIBRARY_BIT_KHR is specified, this pipeline defines a pipeline library which cannot be bound as a ray tracing pipeline directly. Instead, pipeline libraries define common shaders and shader groups which can be included in future pipeline creation.

If pipeline libraries are included in libraries, shaders defined in those libraries are treated as if they were defined as additional entries in pStages, appended in the order they appear in the pLibraries array and in the pStages array when those libraries were defined.

When referencing shader groups in order to obtain a shader group handle, groups defined in those libraries are treated as if they were defined as additional entries in pGroups, appended in the order they appear in the pLibraries array and in the pGroups array when those libraries were defined. The shaders these groups reference are set when the pipeline library is created, referencing those specified in the pipeline library, not in the pipeline that includes it.

If the VkDeferredOperationInfoKHR structure is included in the pNext chain of VkRayTracingPipelineCreateInfoKHR, the operation of this pipeline creation is deferred, as defined in the Deferred Host Operations chapter.

Valid Usage

If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid handle to a ray tracing 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 at least one element of pStages must be VK_SHADER_STAGE_RAYGEN_BIT_KHR
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
layout must be consistent with all shaders specified in pStages
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
flags must not include VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV
If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT or VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT_EXT
maxRecursionDepth must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxRecursionDepth
If flags includes VK_PIPELINE_CREATE_LIBRARY_BIT_KHR, pLibraryInterface must not be NULL
If the libraryCount member of libraries is greater than 0, pLibraryInterface must not be NULL
Each element of the pLibraries member of libraries must have been created with the value of maxRecursionDepth equal to that in this pipeline
Each element of the pLibraries member of libraries must have been created with a layout that is compatible with the layout in this pipeline
Each element of the pLibraries member of libraries must have been created with values of the maxPayloadSize, maxAttributeSize, and maxCallableSize members of pLibraryInterface equal to those in this pipeline
If flags includes VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR, for any element of pGroups with a type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR, the anyHitShader of that element must not be VK_SHADER_UNUSED_KHR
If flags includes VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR, for any element of pGroups with a type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR, the closestHitShader of that element must not be VK_SHADER_UNUSED_KHR
If the rayTracingPrimitiveCulling feature is not enabled, flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR
If the rayTracingPrimitiveCulling feature is not enabled, flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR
If libraries.libraryCount is zero, then stageCount must not be zero
If libraries.libraryCount is zero, then groupCount must not be zero

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RAY_TRACING_PIPELINE_CREATE_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 VkDeferredOperationInfoKHR or VkPipelineCreationFeedbackCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkPipelineCreateFlagBits values
If stageCount is not 0, pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures
If groupCount is not 0, pGroups must be a valid pointer to an array of groupCount valid VkRayTracingShaderGroupCreateInfoKHR structures
libraries must be a valid VkPipelineLibraryCreateInfoKHR structure
If pLibraryInterface is not NULL, pLibraryInterface must be a valid pointer to a valid VkRayTracingPipelineInterfaceCreateInfoKHR structure
layout must be a valid VkPipelineLayout handle
Both of basePipelineHandle, and layout that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

The VkRayTracingShaderGroupCreateInfoNV structure is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkRayTracingShaderGroupCreateInfoNV {
    VkStructureType                   sType;
    const void*                       pNext;
    VkRayTracingShaderGroupTypeKHR    type;
    uint32_t                          generalShader;
    uint32_t                          closestHitShader;
    uint32_t                          anyHitShader;
    uint32_t                          intersectionShader;
} VkRayTracingShaderGroupCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
type is the type of hit group specified in this structure.
generalShader is the index of the ray generation, miss, or callable shader from VkRayTracingPipelineCreateInfoNV::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_NV, and VK_SHADER_UNUSED_NV otherwise.
closestHitShader is the optional index of the closest hit shader from VkRayTracingPipelineCreateInfoNV::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_NV or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_NV, and VK_SHADER_UNUSED_NV otherwise.
anyHitShader is the optional index of the any-hit shader from VkRayTracingPipelineCreateInfoNV::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_NV or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_NV, and VK_SHADER_UNUSED_NV otherwise.
intersectionShader is the index of the intersection shader from VkRayTracingPipelineCreateInfoNV::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_NV, and VK_SHADER_UNUSED_NV otherwise.

Valid Usage

If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_NV then generalShader must be a valid index into VkRayTracingPipelineCreateInfoNV::pStages referring to a shader of VK_SHADER_STAGE_RAYGEN_BIT_NV, VK_SHADER_STAGE_MISS_BIT_NV, or VK_SHADER_STAGE_CALLABLE_BIT_NV
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_NV then closestHitShader, anyHitShader, and intersectionShader must be VK_SHADER_UNUSED_NV
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_NV then intersectionShader must be a valid index into VkRayTracingPipelineCreateInfoNV::pStages referring to a shader of VK_SHADER_STAGE_INTERSECTION_BIT_NV
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_NV then intersectionShader must be VK_SHADER_UNUSED_NV
closestHitShader must be either VK_SHADER_UNUSED_NV or a valid index into VkRayTracingPipelineCreateInfoNV::pStages referring to a shader of VK_SHADER_STAGE_CLOSEST_HIT_BIT_NV
anyHitShader must be either VK_SHADER_UNUSED_NV or a valid index into VkRayTracingPipelineCreateInfoNV::pStages referring to a shader of VK_SHADER_STAGE_ANY_HIT_BIT_NV

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RAY_TRACING_SHADER_GROUP_CREATE_INFO_NV
pNext must be NULL
type must be a valid VkRayTracingShaderGroupTypeKHR value

The VkRayTracingShaderGroupCreateInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkRayTracingShaderGroupCreateInfoKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkRayTracingShaderGroupTypeKHR    type;
    uint32_t                          generalShader;
    uint32_t                          closestHitShader;
    uint32_t                          anyHitShader;
    uint32_t                          intersectionShader;
    const void*                       pShaderGroupCaptureReplayHandle;
} VkRayTracingShaderGroupCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
type is the type of hit group specified in this structure.
generalShader is the index of the ray generation, miss, or callable shader from VkRayTracingPipelineCreateInfoKHR::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR, and VK_SHADER_UNUSED_KHR otherwise.
closestHitShader is the optional index of the closest hit shader from VkRayTracingPipelineCreateInfoKHR::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR, and VK_SHADER_UNUSED_KHR otherwise.
anyHitShader is the optional index of the any-hit shader from VkRayTracingPipelineCreateInfoKHR::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR or VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR, and VK_SHADER_UNUSED_KHR otherwise.
intersectionShader is the index of the intersection shader from VkRayTracingPipelineCreateInfoKHR::pStages in the group if the shader group has type of VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR, and VK_SHADER_UNUSED_KHR otherwise.
pShaderGroupCaptureReplayHandle is an optional pointer to replay information for this shader group. Ignored if VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingShaderGroupHandleCaptureReplay is VK_FALSE.

Valid Usage

If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR then generalShader must be a valid index into VkRayTracingPipelineCreateInfoKHR::pStages referring to a shader of VK_SHADER_STAGE_RAYGEN_BIT_KHR, VK_SHADER_STAGE_MISS_BIT_KHR, or VK_SHADER_STAGE_CALLABLE_BIT_KHR
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR then closestHitShader, anyHitShader, and intersectionShader must be VK_SHADER_UNUSED_KHR
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR then intersectionShader must be a valid index into VkRayTracingPipelineCreateInfoKHR::pStages referring to a shader of VK_SHADER_STAGE_INTERSECTION_BIT_KHR
If type is VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR then intersectionShader must be VK_SHADER_UNUSED_KHR
closestHitShader must be either VK_SHADER_UNUSED_KHR or a valid index into VkRayTracingPipelineCreateInfoKHR::pStages referring to a shader of VK_SHADER_STAGE_CLOSEST_HIT_BIT_KHR
anyHitShader must be either VK_SHADER_UNUSED_KHR or a valid index into VkRayTracingPipelineCreateInfoKHR::pStages referring to a shader of VK_SHADER_STAGE_ANY_HIT_BIT_KHR
If VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingShaderGroupHandleCaptureReplayMixed is VK_FALSE then pShaderGroupCaptureReplayHandle must not be provided if it has not been provided on a previous call to ray tracing pipeline creation
If VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingShaderGroupHandleCaptureReplayMixed is VK_FALSE then the caller must guarantee that no ray tracing pipeline creation commands with pShaderGroupCaptureReplayHandle provided execute simultaneously with ray tracing pipeline creation commands without pShaderGroupCaptureReplayHandle provided

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RAY_TRACING_SHADER_GROUP_CREATE_INFO_KHR
pNext must be NULL
type must be a valid VkRayTracingShaderGroupTypeKHR value

Possible values of type in VkRayTracingShaderGroupCreateInfoKHR are:

// Provided by VK_KHR_ray_tracing
typedef enum VkRayTracingShaderGroupTypeKHR {
    VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR = 0,
    VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR = 1,
    VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR = 2,
  // Provided by VK_NV_ray_tracing
    VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_NV = VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR,
  // Provided by VK_NV_ray_tracing
    VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_NV = VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR,
  // Provided by VK_NV_ray_tracing
    VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_NV = VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR,
} VkRayTracingShaderGroupTypeKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkRayTracingShaderGroupTypeKHR VkRayTracingShaderGroupTypeNV;

VK_RAY_TRACING_SHADER_GROUP_TYPE_GENERAL_KHR indicates a shader group with a single VK_SHADER_STAGE_RAYGEN_BIT_KHR, VK_SHADER_STAGE_MISS_BIT_KHR, or VK_SHADER_STAGE_CALLABLE_BIT_KHR shader in it.
VK_RAY_TRACING_SHADER_GROUP_TYPE_TRIANGLES_HIT_GROUP_KHR specifies a shader group that only hits triangles and must not contain an intersection shader, only closest hit and any-hit shaders.
VK_RAY_TRACING_SHADER_GROUP_TYPE_PROCEDURAL_HIT_GROUP_KHR specifies a shader group that only intersects with custom geometry and must contain an intersection shader and may contain closest hit and any-hit shaders.

Note

For current group types, the hit group type could be inferred from the presence or absence of the intersection shader, but we provide the type explicitly for future hit groups that do not have that property.

The VkRayTracingPipelineInterfaceCreateInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkRayTracingPipelineInterfaceCreateInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           maxPayloadSize;
    uint32_t           maxAttributeSize;
    uint32_t           maxCallableSize;
} VkRayTracingPipelineInterfaceCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
maxPayloadSize is the maximum payload size in bytes used by any shader in the pipeline.
maxAttributeSize is the maximum attribute structure size in bytes used by any shader in the pipeline.
maxCallableSize is the maximum callable data size in bytes used by any shader in the pipeline.

maxPayloadSize is calculated as the maximum number of bytes used by any block declared in the RayPayloadKHR or IncomingRayPayloadKHR storage classes. maxAttributeSize is calculated as the maximum number of bytes used by any block declared in the HitAttributeKHR storage class. maxCallableSize is calculated as the maximum number of bytes used by any block declred in the CallableDataKHR or IncomingCallableDataKHR. As variables in these storage classes do not have explicit offsets, the size should be calculated as if each variable has a scalar alignment equal to the largest scalar alignment of any of the block’s members.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_RAY_TRACING_PIPELINE_INTERFACE_CREATE_INFO_KHR
pNext must be NULL

To query the opaque handles of shaders in the ray tracing pipeline, call:

// Provided by VK_KHR_ray_tracing
VkResult vkGetRayTracingShaderGroupHandlesKHR(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    uint32_t                                    firstGroup,
    uint32_t                                    groupCount,
    size_t                                      dataSize,
    void*                                       pData);

or the equivalent command

// Provided by VK_NV_ray_tracing
VkResult vkGetRayTracingShaderGroupHandlesNV(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    uint32_t                                    firstGroup,
    uint32_t                                    groupCount,
    size_t                                      dataSize,
    void*                                       pData);

device is the logical device containing the ray tracing pipeline.
pipeline is the ray tracing pipeline object containing the shaders.
firstGroup is the index of the first group to retrieve a handle for from the VkRayTracingPipelineCreateInfoKHR::pGroups or VkRayTracingPipelineCreateInfoNV::pGroups array.
groupCount is the number of shader handles to retrieve.
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.

Valid Usage

firstGroup must be less than the number of shader groups in pipeline
The sum of firstGroup and groupCount must be less than or equal to the number of shader groups in pipeline
dataSize must be at least VkPhysicalDeviceRayTracingPropertiesKHR::shaderGroupHandleSize × groupCount
pipeline must have not been created with VK_PIPELINE_CREATE_LIBRARY_BIT_KHR

Valid Usage (Implicit)

device must be a valid VkDevice handle
pipeline must be a valid VkPipeline handle
pData must be a valid pointer to an array of dataSize bytes
dataSize must be greater than 0
pipeline must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

To query the optional capture handle information of shaders in the ray tracing pipeline, call:

// Provided by VK_KHR_ray_tracing
VkResult vkGetRayTracingCaptureReplayShaderGroupHandlesKHR(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    uint32_t                                    firstGroup,
    uint32_t                                    groupCount,
    size_t                                      dataSize,
    void*                                       pData);

device is the logical device containing the ray tracing pipeline.
pipeline is the ray tracing pipeline object containing the shaders.
firstGroup is the index of the first group to retrieve a handle for from the VkRayTracingPipelineCreateInfoKHR::pGroups array.
groupCount is the number of shader handles to retrieve.
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.

Valid Usage

firstGroup must be less than the number of shader groups in pipeline
The sum of firstGroup and groupCount must be less than or equal to the number of shader groups in pipeline
dataSize must be at least VkPhysicalDeviceRayTracingPropertiesKHR::shaderGroupHandleCaptureReplaySize × groupCount
VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingShaderGroupHandleCaptureReplay must be enabled to call this function

Valid Usage (Implicit)

device must be a valid VkDevice handle
pipeline must be a valid VkPipeline handle
pData must be a valid pointer to an array of dataSize bytes
dataSize must be greater than 0
pipeline must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

Ray tracing pipelines can contain more shaders than a graphics or compute pipeline, so to allow parallel compilation of shaders within a pipeline, an application can choose to defer compilation until a later point in time.

To compile a deferred shader in a pipeline call:

// Provided by VK_NV_ray_tracing
VkResult vkCompileDeferredNV(
    VkDevice                                    device,
    VkPipeline                                  pipeline,
    uint32_t                                    shader);

device is the logical device containing the ray tracing pipeline.
pipeline is the ray tracing pipeline object containing the shaders.
shader is the index of the shader to compile.

Valid Usage

pipeline must have been created with VK_PIPELINE_CREATE_DEFER_COMPILE_BIT_NV
shader must not have been called as a deferred compile before

Valid Usage (Implicit)

device must be a valid VkDevice handle
pipeline must be a valid VkPipeline handle
pipeline must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

9.14. Pipeline Creation Feedback

Feedback about the creation of a particular pipeline object can be obtained by adding a VkPipelineCreationFeedbackCreateInfoEXT structure to the pNext chain of VkGraphicsPipelineCreateInfo, VkRayTracingPipelineCreateInfoKHR, VkRayTracingPipelineCreateInfoNV, or VkComputePipelineCreateInfo. The VkPipelineCreationFeedbackCreateInfoEXT structure is defined as:

// Provided by VK_EXT_pipeline_creation_feedback
typedef struct VkPipelineCreationFeedbackCreateInfoEXT {
    VkStructureType                   sType;
    const void*                       pNext;
    VkPipelineCreationFeedbackEXT*    pPipelineCreationFeedback;
    uint32_t                          pipelineStageCreationFeedbackCount;
    VkPipelineCreationFeedbackEXT*    pPipelineStageCreationFeedbacks;
} VkPipelineCreationFeedbackCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
pPipelineCreationFeedback is a pointer to a VkPipelineCreationFeedbackEXT structure.
pipelineStageCreationFeedbackCount is the number of elements in pPipelineStageCreationFeedbacks.
pPipelineStageCreationFeedbacks is a pointer to an array of pipelineStageCreationFeedbackCount VkPipelineCreationFeedbackEXT structures.

An implementation should write pipeline creation feedback to pPipelineCreationFeedback and may write pipeline stage creation feedback to pPipelineStageCreationFeedbacks. An implementation must set or clear the VK_PIPELINE_CREATION_FEEDBACK_VALID_BIT_EXT in VkPipelineCreationFeedbackEXT::flags for pPipelineCreationFeedback and every element of pPipelineStageCreationFeedbacks.

Note

One common scenario for an implementation to skip per-stage feedback is when VK_PIPELINE_CREATION_FEEDBACK_APPLICATION_PIPELINE_CACHE_HIT_BIT_EXT is set in pPipelineCreationFeedback.

When chained to VkRayTracingPipelineCreateInfoKHR, VkRayTracingPipelineCreateInfoNV, or VkGraphicsPipelineCreateInfo, the i element of pPipelineStageCreationFeedbacks corresponds to the i element of VkRayTracingPipelineCreateInfoKHR::pStages, VkRayTracingPipelineCreateInfoNV::pStages, or VkGraphicsPipelineCreateInfo::pStages. When chained to VkComputePipelineCreateInfo, the first element of pPipelineStageCreationFeedbacks corresponds to VkComputePipelineCreateInfo::stage.

Valid Usage

When chained to VkGraphicsPipelineCreateInfo, VkPipelineCreationFeedbackEXT::pipelineStageCreationFeedbackCount must equal VkGraphicsPipelineCreateInfo::stageCount
When chained to VkComputePipelineCreateInfo, VkPipelineCreationFeedbackEXT::pipelineStageCreationFeedbackCount must equal 1
When chained to VkRayTracingPipelineCreateInfoKHR, VkPipelineCreationFeedbackEXT::pipelineStageCreationFeedbackCount must equal VkRayTracingPipelineCreateInfoKHR::stageCount
When chained to VkRayTracingPipelineCreateInfoNV, VkPipelineCreationFeedbackEXT::pipelineStageCreationFeedbackCount must equal VkRayTracingPipelineCreateInfoNV::stageCount

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PIPELINE_CREATION_FEEDBACK_CREATE_INFO_EXT
pPipelineCreationFeedback must be a valid pointer to a VkPipelineCreationFeedbackEXT structure
pPipelineStageCreationFeedbacks must be a valid pointer to an array of pipelineStageCreationFeedbackCount VkPipelineCreationFeedbackEXT structures
pipelineStageCreationFeedbackCount must be greater than 0

The VkPipelineCreationFeedbackEXT structure is defined as:

// Provided by VK_EXT_pipeline_creation_feedback
typedef struct VkPipelineCreationFeedbackEXT {
    VkPipelineCreationFeedbackFlagsEXT    flags;
    uint64_t                              duration;
} VkPipelineCreationFeedbackEXT;

flags is a bitmask of VkPipelineCreationFeedbackFlagBitsEXT providing feedback about the creation of a pipeline or of a pipeline stage.
duration is the duration spent creating a pipeline or pipeline stage in nanoseconds.

If the VK_PIPELINE_CREATION_FEEDBACK_VALID_BIT_EXT is not set in flags, an implementation must not set any other bits in flags, and the values of all other VkPipelineCreationFeedbackEXT data members are undefined.

Possible values of the flags member of VkPipelineCreationFeedbackEXT are:

// Provided by VK_EXT_pipeline_creation_feedback
typedef enum VkPipelineCreationFeedbackFlagBitsEXT {
    VK_PIPELINE_CREATION_FEEDBACK_VALID_BIT_EXT = 0x00000001,
    VK_PIPELINE_CREATION_FEEDBACK_APPLICATION_PIPELINE_CACHE_HIT_BIT_EXT = 0x00000002,
    VK_PIPELINE_CREATION_FEEDBACK_BASE_PIPELINE_ACCELERATION_BIT_EXT = 0x00000004,
} VkPipelineCreationFeedbackFlagBitsEXT;

VK_PIPELINE_CREATION_FEEDBACK_VALID_BIT_EXT indicates that the feedback information is valid.

VK_PIPELINE_CREATION_FEEDBACK_APPLICATION_PIPELINE_CACHE_HIT_BIT_EXT indicates that a readily usable pipeline or pipeline stage was found in the pipelineCache specified by the application in the pipeline creation command.

An implementation should set the VK_PIPELINE_CREATION_FEEDBACK_APPLICATION_PIPELINE_CACHE_HIT_BIT_EXT bit if it was able to avoid the large majority of pipeline or pipeline stage creation work by using the pipelineCache parameter of vkCreateGraphicsPipelines, vkCreateRayTracingPipelinesKHR, vkCreateRayTracingPipelinesNV, or vkCreateComputePipelines. When an implementation sets this bit for the entire pipeline, it may leave it unset for any stage.

Note

Implementations are encouraged to provide a meaningful signal to applications using this bit. The intention is to communicate to the application that the pipeline or pipeline stage was created "as fast as it gets" using the pipeline cache provided by the application. If an implementation uses an internal cache, it is discouraged from setting this bit as the feedback would be unactionable.

VK_PIPELINE_CREATION_FEEDBACK_BASE_PIPELINE_ACCELERATION_BIT_EXT indicates that the base pipeline specified by the basePipelineHandle or basePipelineIndex member of the Vk*PipelineCreateInfo structure was used to accelerate the creation of the pipeline.

An implementation should set the VK_PIPELINE_CREATION_FEEDBACK_BASE_PIPELINE_ACCELERATION_BIT_EXT bit if it was able to avoid a significant amount of work by using the base pipeline.

Note

While "significant amount of work" is subjective, implementations are encouraged to provide a meaningful signal to applications using this bit. For example, a 1% reduction in duration may not warrant setting this bit, while a 50% reduction would.

// Provided by VK_EXT_pipeline_creation_feedback
typedef VkFlags VkPipelineCreationFeedbackFlagsEXT;

VkPipelineCreationFeedbackFlagsEXT is a bitmask type for providing zero or more VkPipelineCreationFeedbackFlagBitsEXT.

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:

// Provided by VK_VERSION_1_0
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 PFN_vkAllocationFunction pointer to an application-defined memory allocation function.
pfnReallocation is a PFN_vkReallocationFunction pointer to an application-defined memory reallocation function.
pfnFree is a PFN_vkFreeFunction pointer to an application-defined memory free function.
pfnInternalAllocation is a PFN_vkInternalAllocationNotification pointer to an application-defined function that is called by the implementation when the implementation makes internal allocations.
pfnInternalFree is a PFN_vkInternalFreeNotification pointer to an application-defined function that is called by the implementation when the implementation frees internal allocations.

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:

// Provided by VK_VERSION_1_0
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 runtime 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:

// Provided by VK_VERSION_1_0
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 a compatible 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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 defining 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:

// Provided by VK_VERSION_1_0
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 a VkValidationCacheEXT or VkPipelineCache object, 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:

// Provided by VK_VERSION_1_0
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 a 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 a 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, or pfnReallocation with zero size, 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:

// Provided by VK_VERSION_1_0
void vkGetPhysicalDeviceMemoryProperties(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties*           pMemoryProperties);

physicalDevice is the handle to the device to query.
pMemoryProperties is a pointer to a 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:

// Provided by VK_VERSION_1_0
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 VK_MAX_MEMORY_TYPES 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 VK_MAX_MEMORY_HEAPS 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
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_CACHED_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD
VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD
VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD
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_COHERENT_BIT_AMD
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD |
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_CACHED_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD |
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD
VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD |
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD
VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT |
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT |
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD |
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD
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_COHERENT_BIT_AMD |
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD

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. If the deviceCoherentMemory feature is enabled, there must be at least one memory type with the VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD 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:

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) ; or
the propertyFlags members of Y includes VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD or VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD and X does not

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 a valid 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 a valid order.

There may be a performance penalty for using device coherent or uncached device memory types, and using these accidentally is undesirable. In order to avoid this, memory types with these properties always appear at the end of the list; but are subject to the same rules otherwise.

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:

// Provided by VK_VERSION_1_1
void vkGetPhysicalDeviceMemoryProperties2(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties2*          pMemoryProperties);

or the equivalent command

// Provided by VK_KHR_get_physical_device_properties2
void vkGetPhysicalDeviceMemoryProperties2KHR(
    VkPhysicalDevice                            physicalDevice,
    VkPhysicalDeviceMemoryProperties2*          pMemoryProperties);

physicalDevice is the handle to the device to query.
pMemoryProperties is a pointer to a 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:

// Provided by VK_VERSION_1_1
typedef struct VkPhysicalDeviceMemoryProperties2 {
    VkStructureType                     sType;
    void*                               pNext;
    VkPhysicalDeviceMemoryProperties    memoryProperties;
} VkPhysicalDeviceMemoryProperties2;

or the equivalent

// Provided by VK_KHR_get_physical_device_properties2
typedef VkPhysicalDeviceMemoryProperties2 VkPhysicalDeviceMemoryProperties2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
memoryProperties is a VkPhysicalDeviceMemoryProperties structure which is populated with the same values as in vkGetPhysicalDeviceMemoryProperties.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_PROPERTIES_2
pNext must be NULL or a pointer to a valid instance of VkPhysicalDeviceMemoryBudgetPropertiesEXT
The sType value of each struct in the pNext chain must be unique

The VkMemoryHeap structure is defined as:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
typedef enum VkMemoryHeapFlagBits {
    VK_MEMORY_HEAP_DEVICE_LOCAL_BIT = 0x00000001,
  // Provided by VK_VERSION_1_1
    VK_MEMORY_HEAP_MULTI_INSTANCE_BIT = 0x00000002,
  // Provided by VK_KHR_device_group_creation
    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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkMemoryHeapFlags;

VkMemoryHeapFlags is a bitmask type for setting a mask of zero or more VkMemoryHeapFlagBits.

The VkMemoryType structure is defined as:

// Provided by VK_VERSION_1_0
typedef struct VkMemoryType {
    VkMemoryPropertyFlags    propertyFlags;
    uint32_t                 heapIndex;
} VkMemoryType;

heapIndex describes which memory heap this memory type corresponds to, and must be less than memoryHeapCount from the VkPhysicalDeviceMemoryProperties structure.
propertyFlags is a bitmask of VkMemoryPropertyFlagBits of properties for this memory type.

Bits which may be set in VkMemoryType::propertyFlags, indicating properties of a memory heap, are:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_1
    VK_MEMORY_PROPERTY_PROTECTED_BIT = 0x00000020,
  // Provided by VK_AMD_device_coherent_memory
    VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD = 0x00000040,
  // Provided by VK_AMD_device_coherent_memory
    VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD = 0x00000080,
} 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.
VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD bit specifies that device accesses to allocations of this memory type are automatically made available and visible.
VK_MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD bit specifies that memory allocated with this type is not cached on the device. Uncached device memory is always device coherent.

For any memory allocated with both the VK_MEMORY_PROPERTY_HOST_COHERENT_BIT and the VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD, host or device accesses also perform automatic memory domain transfer operations, such that writes are always automatically available and visible to both host and device memory domains.

Note

Device coherence is a useful property for certain debugging use cases (e.g. crash analysis, where performing separate coherence actions could mean values are not reported correctly). However, device coherent accesses may be slower than equivalent accesses without device coherence, particularly if they are also device uncached. For device uncached memory in particular, repeated accesses to the same or neighbouring memory locations over a short time period (e.g. within a frame) may be slower than it would be for the equivalent cached memory type. As such, it is generally inadvisable to use device coherent or device uncached memory except when really needed.

// Provided by VK_VERSION_1_0
typedef VkFlags VkMemoryPropertyFlags;

VkMemoryPropertyFlags is a bitmask type for setting a mask of zero or more VkMemoryPropertyFlagBits.

If the VkPhysicalDeviceMemoryBudgetPropertiesEXT structure is included in the pNext chain of VkPhysicalDeviceMemoryProperties2, it is filled with the current memory budgets and usages.

The VkPhysicalDeviceMemoryBudgetPropertiesEXT structure is defined as:

// Provided by VK_EXT_memory_budget
typedef struct VkPhysicalDeviceMemoryBudgetPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    VkDeviceSize       heapBudget[VK_MAX_MEMORY_HEAPS];
    VkDeviceSize       heapUsage[VK_MAX_MEMORY_HEAPS];
} VkPhysicalDeviceMemoryBudgetPropertiesEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
heapBudget is an array of VK_MAX_MEMORY_HEAPS VkDeviceSize values in which memory budgets are returned, with one element for each memory heap. A heap’s budget is a rough estimate of how much memory the process can allocate from that heap before allocations may fail or cause performance degradation. The budget includes any currently allocated device memory.
heapUsage is an array of VK_MAX_MEMORY_HEAPS VkDeviceSize values in which memory usages are returned, with one element for each memory heap. A heap’s usage is an estimate of how much memory the process is currently using in that heap.

The values returned in this structure are not invariant. The heapBudget and heapUsage values must be zero for array elements greater than or equal to VkPhysicalDeviceMemoryProperties::memoryHeapCount. The heapBudget value must be non-zero for array elements less than VkPhysicalDeviceMemoryProperties::memoryHeapCount. The heapBudget value must be less than or equal to VkMemoryHeap::size for each heap.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_MEMORY_BUDGET_PROPERTIES_EXT

A Vulkan device operates on data in device memory via memory objects that are represented in the API by a VkDeviceMemory handle:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDeviceMemory)

To allocate memory objects, call:

// Provided by VK_VERSION_1_0
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 a 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.

The cumulative memory size allocated to a heap can be limited by the size of the specified heap. In such cases, allocated memory is tracked on a per-device and per-heap basis. Some platforms allow overallocation into other heaps. The overallocation behavior can be specified through the VK_AMD_memory_overallocation_behavior extension.

Valid Usage

pAllocateInfo->allocationSize must be less than or equal to VkPhysicalDeviceMemoryProperties::memoryHeaps[memindex].size where memindex = VkPhysicalDeviceMemoryProperties::memoryTypes[pAllocateInfo->memoryTypeIndex].heapIndex as returned by vkGetPhysicalDeviceMemoryProperties for the VkPhysicalDevice that device was created from
pAllocateInfo->memoryTypeIndex must be less than VkPhysicalDeviceMemoryProperties::memoryTypeCount as returned by vkGetPhysicalDeviceMemoryProperties for the VkPhysicalDevice that device was created from
If the deviceCoherentMemory feature is not enabled, pAllocateInfo->memoryTypeIndex must not identify a memory type supporting VK_MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD
There must be less than VkPhysicalDeviceLimits::maxMemoryAllocationCount device memory allocations currently allocated on the device.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_INVALID_EXTERNAL_HANDLE
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR

The VkMemoryAllocateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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

A VkMemoryAllocateInfo structure defines a memory import operation if its pNext chain includes 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

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 is ignored. The implementation must query the size of these allocations from the OS.

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 includes a VkExportMemoryAllocateInfo structure, 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 include a VkMemoryDedicatedAllocateInfo or VkDedicatedAllocationMemoryAllocateInfoNV structure with either its image or buffer member set to a value other than VK_NULL_HANDLE.
If the pNext chain includes a VkExportMemoryAllocateInfo structure, it must not include a VkExportMemoryAllocateInfoNV or VkExportMemoryWin32HandleInfoNV structure
If the pNext chain includes a VkImportMemoryWin32HandleInfoKHR structure, it must not include a VkImportMemoryWin32HandleInfoNV structure
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_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 is a host pointer, the pNext chain must not include a VkDedicatedAllocationMemoryAllocateInfoNV structure with either its image or buffer field set to a value other than VK_NULL_HANDLE
If the parameters define an import operation and the external handle is a host pointer, the pNext chain must not include a VkMemoryDedicatedAllocateInfo structure with either its image or buffer field set to a value other than VK_NULL_HANDLE
If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID, allocationSize must be the size returned by vkGetAndroidHardwareBufferPropertiesANDROID for the Android hardware buffer
If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID, and the pNext chain does not include a VkMemoryDedicatedAllocateInfo structure or VkMemoryDedicatedAllocateInfo::image is VK_NULL_HANDLE, the Android hardware buffer must have a AHardwareBuffer_Desc::format of AHARDWAREBUFFER_FORMAT_BLOB and a AHardwareBuffer_Desc::usage that includes AHARDWAREBUFFER_USAGE_GPU_DATA_BUFFER
If the parameters define an import operation and the external handle type is VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID, memoryTypeIndex must be one of those returned by vkGetAndroidHardwareBufferPropertiesANDROID for the Android hardware buffer
If the parameters do not define an import operation, and the pNext chain includes a VkExportMemoryAllocateInfo structure with VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID included in its handleTypes member, and the pNext chain includes a VkMemoryDedicatedAllocateInfo structure 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 a VkMemoryDedicatedAllocateInfo with image that is not VK_NULL_HANDLE, the Android hardware buffer’s AHardwareBuffer::usage must include at least one of AHARDWAREBUFFER_USAGE_GPU_FRAMEBUFFER or AHARDWAREBUFFER_USAGE_GPU_SAMPLED_IMAGE
If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes a VkMemoryDedicatedAllocateInfo with image that is not VK_NULL_HANDLE, the format of image must be VK_FORMAT_UNDEFINED or the format returned by vkGetAndroidHardwareBufferPropertiesANDROID in VkAndroidHardwareBufferFormatPropertiesANDROID::format for the Android hardware buffer
If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes a VkMemoryDedicatedAllocateInfo structure with image that is not VK_NULL_HANDLE, the width, height, and array layer dimensions of image and the Android hardware buffer’s AHardwareBuffer_Desc must be identical
If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes a VkMemoryDedicatedAllocateInfo structure with image that is not VK_NULL_HANDLE, and the Android hardware buffer’s AHardwareBuffer::usage includes AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE, the image must have a complete mipmap chain
If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes a VkMemoryDedicatedAllocateInfo structure with image that is not VK_NULL_HANDLE, and the Android hardware buffer’s AHardwareBuffer::usage does not include AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE, the image must have exactly one mipmap level
If the parameters define an import operation, the external handle is an Android hardware buffer, and the pNext chain includes a VkMemoryDedicatedAllocateInfo structure with image that is not VK_NULL_HANDLE, each bit set in the usage of image 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 AHardwareBuffer_Desc::usage
If VkMemoryOpaqueCaptureAddressAllocateInfo::opaqueCaptureAddress is not zero, VkMemoryAllocateFlagsInfo::flags must include VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT
If VkMemoryAllocateFlagsInfo::flags includes VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT, the bufferDeviceAddressCaptureReplay feature must be enabled
If VkMemoryAllocateFlagsInfo::flags includes VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT, the bufferDeviceAddress feature must be enabled
If the pNext chain includes a VkImportMemoryHostPointerInfoEXT structure, VkMemoryOpaqueCaptureAddressAllocateInfo::opaqueCaptureAddress must be zero
If the parameters define an import operation, VkMemoryOpaqueCaptureAddressAllocateInfo::opaqueCaptureAddress must be zero

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_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 VkDedicatedAllocationMemoryAllocateInfoNV, VkExportMemoryAllocateInfo, VkExportMemoryAllocateInfoNV, VkExportMemoryWin32HandleInfoKHR, VkExportMemoryWin32HandleInfoNV, VkImportAndroidHardwareBufferInfoANDROID, VkImportMemoryFdInfoKHR, VkImportMemoryHostPointerInfoEXT, VkImportMemoryWin32HandleInfoKHR, VkImportMemoryWin32HandleInfoNV, VkMemoryAllocateFlagsInfo, VkMemoryDedicatedAllocateInfo, VkMemoryOpaqueCaptureAddressAllocateInfo, or VkMemoryPriorityAllocateInfoEXT
The sType value of each struct in the pNext chain must be unique

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:

// Provided by VK_VERSION_1_1
typedef struct VkMemoryDedicatedAllocateInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
    VkBuffer           buffer;
} VkMemoryDedicatedAllocateInfo;

or the equivalent

// Provided by VK_KHR_dedicated_allocation
typedef VkMemoryDedicatedAllocateInfo VkMemoryDedicatedAllocateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 and the memory is not an imported Android Hardware Buffer, 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 and the memory is not an imported Android Hardware Buffer, 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 of non-ignored parameters 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:

// Provided by VK_NV_dedicated_allocation
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 a structure extending this 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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

If the pNext chain includes a VkMemoryPriorityAllocateInfoEXT structure, then that structure includes a priority for the memory.

The VkMemoryPriorityAllocateInfoEXT structure is defined as:

// Provided by VK_EXT_memory_priority
typedef struct VkMemoryPriorityAllocateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    float              priority;
} VkMemoryPriorityAllocateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
priority is a floating-point value between 0 and 1, indicating the priority of the allocation relative to other memory allocations. Larger values are higher priority. The granularity of the priorities is implementation-dependent.

Memory allocations with higher priority may be more likely to stay in device-local memory when the system is under memory pressure.

If this structure is not included, it is as if the priority value were 0.5.

Valid Usage

priority must be between 0 and 1, inclusive

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_PRIORITY_ALLOCATE_INFO_EXT

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:

// Provided by VK_VERSION_1_1
typedef struct VkExportMemoryAllocateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExportMemoryAllocateInfo;

or the equivalent

// Provided by VK_KHR_external_memory
typedef VkExportMemoryAllocateInfo VkExportMemoryAllocateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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

The bits in handleTypes must be supported and compatible, as reported by VkExternalImageFormatProperties or VkExternalBufferProperties

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO
handleTypes must be a valid combination of VkExternalMemoryHandleTypeFlagBits values

To specify additional attributes of NT handles exported from a memory object, add a VkExportMemoryWin32HandleInfoKHR structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkExportMemoryWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_memory_win32
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 a structure extending this 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 VkExportMemoryAllocateInfo is not present in the same pNext chain, this structure is ignored.

If VkExportMemoryAllocateInfo is present in the pNext chain of VkMemoryAllocateInfo with a Windows handleType, but either VkExportMemoryWin32HandleInfoKHR is not present in the pNext chain, or if it is but 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 used depend on the handle type.

For handles of the following types:

VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_TEXTURE_BIT

The implementation must ensure the access rights allow read and write access to the memory.

For handles of the following types:

VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP_BIT VK_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE_BIT

The access rights must be:

GENERIC_ALL

1: https://docs.microsoft.com/en-us/windows/win32/sync/synchronization-object-security-and-access-rights

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, a VkExportMemoryWin32HandleInfoKHR structure must not be included 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:

// Provided by VK_KHR_external_memory_win32
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 a structure extending this 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)

sType must be VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_KHR
If handleType is not 0, handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

To export a Windows handle representing the underlying resources of a Vulkan device memory object, call:

// Provided by VK_KHR_external_memory_win32
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 a 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

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetWin32HandleInfoKHR structure is defined as:

// Provided by VK_KHR_external_memory_win32
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 a structure extending this 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:

// Provided by VK_KHR_external_memory_win32
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

VK_SUCCESS

VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryWin32HandlePropertiesKHR structure returned is defined as:

// Provided by VK_KHR_external_memory_win32
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 a structure extending this structure.
memoryTypeBits is a bitmask containing one bit set for every memory type which the specified windows handle can be imported as.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_WIN32_HANDLE_PROPERTIES_KHR
pNext must be NULL

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:

// Provided by VK_KHR_external_memory_fd
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 a structure extending this 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.

Valid Usage

If handleType is not 0, it must be supported for import, as reported by VkExternalImageFormatProperties or VkExternalBufferProperties
The memory from which fd was exported must have been created on the same underlying physical device as device
If handleType is not 0, it must be defined as a POSIX file descriptor handle
If handleType is not 0, fd must be a valid handle of the type specified by handleType
The memory represented by fd must have been created from a physical device and driver that is compatible with device and handleType, as described in External memory handle types compatibility
fd must obey any requirements listed for handleType in external memory handle types compatibility

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMPORT_MEMORY_FD_INFO_KHR
If handleType is not 0, handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

To export a POSIX file descriptor representing the underlying resources of a Vulkan device memory object, call:

// Provided by VK_KHR_external_memory_fd
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 a 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 an int value

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetFdInfoKHR structure is defined as:

// Provided by VK_KHR_external_memory_fd
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 a structure extending this 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:

// Provided by VK_KHR_external_memory_fd
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 is a pointer to a VkMemoryFdPropertiesKHR structure in which the properties of the handle fd are returned.

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

VK_SUCCESS

VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryFdPropertiesKHR structure returned is defined as:

// Provided by VK_KHR_external_memory_fd
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 a structure extending this structure.
memoryTypeBits is a bitmask containing one bit set for every memory type which the specified file descriptor can be imported as.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_FD_PROPERTIES_KHR
pNext must be NULL

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:

// Provided by VK_EXT_external_memory_host
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 a structure extending this 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 VkExternalMemoryProperties
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)

sType must be VK_STRUCTURE_TYPE_IMPORT_MEMORY_HOST_POINTER_INFO_EXT
handleType must be a valid VkExternalMemoryHandleTypeFlagBits value

To determine the correct parameters to use when importing host pointers, call:

// Provided by VK_EXT_external_memory_host
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.
pMemoryHostPointerProperties is a pointer to a VkMemoryHostPointerPropertiesEXT structure in which the host pointer properties are returned.

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

VK_SUCCESS

VK_ERROR_INVALID_EXTERNAL_HANDLE

The VkMemoryHostPointerPropertiesEXT structure is defined as:

// Provided by VK_EXT_external_memory_host
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 a structure extending this structure.
memoryTypeBits is a bitmask containing one bit set for every memory type which the specified host pointer can be imported as.

The value returned by memoryTypeBits must only include bits that identify memory types which are host visible.

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:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a structure extending this 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

If buffer is not NULL, Android hardware buffers must be supported for import, as reported by VkExternalImageFormatProperties or VkExternalBufferProperties
If buffer is not NULL, it must be a valid Android hardware buffer object with AHardwareBuffer_Desc::usage compatible with Vulkan as described in Android Hardware Buffers

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMPORT_ANDROID_HARDWARE_BUFFER_INFO_ANDROID
buffer must be a valid pointer to an AHardwareBuffer value

To export an Android hardware buffer representing the underlying resources of a Vulkan device memory object, call:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a 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 an AHardwareBuffer value

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkMemoryGetAndroidHardwareBufferInfoANDROID structure is defined as:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a structure extending this 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 VkExportMemoryAllocateInfo::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

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_GET_ANDROID_HARDWARE_BUFFER_INFO_ANDROID
pNext must be NULL
memory must be a valid VkDeviceMemory handle

To determine the memory parameters to use when importing an Android hardware buffer, call:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 is a pointer to a VkAndroidHardwareBufferPropertiesANDROID structure in which the properties of buffer are returned.

Valid Usage

buffer must be a valid Android hardware buffer object with at least one of the AHARDWAREBUFFER_USAGE_GPU_* flags in its AHardwareBuffer_Desc::usage

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

VK_SUCCESS

VK_ERROR_INVALID_EXTERNAL_HANDLE_KHR

The VkAndroidHardwareBufferPropertiesANDROID structure returned is defined as:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a structure extending this 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.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ANDROID_HARDWARE_BUFFER_PROPERTIES_ANDROID
pNext must be NULL or a pointer to a valid instance of VkAndroidHardwareBufferFormatPropertiesANDROID
The sType value of each struct in the pNext chain must be unique

To obtain format properties of an Android hardware buffer, include a VkAndroidHardwareBufferFormatPropertiesANDROID structure in the pNext chain of the VkAndroidHardwareBufferPropertiesANDROID structure passed to vkGetAndroidHardwareBufferPropertiesANDROID. This structure is defined as:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a structure extending this structure.
format is the Vulkan format corresponding to the Android hardware buffer’s format, or VK_FORMAT_UNDEFINED if there is not 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 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′C_BC_R conversion even if the non-external format does not, 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 the identity swizzle.

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:

// Provided by VK_NV_external_memory
typedef struct VkExportMemoryAllocateInfoNV {
    VkStructureType                      sType;
    const void*                          pNext;
    VkExternalMemoryHandleTypeFlagsNV    handleTypes;
} VkExportMemoryAllocateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBitsNV specifying one or more memory handle types that may be exported. Multiple handle types may be requested for the same allocation as long as they are compatible, as reported by vkGetPhysicalDeviceExternalImageFormatPropertiesNV.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO_NV
handleTypes must be a valid combination of VkExternalMemoryHandleTypeFlagBitsNV values

When VkExportMemoryAllocateInfoNV::handleTypes includes VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_BIT_NV, add a VkExportMemoryWin32HandleInfoNV structure 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:

// Provided by VK_NV_external_memory_win32
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 a structure extending this 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

1: https://docs.microsoft.com/en-us/windows/win32/sync/synchronization-object-security-and-access-rights

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:

// Provided by VK_NV_external_memory_win32
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 a structure extending this 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)

sType must be VK_STRUCTURE_TYPE_IMPORT_MEMORY_WIN32_HANDLE_INFO_NV
handleType must be a valid combination of VkExternalMemoryHandleTypeFlagBitsNV values

Bits which can be set in handleType are:

Possible values of VkImportMemoryWin32HandleInfoNV::handleType, specifying the type of an external memory handle, are:

// Provided by VK_NV_external_memory_capabilities
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.

// Provided by VK_NV_external_memory_capabilities
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:

// Provided by VK_NV_external_memory_win32
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 is a pointer to a Windows HANDLE in which the handle is returned.

Valid Usage

handleType must be a flag specified in VkExportMemoryAllocateInfoNV::handleTypes when allocating memory

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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_1
typedef struct VkMemoryAllocateFlagsInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkMemoryAllocateFlags    flags;
    uint32_t                 deviceMask;
} VkMemoryAllocateFlagsInfo;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkMemoryAllocateFlagsInfo VkMemoryAllocateFlagsInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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:

// Provided by VK_VERSION_1_1
typedef enum VkMemoryAllocateFlagBits {
    VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT = 0x00000001,
  // Provided by VK_VERSION_1_2
    VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT = 0x00000002,
  // Provided by VK_VERSION_1_2
    VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT = 0x00000004,
  // Provided by VK_KHR_device_group
    VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT_KHR = VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT,
  // Provided by VK_KHR_buffer_device_address
    VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT_KHR = VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT,
  // Provided by VK_KHR_buffer_device_address
    VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT_KHR = VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT,
} VkMemoryAllocateFlagBits;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkMemoryAllocateFlagBits VkMemoryAllocateFlagBitsKHR;

VK_MEMORY_ALLOCATE_DEVICE_MASK_BIT specifies that memory will be allocated for the devices in VkMemoryAllocateFlagsInfo::deviceMask.
VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT specifies that the memory can be attached to a buffer object created with the VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT bit set in usage, and that the memory handle can be used to retrieve an opaque address via vkGetDeviceMemoryOpaqueCaptureAddress.
VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT specifies that the memory’s address can be saved and reused on a subsequent run (e.g. for trace capture and replay), see VkBufferOpaqueCaptureAddressCreateInfo for more detail.

// Provided by VK_VERSION_1_1
typedef VkFlags VkMemoryAllocateFlags;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkMemoryAllocateFlags VkMemoryAllocateFlagsKHR;

VkMemoryAllocateFlags is a bitmask type for setting a mask of zero or more VkMemoryAllocateFlagBits.

To request a specific device address for a memory allocation, add a VkMemoryOpaqueCaptureAddressAllocateInfo structure to the pNext chain of the VkMemoryAllocateInfo structure. The VkMemoryOpaqueCaptureAddressAllocateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkMemoryOpaqueCaptureAddressAllocateInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint64_t           opaqueCaptureAddress;
} VkMemoryOpaqueCaptureAddressAllocateInfo;

or the equivalent

// Provided by VK_KHR_buffer_device_address
typedef VkMemoryOpaqueCaptureAddressAllocateInfo VkMemoryOpaqueCaptureAddressAllocateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
opaqueCaptureAddress is the opaque capture address requested for the memory allocation.

If opaqueCaptureAddress is zero, no specific address is requested.

If opaqueCaptureAddress is not zero, it should be an address retrieved from vkGetDeviceMemoryOpaqueCaptureAddress on an identically created memory allocation on the same implementation.

Note

In most cases, it is expected that a non-zero opaqueAddress is an address retrieved from vkGetDeviceMemoryOpaqueCaptureAddress on an identically created memory allocation. If this is not the case, it likely that VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS errors will occur.

This is, however, not a strict requirement because trace capture/replay tools may need to adjust memory allocation parameters for imported memory.

If this structure is not present, it is as if opaqueCaptureAddress is zero.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_OPAQUE_CAPTURE_ADDRESS_ALLOCATE_INFO

To free a memory object, call:

// Provided by VK_VERSION_1_0
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. Memory can be freed whilst still bound to resources, but those resources must not be used afterwards. 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:

// Provided by VK_VERSION_1_0
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 is a pointer to a void * variable 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.

After a successful call to vkMapMemory the memory object memory is considered to be currently host mapped.

Note

It is an application error to call vkMapMemory on a memory object that is already host 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 host mapped, 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 host 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_MEMORY_MAP_FAILED

// Provided by VK_VERSION_1_0
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.

Note

While memory objects imported from a handle type of VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_ALLOCATION_BIT_EXT or VK_EXTERNAL_MEMORY_HANDLE_TYPE_HOST_MAPPED_FOREIGN_MEMORY_BIT_EXT are inherently mapped to host address space, they are not considered to be host mapped device memory unless they are explicitly host mapped using vkMapMemory. That means flushing or invalidating host caches with respect to host accesses performed on such memory through the original host pointer specified at import time is the responsibility of the application and must be performed with appropriate synchronization primitives provided by the platform which are outside the scope of Vulkan. vkFlushMappedMemoryRanges and vkInvalidateMappedMemoryRanges, however, can still be used on such memory objects to synchronize host accesses performed through the host pointer of the host mapped device memory range returned by vkMapMemory.

To flush ranges of non-coherent memory from the host caches, call:

// Provided by VK_VERSION_1_0
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 are made available to the host memory domain, such that they can be made available to the device memory domain via memory domain operations using 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 host 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 host 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

To invalidate ranges of non-coherent memory from the host caches, call:

// Provided by VK_VERSION_1_0
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 available to the host memory domain using 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 host mapped, or the last time it was invalidated.

Note

Mapping non-coherent memory does not implicitly invalidate that memory.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkMappedMemoryRange structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 host 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:

// Provided by VK_VERSION_1_0
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 host 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:

// Provided by VK_VERSION_1_0
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
- Protected queue operations

Protected Memory Access Rules

If VkPhysicalDeviceProtectedMemoryProperties::protectedNoFault is VK_FALSE, applications must not perform any of the following operations:

Write to unprotected memory within protected queue operations.
Access protected memory within protected queue operations other than in framebuffer-space pipeline stages, the compute shader stage, or the transfer stage.
Perform a query within protected queue operations.
Execute an indirect command within protected queue operations.

If VkPhysicalDeviceProtectedMemoryProperties::protectedNoFault is VK_TRUE, these operations are valid, but reads will return undefined values, and writes will either be dropped or store undefined values.

Whether these operations are valid or not, or if any other invalid usage is performed, the implementation must guarantee that:

Protected device memory must never be visible to the host.
Values written to unprotected device memory must not be a function of values from protected memory.

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. An VkImage or VkBuffer can be bound to the imported or exported VkDeviceMemory object if it is created with VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID.

To remove an unnecessary compile-time dependency, an incomplete type definition of AHardwareBuffer is provided in the Vulkan headers:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
struct AHardwareBuffer;

The actual AHardwareBuffer type is defined in Android NDK headers.

Note

The NDK format, usage, and size/dimensions of an AHardwareBuffer object can be obtained with the AHardwareBuffer_describe function. While Android hardware buffers can be imported to or exported from Vulkan without using that function, valid usage and implementation behavior is defined in terms of the AHardwareBuffer_Desc properties it returns.

Android hardware buffer objects are reference-counted using Android NDK functions outside of 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 that 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 the 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 supports Vulkan usages and creation flags in the usage equivalence table which do not have Android hardware buffer equivalents.

An VkAndroidHardwareBufferUsageANDROID structure can be included in 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 an 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 an 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 may represent images using implementation-specific formats, layouts, color models, etc., which do not 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′C_BC_R conversion enabled, and must have optimal tiling.

Images that will be backed by an Android hardware buffer can use an external format by setting VkImageCreateInfo::format to VK_FORMAT_UNDEFINED and including a VkExternalFormatANDROID structure 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 a VkAndroidHardwareBufferFormatPropertiesANDROID structure 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 does.

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` ¹	`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_FRAMEBUFFER` ³	`VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT`
`AHARDWAREBUFFER_USAGE_GPU_FRAMEBUFFER` ³	`VK_IMAGE_USAGE_DEPTH_STENCIL_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`

1: Vulkan does not differentiate between AHARDWAREBUFFER_FORMAT_R8G8B8A8_UNORM and AHARDWAREBUFFER_FORMAT_R8G8B8X8_UNORM: they both behave as VK_FORMAT_R8G8B8A8_UNORM. After an external entity writes to a AHARDWAREBUFFER_FORMAT_R8G8B8X8_UNORM Android hardware buffer, the values read by Vulkan from the X/A channel are undefined. To emulate the traditional behavior of the X channel during sampling or blending, applications should use VK_COMPONENT_SWIZZLE_ONE in image view component mappings and VK_BLEND_FACTOR_ONE in color blend factors. There is no way to avoid copying these undefined values when copying from such an image to another image or buffer.
2: The AHARDWAREBUFFER_USAGE_GPU_MIPMAP_COMPLETE flag does not correspond to a Vulkan image usage or creation flag. Instead, its presence indicates that the Android hardware buffer contains a complete mipmap chain, and its absence indicates that the Android hardware buffer contains only a single mip level.
3: Only image usages valid for the format are valid. It would be invalid to take a Android Hardware Buffer with a format of AHARDWAREBUFFER_FORMAT_R8G8B8A8_UNORM that has a AHARDWAREBUFFER_USAGE_GPU_FRAMEBUFFER usage and try to create an image with VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT.

Note

When using VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT with Android hardware buffer images, applications should use VkImageFormatListCreateInfo 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 can 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 both 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:

// Provided by VK_VERSION_1_1
void vkGetDeviceGroupPeerMemoryFeatures(
    VkDevice                                    device,
    uint32_t                                    heapIndex,
    uint32_t                                    localDeviceIndex,
    uint32_t                                    remoteDeviceIndex,
    VkPeerMemoryFeatureFlags*                   pPeerMemoryFeatures);

or the equivalent command

// Provided by VK_KHR_device_group
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 VkPeerMemoryFeatureFlags bitmask 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

Bits which may be set in the value returned for vkGetDeviceGroupPeerMemoryFeatures::pPeerMemoryFeatures, indicating the supported peer memory features, are:

// Provided by VK_VERSION_1_1
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,
  // Provided by VK_KHR_device_group
    VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT_KHR = VK_PEER_MEMORY_FEATURE_COPY_SRC_BIT,
  // Provided by VK_KHR_device_group
    VK_PEER_MEMORY_FEATURE_COPY_DST_BIT_KHR = VK_PEER_MEMORY_FEATURE_COPY_DST_BIT,
  // Provided by VK_KHR_device_group
    VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT_KHR = VK_PEER_MEMORY_FEATURE_GENERIC_SRC_BIT,
  // Provided by VK_KHR_device_group
    VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT_KHR = VK_PEER_MEMORY_FEATURE_GENERIC_DST_BIT,
} VkPeerMemoryFeatureFlagBits;

or the equivalent

// Provided by VK_KHR_device_group
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.

// Provided by VK_VERSION_1_1
typedef VkFlags VkPeerMemoryFeatureFlags;

or the equivalent

// Provided by VK_KHR_device_group
typedef VkPeerMemoryFeatureFlags VkPeerMemoryFeatureFlagsKHR;

VkPeerMemoryFeatureFlags is a bitmask type for setting a mask of zero or more VkPeerMemoryFeatureFlagBits.

To query a 64-bit opaque capture address value from a memory object, call:

// Provided by VK_VERSION_1_2
uint64_t vkGetDeviceMemoryOpaqueCaptureAddress(
    VkDevice                                    device,
    const VkDeviceMemoryOpaqueCaptureAddressInfo* pInfo);

or the equivalent command

// Provided by VK_KHR_buffer_device_address
uint64_t vkGetDeviceMemoryOpaqueCaptureAddressKHR(
    VkDevice                                    device,
    const VkDeviceMemoryOpaqueCaptureAddressInfo* pInfo);

device is the logical device that the memory object was allocated on.
pInfo is a pointer to a VkDeviceMemoryOpaqueCaptureAddressInfo structure specifying the memory object to retrieve an address for.

The 64-bit return value is an opaque address representing the start of pInfo->memory.

If the memory object was allocated with a non-zero value of VkMemoryOpaqueCaptureAddressAllocateInfo::opaqueCaptureAddress, the return value must be the same address.

Note

The expected usage for these opaque addresses is only for trace capture/replay tools to store these addresses in a trace and subsequently specify them during replay.

Valid Usage

The bufferDeviceAddress feature must be enabled
If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkDeviceMemoryOpaqueCaptureAddressInfo structure

The VkDeviceMemoryOpaqueCaptureAddressInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkDeviceMemoryOpaqueCaptureAddressInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkDeviceMemory     memory;
} VkDeviceMemoryOpaqueCaptureAddressInfo;

or the equivalent

// Provided by VK_KHR_buffer_device_address
typedef VkDeviceMemoryOpaqueCaptureAddressInfo VkDeviceMemoryOpaqueCaptureAddressInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
memory specifies the memory whose address is being queried.

Valid Usage

memory must have been allocated with VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DEVICE_MEMORY_OPAQUE_CAPTURE_ADDRESS_INFO
pNext must be NULL
memory must be a valid VkDeviceMemory handle

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkBuffer)

To create buffers, call:

// Provided by VK_VERSION_1_0
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 a VkBufferCreateInfo structure containing parameters affecting creation of the buffer.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pBuffer is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR

The VkBufferCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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).

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 includes a VkExternalMemoryBufferCreateInfo structure, its handleTypes member must only contain bits that are also in VkExternalBufferProperties::externalMemoryProperties.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 includes a VkDedicatedAllocationBufferCreateInfoNV structure, 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
If VkBufferDeviceAddressCreateInfoEXT::deviceAddress is not zero, flags must include VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT
If VkBufferOpaqueCaptureAddressCreateInfo::opaqueCaptureAddress is not zero, flags must include VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT
If flags includes VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT, the bufferDeviceAddressCaptureReplay or VkPhysicalDeviceBufferDeviceAddressFeaturesEXT::bufferDeviceAddressCaptureReplay feature must be enabled

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BUFFER_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 VkBufferDeviceAddressCreateInfoEXT, VkBufferOpaqueCaptureAddressCreateInfo, VkDedicatedAllocationBufferCreateInfoNV, or VkExternalMemoryBufferCreateInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkBufferCreateFlagBits values
usage must be a valid combination of VkBufferUsageFlagBits values
usage must not be 0
sharingMode must be a valid VkSharingMode value

Bits which can be set in VkBufferCreateInfo::usage, specifying usage behavior of a buffer, are:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_2
    VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT = 0x00020000,
  // Provided by VK_EXT_transform_feedback
    VK_BUFFER_USAGE_TRANSFORM_FEEDBACK_BUFFER_BIT_EXT = 0x00000800,
  // Provided by VK_EXT_transform_feedback
    VK_BUFFER_USAGE_TRANSFORM_FEEDBACK_COUNTER_BUFFER_BIT_EXT = 0x00001000,
  // Provided by VK_EXT_conditional_rendering
    VK_BUFFER_USAGE_CONDITIONAL_RENDERING_BIT_EXT = 0x00000200,
  // Provided by VK_KHR_ray_tracing
    VK_BUFFER_USAGE_RAY_TRACING_BIT_KHR = 0x00000400,
  // Provided by VK_NV_ray_tracing
    VK_BUFFER_USAGE_RAY_TRACING_BIT_NV = VK_BUFFER_USAGE_RAY_TRACING_BIT_KHR,
  // Provided by VK_EXT_buffer_device_address
    VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT_EXT = VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT,
  // Provided by VK_KHR_buffer_device_address
    VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT_KHR = VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT,
} 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, vkCmdDrawMeshTasksIndirectNV, vkCmdDrawMeshTasksIndirectCountNV, or vkCmdDispatchIndirect. It is also suitable for passing as the buffer member of VkIndirectCommandsStreamNV, or sequencesCountBuffer or sequencesIndexBuffer or preprocessedBuffer member of VkGeneratedCommandsInfoNV
VK_BUFFER_USAGE_CONDITIONAL_RENDERING_BIT_EXT specifies that the buffer is suitable for passing as the buffer parameter to vkCmdBeginConditionalRenderingEXT.
VK_BUFFER_USAGE_TRANSFORM_FEEDBACK_BUFFER_BIT_EXT specifies that the buffer is suitable for using for binding as a transform feedback buffer with vkCmdBindTransformFeedbackBuffersEXT.
VK_BUFFER_USAGE_TRANSFORM_FEEDBACK_COUNTER_BUFFER_BIT_EXT specifies that the buffer is suitable for using as a counter buffer with vkCmdBeginTransformFeedbackEXT and vkCmdEndTransformFeedbackEXT.
VK_BUFFER_USAGE_RAY_TRACING_BIT_KHR specifies that the buffer is suitable for use in vkCmdTraceRaysKHR and vkCmdBuildAccelerationStructureKHR.
VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT specifies that the buffer can be used to retrieve a buffer device address via vkGetBufferDeviceAddress and use that address to access the buffer’s memory from a shader.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
typedef enum VkBufferCreateFlagBits {
    VK_BUFFER_CREATE_SPARSE_BINDING_BIT = 0x00000001,
    VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT = 0x00000002,
    VK_BUFFER_CREATE_SPARSE_ALIASED_BIT = 0x00000004,
  // Provided by VK_VERSION_1_1
    VK_BUFFER_CREATE_PROTECTED_BIT = 0x00000008,
  // Provided by VK_VERSION_1_2
    VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT = 0x00000010,
  // Provided by VK_EXT_buffer_device_address
    VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT_EXT = VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT,
  // Provided by VK_KHR_buffer_device_address
    VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT_KHR = VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT,
} 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.
VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT specifies that the buffer’s address can be saved and reused on a subsequent run (e.g. for trace capture and replay), see VkBufferOpaqueCaptureAddressCreateInfo for more detail.

See Sparse Resource Features and Physical Device Features for details of the sparse memory features supported on a device.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_NV_dedicated_allocation
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 a structure extending this 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:

// Provided by VK_VERSION_1_1
typedef struct VkExternalMemoryBufferCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExternalMemoryBufferCreateInfo;

or the equivalent

// Provided by VK_KHR_external_memory
typedef VkExternalMemoryBufferCreateInfo VkExternalMemoryBufferCreateInfoKHR;

Note

A VkExternalMemoryBufferCreateInfo structure must be included in the creation parameters for a buffer that will be bound to memory that is either exported or imported.

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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_BUFFER_CREATE_INFO
handleTypes must be a valid combination of VkExternalMemoryHandleTypeFlagBits values

To request a specific device address for a buffer, add a VkBufferOpaqueCaptureAddressCreateInfo structure to the pNext chain of the VkBufferCreateInfo structure. The VkBufferOpaqueCaptureAddressCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkBufferOpaqueCaptureAddressCreateInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint64_t           opaqueCaptureAddress;
} VkBufferOpaqueCaptureAddressCreateInfo;

or the equivalent

// Provided by VK_KHR_buffer_device_address
typedef VkBufferOpaqueCaptureAddressCreateInfo VkBufferOpaqueCaptureAddressCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
opaqueCaptureAddress is the opaque capture address requested for the buffer.

If opaqueCaptureAddress is zero, no specific address is requested.

If opaqueCaptureAddress is not zero, then it should be an address retrieved from vkGetBufferOpaqueCaptureAddress for an identically created buffer on the same implementation.

If this structure is not present, it is as if opaqueCaptureAddress is zero.

Apps should avoid creating buffers with app-provided addresses and implementation-provided addresses in the same process, to reduce the likelihood of VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS errors.

Note

The expected usage for this is that a trace capture/replay tool will add the VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT flag to all buffers that use VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT, and during capture will save the queried opaque device addresses in the trace. During replay, the buffers will be created specifying the original address so any address values stored in the trace data will remain valid.

Implementations are expected to separate such buffers in the GPU address space so normal allocations will avoid using these addresses. Apps/tools should avoid mixing app-provided and implementation-provided addresses for buffers created with VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT, to avoid address space allocation conflicts.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BUFFER_OPAQUE_CAPTURE_ADDRESS_CREATE_INFO

Alternatively, to request a specific device address for a buffer, add a VkBufferDeviceAddressCreateInfoEXT structure to the pNext chain of the VkBufferCreateInfo structure. The VkBufferDeviceAddressCreateInfoEXT structure is defined as:

// Provided by VK_EXT_buffer_device_address
typedef struct VkBufferDeviceAddressCreateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkDeviceAddress    deviceAddress;
} VkBufferDeviceAddressCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
deviceAddress is the device address requested for the buffer.

If deviceAddress is zero, no specific address is requested.

If deviceAddress is not zero, then it must be an address retrieved from an identically created buffer on the same implementation. The buffer must also be bound to an identically created VkDeviceMemory object.

If this structure is not present, it is as if deviceAddress is zero.

Apps should avoid creating buffers with app-provided addresses and implementation-provided addresses in the same process, to reduce the likelihood of VK_ERROR_INVALID_DEVICE_ADDRESS_EXT errors.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BUFFER_DEVICE_ADDRESS_CREATE_INFO_EXT

To destroy a buffer, call:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkBufferView)

To create a buffer view, call:

// Provided by VK_VERSION_1_0
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 a VkBufferViewCreateInfo structure containing parameters to be used to create the buffer.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pView is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkBufferViewCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 texel block size of format, the nearest smaller multiple is used.

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 an integer multiple of the texel block size of format
If range is not equal to VK_WHOLE_SIZE, the number of texel buffer elements given by (⌊range / (texel block size)⌋ × (texels per block)) where texel block size and texels per block are as defined in the Compatible Formats table for 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
If range is equal to VK_WHOLE_SIZE, the number of texel buffer elements given by (⌊(size - offset) / (texel block size)⌋ × (texels per block)) where size is the size of buffer, and texel block size and texels per block are as defined in the Compatible Formats table for format, must be less than or equal to VkPhysicalDeviceLimits::maxTexelBufferElements
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
If the texelBufferAlignment feature is not enabled, offset must be a multiple of VkPhysicalDeviceLimits::minTexelBufferOffsetAlignment
If the texelBufferAlignment feature is enabled and if buffer was created with usage containing VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT, offset must be a multiple of the lesser of VkPhysicalDeviceTexelBufferAlignmentPropertiesEXT::storageTexelBufferOffsetAlignmentBytes or, if VkPhysicalDeviceTexelBufferAlignmentPropertiesEXT::storageTexelBufferOffsetSingleTexelAlignment is VK_TRUE, the size of a texel of the requested format. If the size of a texel is a multiple of three bytes, then the size of a single component of format is used instead
If the texelBufferAlignment feature is enabled and if buffer was created with usage containing VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT, offset must be a multiple of the lesser of VkPhysicalDeviceTexelBufferAlignmentPropertiesEXT::uniformTexelBufferOffsetAlignmentBytes or, if VkPhysicalDeviceTexelBufferAlignmentPropertiesEXT::uniformTexelBufferOffsetSingleTexelAlignment is VK_TRUE, the size of a texel of the requested format. If the size of a texel is a multiple of three bytes, then the size of a single component of format is used instead

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

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkImage)

To create images, call:

// Provided by VK_VERSION_1_0
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 a VkImageCreateInfo structure containing parameters to be used to create the image.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pImage is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkImageCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 texel blocks 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 a VkSampleCountFlagBits specifying the number of samples per texel.
tiling is a VkImageTiling value specifying the tiling arrangement of the texel blocks 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

Images created with a format from one of those listed in Formats requiring sampler Y′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views have further restrictions on their limits and capabilities compared to images created with other formats. Creation of images with a format requiring Y′C_BC_R conversion may not be supported unless other parameters meet all of the constraints:

imageType is VK_IMAGE_TYPE_2D
mipLevels is 1
arrayLayers is 1
samples is VK_SAMPLE_COUNT_1_BIT

Implementations may support additional limits and capabilities beyond those listed above.

To determine the set of valid usage bits for a given format, call vkGetPhysicalDeviceFormatProperties.

If the size of the resultant image would exceed maxResourceSize, then vkCreateImage must fail and return VK_ERROR_OUT_OF_DEVICE_MEMORY. This failure may occur even when all image creation parameters satisfy their valid usage requirements.

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).

Image Creation Limits

Valid values for some image creation parameters are limited by a numerical upper bound or by inclusion in a bitset. For example, VkImageCreateInfo::arrayLayers is limited by imageCreateMaxArrayLayers, defined below; and VkImageCreateInfo::samples is limited by imageCreateSampleCounts, also defined below.

Several limiting values are defined below, as well as assisting values from which the limiting values are derived. The limiting values are referenced by the relevant valid usage statements of VkImageCreateInfo.

Let uint64_t imageCreateDrmFormatModifiers[] be the set of Linux DRM format modifiers that the resultant image may have.
- If tiling is not VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then imageCreateDrmFormatModifiers is empty.
- If VkImageCreateInfo::pNext contains VkImageDrmFormatModifierExplicitCreateInfoEXT, then imageCreateDrmFormatModifiers contains exactly one modifier, VkImageDrmFormatModifierExplicitCreateInfoEXT::drmFormatModifier.
- If VkImageCreateInfo::pNext contains VkImageDrmFormatModifierListCreateInfoEXT, then imageCreateDrmFormatModifiers contains the entire array VkImageDrmFormatModifierListCreateInfoEXT::pDrmFormatModifiers.
Let VkBool32 imageCreateMaybeLinear indicate if the resultant image may be linear.
- If tiling is VK_IMAGE_TILING_LINEAR, then imageCreateMaybeLinear is true.
- If tiling is VK_IMAGE_TILING_OPTIMAL, then imageCreateMaybeLinear is false.
- If tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then imageCreateMaybeLinear_ is true if and only if imageCreateDrmFormatModifiers contains DRM_FORMAT_MOD_LINEAR.
Let VkFormatFeatureFlags imageCreateFormatFeatures be the set of valid format features available during image creation.
- If tiling is VK_IMAGE_TILING_LINEAR, then imageCreateFormatFeatures is the value of VkFormatProperties::linearTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties with parameter format equal to VkImageCreateInfo::format.
- If tiling is VK_IMAGE_TILING_OPTIMAL, and if the pNext chain includes no VkExternalFormatANDROID structure with non-zero externalFormat, then imageCreateFormatFeatures is value of VkFormatProperties::optimalTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties with parameter format equal to VkImageCreateInfo::format.
- If tiling is VK_IMAGE_TILING_OPTIMAL, and if the pNext chain includes a VkExternalFormatANDROID structure with non-zero externalFormat, then imageCreateFormatFeatures is the value of VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures obtained by vkGetAndroidHardwareBufferPropertiesANDROID with a matching externalFormat value.
- If tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then the value of imageCreateFormatFeatures is found by calling vkGetPhysicalDeviceFormatProperties2 with VkImageFormatProperties::format equal to VkImageCreateInfo::format and with VkDrmFormatModifierPropertiesListEXT chained into VkImageFormatProperties2; by collecting all members of the returned array VkDrmFormatModifierPropertiesListEXT::pDrmFormatModifierProperties whose drmFormatModifier belongs to imageCreateDrmFormatModifiers; and by taking the bitwise intersection, over the collected array members, of drmFormatModifierTilingFeatures. (The resultant imageCreateFormatFeatures may be empty).
Let VkImageFormatProperties2 imageCreateImageFormatPropertiesList[] be defined as follows.
- If VkImageCreateInfo::pNext contains no VkExternalFormatANDROID structure with non-zero externalFormat, then imageCreateImageFormatPropertiesList is the list of structures obtained by calling vkGetPhysicalDeviceImageFormatProperties2, possibly multiple times, as follows:
  - The parameters VkPhysicalDeviceImageFormatInfo2::format, imageType, tiling, usage, and flags must be equal to those in VkImageCreateInfo.
  - If VkImageCreateInfo::pNext contains a VkExternalMemoryImageCreateInfo structure whose handleTypes is not 0, then VkPhysicalDeviceImageFormatInfo2::pNext must contain a VkPhysicalDeviceExternalImageFormatInfo structure whose handleType is not 0; and vkGetPhysicalDeviceImageFormatProperties2 must be called for each handle type in VkExternalMemoryImageCreateInfo::handleTypes, successively setting VkPhysicalDeviceExternalImageFormatInfo::handleType on each call.
  - If VkImageCreateInfo::pNext contains no VkExternalMemoryImageCreateInfo structure, or contains a structure whose handleTypes is 0, then VkPhysicalDeviceImageFormatInfo2::pNext must either contain no VkPhysicalDeviceExternalImageFormatInfo structure, or contain a structure whose handleType is 0.
  - If tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then VkPhysicalDeviceImageFormatInfo2::pNext must contain a VkPhysicalDeviceImageDrmFormatModifierInfoEXT structure where sharingMode is equal to VkImageCreateInfo::sharingMode; and, if sharingMode is VK_SHARING_MODE_CONCURRENT, then queueFamilyIndexCount and pQueueFamilyIndices must be equal to those in VkImageCreateInfo; and, if flags contains VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT, then the VkImageFormatListCreateInfo structure included in the pNext chain of VkPhysicalDeviceImageFormatInfo2 must be equivalent to the one included in the pNext chain of VkImageCreateInfo; and vkGetPhysicalDeviceImageFormatProperties2 must be called for each modifier in imageCreateDrmFormatModifiers, successively setting VkPhysicalDeviceImageDrmFormatModifierInfoEXT::drmFormatModifier on each call.
  - If tiling is not VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then VkPhysicalDeviceImageFormatInfo2::pNext must contain no VkPhysicalDeviceImageDrmFormatModifierInfoEXT structure.
  - If any call to vkGetPhysicalDeviceImageFormatProperties2 returns an error, then imageCreateImageFormatPropertiesList is defined to be the empty list.
- If VkImageCreateInfo::pNext contains a VkExternalFormatANDROID structure with non-zero externalFormat, then imageCreateImageFormatPropertiesList contains a single element where:
  - VkImageFormatProperties::maxMipLevels is ⌊log₂(max(extent.width, extent.height, extent.depth))⌋ + 1.
  - VkImageFormatProperties::maxArrayLayers is VkPhysicalDeviceLimits::maxImageArrayLayers.
  - Each component of VkImageFormatProperties::maxExtent is VkPhysicalDeviceLimits::maxImageDimension2D.
  - VkImageFormatProperties::sampleCounts contains exactly VK_SAMPLE_COUNT_1_BIT.
Let uint32_t imageCreateMaxMipLevels be the minimum value of VkImageFormatProperties::maxMipLevels in imageCreateImageFormatPropertiesList. The value is undefined if imageCreateImageFormatPropertiesList is empty.
Let uint32_t imageCreateMaxArrayLayers be the minimum value of VkImageFormatProperties::maxArrayLayers in imageCreateImageFormatPropertiesList. The value is undefined if imageCreateImageFormatPropertiesList is empty.
Let VkExtent3D imageCreateMaxExtent be the component-wise minimum over all VkImageFormatProperties::maxExtent values in imageCreateImageFormatPropertiesList. The value is undefined if imageCreateImageFormatPropertiesList is empty.
Let VkSampleCountFlags imageCreateSampleCounts be the intersection of each VkImageFormatProperties::sampleCounts in imageCreateImageFormatPropertiesList. The value is undefined if imageCreateImageFormatPropertiesList is empty.

Valid Usage

Each of the following values (as described in Image Creation Limits) must not be undefined imageCreateMaxMipLevels, imageCreateMaxArrayLayers, imageCreateMaxExtent, and imageCreateSampleCounts
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 pNext chain includes a VkExternalFormatANDROID structure, and its externalFormat member is non-zero the format must be VK_FORMAT_UNDEFINED
If the pNext chain does not include a VkExternalFormatANDROID structure, or does and its externalFormat member is 0, the 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_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT, imageType must be VK_IMAGE_TYPE_2D
If flags contains VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT, imageType must be VK_IMAGE_TYPE_3D
extent.width must be less than or equal to imageCreateMaxExtent.width (as defined in Image Creation Limits)
extent.height must be less than or equal to imageCreateMaxExtent.height (as defined in Image Creation Limits)
extent.depth must be less than or equal to imageCreateMaxExtent.depth (as defined in Image Creation Limits)
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_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 the number of levels in the complete mipmap chain based on extent.width, extent.height, and extent.depth
mipLevels must be less than or equal to imageCreateMaxMipLevels (as defined in Image Creation Limits)
arrayLayers must be less than or equal to imageCreateMaxArrayLayers (as defined in Image Creation Limits)
If imageType is VK_IMAGE_TYPE_3D, arrayLayers must be 1
If samples is not VK_SAMPLE_COUNT_1_BIT, then imageType must be VK_IMAGE_TYPE_2D, flags must not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT, mipLevels must be equal to 1, and imageCreateMaybeLinear (as defined in Image Creation Limits) must be false,
If samples is not VK_SAMPLE_COUNT_1_BIT, usage must not contain VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT
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_FRAGMENT_DENSITY_MAP_BIT_EXT, extent.width must be less than or equal to $⌈ \frac{m a x F r a m e b u f f e r W i d t h}{m i n F r a g m e n t D e n s i t y T e x e l S i z e _{w i d t h}} ⌉$
If usage includes VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT, extent.height must be less than or equal to $⌈ \frac{m a x F r a m e b u f f e r H e i g h t}{m i n F r a g m e n t D e n s i t y T e x e l S i z e _{h e i g h t}} ⌉$
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 imageCreateSampleCounts (as defined in Image Creation Limits)
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 tiling is VK_IMAGE_TILING_LINEAR, flags must not contain VK_IMAGE_CREATE_SPARSE_RESIDENCY_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 includes a VkExternalMemoryImageCreateInfoNV structure, it must not contain a VkExternalMemoryImageCreateInfo structure
If the pNext chain includes a VkExternalMemoryImageCreateInfo structure, 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 a VkPhysicalDeviceExternalImageFormatInfo structure included in the pNext chain, with a handleType equal to any one of the handle types specified in VkExternalMemoryImageCreateInfo::handleTypes
If the pNext chain includes a VkExternalMemoryImageCreateInfoNV structure, 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 imageCreateMaybeLinear (as defined in Image Creation Limits) must be false
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′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views, then mipLevels must be 1
If the image format is one of those listed in Formats requiring sampler Y′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views, samples must be VK_SAMPLE_COUNT_1_BIT
If the image format is one of those listed in Formats requiring sampler Y′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views, imageType must be VK_IMAGE_TYPE_2D
If the image format is one of those listed in Formats requiring sampler Y′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views, and the ycbcrImageArrays feature is not enabled, arrayLayers must be 1
If format is a multi-planar format, and if imageCreateFormatFeatures (as defined in Image Creation Limits) does not contain VK_FORMAT_FEATURE_DISJOINT_BIT, then 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 tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then the pNext chain must include exactly one of VkImageDrmFormatModifierListCreateInfoEXT or VkImageDrmFormatModifierExplicitCreateInfoEXT structures
If the pNext chain includes a VkImageDrmFormatModifierListCreateInfoEXT or VkImageDrmFormatModifierExplicitCreateInfoEXT structure, then tiling must be VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT
If tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT and flags contains VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT, then the pNext chain must include a VkImageFormatListCreateInfo structure with non-zero viewFormatCount
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
If the pNext chain includes a VkExternalMemoryImageCreateInfo structure whose handleTypes member includes VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID, mipLevels must either be 1 or equal to the number of levels in the complete mipmap chain based on extent.width, extent.height, and extent.depth
If the pNext chain includes a VkExternalFormatANDROID structure whose externalFormat member is not 0, flags must not include VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT
If the pNext chain includes a VkExternalFormatANDROID structure whose externalFormat member is not 0, usage must not include any usages except VK_IMAGE_USAGE_SAMPLED_BIT
If the pNext chain includes a VkExternalFormatANDROID structure whose externalFormat member is not 0, tiling must be VK_IMAGE_TILING_OPTIMAL
If format is a depth-stencil format, usage includes VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, and the pNext chain includes a VkImageStencilUsageCreateInfo structure, then its VkImageStencilUsageCreateInfo::stencilUsage member must also include VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If format is a depth-stencil format, usage does not include VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, and the pNext chain includes a VkImageStencilUsageCreateInfo structure, then its VkImageStencilUsageCreateInfo::stencilUsage member must also not include VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT
If format is a depth-stencil format, usage includes VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, and the pNext chain includes a VkImageStencilUsageCreateInfo structure, then its VkImageStencilUsageCreateInfo::stencilUsage member must also include VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT
If format is a depth-stencil format, usage does not include VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, and the pNext chain includes a VkImageStencilUsageCreateInfo structure, then its VkImageStencilUsageCreateInfo::stencilUsage member must also not include VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT
If Format is a depth-stencil format and the pNext chain includes a VkImageStencilUsageCreateInfo structure with its stencilUsage member including VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, extent.width must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferWidth
If format is a depth-stencil format and the pNext chain includes a VkImageStencilUsageCreateInfo structure with its stencilUsage member including VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, extent.height must be less than or equal to VkPhysicalDeviceLimits::maxFramebufferHeight
If the multisampled storage images feature is not enabled, format is a depth-stencil format and the pNext chain includes a VkImageStencilUsageCreateInfo structure with its stencilUsage including VK_IMAGE_USAGE_STORAGE_BIT, samples must be VK_SAMPLE_COUNT_1_BIT
If flags contains VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV, imageType must be VK_IMAGE_TYPE_2D or VK_IMAGE_TYPE_3D
If flags contains VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV, it must not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT and the format must not be a depth/stencil format
If flags contains VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV and imageType is VK_IMAGE_TYPE_2D, extent.width and extent.height must be greater than 1
If flags contains VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV and imageType is VK_IMAGE_TYPE_3D, extent.width, extent.height, and extent.depth must be greater than 1
If usage includes VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, imageType must be VK_IMAGE_TYPE_2D
If usage includes VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, samples must be VK_SAMPLE_COUNT_1_BIT
If usage includes VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, tiling must be VK_IMAGE_TILING_OPTIMAL
If flags contains VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT, tiling must be VK_IMAGE_TILING_OPTIMAL
If flags contains VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT, imageType must be VK_IMAGE_TYPE_2D
If flags contains VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT, flags must not contain VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT
If flags contains VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT, mipLevels must be 1

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_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 VkDedicatedAllocationImageCreateInfoNV, VkExternalFormatANDROID, VkExternalMemoryImageCreateInfo, VkExternalMemoryImageCreateInfoNV, VkImageDrmFormatModifierExplicitCreateInfoEXT, VkImageDrmFormatModifierListCreateInfoEXT, VkImageFormatListCreateInfo, VkImageStencilUsageCreateInfo, or VkImageSwapchainCreateInfoKHR
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkImageCreateFlagBits values
imageType must be a valid VkImageType value
format must be a valid VkFormat value
samples must be a valid VkSampleCountFlagBits value
tiling must be a valid VkImageTiling value
usage must be a valid combination of VkImageUsageFlagBits values
usage must not be 0
sharingMode must be a valid VkSharingMode value
initialLayout must be a valid VkImageLayout value

The VkImageStencilUsageCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkImageStencilUsageCreateInfo {
    VkStructureType      sType;
    const void*          pNext;
    VkImageUsageFlags    stencilUsage;
} VkImageStencilUsageCreateInfo;

or the equivalent

// Provided by VK_EXT_separate_stencil_usage
typedef VkImageStencilUsageCreateInfo VkImageStencilUsageCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
stencilUsage is a bitmask of VkImageUsageFlagBits describing the intended usage of the stencil aspect of the image.

If the pNext chain of VkImageCreateInfo includes a VkImageStencilUsageCreateInfo structure, then that structure includes the usage flags specific to the stencil aspect of the image for an image with a depth-stencil format.

This structure specifies image usages which only apply to the stencil aspect of a depth/stencil format image. When this structure is included in the pNext chain of VkImageCreateInfo, the stencil aspect of the image must only be used as specified by stencilUsage. When this structure is not included in the pNext chain of VkImageCreateInfo, the stencil aspect of an image must only be used as specified VkImageCreateInfo::usage. Use of other aspects of an image are unaffected by this structure.

This structure can also be included in the pNext chain of VkPhysicalDeviceImageFormatInfo2 to query additional capabilities specific to image creation parameter combinations including a separate set of usage flags for the stencil aspect of the image using vkGetPhysicalDeviceImageFormatProperties2. When this structure is not included in the pNext chain of VkPhysicalDeviceImageFormatInfo2 then the implicit value of stencilUsage matches that of VkPhysicalDeviceImageFormatInfo2::usage.

Valid Usage

If stencilUsage includes VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT, it must not include bits other than VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT or VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_STENCIL_USAGE_CREATE_INFO
stencilUsage must be a valid combination of VkImageUsageFlagBits values
stencilUsage must not be 0

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:

// Provided by VK_NV_dedicated_allocation
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 a structure extending this 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:

// Provided by VK_VERSION_1_1
typedef struct VkExternalMemoryImageCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    VkExternalMemoryHandleTypeFlags    handleTypes;
} VkExternalMemoryImageCreateInfo;

or the equivalent

// Provided by VK_KHR_external_memory
typedef VkExternalMemoryImageCreateInfo VkExternalMemoryImageCreateInfoKHR;

Note

A VkExternalMemoryImageCreateInfo structure must be included in the creation parameters for an image that will be bound to memory that is either exported or imported.

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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:

// Provided by VK_NV_external_memory
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 a structure extending this structure.
handleTypes is a bitmask of VkExternalMemoryHandleTypeFlagBitsNV specifying one or more external memory handle types.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_EXTERNAL_MEMORY_IMAGE_CREATE_INFO_NV
handleTypes must be a valid combination of VkExternalMemoryHandleTypeFlagBitsNV values

To create an image with an external format, add a VkExternalFormatANDROID structure in the pNext chain of VkImageCreateInfo. VkExternalFormatANDROID is defined as:

// Provided by VK_ANDROID_external_memory_android_hardware_buffer
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 a structure extending this 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.

Valid Usage

externalFormat must be 0 or a value returned in the externalFormat member of VkAndroidHardwareBufferFormatPropertiesANDROID by an earlier call to vkGetAndroidHardwareBufferPropertiesANDROID

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:

// Provided by VK_KHR_swapchain with VK_VERSION_1_1, VK_KHR_device_group with VK_KHR_swapchain
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 a structure extending this structure.
swapchain is VK_NULL_HANDLE or a handle of a swapchain that the image will be bound to.

Valid Usage

If swapchain is not VK_NULL_HANDLE, the fields of VkImageCreateInfo must match the implied image creation parameters of the swapchain

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 VkImageFormatListCreateInfo structure, then that structure contains a list of all formats that can be used when creating views of this image.

The VkImageFormatListCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkImageFormatListCreateInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           viewFormatCount;
    const VkFormat*    pViewFormats;
} VkImageFormatListCreateInfo;

or the equivalent

// Provided by VK_KHR_image_format_list
typedef VkImageFormatListCreateInfo VkImageFormatListCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 VkImageFormatListCreateInfo structure were not included in the pNext list of VkImageCreateInfo.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_FORMAT_LIST_CREATE_INFO
If viewFormatCount is not 0, pViewFormats must be a valid pointer to an array of viewFormatCount valid VkFormat values

If the pNext chain of VkImageCreateInfo includes a VkImageDrmFormatModifierListCreateInfoEXT structure, then the image will be created with one of the Linux DRM format modifiers listed in the structure. The choice of modifier is implementation-dependent.

The VkImageDrmFormatModifierListCreateInfoEXT structure is defined as:

// Provided by VK_EXT_image_drm_format_modifier
typedef struct VkImageDrmFormatModifierListCreateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           drmFormatModifierCount;
    const uint64_t*    pDrmFormatModifiers;
} VkImageDrmFormatModifierListCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
drmFormatModifierCount is the length of the pDrmFormatModifiers array.
pDrmFormatModifiers is a pointer to an array of Linux DRM format modifiers.

Valid Usage

Each modifier in pDrmFormatModifiers must be compatible with the parameters in VkImageCreateInfo and its pNext chain, as determined by querying VkPhysicalDeviceImageFormatInfo2 extended with VkPhysicalDeviceImageDrmFormatModifierInfoEXT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_DRM_FORMAT_MODIFIER_LIST_CREATE_INFO_EXT
pDrmFormatModifiers must be a valid pointer to an array of drmFormatModifierCount uint64_t values
drmFormatModifierCount must be greater than 0

If the pNext chain of VkImageCreateInfo includes a VkImageDrmFormatModifierExplicitCreateInfoEXT structure, then the image will be created with the Linux DRM format modifier and memory layout defined by the structure.

The VkImageDrmFormatModifierExplicitCreateInfoEXT structure is defined as:

// Provided by VK_EXT_image_drm_format_modifier
typedef struct VkImageDrmFormatModifierExplicitCreateInfoEXT {
    VkStructureType               sType;
    const void*                   pNext;
    uint64_t                      drmFormatModifier;
    uint32_t                      drmFormatModifierPlaneCount;
    const VkSubresourceLayout*    pPlaneLayouts;
} VkImageDrmFormatModifierExplicitCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
drmFormatModifier is the Linux DRM format modifier with which the image will be created.
drmFormatModifierPlaneCount is the number of memory planes in the image (as reported by VkDrmFormatModifierPropertiesEXT) as well as the length of the pPlaneLayouts array.
pPlaneLayouts is a pointer to an array of VkSubresourceLayout structures describing the image’s memory planes.

The i^th member of pPlaneLayouts describes the layout of the image’s i^th memory plane (that is, VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT). In each element of pPlaneLayouts, the implementation must ignore size. The implementation calculates the size of each plane, which the application can query with vkGetImageSubresourceLayout.

When creating an image with VkImageDrmFormatModifierExplicitCreateInfoEXT, it is the application’s responsibility to satisfy all valid usage requirements. However, the implementation must validate that the provided pPlaneLayouts, when combined with the provided drmFormatModifier and other creation parameters in VkImageCreateInfo and its pNext chain, produce a valid image. (This validation is necessarily implementation-dependent and outside the scope of Vulkan, and therefore not described by valid usage requirements). If this validation fails, then vkCreateImage returns VK_ERROR_INVALID_DRM_FORMAT_MODIFIER_PLANE_LAYOUT_EXT.

Valid Usage

drmFormatModifier must be compatible with the parameters in VkImageCreateInfo and its pNext chain, as determined by querying VkPhysicalDeviceImageFormatInfo2 extended with VkPhysicalDeviceImageDrmFormatModifierInfoEXT
drmFormatModifierPlaneCount must be equal to the VkDrmFormatModifierPropertiesEXT::drmFormatModifierPlaneCount associated with VkImageCreateInfo::format and drmFormatModifier, as found by querying VkDrmFormatModifierPropertiesListEXT
For each element of pPlaneLayouts, size must be 0
For each element of pPlaneLayouts, arrayPitch must be 0 if VkImageCreateInfo::arrayLayers is 1
For each element of pPlaneLayouts, depthPitch must be 0 if VkImageCreateInfo::extent.depth is 1

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_DRM_FORMAT_MODIFIER_EXPLICIT_CREATE_INFO_EXT
If drmFormatModifierPlaneCount is not 0, pPlaneLayouts must be a valid pointer to an array of drmFormatModifierPlaneCount VkSubresourceLayout structures

Bits which can be set in VkImageCreateInfo::usage, specifying intended usage of an image, are:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_NV_shading_rate_image
    VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV = 0x00000100,
  // Provided by VK_EXT_fragment_density_map
    VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT = 0x00000200,
} 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 or depth/stencil resolve 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.
VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV specifies that the image can be used to create a VkImageView suitable for use as a shading rate image.
VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT specifies that the image can be used to create a VkImageView suitable for use as a fragment density map image.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_ALIAS_BIT = 0x00000400,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT = 0x00000040,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT = 0x00000020,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT = 0x00000080,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_EXTENDED_USAGE_BIT = 0x00000100,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_PROTECTED_BIT = 0x00000800,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_CREATE_DISJOINT_BIT = 0x00000200,
  // Provided by VK_NV_corner_sampled_image
    VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV = 0x00002000,
  // Provided by VK_EXT_sample_locations
    VK_IMAGE_CREATE_SAMPLE_LOCATIONS_COMPATIBLE_DEPTH_BIT_EXT = 0x00001000,
  // Provided by VK_EXT_fragment_density_map
    VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT = 0x00004000,
  // Provided by VK_KHR_device_group with VK_KHR_bind_memory2
    VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT_KHR = VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT,
  // Provided by VK_KHR_maintenance1
    VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT_KHR = VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT,
  // Provided by VK_KHR_maintenance2
    VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT_KHR = VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT,
  // Provided by VK_KHR_maintenance2
    VK_IMAGE_CREATE_EXTENDED_USAGE_BIT_KHR = VK_IMAGE_CREATE_EXTENDED_USAGE_BIT,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_IMAGE_CREATE_DISJOINT_BIT_KHR = VK_IMAGE_CREATE_DISJOINT_BIT,
  // Provided by VK_KHR_bind_memory2
    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.
VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV specifies that the image is a corner-sampled image.
VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT specifies that an image can be in a subsampled format which may be more optimal when written as an attachment by a render pass that has a fragment density map attachment. Accessing a subsampled image has additional considerations:
- Image data read as an image sampler will have undefined values if the sampler was not created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT or was not sampled through the use of a combined image sampler with an immutable sampler in VkDescriptorSetLayoutBinding.
- Image data read with an input attachment will have undefined values if the contents were not written as an attachment in an earlier subpass of the same render pass.
- Image data read as an image sampler in the fragment shader will be additionally be read by the device during VK_PIPELINE_STAGE_VERTEX_SHADER_BIT if VkPhysicalDeviceFragmentDensityMap2PropertiesEXT::subsampledCoarseReconstructionEarlyAccess is VK_TRUE and the sampler was created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_COARSE_RECONSTRUCTION_BIT_EXT.
- Image data read with load operations are resampled to the fragment density of the render pass if VkPhysicalDeviceFragmentDensityMap2PropertiesEXT::subsampledLoads is VK_TRUE. Otherwise, values of image data are undefined.
- Image contents outside of the render area take on undefined values if the image is stored as a render pass attachment.

See Sparse Resource Features and Sparse Physical Device Features for more details.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 texel blocks in an image, are:

// Provided by VK_VERSION_1_0
typedef enum VkImageTiling {
    VK_IMAGE_TILING_OPTIMAL = 0,
    VK_IMAGE_TILING_LINEAR = 1,
  // Provided by VK_EXT_image_drm_format_modifier
    VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT = 1000158000,
} 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).
VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT indicates that the image’s tiling is defined by a Linux DRM format modifier. The modifier is specified at image creation with VkImageDrmFormatModifierListCreateInfoEXT or VkImageDrmFormatModifierExplicitCreateInfoEXT, and can be queried with vkGetImageDrmFormatModifierPropertiesEXT.

To query the memory layout of an image subresource, call:

// Provided by VK_VERSION_1_0
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 is a pointer to a VkSubresourceLayout structure in which the layout is returned.

If the image is linear, then the returned layout is valid for host access.

If the image’s tiling is VK_IMAGE_TILING_LINEAR and its format is a multi-planar format, then vkGetImageSubresourceLayout describes one format plane of the image. If the image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then vkGetImageSubresourceLayout describes one memory plane of the image. If the image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT and the image is non-linear, then the returned layout has an implementation-dependent meaning; the vendor of the image’s DRM format modifier may provide documentation that explains how to interpret the returned layout.

vkGetImageSubresourceLayout is invariant for the lifetime of a single image. However, the subresource layout of images in Android hardware buffer external memory is not known until the image has been bound to memory, so applications must not call vkGetImageSubresourceLayout for such an image before it has been bound.

Valid Usage

image must have been created with tiling equal to VK_IMAGE_TILING_LINEAR or VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT
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 tiling of the image is VK_IMAGE_TILING_LINEAR and its format 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 tiling of the image is VK_IMAGE_TILING_LINEAR and its format 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
If the tiling of the image is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then the aspectMask member of pSubresource must be VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT and the index i must be less than the VkDrmFormatModifierPropertiesEXT::drmFormatModifierPlaneCount associated with the image’s format and VkImageDrmFormatModifierPropertiesEXT::drmFormatModifier

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:

// Provided by VK_VERSION_1_0
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)

aspectMask must be a valid combination of VkImageAspectFlagBits values
aspectMask must not be 0

Information about the layout of the image subresource is returned in a VkSubresourceLayout structure:

// Provided by VK_VERSION_1_0
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 or the plane 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.

If the image is linear, then 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;

The value of arrayPitch is undefined for images that were not created as arrays. depthPitch is defined only for 3D images.

If the image has a single-plane color format and its tiling is VK_IMAGE_TILING_LINEAR , then the aspectMask member of VkImageSubresource must be VK_IMAGE_ASPECT_COLOR_BIT.

If the image has a depth/stencil format and its tiling is VK_IMAGE_TILING_LINEAR , then 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.

If the image has a multi-planar format and its tiling is VK_IMAGE_TILING_LINEAR , then 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. If the image is disjoint, then the offset is relative to the base address of the plane. If the image is non-disjoint, then the offset is relative to the base address of the image.

If the image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then the aspectMask member of VkImageSubresource must be one of VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT, where the maximum allowed plane index i is defined by the VkDrmFormatModifierPropertiesEXT::drmFormatModifierPlaneCount associated with the image’s VkImageCreateInfo::format and modifier. The memory range used by the subresource is described by offset and size. If the image is disjoint, then the offset is relative to the base address of the memory plane. If the image is non-disjoint, then the offset is relative to the base address of the image. If the image is non-linear, then rowPitch, arrayPitch, and depthPitch have an implementation-dependent meaning.

If an image was created with VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then the image has a Linux DRM format modifier. To query the modifier, call:

// Provided by VK_EXT_image_drm_format_modifier
VkResult vkGetImageDrmFormatModifierPropertiesEXT(
    VkDevice                                    device,
    VkImage                                     image,
    VkImageDrmFormatModifierPropertiesEXT*      pProperties);

device is the logical device that owns the image.
image is the queried image.
pProperties will return properties of the image’s DRM format modifier.

Valid Usage

image must have been created with tiling equal to VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT

Valid Usage (Implicit)

device must be a valid VkDevice handle
image must be a valid VkImage handle
pProperties must be a valid pointer to a VkImageDrmFormatModifierPropertiesEXT structure
image must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

The VkImageDrmFormatModifierPropertiesEXT structure is defined as:

// Provided by VK_EXT_image_drm_format_modifier
typedef struct VkImageDrmFormatModifierPropertiesEXT {
    VkStructureType    sType;
    void*              pNext;
    uint64_t           drmFormatModifier;
} VkImageDrmFormatModifierPropertiesEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
drmFormatModifier returns the image’s Linux DRM format modifier.

If the image was created with VkImageDrmFormatModifierListCreateInfoEXT, then the returned drmFormatModifier must belong to the list of modifiers provided at time of image creation in VkImageDrmFormatModifierListCreateInfoEXT::pDrmFormatModifiers. If the image was created with VkImageDrmFormatModifierExplicitCreateInfoEXT, then the returned drmFormatModifier must be the modifier provided at time of image creation in VkImageDrmFormatModifierExplicitCreateInfoEXT::drmFormatModifier.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_DRM_FORMAT_MODIFIER_PROPERTIES_EXT
pNext must be NULL

To destroy an image, call:

// Provided by VK_VERSION_1_0
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

11.3.1. Image Format Features

Valid uses of a VkImage may depend on the image’s format features, defined below. Such constraints are documented in the affected valid usage statement.

If the image was created with VK_IMAGE_TILING_LINEAR, then its set of format features is the value of VkFormatProperties::linearTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties on the same format as VkImageCreateInfo::format.
If the image was created with VK_IMAGE_TILING_OPTIMAL, but without an Android hardware buffer external format, then its set of format features is the value of VkFormatProperties::optimalTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties on the same format as VkImageCreateInfo::format.
If the image was created with an Android hardware buffer external format, then its set of format features is the value of VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures found by calling vkGetAndroidHardwareBufferPropertiesANDROID on the Android hardware buffer that was imported to the VkDeviceMemory to which the image is bound.
If the image was created with VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then:
- The image’s DRM format modifier is the value of VkImageDrmFormatModifierListCreateInfoEXT::drmFormatModifier found by calling vkGetImageDrmFormatModifierPropertiesEXT.
- Let VkDrmFormatModifierPropertiesListEXT::pDrmFormatModifierProperties be the array found by calling vkGetPhysicalDeviceFormatProperties2 on the same format as VkImageCreateInfo::format.
- Let VkDrmFormatModifierPropertiesEXT prop be an array element whose drmFormatModifier member is the value of the image’s DRM format modifier.
- Then the image set of format features is the value of taking the bitwise intersection over the collected prop::drmFormatModifierTilingFeatures.

11.3.2. Corner-Sampled Images

A corner-sampled image is an image where unnormalized texel coordinates are centered on integer values rather than half-integer values.

A corner-sampled image has a number of differences compared to conventional texture image:

Texels are centered on integer coordinates. See Unnormalized Texel Coordinate Operations
Normalized coordinates are scaled using coord × (dim - 1) rather than coord × dim, where dim is the size of one dimension of the image. See normalized texel coordinate transform.
Partial derivatives are scaled using coord × (dim - 1) rather than coord × dim. See Scale Factor Operation.
Calculation of the next higher lod size goes according to ⌈dim / 2⌉ rather than ⌊dim / 2⌋. See Image Miplevel Sizing.
The minimum level size is 2x2 for 2D images and 2x2x2 for 3D images. See Image Miplevel Sizing.

Corner-sampling is only supported for 2D and 3D images. When sampling a corner-sampled image, the sampler addressing mode must be VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE. Corner-sampled images are not supported as cubemaps or depth/stencil images.

11.3.3. Image Miplevel Sizing

A complete mipmap chain is the full set of miplevels, from the largest miplevel provided, down to the minimum miplevel size.

Conventional Images

For conventional images, the dimensions of each successive miplevel, n+1, are:

: width_n+1 = max(⌊width_n/2⌋, 1)
: height_n+1 = max(⌊height_n/2⌋, 1)
: depth_n+1 = max(⌊depth_n/2⌋, 1)

where width_n, height_n, and depth_n are the dimensions of the next larger miplevel, n.

The minimum miplevel size is:

1 for one-dimensional images,
1x1 for two-dimensional images, and
1x1x1 for three-dimensional images.

The number of levels in a complete mipmap chain is:

: ⌊log₂(max(width₀, height₀, depth₀))⌋ + 1

where width₀, height₀, and depth₀ are the dimensions of the largest (most detailed) miplevel, 0.

Corner-Sampled Images

For corner-sampled images, the dimensions of each successive miplevel, n+1, are:

: width_n+1 = max(⌈width_n/2⌉, 2)
: height_n+1 = max(⌈height_n/2⌉, 2)
: depth_n+1 = max(⌈depth_n/2⌉, 2)

where width_n, height_n, and depth_n are the dimensions of the next larger miplevel, n.

The minimum miplevel size is:

2x2 for two-dimensional images, and
2x2x2 for three-dimensional images.

The number of levels in a complete mipmap chain is:

: ⌈log₂(max(width₀, height₀, depth₀))⌉

where width₀, height₀, and depth₀ are the dimensions of the largest (most detailed) miplevel, 0.

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 linear images 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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL = 1000117000,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL = 1000117001,
  // Provided by VK_VERSION_1_2
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL = 1000241000,
  // Provided by VK_VERSION_1_2
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL = 1000241001,
  // Provided by VK_VERSION_1_2
    VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL = 1000241002,
  // Provided by VK_VERSION_1_2
    VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL = 1000241003,
  // Provided by VK_KHR_swapchain
    VK_IMAGE_LAYOUT_PRESENT_SRC_KHR = 1000001002,
  // Provided by VK_KHR_shared_presentable_image
    VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR = 1000111000,
  // Provided by VK_NV_shading_rate_image
    VK_IMAGE_LAYOUT_SHADING_RATE_OPTIMAL_NV = 1000164003,
  // Provided by VK_EXT_fragment_density_map
    VK_IMAGE_LAYOUT_FRAGMENT_DENSITY_MAP_OPTIMAL_EXT = 1000218000,
  // Provided by VK_KHR_maintenance2
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL,
  // Provided by VK_KHR_maintenance2
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL,
  // Provided by VK_KHR_separate_depth_stencil_layouts
    VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL,
  // Provided by VK_KHR_separate_depth_stencil_layouts
    VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL_KHR = VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL,
  // Provided by VK_KHR_separate_depth_stencil_layouts
    VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL_KHR = VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL,
  // Provided by VK_KHR_separate_depth_stencil_layouts
    VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL_KHR = VK_IMAGE_LAYOUT_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 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 specifies a layout for both the depth and stencil aspects of a depth/stencil format image allowing read and write access as a depth/stencil attachment. It is equivalent to VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL and VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL.
VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL specifies a layout for both the depth and stencil aspects of a depth/stencil format image allowing read only access as a depth/stencil attachment or in shaders. It is equivalent to VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL and VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL.
VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL specifies a layout for depth/stencil format images allowing read and write access to the stencil aspect as a stencil attachment, and read only access to the depth aspect as a depth attachment or in shaders. It is equivalent to VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL and VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL.
VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL specifies a layout for depth/stencil format images allowing read and write access to the depth aspect as a depth attachment, and read only access to the stencil aspect as a stencil attachment or in shaders. It is equivalent to VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL and VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL.
VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_OPTIMAL specifies a layout for the depth aspect of a depth/stencil format image allowing read and write access as a depth attachment.
VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_OPTIMAL specifies a layout for the depth aspect of a depth/stencil format image allowing read-only access as a depth attachment or in shaders.
VK_IMAGE_LAYOUT_STENCIL_ATTACHMENT_OPTIMAL specifies a layout for the stencil aspect of a depth/stencil format image allowing read and write access as a stencil attachment.
VK_IMAGE_LAYOUT_STENCIL_READ_ONLY_OPTIMAL specifies a layout for the stencil aspect of a depth/stencil format image allowing read-only access as a stencil attachment or in shaders.
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.
VK_IMAGE_LAYOUT_SHADING_RATE_OPTIMAL_NV must only be used as a read-only shading-rate-image. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV usage bit enabled.
VK_IMAGE_LAYOUT_FRAGMENT_DENSITY_MAP_OPTIMAL_EXT must only be used as a fragment density map attachment in a VkRenderPass. This layout is valid only for image subresources of images created with the VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT usage bit enabled.

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).

11.4.1. Image Layout Matching Rules

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 exactly the layout specified via the API controlling those accesses , except in case of accesses to an image with a depth/stencil format performed through descriptors referring to only a single aspect of the image, where the following relaxed matching rules apply:

Descriptors referring just to the depth aspect of a depth/stencil image only need to match in the image layout of the depth aspect, thus VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL and VK_IMAGE_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_OPTIMAL are considered to match.
Descriptors referring just to the stencil aspect of a depth/stencil image only need to match in the image layout of the stencil aspect, thus VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL and VK_IMAGE_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_OPTIMAL are considered to match .

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 return undefined values.

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkImageView)

The types of image views that can be created are:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 a 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 is a pointer to a VkImageView handle in which the resulting image view object is returned.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkImageViewCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkImageViewCreateFlagBits describing additional parameters of the image view.
image is a VkImage on which the view will be created.
viewType is a VkImageViewType value specifying the type of the image view.
format is a VkFormat describing the format and type used to interpret texel blocks 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.

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. If the image was has a depth-stencil format and was created with a VkImageStencilUsageCreateInfo structure included in the pNext chain of VkImageCreateInfo, the usage is calculated based on the subresource.aspectMask provided:

If aspectMask includes only VK_IMAGE_ASPECT_STENCIL_BIT, the implicit usage is equal to VkImageStencilUsageCreateInfo::stencilUsage.
If aspectMask includes only VK_IMAGE_ASPECT_DEPTH_BIT, the implicit usage is equal to VkImageCreateInfo::usage.
If both aspects are included in aspectMask, the implicit usage is equal to the intersection of VkImageCreateInfo::usage and VkImageStencilUsageCreateInfo::stencilUsage. The implicit usage can be overriden by adding a VkImageViewUsageCreateInfo structure to the pNext chain.

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 -2^b may be changed to -2^b + 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.

The VkComponentMapping components member describes a remapping from components of the image to components of the vector returned by shader image instructions. This remapping must be the identity swizzle for storage image descriptors, input attachment descriptors, framebuffer attachments, and any VkImageView used with a combined image sampler that enables sampler Y’C_BC_R conversion.

If the image view is to be used with a sampler which supports sampler Y′C_BC_R conversion, an identically defined object of type VkSamplerYcbcrConversion to that used to create the sampler must be passed to vkCreateImageView in a VkSamplerYcbcrConversionInfo included in the pNext chain of VkImageViewCreateInfo. Conversely, if a VkSamplerYcbcrConversion object is passed to vkCreateImageView, an identically defined VkSamplerYcbcrConversion object must be used when sampling the image.

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′C_BC_R 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′C_BC_R 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 (u_plane, v_plane) unnormalized coordinates of the reduced-resolution plane, and these coordinates access the same memory locations as the (u_color, v_color) unnormalized coordinates of the color aspect for which chroma reconstruction operations operate on the same (u_plane, v_plane) or (i_plane, j_plane) 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
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, VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, or VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT
The format features of the resultant image view must contain at least one bit
If usage contains VK_IMAGE_USAGE_SAMPLED_BIT, then the format features of the resultant image view must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT
If usage contains VK_IMAGE_USAGE_STORAGE_BIT, then the image view’s format features must contain VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT
If usage contains VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, then the image view’s format features must contain VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
If usage contains VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, then the image view’s format features must contain VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
If usage contains VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT, then the image view’s format features must contain at least one of VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT or VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT
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 was created with usage containing VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT, subresourceRange.levelCount must be 1
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 depth computed from baseMipLevel and extent.depth specified in VkImageCreateInfo when image was created, according to the formula defined in Image Miplevel Sizing
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 depth computed from baseMipLevel and extent.depth specified in VkImageCreateInfo when image was created, according to the formula defined in Image Miplevel Sizing
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 VkImageFormatListCreateInfo structure was included in the pNext chain of the VkImageCreateInfo structure used when creating image and VkImageFormatListCreateInfo::viewFormatCount is not zero then format must be one of the formats in VkImageFormatListCreateInfo::pViewFormats
If a VkImageFormatListCreateInfo structure was included in the pNext chain of the VkImageCreateInfo structure used when creating image and VkImageFormatListCreateInfo::viewFormatCount is not zero then all of the formats in VkImageFormatListCreateInfo::pViewFormats must be compatible with the format as described in the compatibility table
If flags dose not contain VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT and the pNext chain include a VkImageFormatListCreateInfo structure then VkImageFormatListCreateInfo::viewFormatCount must be 0 or 1
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 includes a VkSamplerYcbcrConversionInfo structure with a conversion value other than VK_NULL_HANDLE, all members of components must have the identity swizzle
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
If image has an external format, the pNext chain must include a VkSamplerYcbcrConversionInfo structure with a conversion object created with the same external format as image
If image has an external format, all members of components must be the identity swizzle
If image was created with usage containing VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, viewType must be VK_IMAGE_VIEW_TYPE_2D or VK_IMAGE_VIEW_TYPE_2D_ARRAY
If image was created with usage containing VK_IMAGE_USAGE_SHADING_RATE_IMAGE_BIT_NV, format must be VK_FORMAT_R8_UINT
If dynamic fragment density map feature is not enabled, flags must not contain VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT
If deferred fragment density map feature is not enabled, flags must not contain VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DEFERRED_BIT_EXT
If flags contains VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DEFERRED_BIT_EXT, flags must not contain VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT
If image was created with flags containing VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT and usage containing VK_IMAGE_USAGE_SAMPLED_BIT, subresourceRange.layerCount must be less than or equal to VkPhysicalDeviceFragmentDensityMap2PropertiesEXT::maxSubsampledArrayLayers
If flags does not contain VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT and image was created with usage containing VK_IMAGE_USAGE_FRAGMENT_DENSITY_MAP_BIT_EXT, its flags must not contain any of VK_IMAGE_CREATE_PROTECTED_BIT, VK_IMAGE_CREATE_SPARSE_BINDING_BIT, VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT, or VK_IMAGE_CREATE_SPARSE_ALIASED_BIT
If the pNext chain includes a VkImageViewUsageCreateInfo structure, and image was not created with a VkImageStencilUsageCreateInfo structure included in the pNext chain of VkImageCreateInfo, its usage member must not include any bits that were not set in the usage member of the VkImageCreateInfo structure used to create image
If the pNext chain includes a VkImageViewUsageCreateInfo structure, image was created with a VkImageStencilUsageCreateInfo structure included in the pNext chain of VkImageCreateInfo, and subResourceRange.aspectMask includes VK_IMAGE_ASPECT_STENCIL_BIT, the usage member of the VkImageViewUsageCreateInfo instance must not include any bits that were not set in the usage member of the VkImageStencilUsageCreateInfo structure used to create image
If the pNext chain includes a VkImageViewUsageCreateInfo structure, image was created with a VkImageStencilUsageCreateInfo structure included in the pNext chain of VkImageCreateInfo, and subResourceRange.aspectMask includes bits other than VK_IMAGE_ASPECT_STENCIL_BIT, the usage member of the VkImageViewUsageCreateInfo structure must not include any bits that were not set in the usage member of the VkImageCreateInfo structure used to create image
If viewType is VK_IMAGE_VIEW_TYPE_CUBE and subresourceRange.layerCount is not VK_REMAINING_ARRAY_LAYERS, subresourceRange.layerCount must be 6
If viewType is VK_IMAGE_VIEW_TYPE_CUBE_ARRAY and subresourceRange.layerCount is not VK_REMAINING_ARRAY_LAYERS, subresourceRange.layerCount must be a multiple of 6
If viewType is VK_IMAGE_VIEW_TYPE_CUBE and subresourceRange.layerCount is VK_REMAINING_ARRAY_LAYERS, the remaining number of layers must be 6
If viewType is VK_IMAGE_VIEW_TYPE_CUBE_ARRAY and subresourceRange.layerCount is VK_REMAINING_ARRAY_LAYERS, the remaining number of layers must be a multiple of 6

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 VkImageViewASTCDecodeModeEXT, VkImageViewUsageCreateInfo, or VkSamplerYcbcrConversionInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkImageViewCreateFlagBits values
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

Bits which can be set in VkImageViewCreateInfo::flags, specifying additional parameters of an image, are:

// Provided by VK_VERSION_1_0
typedef enum VkImageViewCreateFlagBits {
  // Provided by VK_EXT_fragment_density_map
    VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT = 0x00000001,
  // Provided by VK_EXT_fragment_density_map2
    VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DEFERRED_BIT_EXT = 0x00000002,
} VkImageViewCreateFlagBits;

VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DYNAMIC_BIT_EXT specifies that the fragment density map will be read by device during VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
VK_IMAGE_VIEW_CREATE_FRAGMENT_DENSITY_MAP_DEFERRED_BIT_EXT specifies that the fragment density map will be read by the host during VkEndCommandBuffer for the primary command buffer that the render pass is recorded into

// Provided by VK_VERSION_1_0
typedef VkFlags VkImageViewCreateFlags;

VkImageViewCreateFlags is a bitmask type for setting a mask of zero or more VkImageViewCreateFlagBits.

The set of usages for the created image view can be restricted compared to the parent image’s usage flags by adding a VkImageViewUsageCreateInfo structure to the pNext chain of VkImageViewCreateInfo.

The VkImageViewUsageCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkImageViewUsageCreateInfo {
    VkStructureType      sType;
    const void*          pNext;
    VkImageUsageFlags    usage;
} VkImageViewUsageCreateInfo;

or the equivalent

// Provided by VK_KHR_maintenance2
typedef VkImageViewUsageCreateInfo VkImageViewUsageCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 (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:

// Provided by VK_VERSION_1_0
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, the last cube map in the array must not be accessed.

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 image view 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 image view 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 image view 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.

When creating a VkImageView, if sampler Y′C_BC_R 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′C_BC_R 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
aspectMask must not include VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT for any index i

Valid Usage (Implicit)

aspectMask must be a valid combination of VkImageAspectFlagBits values
aspectMask must not be 0

Bits which can be set in an aspect mask to specify aspects of an image for purposes such as identifying a subresource, are:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_ASPECT_PLANE_0_BIT = 0x00000010,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_ASPECT_PLANE_1_BIT = 0x00000020,
  // Provided by VK_VERSION_1_1
    VK_IMAGE_ASPECT_PLANE_2_BIT = 0x00000040,
  // Provided by VK_EXT_image_drm_format_modifier
    VK_IMAGE_ASPECT_MEMORY_PLANE_0_BIT_EXT = 0x00000080,
  // Provided by VK_EXT_image_drm_format_modifier
    VK_IMAGE_ASPECT_MEMORY_PLANE_1_BIT_EXT = 0x00000100,
  // Provided by VK_EXT_image_drm_format_modifier
    VK_IMAGE_ASPECT_MEMORY_PLANE_2_BIT_EXT = 0x00000200,
  // Provided by VK_EXT_image_drm_format_modifier
    VK_IMAGE_ASPECT_MEMORY_PLANE_3_BIT_EXT = 0x00000400,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_IMAGE_ASPECT_PLANE_0_BIT_KHR = VK_IMAGE_ASPECT_PLANE_0_BIT,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_IMAGE_ASPECT_PLANE_1_BIT_KHR = VK_IMAGE_ASPECT_PLANE_1_BIT,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    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.
VK_IMAGE_ASPECT_PLANE_0_BIT specifies plane 0 of a multi-planar image format.
VK_IMAGE_ASPECT_PLANE_1_BIT specifies plane 1 of a multi-planar image format.
VK_IMAGE_ASPECT_PLANE_2_BIT specifies plane 2 of a multi-planar image format.
VK_IMAGE_ASPECT_MEMORY_PLANE_0_BIT_EXT specifies memory plane 0.
VK_IMAGE_ASPECT_MEMORY_PLANE_1_BIT_EXT specifies memory plane 1.
VK_IMAGE_ASPECT_MEMORY_PLANE_2_BIT_EXT specifies memory plane 2.
VK_IMAGE_ASPECT_MEMORY_PLANE_3_BIT_EXT specifies memory plane 3.

// Provided by VK_VERSION_1_0
typedef VkFlags VkImageAspectFlags;

VkImageAspectFlags is a bitmask type for setting a mask of zero or more VkImageAspectFlagBits.

The VkComponentMapping structure is defined as:

// Provided by VK_VERSION_1_0
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)

r must be a valid VkComponentSwizzle value
g must be a valid VkComponentSwizzle value
b must be a valid VkComponentSwizzle value
a must be a valid VkComponentSwizzle value

Possible values of the members of VkComponentMapping, specifying the component values placed in each component of the output vector, are:

// Provided by VK_VERSION_1_0
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`

If the pNext list includes a VkImageViewASTCDecodeModeEXT structure, then that structure includes a parameter specifying the decode mode for image views using ASTC compressed formats.

The VkImageViewASTCDecodeModeEXT structure is defined as:

// Provided by VK_EXT_astc_decode_mode
typedef struct VkImageViewASTCDecodeModeEXT {
    VkStructureType    sType;
    const void*        pNext;
    VkFormat           decodeMode;
} VkImageViewASTCDecodeModeEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
decodeMode is the intermediate format used to decode ASTC compressed formats.

Valid Usage

decodeMode must be one of VK_FORMAT_R16G16B16A16_SFLOAT, VK_FORMAT_R8G8B8A8_UNORM, or VK_FORMAT_E5B9G9R9_UFLOAT_PACK32
If the decodeModeSharedExponent feature is not enabled, decodeMode must not be VK_FORMAT_E5B9G9R9_UFLOAT_PACK32
If decodeMode is VK_FORMAT_R8G8B8A8_UNORM the image view must not include blocks using any of the ASTC HDR modes
format of the image view must be one of the ASTC Compressed Image Formats

If format uses sRGB encoding then the decodeMode has no effect.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_VIEW_ASTC_DECODE_MODE_EXT
decodeMode must be a valid VkFormat value

To destroy an image view, call:

// Provided by VK_VERSION_1_0
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

To get the handle for an image view, call:

// Provided by VK_NVX_image_view_handle
uint32_t vkGetImageViewHandleNVX(
    VkDevice                                    device,
    const VkImageViewHandleInfoNVX*             pInfo);

device is the logical device that owns the image view.
pInfo describes the image view to query and type of handle.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkImageViewHandleInfoNVX structure

The VkImageViewHandleInfoNVX structure is defined as:

// Provided by VK_NVX_image_view_handle
typedef struct VkImageViewHandleInfoNVX {
    VkStructureType     sType;
    const void*         pNext;
    VkImageView         imageView;
    VkDescriptorType    descriptorType;
    VkSampler           sampler;
} VkImageViewHandleInfoNVX;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
imageView is the image view to query.
descriptorType is the type of descriptor for which to query a handle.
sampler is the sampler to combine with the image view when generating the handle.

Valid Usage

descriptorType must be VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE, VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER
sampler must be a valid VkSampler if descriptorType is VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER
If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE or VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, the image that imageView was created from must have been created with the VK_IMAGE_USAGE_SAMPLED_BIT usage bit set
If descriptorType is VK_DESCRIPTOR_TYPE_STORAGE_IMAGE, the image that imageView was created from must have been created with the VK_IMAGE_USAGE_STORAGE_BIT usage bit set

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_VIEW_HANDLE_INFO_NVX
pNext must be NULL
imageView must be a valid VkImageView handle
descriptorType must be a valid VkDescriptorType value
If sampler is not VK_NULL_HANDLE, sampler must be a valid VkSampler handle
Both of imageView, and sampler that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

To get the device address for an image view, call:

// Provided by VK_NVX_image_view_handle
VkResult vkGetImageViewAddressNVX(
    VkDevice                                    device,
    VkImageView                                 imageView,
    VkImageViewAddressPropertiesNVX*            pProperties);

device is the logical device that owns the image view.
imageView is a handle to the image view.
pProperties contains the device address and size when the call returns.

Valid Usage (Implicit)

device must be a valid VkDevice handle
imageView must be a valid VkImageView handle
pProperties must be a valid pointer to a VkImageViewAddressPropertiesNVX structure
imageView must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_UNKNOWN

The VkImageViewAddressPropertiesNVX structure is defined as:

// Provided by VK_NVX_image_view_handle
typedef struct VkImageViewAddressPropertiesNVX {
    VkStructureType    sType;
    void*              pNext;
    VkDeviceAddress    deviceAddress;
    VkDeviceSize       size;
} VkImageViewAddressPropertiesNVX;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
deviceAddress is the device address of the image view.
size is the size in bytes of the image view device memory.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_IMAGE_VIEW_ADDRESS_PROPERTIES_NVX
pNext must be NULL

11.5.1. Image View Format Features

Valid uses of a VkImageView may depend on the image view’s format features, defined below. Such constraints are documented in the affected valid usage statement.

If VkImageViewCreateInfo::image was created with VK_IMAGE_TILING_LINEAR, then the image view’s set of format features is the value of VkFormatProperties::linearTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties on the same format as VkImageViewCreateInfo::format.
If VkImageViewCreateInfo::image was created with VK_IMAGE_TILING_OPTIMAL, but without an Android hardware buffer external format, then the image view’s set of format features is the value of VkFormatProperties::optimalTilingFeatures found by calling vkGetPhysicalDeviceFormatProperties on the same format as VkImageViewCreateInfo::format.
If VkImageViewCreateInfo::image was created with an Android hardware buffer external format, then the image views’s set of format features is the value of VkAndroidHardwareBufferFormatPropertiesANDROID::formatFeatures found by calling vkGetAndroidHardwareBufferPropertiesANDROID on the Android hardware buffer that was imported to the VkDeviceMemory to which the VkImageViewCreateInfo::image is bound.
If VkImageViewCreateInfo::image was created with VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then:
- The image’s DRM format modifier is the value of VkImageDrmFormatModifierListCreateInfoEXT::drmFormatModifier found by calling vkGetImageDrmFormatModifierPropertiesEXT.
- Let VkDrmFormatModifierPropertiesListEXT::pDrmFormatModifierProperties be the array found by calling vkGetPhysicalDeviceFormatProperties2 on the same format as VkImageViewCreateInfo::format.
- Let VkDrmFormatModifierPropertiesEXT prop be an array element whose drmFormatModifier member is the value of the image’s DRM format modifier.
- Then the image view’s set of format features is the value of taking the bitwise intersection, over the collected prop::drmFormatModifierTilingFeatures.

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:

// Provided by VK_VERSION_1_0
void vkGetBufferMemoryRequirements(
    VkDevice                                    device,
    VkBuffer                                    buffer,
    VkMemoryRequirements*                       pMemoryRequirements);

device is the logical device that owns the buffer.
buffer is the buffer to query.
pMemoryRequirements is a pointer to a 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:

// Provided by VK_VERSION_1_0
void vkGetImageMemoryRequirements(
    VkDevice                                    device,
    VkImage                                     image,
    VkMemoryRequirements*                       pMemoryRequirements);

device is the logical device that owns the image.
image is the image to query.
pMemoryRequirements is a pointer to a 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
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
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:

// Provided by VK_VERSION_1_0
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 applications must not call vkGetImageMemoryRequirements or vkGetImageMemoryRequirements2 with such an VkImage before it has been bound to memory. When importing Android hardware buffer memory, the allocationSize can be determined by calling vkGetAndroidHardwareBufferPropertiesANDROID. When allocating new memory for a VkImage that can be exported to an 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.

When allocating new memory for a VkBuffer that can be exported to an Android hardware buffer an application may still call vkGetBufferMemoryRequirements or vkGetBufferMemoryRequirements2 with VkBuffer before it has been bound to memory.

If the resource being queried was created with the 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 external memory handle type, the value of size has no meaning and should be ignored.

The implementation guarantees certain properties about the memory requirements returned by vkGetBufferMemoryRequirements2, vkGetImageMemoryRequirements2, 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 linear image, 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.
The alignment member satisfies the buffer descriptor offset alignment requirements associated with the VkBuffer’s usage:
- If usage included VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT or VK_BUFFER_USAGE_STORAGE_TEXEL_BUFFER_BIT, alignment must be an integer multiple of VkPhysicalDeviceLimits::minTexelBufferOffsetAlignment.
- If usage included VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT, alignment must be an integer multiple of VkPhysicalDeviceLimits::minUniformBufferOffsetAlignment.
- If usage included VK_BUFFER_USAGE_STORAGE_BUFFER_BIT, alignment must be an integer multiple of VkPhysicalDeviceLimits::minStorageBufferOffsetAlignment.
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 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:

// Provided by VK_VERSION_1_1
void vkGetBufferMemoryRequirements2(
    VkDevice                                    device,
    const VkBufferMemoryRequirementsInfo2*      pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

or the equivalent command

// Provided by VK_KHR_get_memory_requirements2
void vkGetBufferMemoryRequirements2KHR(
    VkDevice                                    device,
    const VkBufferMemoryRequirementsInfo2*      pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

device is the logical device that owns the buffer.
pInfo is a pointer to a VkBufferMemoryRequirementsInfo2 structure containing parameters required for the memory requirements query.
pMemoryRequirements is a pointer to a 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:

// Provided by VK_VERSION_1_1
typedef struct VkBufferMemoryRequirementsInfo2 {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           buffer;
} VkBufferMemoryRequirementsInfo2;

or the equivalent

// Provided by VK_KHR_get_memory_requirements2
typedef VkBufferMemoryRequirementsInfo2 VkBufferMemoryRequirementsInfo2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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:

// Provided by VK_VERSION_1_1
void vkGetImageMemoryRequirements2(
    VkDevice                                    device,
    const VkImageMemoryRequirementsInfo2*       pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

or the equivalent command

// Provided by VK_KHR_get_memory_requirements2
void vkGetImageMemoryRequirements2KHR(
    VkDevice                                    device,
    const VkImageMemoryRequirementsInfo2*       pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

device is the logical device that owns the image.
pInfo is a pointer to a VkImageMemoryRequirementsInfo2 structure containing parameters required for the memory requirements query.
pMemoryRequirements is a pointer to a 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:

// Provided by VK_VERSION_1_1
typedef struct VkImageMemoryRequirementsInfo2 {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
} VkImageMemoryRequirementsInfo2;

or the equivalent

// Provided by VK_KHR_get_memory_requirements2
typedef VkImageMemoryRequirementsInfo2 VkImageMemoryRequirementsInfo2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
image is the image to query.

Valid Usage

If image was created with a multi-planar format and the VK_IMAGE_CREATE_DISJOINT_BIT flag, there must be a VkImagePlaneMemoryRequirementsInfo included in the pNext chain of the VkImageMemoryRequirementsInfo2 structure
If image was created with VK_IMAGE_CREATE_DISJOINT_BIT and with VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then there must be a VkImagePlaneMemoryRequirementsInfo included in the pNext chain of the VkImageMemoryRequirementsInfo2 structure
If image was not created with the VK_IMAGE_CREATE_DISJOINT_BIT flag, there must not be a VkImagePlaneMemoryRequirementsInfo included in the pNext chain of the VkImageMemoryRequirementsInfo2 structure
If image was created with a single-plane format and with any tiling other than VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then there must not be a VkImagePlaneMemoryRequirementsInfo included in the pNext chain of the VkImageMemoryRequirementsInfo2 structure
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)

sType must be VK_STRUCTURE_TYPE_IMAGE_MEMORY_REQUIREMENTS_INFO_2
pNext must be NULL or a pointer to a valid instance of VkImagePlaneMemoryRequirementsInfo
The sType value of each struct in the pNext chain must be unique
image must be a valid VkImage handle

To determine the memory requirements for a plane of a disjoint image, add a VkImagePlaneMemoryRequirementsInfo structure to the pNext chain of the VkImageMemoryRequirementsInfo2 structure.

The VkImagePlaneMemoryRequirementsInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkImagePlaneMemoryRequirementsInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkImageAspectFlagBits    planeAspect;
} VkImagePlaneMemoryRequirementsInfo;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
typedef VkImagePlaneMemoryRequirementsInfo VkImagePlaneMemoryRequirementsInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
planeAspect is the aspect corresponding to the image plane to query.

Valid Usage

If the image’s tiling is VK_IMAGE_TILING_LINEAR or VK_IMAGE_TILING_OPTIMAL, then planeAspect must be a single valid format plane for the image (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)
If the image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then planeAspect must be a single valid memory plane for the image (that is, aspectMask must specify a plane index that is less than the VkDrmFormatModifierPropertiesEXT::drmFormatModifierPlaneCount associated with the image’s format and VkImageDrmFormatModifierPropertiesEXT::drmFormatModifier)

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:

// Provided by VK_VERSION_1_1
typedef struct VkMemoryRequirements2 {
    VkStructureType         sType;
    void*                   pNext;
    VkMemoryRequirements    memoryRequirements;
} VkMemoryRequirements2;

or the equivalent

// Provided by VK_KHR_get_memory_requirements2, VK_NV_ray_tracing
typedef VkMemoryRequirements2 VkMemoryRequirements2KHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
memoryRequirements is a VkMemoryRequirements structure describing the memory requirements of the resource.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_MEMORY_REQUIREMENTS_2
pNext must be NULL or a pointer to a valid instance of VkMemoryDedicatedRequirements
The sType value of each struct in the pNext chain must be unique

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:

// Provided by VK_VERSION_1_1
typedef struct VkMemoryDedicatedRequirements {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           prefersDedicatedAllocation;
    VkBool32           requiresDedicatedAllocation;
} VkMemoryDedicatedRequirements;

or the equivalent

// Provided by VK_KHR_dedicated_allocation
typedef VkMemoryDedicatedRequirements VkMemoryDedicatedRequirementsKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 included a VkExternalMemoryBufferCreateInfo structure, 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.

The pNext chain of VkImageCreateInfo for the call to vkCreateImage used to create the image being queried included a VkExternalMemoryImageCreateInfo structure, 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.

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:

// Provided by VK_VERSION_1_0
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
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 a VkMemoryDedicatedAllocateInfo structure 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
If the value of VkExportMemoryAllocateInfo::handleTypes used to allocate memory is not 0, it must include at least one of the handles set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If memory was created by a memory import operation, that is not VkImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value, the external handle type of the imported memory must also have been set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If memory was created with the VkImportAndroidHardwareBufferInfoANDROID memory import operation with a non-NULL buffer value, VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must also have been set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If the VkPhysicalDeviceBufferDeviceAddressFeatures::bufferDeviceAddress feature is enabled and buffer was created with the VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT bit set, memory must have been allocated with the VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT bit set

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR

To attach memory to buffer objects for one or more buffers at a time, call:

// Provided by VK_VERSION_1_1
VkResult vkBindBufferMemory2(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindBufferMemoryInfo*               pBindInfos);

or the equivalent command

// Provided by VK_KHR_bind_memory2
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 bindInfoCount VkBindBufferMemoryInfo structures 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR

VkBindBufferMemoryInfo contains members corresponding to the parameters of vkBindBufferMemory.

The VkBindBufferMemoryInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkBindBufferMemoryInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           buffer;
    VkDeviceMemory     memory;
    VkDeviceSize       memoryOffset;
} VkBindBufferMemoryInfo;

or the equivalent

// Provided by VK_KHR_bind_memory2
typedef VkBindBufferMemoryInfo VkBindBufferMemoryInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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
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 a VkMemoryDedicatedAllocateInfo structure 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 buffer and memoryOffset must be zero
If the pNext chain includes a VkBindBufferMemoryDeviceGroupInfo structure, all instances of memory specified by VkBindBufferMemoryDeviceGroupInfo::pDeviceIndices must have been allocated
If the value of VkExportMemoryAllocateInfo::handleTypes used to allocate memory is not 0, it must include at least one of the handles set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If memory was created by a memory import operation, that is not VkImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value, the external handle type of the imported memory must also have been set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If memory was created with the VkImportAndroidHardwareBufferInfoANDROID memory import operation with a non-NULL buffer value, VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must also have been set in VkExternalMemoryBufferCreateInfo::handleTypes when buffer was created
If the VkPhysicalDeviceBufferDeviceAddressFeaturesKHR::bufferDeviceAddress feature is enabled and buffer was created with the VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT_KHR bit set, memory must have been allocated with the VK_MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT_KHR bit set

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
The sType value of each struct in the pNext chain must be unique
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

// Provided by VK_VERSION_1_1
typedef struct VkBindBufferMemoryDeviceGroupInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           deviceIndexCount;
    const uint32_t*    pDeviceIndices;
} VkBindBufferMemoryDeviceGroupInfo;

or the equivalent

// Provided by VK_KHR_device_group with VK_KHR_bind_memory2
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 a structure extending this 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:

// Provided by VK_VERSION_1_0
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 difference of the size of memory and memoryOffset must be greater than or equal to the size member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements with the same image
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 dedicated allocation image aliasing feature is not enabled, and the VkMemoryAllocateInfo provided when memory was allocated included a VkMemoryDedicatedAllocateInfo structure in its pNext chain, and VkMemoryDedicatedAllocateInfo::image was not VK_NULL_HANDLE, then image must equal VkMemoryDedicatedAllocateInfo::image and memoryOffset must be zero
If the dedicated allocation image aliasing feature is enabled, and the VkMemoryAllocateInfo provided when memory was allocated included a VkMemoryDedicatedAllocateInfo structure in its pNext chain, and VkMemoryDedicatedAllocateInfo::image was not VK_NULL_HANDLE, then memoryOffset must be zero, and image must be either equal to VkMemoryDedicatedAllocateInfo::image or an image that was created using the same parameters in VkImageCreateInfo, with the exception that extent and arrayLayers may differ subject to the following restrictions: every dimension in the extent parameter of the image being bound must be equal to or smaller than the original image for which the allocation was created; and the arrayLayers parameter of the image being bound must be equal to or smaller than the original image for which the allocation was created
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
If the value of VkExportMemoryAllocateInfo::handleTypes used to allocate memory is not 0, it must include at least one of the handles set in VkExternalMemoryImageCreateInfo::handleTypes when image was created
If memory was created by a memory import operation, that is not VkImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value, the external handle type of the imported memory must also have been set in VkExternalMemoryImageCreateInfo::handleTypes when image was created
If memory was created with the VkImportAndroidHardwareBufferInfoANDROID memory import operation with a non-NULL buffer value, VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must also have been set in VkExternalMemoryImageCreateInfo::handleTypes when image was created

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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_1
VkResult vkBindImageMemory2(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindImageMemoryInfo*                pBindInfos);

or the equivalent command

// Provided by VK_KHR_bind_memory2
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 VkBindImageMemoryInfo structures, describing images and memory to bind.

On some implementations, it may be more efficient to batch memory bindings into a single command.

Valid Usage

If any VkBindImageMemoryInfo::image was created with VK_IMAGE_CREATE_DISJOINT_BIT then all planes of VkBindImageMemoryInfo::image must be bound individually in separate pBindInfos
pBindInfos must not refer to the same image subresource more than once

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

VK_SUCCESS

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:

// Provided by VK_VERSION_1_1
typedef struct VkBindImageMemoryInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkImage            image;
    VkDeviceMemory     memory;
    VkDeviceSize       memoryOffset;
} VkBindImageMemoryInfo;

or the equivalent

// Provided by VK_KHR_bind_memory2
typedef VkBindImageMemoryInfo VkBindImageMemoryInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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

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
If the pNext chain does not include a VkBindImagePlaneMemoryInfo structure, 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 vkGetImageMemoryRequirements2 with image
If the pNext chain does not include a VkBindImagePlaneMemoryInfo structure, memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements2 with image
If the pNext chain does not include a VkBindImagePlaneMemoryInfo structure, the difference of the size of memory and memoryOffset must be greater than or equal to the size member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements2 with the same image
If the pNext chain includes a VkBindImagePlaneMemoryInfo structure, image must have been created with the VK_IMAGE_CREATE_DISJOINT_BIT bit set
If the pNext chain includes a VkBindImagePlaneMemoryInfo structure, 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 vkGetImageMemoryRequirements2 with image and where VkBindImagePlaneMemoryInfo::planeAspect corresponds to the VkImagePlaneMemoryRequirementsInfo::planeAspect in the VkImageMemoryRequirementsInfo2 structure’s pNext chain
If the pNext chain includes a VkBindImagePlaneMemoryInfo structure, memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements2 with image and where VkBindImagePlaneMemoryInfo::planeAspect corresponds to the VkImagePlaneMemoryRequirementsInfo::planeAspect in the VkImageMemoryRequirementsInfo2 structure’s pNext chain
If the pNext chain includes a VkBindImagePlaneMemoryInfo structure, the difference of the size of memory and memoryOffset must be greater than or equal to the size member of the VkMemoryRequirements structure returned from a call to vkGetImageMemoryRequirements2 with the same image and where VkBindImagePlaneMemoryInfo::planeAspect corresponds to the VkImagePlaneMemoryRequirementsInfo::planeAspect in the VkImageMemoryRequirementsInfo2 structure’s pNext chain
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 and memoryOffset must be zero
If the dedicated allocation image aliasing feature is not enabled, and the VkMemoryAllocateInfo provided when memory was allocated included a VkMemoryDedicatedAllocateInfo structure in its pNext chain, and VkMemoryDedicatedAllocateInfo::image was not VK_NULL_HANDLE, then image must equal VkMemoryDedicatedAllocateInfo::image and memoryOffset must be zero
If the dedicated allocation image aliasing feature is enabled, and the VkMemoryAllocateInfo provided when memory was allocated included a VkMemoryDedicatedAllocateInfo structure in its pNext chain, and VkMemoryDedicatedAllocateInfo::image was not VK_NULL_HANDLE, then memoryOffset must be zero, and image must be either equal to VkMemoryDedicatedAllocateInfo::image or an image that was created using the same parameters in VkImageCreateInfo, with the exception that extent and arrayLayers may differ subject to the following restrictions: every dimension in the extent parameter of the image being bound must be equal to or smaller than the original image for which the allocation was created; and the arrayLayers parameter of the image being bound must be equal to or smaller than the original image for which the allocation was created
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 image and memoryOffset must be zero
If the pNext chain includes a VkBindImageMemoryDeviceGroupInfo structure, all instances of memory specified by VkBindImageMemoryDeviceGroupInfo::pDeviceIndices must have been allocated
If the pNext chain includes a VkBindImageMemoryDeviceGroupInfo structure, and VkBindImageMemoryDeviceGroupInfo::splitInstanceBindRegionCount is not zero, then image must have been created with the VK_IMAGE_CREATE_SPLIT_INSTANCE_BIND_REGIONS_BIT bit set
If the pNext chain includes a VkBindImageMemoryDeviceGroupInfo structure, all elements of VkBindImageMemoryDeviceGroupInfo::pSplitInstanceBindRegions must be valid rectangles contained within the dimensions of image
If the pNext chain includes a VkBindImageMemoryDeviceGroupInfo structure, the union of the areas of all elements of VkBindImageMemoryDeviceGroupInfo::pSplitInstanceBindRegions that correspond to the same instance of image must cover the entire image
If image was created with a valid swapchain handle in VkImageSwapchainCreateInfoKHR::swapchain, then the pNext chain must include a VkBindImageMemorySwapchainInfoKHR structure containing the same swapchain handle
If the pNext chain includes a VkBindImageMemorySwapchainInfoKHR structure, memory must be VK_NULL_HANDLE
If the pNext chain does not include a VkBindImageMemorySwapchainInfoKHR structure, memory must be a valid VkDeviceMemory handle
If the value of VkExportMemoryAllocateInfo::handleTypes used to allocate memory is not 0, it must include at least one of the handles set in VkExternalMemoryImageCreateInfo::handleTypes when image was created
If memory was created by a memory import operation, that is not VkImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value, the external handle type of the imported memory must also have been set in VkExternalMemoryImageCreateInfo::handleTypes when image was created
If memory was created with the VkImportAndroidHardwareBufferInfoANDROID memory import operation with a non-NULL buffer value, VK_EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must also have been set in VkExternalMemoryImageCreateInfo::handleTypes when image was created

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
The sType value of each struct in the pNext chain must be unique
image must be a valid VkImage handle
Both of image, and memory that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

// Provided by VK_VERSION_1_1
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

// Provided by VK_KHR_device_group with VK_KHR_bind_memory2
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 a structure extending this 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 VkRect2D structures 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:

// Provided by VK_KHR_swapchain with VK_VERSION_1_1, VK_KHR_device_group with VK_KHR_swapchain
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 a structure extending this 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, add a VkBindImagePlaneMemoryInfo structure to the pNext chain of VkBindImageMemoryInfo.

The VkBindImagePlaneMemoryInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkBindImagePlaneMemoryInfo {
    VkStructureType          sType;
    const void*              pNext;
    VkImageAspectFlagBits    planeAspect;
} VkBindImagePlaneMemoryInfo;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
typedef VkBindImagePlaneMemoryInfo VkBindImagePlaneMemoryInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
planeAspect is the aspect of the disjoint image plane to bind.

Valid Usage

If the image’s tiling is VK_IMAGE_TILING_LINEAR or VK_IMAGE_TILING_OPTIMAL, then planeAspect must be a single valid format plane for the image (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)
If the image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then planeAspect must be a single valid memory plane for the image (that is, aspectMask must specify a plane index that is less than the VkDrmFormatModifierPropertiesEXT::drmFormatModifierPlaneCount associated with the image’s format and VkImageDrmFormatModifierPropertiesEXT::drmFormatModifier)

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.

Buffer and image objects are created with a sharing mode controlling how they can be accessed from queues. The supported sharing modes are:

// Provided by VK_VERSION_1_0
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.

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

Release exclusive ownership from the source instance or API.
Ensure the release operation has completed using semaphores or fences.
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, vkBindAccelerationStructureMemoryKHR, vkBindAccelerationStructureMemoryNV, 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, resource_A and resource_B, bound respectively to memory range_A and range_B. Let paddedRange_A and paddedRange_B be, respectively, range_A and range_B 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 range_A and range_B. If one resource is linear and the other is non-linear, then the resources alias the memory in the intersection of paddedRange_A and paddedRange_B.

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 a non-linear, 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 a 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.
If either image’s tiling is VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT, then both images must be linear.
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.

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.

11.9. Acceleration Structures

Acceleration structures are an opaque structure that is built by the implementation to more efficiently perform spatial queries on the provided geometric data. For this extension, an acceleration structure is either a top-level acceleration structure containing a set of bottom-level acceleration structures or a bottom-level acceleration structure containing either a set of axis-aligned bounding boxes for custom geometry or a set of triangles.

Each instance in the top-level acceleration structure contains a reference to a bottom-level acceleration structure as well as an instance transform plus information required to index into the shader bindings. The top-level acceleration structure is what is bound to the acceleration descriptor to trace inside the shader in the ray tracing pipeline.

Acceleration structures are represented by VkAccelerationStructureKHR handles:

// Provided by VK_KHR_ray_tracing
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkAccelerationStructureKHR)

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkAccelerationStructureKHR VkAccelerationStructureNV;

To create acceleration structures, call:

// Provided by VK_NV_ray_tracing
VkResult vkCreateAccelerationStructureNV(
    VkDevice                                    device,
    const VkAccelerationStructureCreateInfoNV*  pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkAccelerationStructureNV*                  pAccelerationStructure);

device is the logical device that creates the buffer object.
pCreateInfo is a pointer to a VkAccelerationStructureCreateInfoNV structure containing parameters affecting creation of the acceleration structure.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pAccelerationStructure is a pointer to a VkAccelerationStructureNV handle in which the resulting acceleration structure object is returned.

Similar to other objects in Vulkan, the acceleration structure creation merely creates an object with a specific “shape” as specified by the information in VkAccelerationStructureInfoNV and compactedSize in pCreateInfo. Populating the data in the object after allocating and binding memory is done with vkCmdBuildAccelerationStructureNV and vkCmdCopyAccelerationStructureNV.

Acceleration structure creation uses the count and type information from the geometries, but does not use the data references in the structures.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pCreateInfo must be a valid pointer to a valid VkAccelerationStructureCreateInfoNV structure
If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure
pAccelerationStructure must be a valid pointer to a VkAccelerationStructureNV handle

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY

The VkAccelerationStructureCreateInfoNV structure is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkAccelerationStructureCreateInfoNV {
    VkStructureType                  sType;
    const void*                      pNext;
    VkDeviceSize                     compactedSize;
    VkAccelerationStructureInfoNV    info;
} VkAccelerationStructureCreateInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
compactedSize is the size from the result of vkCmdWriteAccelerationStructuresPropertiesNV if this acceleration structure is going to be the target of a compacting copy.
info is the VkAccelerationStructureInfoNV structure specifying further parameters of the created acceleration structure.

Valid Usage

If compactedSize is not 0 then both info.geometryCount and info.instanceCount must be 0

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_CREATE_INFO_NV
pNext must be NULL
info must be a valid VkAccelerationStructureInfoNV structure

The VkAccelerationStructureInfoNV structure is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkAccelerationStructureInfoNV {
    VkStructureType                        sType;
    const void*                            pNext;
    VkAccelerationStructureTypeNV          type;
    VkBuildAccelerationStructureFlagsNV    flags;
    uint32_t                               instanceCount;
    uint32_t                               geometryCount;
    const VkGeometryNV*                    pGeometries;
} VkAccelerationStructureInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
type is a VkAccelerationStructureTypeNV value specifying the type of acceleration structure that will be created.
flags is a bitmask of VkBuildAccelerationStructureFlagBitsNV specifying additional parameters of the acceleration structure.
instanceCount specifies the number of instances that will be in the new acceleration structure.
geometryCount specifies the number of geometries that will be in the new acceleration structure.
pGeometries is a pointer to an array of geometryCount VkGeometryNV structures containing the scene data being passed into the acceleration structure.

VkAccelerationStructureInfoNV contains information that is used both for acceleration structure creation with vkCreateAccelerationStructureNV and in combination with the actual geometric data to build the acceleration structure with vkCmdBuildAccelerationStructureNV.

Valid Usage

geometryCount must be less than or equal to VkPhysicalDeviceRayTracingPropertiesNV::maxGeometryCount
instanceCount must be less than or equal to VkPhysicalDeviceRayTracingPropertiesNV::maxInstanceCount
The total number of triangles in all geometries must be less than or equal to VkPhysicalDeviceRayTracingPropertiesNV::maxTriangleCount
If type is VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_NV then geometryCount must be 0
If type is VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_NV then instanceCount must be 0
If type is VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_NV then the geometryType member of each geometry in pGeometries must be the same
If flags has the VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_NV bit set, then it must not have the VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_NV bit set
scratch must have been created with VK_BUFFER_USAGE_RAY_TRACING_BIT_NV usage flag
If instanceData is not VK_NULL_HANDLE, instanceData must have been created with VK_BUFFER_USAGE_RAY_TRACING_BIT_NV usage flag

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_INFO_NV
pNext must be NULL
type must be a valid VkAccelerationStructureTypeNV value
flags must be a valid combination of VkBuildAccelerationStructureFlagBitsNV values
If geometryCount is not 0, pGeometries must be a valid pointer to an array of geometryCount valid VkGeometryNV structures

To create an acceleration structure, call:

// Provided by VK_KHR_ray_tracing
VkResult vkCreateAccelerationStructureKHR(
    VkDevice                                    device,
    const VkAccelerationStructureCreateInfoKHR* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkAccelerationStructureKHR*                 pAccelerationStructure);

device is the logical device that creates the buffer object.
pCreateInfo is a pointer to a VkAccelerationStructureCreateInfoKHR structure containing parameters affecting creation of the acceleration structure.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pAccelerationStructure is a pointer to a VkAccelerationStructureKHR handle in which the resulting acceleration structure object is returned.

Similar to other objects in Vulkan, the acceleration structure creation merely creates an object with a specific “shape”. The type and quantity of geometry that can be built into an acceleration structure is determined by the parameters of VkAccelerationStructureCreateInfoKHR.

Populating the data in the object after allocating and binding memory is done with commands such as vkCmdBuildAccelerationStructureKHR, vkBuildAccelerationStructureKHR, vkCmdCopyAccelerationStructureKHR, and vkCopyAccelerationStructureKHR.

The input buffers passed to acceleration structure build commands will be referenced by the implementation for the duration of the command. After the command completes, the acceleration structure may hold a reference to any acceleration structure specified by an active instance contained therein. Apart from this referencing, acceleration structures must be fully self-contained. The application may re-use or free any memory which was used by the command as an input or as scratch without affecting the results of ray traversal.

Valid Usage

The rayTracing or rayQuery feature must be enabled
If VkAccelerationStructureCreateInfoKHR::deviceAddress is not zero, the rayTracingAccelerationStructureCaptureReplay feature must be enabled
If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

device must be a valid VkDevice handle
pCreateInfo must be a valid pointer to a valid VkAccelerationStructureCreateInfoKHR structure
If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure
pAccelerationStructure must be a valid pointer to a VkAccelerationStructureKHR handle

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR

The VkAccelerationStructureCreateInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkAccelerationStructureCreateInfoKHR {
    VkStructureType                                            sType;
    const void*                                                pNext;
    VkDeviceSize                                               compactedSize;
    VkAccelerationStructureTypeKHR                             type;
    VkBuildAccelerationStructureFlagsKHR                       flags;
    uint32_t                                                   maxGeometryCount;
    const VkAccelerationStructureCreateGeometryTypeInfoKHR*    pGeometryInfos;
    VkDeviceAddress                                            deviceAddress;
} VkAccelerationStructureCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
compactedSize is the size from the result of vkCmdWriteAccelerationStructuresPropertiesKHR if this acceleration structure is going to be the target of a compacting copy.
type is a VkAccelerationStructureTypeKHR value specifying the type of acceleration structure that will be created.
flags is a bitmask of VkBuildAccelerationStructureFlagBitsKHR specifying additional parameters of the acceleration structure.
maxGeometryCount specifies the number of geometries that will be in the new acceleration structure.
pGeometryInfos is an array of maxGeometryCount VkAccelerationStructureCreateGeometryTypeInfoKHR structures, which describe the maximum size and format of the data that will be built into the acceleration structure.
deviceAddress is the device address requested for the acceleration structure if the rayTracingAccelerationStructureCaptureReplay feature is being used.

If deviceAddress is zero, no specific address is requested.

If deviceAddress is not zero, deviceAddress must be an address retrieved from an identically created acceleration structure on the same implementation. The acceleration structure must also be bound to an identically created VkDeviceMemory object.

Apps should avoid creating acceleration structures with app-provided addresses and implementation-provided addresses in the same process, to reduce the likelihood of VK_ERROR_INVALID_OPAQUE_CAPTURE_ADDRESS_KHR errors.

Valid Usage

If compactedSize is not 0 then maxGeometryCount must be 0
If compactedSize is 0 then maxGeometryCount must not be 0
If type is VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR then maxGeometryCount must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxGeometryCount
If type is VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR then the maxPrimitiveCount member of each element of the pGeometryInfos array must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxInstanceCount
The total number of triangles in all geometries must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxPrimitiveCount
The total number of AABBs in all geometries must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxPrimitiveCount
If type is VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR and compactedSize is 0, maxGeometryCount must be 1
If type is VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR and compactedSize is 0, the geometryType member of elements of pGeometryInfos must be VK_GEOMETRY_TYPE_INSTANCES_KHR
If type is VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR and compactedSize is 0, the geometryType member of elements of pGeometryInfos must not be VK_GEOMETRY_TYPE_INSTANCES_KHR
If type is VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR then the geometryType member of each geometry in pGeometryInfos must be the same
If flags has the VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_KHR bit set, then it must not have the VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_KHR bit set
If deviceAddress is not 0, VkPhysicalDeviceRayTracingFeaturesKHR::rayTracingAccelerationStructureCaptureReplay must be VK_TRUE

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_CREATE_INFO_KHR
pNext must be NULL
type must be a valid VkAccelerationStructureTypeKHR value
flags must be a valid combination of VkBuildAccelerationStructureFlagBitsKHR values
If maxGeometryCount is not 0, pGeometryInfos must be a valid pointer to an array of maxGeometryCount valid VkAccelerationStructureCreateGeometryTypeInfoKHR structures

The VkAccelerationStructureCreateGeometryTypeInfoKHR structure specifies the shape of geometries that will be built into an acceleration structure and is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkAccelerationStructureCreateGeometryTypeInfoKHR {
    VkStructureType      sType;
    const void*          pNext;
    VkGeometryTypeKHR    geometryType;
    uint32_t             maxPrimitiveCount;
    VkIndexType          indexType;
    uint32_t             maxVertexCount;
    VkFormat             vertexFormat;
    VkBool32             allowsTransforms;
} VkAccelerationStructureCreateGeometryTypeInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
geometryType is a VkGeometryTypeKHR that describes the type of an acceleration structure geometry.
maxPrimitiveCount describes the maximum number of primitives that can be built into an acceleration structure geometry.
indexType is a VkIndexType that describes the index type used to build this geometry when geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR.
maxVertexCount describes the maximum vertex count that can be used to build an acceleration structure geometry when geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR.
vertexFormat is a VkFormat that describes the vertex format used to build this geometry when geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR.
allowsTransforms indicates whether transform data can be used by this acceleration structure or not, when geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR.

When geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR:

if indexType is VK_INDEX_TYPE_NONE_KHR, then this structure describes a set of triangles.
if indexType is not VK_INDEX_TYPE_NONE_KHR, then this structure describes a set of indexed triangles.

Valid Usage

If geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR, vertexFormat must support the VK_FORMAT_FEATURE_ACCELERATION_STRUCTURE_VERTEX_BUFFER_BIT_KHR in VkFormatProperties::bufferFeatures as returned by vkGetPhysicalDeviceFormatProperties2
If geometryType is VK_GEOMETRY_TYPE_TRIANGLES_KHR, indexType must be VK_INDEX_TYPE_UINT16, VK_INDEX_TYPE_UINT32, or VK_INDEX_TYPE_NONE_KHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_CREATE_GEOMETRY_TYPE_INFO_KHR
pNext must be NULL
geometryType must be a valid VkGeometryTypeKHR value
indexType must be a valid VkIndexType value
If vertexFormat is not 0, vertexFormat must be a valid VkFormat value

Values which can be set in VkAccelerationStructureCreateInfoKHR::type or VkAccelerationStructureInfoNV::type specifying the type of acceleration structure, are:

// Provided by VK_KHR_ray_tracing
typedef enum VkAccelerationStructureTypeKHR {
    VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR = 0,
    VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR = 1,
  // Provided by VK_NV_ray_tracing
    VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_NV = VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR,
  // Provided by VK_NV_ray_tracing
    VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_NV = VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR,
} VkAccelerationStructureTypeKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkAccelerationStructureTypeKHR VkAccelerationStructureTypeNV;

VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR is a top-level acceleration structure containing instance data referring to bottom-level acceleration structures.
VK_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL_KHR is a bottom-level acceleration structure containing the AABBs or geometry to be intersected.

Bits which can be set in VkAccelerationStructureCreateInfoKHR::flags or VkAccelerationStructureInfoNV::flags specifying additional parameters for acceleration structure builds, are:

// Provided by VK_KHR_ray_tracing
typedef enum VkBuildAccelerationStructureFlagBitsKHR {
    VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_UPDATE_BIT_KHR = 0x00000001,
    VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_COMPACTION_BIT_KHR = 0x00000002,
    VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_KHR = 0x00000004,
    VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_KHR = 0x00000008,
    VK_BUILD_ACCELERATION_STRUCTURE_LOW_MEMORY_BIT_KHR = 0x00000010,
  // Provided by VK_NV_ray_tracing
    VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_UPDATE_BIT_NV = VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_UPDATE_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_COMPACTION_BIT_NV = VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_COMPACTION_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_NV = VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_NV = VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_BUILD_ACCELERATION_STRUCTURE_LOW_MEMORY_BIT_NV = VK_BUILD_ACCELERATION_STRUCTURE_LOW_MEMORY_BIT_KHR,
} VkBuildAccelerationStructureFlagBitsKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkBuildAccelerationStructureFlagBitsKHR VkBuildAccelerationStructureFlagBitsNV;

VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_UPDATE_BIT_KHR indicates that the specified acceleration structure can be updated with update of VK_TRUE in vkCmdBuildAccelerationStructureKHR or vkCmdBuildAccelerationStructureNV .
VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_COMPACTION_BIT_KHR indicates that the specified acceleration structure can act as the source for a copy acceleration structure command with mode of VK_COPY_ACCELERATION_STRUCTURE_MODE_COMPACT_KHR to produce a compacted acceleration structure.
VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_TRACE_BIT_KHR indicates that the given acceleration structure build should prioritize trace performance over build time.
VK_BUILD_ACCELERATION_STRUCTURE_PREFER_FAST_BUILD_BIT_KHR indicates that the given acceleration structure build should prioritize build time over trace performance.
VK_BUILD_ACCELERATION_STRUCTURE_LOW_MEMORY_BIT_KHR indicates that this acceleration structure should minimize the size of the scratch memory and the final result build, potentially at the expense of build time or trace performance.

Note

VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_UPDATE_BIT_KHR and VK_BUILD_ACCELERATION_STRUCTURE_ALLOW_COMPACTION_BIT_KHR may take more time and memory than a normal build, and so should only be used when those features are needed.

// Provided by VK_KHR_ray_tracing
typedef VkFlags VkBuildAccelerationStructureFlagsKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkBuildAccelerationStructureFlagsKHR VkBuildAccelerationStructureFlagsNV;

VkBuildAccelerationStructureFlagsKHR is a bitmask type for setting a mask of zero or more VkBuildAccelerationStructureFlagBitsKHR.

The VkGeometryNV structure describes geometry in a bottom-level acceleration structure and is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkGeometryNV {
    VkStructureType       sType;
    const void*           pNext;
    VkGeometryTypeKHR     geometryType;
    VkGeometryDataNV      geometry;
    VkGeometryFlagsKHR    flags;
} VkGeometryNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
geometryType specifies the VkGeometryTypeKHR which this geometry refers to.
geometry contains the geometry data as described in VkGeometryDataNV.
flags has VkGeometryFlagBitsKHR describing options for this geometry.

Valid Usage

geometryType must be VK_GEOMETRY_TYPE_TRIANGLES_NV or VK_GEOMETRY_TYPE_AABBS_NV

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GEOMETRY_NV
pNext must be NULL
geometryType must be a valid VkGeometryTypeKHR value
geometry must be a valid VkGeometryDataNV structure
flags must be a valid combination of VkGeometryFlagBitsKHR values

Geometry types are specified by VkGeometryTypeKHR, which takes values:

// Provided by VK_KHR_ray_tracing
typedef enum VkGeometryTypeKHR {
    VK_GEOMETRY_TYPE_TRIANGLES_KHR = 0,
    VK_GEOMETRY_TYPE_AABBS_KHR = 1,
  // Provided by VK_KHR_ray_tracing
    VK_GEOMETRY_TYPE_INSTANCES_KHR = 1000150000,
  // Provided by VK_NV_ray_tracing
    VK_GEOMETRY_TYPE_TRIANGLES_NV = VK_GEOMETRY_TYPE_TRIANGLES_KHR,
  // Provided by VK_NV_ray_tracing
    VK_GEOMETRY_TYPE_AABBS_NV = VK_GEOMETRY_TYPE_AABBS_KHR,
} VkGeometryTypeKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkGeometryTypeKHR VkGeometryTypeNV;

VK_GEOMETRY_TYPE_TRIANGLES_KHR specifies a geometry type consisting of triangles.
VK_GEOMETRY_TYPE_AABBS_KHR specifies a geometry type consisting of axis-aligned bounding boxes.
VK_GEOMETRY_TYPE_INSTANCES_KHR specifies a geometry type consisting of acceleration structure instances.

Bits specifying additional parameters for geometries in acceleration structure builds, are:

// Provided by VK_KHR_ray_tracing
typedef enum VkGeometryFlagBitsKHR {
    VK_GEOMETRY_OPAQUE_BIT_KHR = 0x00000001,
    VK_GEOMETRY_NO_DUPLICATE_ANY_HIT_INVOCATION_BIT_KHR = 0x00000002,
  // Provided by VK_NV_ray_tracing
    VK_GEOMETRY_OPAQUE_BIT_NV = VK_GEOMETRY_OPAQUE_BIT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_GEOMETRY_NO_DUPLICATE_ANY_HIT_INVOCATION_BIT_NV = VK_GEOMETRY_NO_DUPLICATE_ANY_HIT_INVOCATION_BIT_KHR,
} VkGeometryFlagBitsKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkGeometryFlagBitsKHR VkGeometryFlagBitsNV;

VK_GEOMETRY_OPAQUE_BIT_KHR indicates that this geometry does not invoke the any-hit shaders even if present in a hit group.
VK_GEOMETRY_NO_DUPLICATE_ANY_HIT_INVOCATION_BIT_KHR indicates that the implementation must only call the any-hit shader a single time for each primitive in this geometry. If this bit is absent an implementation may invoke the any-hit shader more than once for this geometry.

// Provided by VK_KHR_ray_tracing
typedef VkFlags VkGeometryFlagsKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkGeometryFlagsKHR VkGeometryFlagsNV;

VkGeometryFlagsKHR is a bitmask type for setting a mask of zero or more VkGeometryFlagBitsKHR.

The VkGeometryDataNV structure specifes geometry in a bottom-level acceleration structure and is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkGeometryDataNV {
    VkGeometryTrianglesNV    triangles;
    VkGeometryAABBNV         aabbs;
} VkGeometryDataNV;

triangles contains triangle data if VkGeometryNV::geometryType is VK_GEOMETRY_TYPE_TRIANGLES_NV.
aabbs contains axis-aligned bounding box data if VkGeometryNV::geometryType is VK_GEOMETRY_TYPE_AABBS_NV.

Valid Usage (Implicit)

triangles must be a valid VkGeometryTrianglesNV structure
aabbs must be a valid VkGeometryAABBNV structure

The VkGeometryTrianglesNV structure specifies triangle geometry in a bottom-level acceleration structure and is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkGeometryTrianglesNV {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           vertexData;
    VkDeviceSize       vertexOffset;
    uint32_t           vertexCount;
    VkDeviceSize       vertexStride;
    VkFormat           vertexFormat;
    VkBuffer           indexData;
    VkDeviceSize       indexOffset;
    uint32_t           indexCount;
    VkIndexType        indexType;
    VkBuffer           transformData;
    VkDeviceSize       transformOffset;
} VkGeometryTrianglesNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
vertexData is the buffer containing vertex data for this geometry.
vertexOffset is the offset in bytes within vertexData containing vertex data for this geometry.
vertexCount is the number of valid vertices.
vertexStride is the stride in bytes between each vertex.
vertexFormat is a VkFormat describing the format of each vertex element.
indexData is the buffer containing index data for this geometry.
indexOffset is the offset in bytes within indexData containing index data for this geometry.
indexCount is the number of indices to include in this geometry.
indexType is a VkIndexType describing the format of each index.
transformData is an optional buffer containing an VkTransformMatrixNV structure defining a transformation to be applied to this geometry.
transformOffset is the offset in bytes in transformData of the transform information described above.

If indexType is VK_INDEX_TYPE_NONE_NV, then this structure describes a set of triangles determined by vertexCount. Otherwise, this structure describes a set of indexed triangles determined by indexCount.

Valid Usage

vertexOffset must be less than the size of vertexData
vertexOffset must be a multiple of the component size of vertexFormat
vertexFormat must be one of VK_FORMAT_R32G32B32_SFLOAT, VK_FORMAT_R32G32_SFLOAT, VK_FORMAT_R16G16B16_SFLOAT, VK_FORMAT_R16G16_SFLOAT, VK_FORMAT_R16G16_SNORM, or VK_FORMAT_R16G16B16_SNORM
indexOffset must be less than the size of indexData
indexOffset must be a multiple of the element size of indexType
indexType must be VK_INDEX_TYPE_UINT16, VK_INDEX_TYPE_UINT32, or VK_INDEX_TYPE_NONE_NV
indexData must be VK_NULL_HANDLE if indexType is VK_INDEX_TYPE_NONE_NV
indexData must be a valid VkBuffer handle if indexType is not VK_INDEX_TYPE_NONE_NV
indexCount must be 0 if indexType is VK_INDEX_TYPE_NONE_NV
transformOffset must be less than the size of transformData
transformOffset must be a multiple of 16

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GEOMETRY_TRIANGLES_NV
pNext must be NULL
If vertexData is not VK_NULL_HANDLE, vertexData must be a valid VkBuffer handle
vertexFormat must be a valid VkFormat value
If indexData is not VK_NULL_HANDLE, indexData must be a valid VkBuffer handle
indexType must be a valid VkIndexType value
If transformData is not VK_NULL_HANDLE, transformData must be a valid VkBuffer handle
Each of indexData, transformData, and vertexData that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

The VkGeometryAABBNV structure specifies axis-aligned bounding box geometry in a bottom-level acceleration structure, and is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkGeometryAABBNV {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           aabbData;
    uint32_t           numAABBs;
    uint32_t           stride;
    VkDeviceSize       offset;
} VkGeometryAABBNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
aabbData is the buffer containing axis-aligned bounding box data.
numAABBs is the number of AABBs in this geometry.
stride is the stride in bytes between AABBs in aabbData.
offset is the offset in bytes of the first AABB in aabbData.

The AABB data in memory is six 32-bit floats consisting of the minimum x, y, and z values followed by the maximum x, y, and z values.

Valid Usage

offset must be less than the size of aabbData
offset must be a multiple of 8
stride must be a multiple of 8

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_GEOMETRY_AABB_NV
pNext must be NULL
If aabbData is not VK_NULL_HANDLE, aabbData must be a valid VkBuffer handle

To destroy an acceleration structure, call:

// Provided by VK_KHR_ray_tracing
void vkDestroyAccelerationStructureKHR(
    VkDevice                                    device,
    VkAccelerationStructureKHR                  accelerationStructure,
    const VkAllocationCallbacks*                pAllocator);

or the equivalent command

// Provided by VK_NV_ray_tracing
void vkDestroyAccelerationStructureNV(
    VkDevice                                    device,
    VkAccelerationStructureKHR                  accelerationStructure,
    const VkAllocationCallbacks*                pAllocator);

device is the logical device that destroys the buffer.
accelerationStructure is the acceleration structure to destroy.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.

Valid Usage

All submitted commands that refer to accelerationStructure must have completed execution
If VkAllocationCallbacks were provided when accelerationStructure was created, a compatible set of callbacks must be provided here
If no VkAllocationCallbacks were provided when accelerationStructure was created, pAllocator must be NULL

Valid Usage (Implicit)

device must be a valid VkDevice handle
If accelerationStructure is not VK_NULL_HANDLE, accelerationStructure must be a valid VkAccelerationStructureKHR handle
If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure
If accelerationStructure is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization

Host access to accelerationStructure must be externally synchronized

An acceleration structure has memory requirements for the structure object itself, scratch space for the build, and scratch space for the update.

Scratch space is allocated as a VkBuffer, so for VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_NV and VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_NV the pMemoryRequirements->alignment and pMemoryRequirements->memoryTypeBits values returned by this call must be filled with zero, and should be ignored by the application.

To query the memory requirements call:

// Provided by VK_NV_ray_tracing
void vkGetAccelerationStructureMemoryRequirementsNV(
    VkDevice                                    device,
    const VkAccelerationStructureMemoryRequirementsInfoNV* pInfo,
    VkMemoryRequirements2KHR*                   pMemoryRequirements);

device is the logical device on which the acceleration structure was created.
pInfo specifies the acceleration structure to get memory requirements for.
pMemoryRequirements returns the requested acceleration structure memory requirements.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkAccelerationStructureMemoryRequirementsInfoNV structure
pMemoryRequirements must be a valid pointer to a VkMemoryRequirements2KHR structure

The VkAccelerationStructureMemoryRequirementsInfoNV structure is defined as:

// Provided by VK_NV_ray_tracing
typedef struct VkAccelerationStructureMemoryRequirementsInfoNV {
    VkStructureType                                    sType;
    const void*                                        pNext;
    VkAccelerationStructureMemoryRequirementsTypeNV    type;
    VkAccelerationStructureNV                          accelerationStructure;
} VkAccelerationStructureMemoryRequirementsInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
type selects the type of memory requirement being queried. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_NV returns the memory requirements for the object itself. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_NV returns the memory requirements for the scratch memory when doing a build. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_NV returns the memory requirements for the scratch memory when doing an update.
accelerationStructure is the acceleration structure to be queried for memory requirements.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_INFO_NV
pNext must be NULL
type must be a valid VkAccelerationStructureMemoryRequirementsTypeNV value
accelerationStructure must be a valid VkAccelerationStructureNV handle

An acceleration structure has memory requirements for the structure object itself, scratch space for the build, and scratch space for the update.

Scratch space is allocated as a VkBuffer, so for VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_KHR and VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_KHR the pMemoryRequirements->alignment and pMemoryRequirements->memoryTypeBits values returned by this call must be filled with zero, and should be ignored by the application.

To query the memory requirements call:

// Provided by VK_KHR_ray_tracing
void vkGetAccelerationStructureMemoryRequirementsKHR(
    VkDevice                                    device,
    const VkAccelerationStructureMemoryRequirementsInfoKHR* pInfo,
    VkMemoryRequirements2*                      pMemoryRequirements);

device is the logical device on which the acceleration structure was created.
pInfo specifies the acceleration structure to get memory requirements for.
pMemoryRequirements returns the requested acceleration structure memory requirements.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkAccelerationStructureMemoryRequirementsInfoKHR structure
pMemoryRequirements must be a valid pointer to a VkMemoryRequirements2 structure

The VkAccelerationStructureMemoryRequirementsInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkAccelerationStructureMemoryRequirementsInfoKHR {
    VkStructureType                                     sType;
    const void*                                         pNext;
    VkAccelerationStructureMemoryRequirementsTypeKHR    type;
    VkAccelerationStructureBuildTypeKHR                 buildType;
    VkAccelerationStructureKHR                          accelerationStructure;
} VkAccelerationStructureMemoryRequirementsInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
buildType selects the build types whose memory requirements are being queried.
type selects the type of memory requirement being queried. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR returns the memory requirements for the object itself. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_KHR returns the memory requirements for the scratch memory when doing a build. VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_KHR returns the memory requirements for the scratch memory when doing an update.
accelerationStructure is the acceleration structure to be queried for memory requirements.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_INFO_KHR
pNext must be NULL
type must be a valid VkAccelerationStructureMemoryRequirementsTypeKHR value
buildType must be a valid VkAccelerationStructureBuildTypeKHR value
accelerationStructure must be a valid VkAccelerationStructureKHR handle

Possible values of type in VkAccelerationStructureMemoryRequirementsInfoKHR are:

// Provided by VK_KHR_ray_tracing
typedef enum VkAccelerationStructureMemoryRequirementsTypeKHR {
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR = 0,
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_KHR = 1,
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_KHR = 2,
  // Provided by VK_NV_ray_tracing
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_NV = VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR,
  // Provided by VK_NV_ray_tracing
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_NV = VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_KHR,
  // Provided by VK_NV_ray_tracing
    VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_NV = VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_KHR,
} VkAccelerationStructureMemoryRequirementsTypeKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkAccelerationStructureMemoryRequirementsTypeKHR VkAccelerationStructureMemoryRequirementsTypeNV;

VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR requests the memory requirement for the VkAccelerationStructureKHR backing store.
VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_BUILD_SCRATCH_KHR requests the memory requirement for scratch space during the initial build.
VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_UPDATE_SCRATCH_KHR requests the memory requirement for scratch space during an update.

Possible values of buildType in VkAccelerationStructureMemoryRequirementsInfoKHR are:

// Provided by VK_KHR_ray_tracing
typedef enum VkAccelerationStructureBuildTypeKHR {
    VK_ACCELERATION_STRUCTURE_BUILD_TYPE_HOST_KHR = 0,
    VK_ACCELERATION_STRUCTURE_BUILD_TYPE_DEVICE_KHR = 1,
    VK_ACCELERATION_STRUCTURE_BUILD_TYPE_HOST_OR_DEVICE_KHR = 2,
} VkAccelerationStructureBuildTypeKHR;

VK_ACCELERATION_STRUCTURE_BUILD_TYPE_HOST_KHR requests the memory requirement for operations performed by the host.
VK_ACCELERATION_STRUCTURE_BUILD_TYPE_DEVICE_KHR requests the memory requirement for operations performed by the device.
VK_ACCELERATION_STRUCTURE_BUILD_TYPE_HOST_OR_DEVICE_KHR requests the memory requirement for operations performed by either the host, or the device.

The implementation guarantees certain properties about the memory requirements returned by vkGetAccelerationStructureMemoryRequirementsKHR called with a type of VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR:

The memoryTypeBits member always contains at least one bit set.
The memoryTypeBits member is identical for all VkAccelerationStructureKHR objects created with the same value for the type and flags members in the VkAccelerationStructureCreateInfoKHR structure passed to vkCreateAccelerationStructureKHR and with the same value for buildType in VkAccelerationStructureMemoryRequirementsInfoKHR passed to vkGetAccelerationStructureMemoryRequirementsKHR.

To attach memory to one or more acceleration structures at a time, call:

// Provided by VK_KHR_ray_tracing
VkResult vkBindAccelerationStructureMemoryKHR(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindAccelerationStructureMemoryInfoKHR* pBindInfos);

or the equivalent command

// Provided by VK_NV_ray_tracing
VkResult vkBindAccelerationStructureMemoryNV(
    VkDevice                                    device,
    uint32_t                                    bindInfoCount,
    const VkBindAccelerationStructureMemoryInfoKHR* pBindInfos);

device is the logical device that owns the acceleration structures and memory.
bindInfoCount is the number of elements in pBindInfos.
pBindInfos is a pointer to an array of VkBindAccelerationStructureMemoryInfoKHR structures describing acceleration structures and memory to bind.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pBindInfos must be a valid pointer to an array of bindInfoCount valid VkBindAccelerationStructureMemoryInfoKHR structures
bindInfoCount must be greater than 0

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkBindAccelerationStructureMemoryInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkBindAccelerationStructureMemoryInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    VkAccelerationStructureKHR    accelerationStructure;
    VkDeviceMemory                memory;
    VkDeviceSize                  memoryOffset;
    uint32_t                      deviceIndexCount;
    const uint32_t*               pDeviceIndices;
} VkBindAccelerationStructureMemoryInfoKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkBindAccelerationStructureMemoryInfoKHR VkBindAccelerationStructureMemoryInfoNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
accelerationStructure is the acceleration structure 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 that is to be bound to the acceleration structure. The number of bytes returned in the VkMemoryRequirements::size member in memory, starting from memoryOffset bytes, will be bound to the specified acceleration structure.
deviceIndexCount is the number of elements in pDeviceIndices.
pDeviceIndices is a pointer to an array of device indices.

Valid Usage

accelerationStructure must not already be backed by a memory object
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 vkGetAccelerationStructureMemoryRequirementsKHR with accelerationStructure and type of VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR
memoryOffset must be an integer multiple of the alignment member of the VkMemoryRequirements structure returned from a call to vkGetAccelerationStructureMemoryRequirementsKHR with accelerationStructure and type of VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR
The size member of the VkMemoryRequirements structure returned from a call to vkGetAccelerationStructureMemoryRequirementsKHR with accelerationStructure and type of VK_ACCELERATION_STRUCTURE_MEMORY_REQUIREMENTS_TYPE_OBJECT_KHR must be less than or equal to the size of memory minus memoryOffset

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BIND_ACCELERATION_STRUCTURE_MEMORY_INFO_KHR
pNext must be NULL
accelerationStructure must be a valid VkAccelerationStructureKHR handle
memory must be a valid VkDeviceMemory handle
If deviceIndexCount is not 0, pDeviceIndices must be a valid pointer to an array of deviceIndexCount uint32_t values
Both of accelerationStructure, and memory must have been created, allocated, or retrieved from the same VkDevice

To allow constructing geometry instances with device code if desired, we need to be able to query a opaque handle for an acceleration structure. This handle is a value of 8 bytes. To get this handle, call:

// Provided by VK_NV_ray_tracing
VkResult vkGetAccelerationStructureHandleNV(
    VkDevice                                    device,
    VkAccelerationStructureKHR                  accelerationStructure,
    size_t                                      dataSize,
    void*                                       pData);

device is the logical device that owns the acceleration structures.
accelerationStructure is the acceleration structure.
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.

Valid Usage

dataSize must be large enough to contain the result of the query, as described above
accelerationStructure must be bound completely and contiguously to a single VkDeviceMemory object via vkBindAccelerationStructureMemoryKHR

Valid Usage (Implicit)

device must be a valid VkDevice handle
accelerationStructure must be a valid VkAccelerationStructureKHR handle
pData must be a valid pointer to an array of dataSize bytes
dataSize must be greater than 0
accelerationStructure must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

To query the 64-bit device address for an acceleration structure, call:

// Provided by VK_KHR_ray_tracing
VkDeviceAddress vkGetAccelerationStructureDeviceAddressKHR(
    VkDevice                                    device,
    const VkAccelerationStructureDeviceAddressInfoKHR* pInfo);

device is the logical device that the accelerationStructure was created on.
pInfo is a pointer to a VkAccelerationStructureDeviceAddressInfoKHR structure specifying the acceleration structure to retrieve an address for.

The 64-bit return value is an address of the acceleration structure, which can be used for device and shader operations that involve acceleration structures, such as ray traversal and acceleration structure building.

If the acceleration structure was created with a non-zero value of VkAccelerationStructureCreateInfoKHR::deviceAddress the return value will be the same address.

Valid Usage

If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkAccelerationStructureDeviceAddressInfoKHR structure

The VkAccelerationStructureDeviceAddressInfoKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkAccelerationStructureDeviceAddressInfoKHR {
    VkStructureType               sType;
    const void*                   pNext;
    VkAccelerationStructureKHR    accelerationStructure;
} VkAccelerationStructureDeviceAddressInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
accelerationStructure specifies the acceleration structure whose address is being queried.

Valid Usage

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACCELERATION_STRUCTURE_DEVICE_ADDRESS_INFO_KHR
pNext must be NULL
accelerationStructure must be a valid VkAccelerationStructureKHR handle

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSampler)

To create a sampler object, call:

// Provided by VK_VERSION_1_0
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 a VkSamplerCreateInfo structure specifying the state of the sampler object.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pSampler is a pointer to a VkSampler handle in which the resulting sampler object is returned.

Valid Usage

There must be less than VkPhysicalDeviceLimits::maxSamplerAllocationCount VkSampler objects currently created on the device.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_TOO_MANY_OBJECTS

The VkSamplerCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
flags is a bitmask of VkSamplerCreateFlagBits describing additional parameters of the sampler.
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.
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, 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
maxLod must be greater than or equal to minLod
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′C_BC_R conversion is enabled and the potential format features of the sampler Y′C_BC_R conversion do not support VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT, minFilter and magFilter must be equal to the sampler Y′C_BC_R 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′C_BC_R 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 if sampler Y′C_BC_R conversion is enabled
If samplerMirrorClampToEdge is not enabled, and 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_EXT, anisotropyEnable must be VK_FALSE
If compareEnable is VK_TRUE, the reductionMode member of VkSamplerReductionModeCreateInfo must be VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then minFilter and magFilter must be equal
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then mipmapMode must be VK_SAMPLER_MIPMAP_MODE_NEAREST
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then minLod and maxLod must be zero
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then addressModeU and addressModeV must each be either VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE or VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then anisotropyEnable must be VK_FALSE
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then compareEnable must be VK_FALSE
If flags includes VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT, then unnormalizedCoordinates must be VK_FALSE
If borderColor is set to one of VK_BORDER_COLOR_FLOAT_CUSTOM_EXT or VK_BORDER_COLOR_INT_CUSTOM_EXT, then a VkSamplerCustomBorderColorCreateInfoEXT must be present in the pNext chain
If the customBorderColors feature is not enabled, borderColor must not be set to VK_BORDER_COLOR_FLOAT_CUSTOM_EXT or VK_BORDER_COLOR_INT_CUSTOM_EXT
The maximum number of samplers with custom border colors which can be simultaneously created on a device is implementation-dependent and specified by the maxCustomBorderColorSamplers member of the VkPhysicalDeviceCustomBorderColorPropertiesEXT structure

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SAMPLER_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 VkSamplerCustomBorderColorCreateInfoEXT, VkSamplerReductionModeCreateInfo, or VkSamplerYcbcrConversionInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkSamplerCreateFlagBits values
magFilter must be a valid VkFilter value
minFilter must be a valid VkFilter value
mipmapMode must be a valid VkSamplerMipmapMode value
addressModeU must be a valid VkSamplerAddressMode value
addressModeV must be a valid VkSamplerAddressMode value
addressModeW must be a valid VkSamplerAddressMode value

Bits which can be set in VkSamplerCreateInfo::flags, specifying additional parameters of a sampler, are:

// Provided by VK_VERSION_1_0
typedef enum VkSamplerCreateFlagBits {
  // Provided by VK_EXT_fragment_density_map
    VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT = 0x00000001,
  // Provided by VK_EXT_fragment_density_map
    VK_SAMPLER_CREATE_SUBSAMPLED_COARSE_RECONSTRUCTION_BIT_EXT = 0x00000002,
} VkSamplerCreateFlagBits;

VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT specifies that the sampler will read from an image created with flags containing VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT.
VK_SAMPLER_CREATE_SUBSAMPLED_COARSE_RECONSTRUCTION_BIT_EXT specifies that the implementation may use approximations when reconstructing a full color value for texture access from a subsampled image.

Note

The approximations used when VK_SAMPLER_CREATE_SUBSAMPLED_COARSE_RECONSTRUCTION_BIT_EXT is specified are implementation defined. Some implementations may interpolate between fragment density levels in a subsampled image. In that case, this bit may be used to decide whether the interpolation factors are calculated per fragment or at a coarser granularity.

// Provided by VK_VERSION_1_0
typedef VkFlags VkSamplerCreateFlags;

VkSamplerCreateFlags is a bitmask type for setting a mask of zero or more VkSamplerCreateFlagBits.

The VkSamplerReductionModeCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkSamplerReductionModeCreateInfo {
    VkStructureType           sType;
    const void*               pNext;
    VkSamplerReductionMode    reductionMode;
} VkSamplerReductionModeCreateInfo;

or the equivalent

// Provided by VK_EXT_sampler_filter_minmax
typedef VkSamplerReductionModeCreateInfo VkSamplerReductionModeCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
reductionMode is a VkSamplerReductionMode value controlling how texture filtering combines texel values.

If the pNext chain of VkSamplerCreateInfo includes a VkSamplerReductionModeCreateInfo structure, then that structure includes a mode 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.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SAMPLER_REDUCTION_MODE_CREATE_INFO
reductionMode must be a valid VkSamplerReductionMode value

Reduction modes are specified by VkSamplerReductionMode, which takes values:

// Provided by VK_VERSION_1_2
typedef enum VkSamplerReductionMode {
    VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE = 0,
    VK_SAMPLER_REDUCTION_MODE_MIN = 1,
    VK_SAMPLER_REDUCTION_MODE_MAX = 2,
  // Provided by VK_EXT_sampler_filter_minmax
    VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT = VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE,
  // Provided by VK_EXT_sampler_filter_minmax
    VK_SAMPLER_REDUCTION_MODE_MIN_EXT = VK_SAMPLER_REDUCTION_MODE_MIN,
  // Provided by VK_EXT_sampler_filter_minmax
    VK_SAMPLER_REDUCTION_MODE_MAX_EXT = VK_SAMPLER_REDUCTION_MODE_MAX,
} VkSamplerReductionMode;

or the equivalent

// Provided by VK_EXT_sampler_filter_minmax
typedef VkSamplerReductionMode VkSamplerReductionModeEXT;

VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE 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 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 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:

// Provided by VK_VERSION_1_0
typedef enum VkFilter {
    VK_FILTER_NEAREST = 0,
    VK_FILTER_LINEAR = 1,
  // Provided by VK_IMG_filter_cubic
    VK_FILTER_CUBIC_IMG = 1000015000,
  // Provided by VK_EXT_filter_cubic
    VK_FILTER_CUBIC_EXT = VK_FILTER_CUBIC_IMG,
} VkFilter;

VK_FILTER_NEAREST specifies nearest filtering.
VK_FILTER_LINEAR specifies linear filtering.
VK_FILTER_CUBIC_EXT 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:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_VERSION_1_2, VK_KHR_sampler_mirror_clamp_to_edge
    VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE = 4,
    VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE_KHR = VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE,
} 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 samplerMirrorClampToEdge is enabled, or 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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_EXT_custom_border_color
    VK_BORDER_COLOR_FLOAT_CUSTOM_EXT = 1000287003,
  // Provided by VK_EXT_custom_border_color
    VK_BORDER_COLOR_INT_CUSTOM_EXT = 1000287004,
} 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.
VK_BORDER_COLOR_FLOAT_CUSTOM_EXT indicates that a VkSamplerCustomBorderColorCreateInfoEXT structure is present in the VkSamplerCreateInfo::pNext chain which contains the color data in floating-point format.
VK_BORDER_COLOR_INT_CUSTOM_EXT indicates that a VkSamplerCustomBorderColorCreateInfoEXT structure is present in the VkSamplerCreateInfo::pNext chain which contains the color data in integer format.

These colors are described in detail in Texel Replacement.

To destroy a sampler, call:

// Provided by VK_VERSION_1_0
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′C_BC_R conversion

To create a sampler with Y′C_BC_R conversion enabled, add a VkSamplerYcbcrConversionInfo structure to the pNext chain of the VkSamplerCreateInfo structure. To create a sampler Y′C_BC_R 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′C_BC_R conversion for VK_IMAGE_ASPECT_COLOR_BIT image views , or if the image view has an external format .

The VkSamplerYcbcrConversionInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkSamplerYcbcrConversionInfo {
    VkStructureType             sType;
    const void*                 pNext;
    VkSamplerYcbcrConversion    conversion;
} VkSamplerYcbcrConversionInfo;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
typedef VkSamplerYcbcrConversionInfo VkSamplerYcbcrConversionInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
conversion is a VkSamplerYcbcrConversion handle created with vkCreateSamplerYcbcrConversion.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO
conversion must be a valid VkSamplerYcbcrConversion handle

A sampler Y′C_BC_R conversion is an opaque representation of a device-specific sampler Y′C_BC_R conversion description, represented as a VkSamplerYcbcrConversion handle:

// Provided by VK_VERSION_1_1
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkSamplerYcbcrConversion)

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
typedef VkSamplerYcbcrConversion VkSamplerYcbcrConversionKHR;

To create a VkSamplerYcbcrConversion, call:

// Provided by VK_VERSION_1_1
VkResult vkCreateSamplerYcbcrConversion(
    VkDevice                                    device,
    const VkSamplerYcbcrConversionCreateInfo*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSamplerYcbcrConversion*                   pYcbcrConversion);

or the equivalent command

// Provided by VK_KHR_sampler_ycbcr_conversion
VkResult vkCreateSamplerYcbcrConversionKHR(
    VkDevice                                    device,
    const VkSamplerYcbcrConversionCreateInfo*   pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkSamplerYcbcrConversion*                   pYcbcrConversion);

device is the logical device that creates the sampler Y′C_BC_R conversion.
pCreateInfo is a pointer to a VkSamplerYcbcrConversionCreateInfo structure specifying the requested sampler Y′C_BC_R conversion.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pYcbcrConversion is a pointer to a VkSamplerYcbcrConversion handle in which the resulting sampler Y′C_BC_R conversion is returned.

The interpretation of the configured sampler Y′C_BC_R conversion is described in more detail in the description of sampler Y′C_BC_R conversion in the Image Operations chapter.

Valid Usage

The sampler Y′C_BC_R conversion feature must be enabled

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkSamplerYcbcrConversionCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
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

// Provided by VK_KHR_sampler_ycbcr_conversion
typedef VkSamplerYcbcrConversionCreateInfo VkSamplerYcbcrConversionCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 not downsampled horizontally.
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 format supports VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_BIT the forceExplicitReconstruction value behaves as if it was set to VK_TRUE.

If the pNext chain includes a VkExternalFormatANDROID structure with non-zero externalFormat member, the sampler Y′C_BC_R 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
If an external format conversion is not being created, format must represent unsigned normalized values (i.e. the format must be a UNORM format)
The potential format features of the sampler Y′C_BC_R conversion must support VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT or VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT
If the potential format features of the sampler Y′C_BC_R conversion do not support VK_FORMAT_FEATURE_COSITED_CHROMA_SAMPLES_BIT, xChromaOffset and yChromaOffset must not be VK_CHROMA_LOCATION_COSITED_EVEN if the corresponding channels are downsampled
If the potential format features of the sampler Y′C_BC_R conversion do not support VK_FORMAT_FEATURE_MIDPOINT_CHROMA_SAMPLES_BIT, xChromaOffset and yChromaOffset must not be VK_CHROMA_LOCATION_MIDPOINT if the corresponding channels are downsampled
If the format has a _422 or _420 suffix, then components.g must be the identity swizzle
If the format has a _422 or _420 suffix, then components.a must be the identity swizzle, VK_COMPONENT_SWIZZLE_ONE, or VK_COMPONENT_SWIZZLE_ZERO
If the format has a _422 or _420 suffix, then components.r must be the identity swizzle or VK_COMPONENT_SWIZZLE_B
If the format has a _422 or _420 suffix, then components.b must be the identity swizzle or VK_COMPONENT_SWIZZLE_R
If the format has a _422 or _420 suffix, and if either components.r or components.b is the identity swizzle, both values must be the identity swizzle
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 ycbcrRange is VK_SAMPLER_YCBCR_RANGE_ITU_NARROW then the R, G and B channels obtained by applying the component swizzle to format must each have a bit-depth greater than or equal to 8
If the potential format features of the sampler Y′C_BC_R conversion do not support VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_CHROMA_RECONSTRUCTION_EXPLICIT_FORCEABLE_BIT forceExplicitReconstruction must be VK_FALSE
If the potential format features of the sampler Y′C_BC_R conversion do not support VK_FORMAT_FEATURE_SAMPLED_IMAGE_YCBCR_CONVERSION_LINEAR_FILTER_BIT, chromaFilter must not be VK_FILTER_LINEAR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_CREATE_INFO
pNext must be NULL or a pointer to a valid instance of VkExternalFormatANDROID
The sType value of each struct in the pNext chain must be unique
format must be a valid VkFormat value
ycbcrModel must be a valid VkSamplerYcbcrModelConversion value
ycbcrRange must be a valid VkSamplerYcbcrRange value
components must be a valid VkComponentMapping structure
xChromaOffset must be a valid VkChromaLocation value
yChromaOffset must be a valid VkChromaLocation value
chromaFilter must be a valid VkFilter value

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′C_BC_R conversion in the Image Operations chapter.

VkSamplerYcbcrModelConversion defines the conversion from the source color model to the shader color model. Possible values are:

// Provided by VK_VERSION_1_1
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,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_RGB_IDENTITY,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_IDENTITY,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020_KHR = VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020,
} VkSamplerYcbcrModelConversion;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
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′C_BC_R.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709 specifies the color model conversion from Y′C_BC_R to R′G′B′ defined in BT.709 and described in the “BT.709 Y’C_BC_R conversion” section of the Khronos Data Format Specification.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601 specifies the color model conversion from Y′C_BC_R to R′G′B′ defined in BT.601 and described in the “BT.601 Y’C_BC_R conversion” section of the Khronos Data Format Specification.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020 specifies the color model conversion from Y′C_BC_R to R′G′B′ defined in BT.2020 and described in the “BT.2020 Y’C_BC_R 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′C_BC_R range expansion and model conversion:

the Y (Y′ luma) channel corresponds to the G channel of an RGB image.
the CB (C_B or “U” blue color difference) channel corresponds to the B channel of an RGB image.
the CR (C_R 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:

// Provided by VK_VERSION_1_1
typedef enum VkSamplerYcbcrRange {
    VK_SAMPLER_YCBCR_RANGE_ITU_FULL = 0,
    VK_SAMPLER_YCBCR_RANGE_ITU_NARROW = 1,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_RANGE_ITU_FULL_KHR = VK_SAMPLER_YCBCR_RANGE_ITU_FULL,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_SAMPLER_YCBCR_RANGE_ITU_NARROW_KHR = VK_SAMPLER_YCBCR_RANGE_ITU_NARROW,
} VkSamplerYcbcrRange;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
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′C_BC_R 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 defines the location of downsampled chroma channel samples relative to the luma samples, and is defined as:

// Provided by VK_VERSION_1_1
typedef enum VkChromaLocation {
    VK_CHROMA_LOCATION_COSITED_EVEN = 0,
    VK_CHROMA_LOCATION_MIDPOINT = 1,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_CHROMA_LOCATION_COSITED_EVEN_KHR = VK_CHROMA_LOCATION_COSITED_EVEN,
  // Provided by VK_KHR_sampler_ycbcr_conversion
    VK_CHROMA_LOCATION_MIDPOINT_KHR = VK_CHROMA_LOCATION_MIDPOINT,
} VkChromaLocation;

or the equivalent

// Provided by VK_KHR_sampler_ycbcr_conversion
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′C_BC_R conversion, call:

// Provided by VK_VERSION_1_1
void vkDestroySamplerYcbcrConversion(
    VkDevice                                    device,
    VkSamplerYcbcrConversion                    ycbcrConversion,
    const VkAllocationCallbacks*                pAllocator);

or the equivalent command

// Provided by VK_KHR_sampler_ycbcr_conversion
void vkDestroySamplerYcbcrConversionKHR(
    VkDevice                                    device,
    VkSamplerYcbcrConversion                    ycbcrConversion,
    const VkAllocationCallbacks*                pAllocator);

device is the logical device that destroys the Y′C_BC_R 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

In addition to the predefined border color values, applications can provide a custom border color value by including the VkSamplerCustomBorderColorCreateInfoEXT structure in the VkSamplerCreateInfo::pNext chain.

The VkSamplerCustomBorderColorCreateInfoEXT structure is defined as:

// Provided by VK_EXT_custom_border_color
typedef struct VkSamplerCustomBorderColorCreateInfoEXT {
    VkStructureType      sType;
    const void*          pNext;
    VkClearColorValue    customBorderColor;
    VkFormat             format;
} VkSamplerCustomBorderColorCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
customBorderColor is a VkClearColorValue representing the desired custom sampler border color.
format is a VkFormat representing the format of the sampled image view(s). This field may be VK_FORMAT_UNDEFINED if the customBorderColorWithoutFormat feature is enabled.

Valid Usage

If provided format is not VK_FORMAT_UNDEFINED then the VkSamplerCreateInfo::borderColor type must match the sampled type of the provided format, as shown in the SPIR-V Sampled Type column of the Interpretation of Numeric Format table
If the customBorderColorWithoutFormat feature is not enabled then format must not be VK_FORMAT_UNDEFINED
If the sampler is used to sample an image view of VK_FORMAT_B4G4R4A4_UNORM_PACK16, VK_FORMAT_B5G6R5_UNORM_PACK16, or VK_FORMAT_B5G5R5A1_UNORM_PACK16 format then format must not be VK_FORMAT_UNDEFINED

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_SAMPLER_CUSTOM_BORDER_COLOR_CREATE_INFO_EXT
format must be a valid VkFormat value

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.

Shaders can also access buffers without going through descriptors by using Physical Storage Buffer Access to access them through 64-bit addresses.

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 views whose format features contain VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT.

Stores to storage images are supported in compute shaders for image views whose format features contain VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT.

Atomic operations on storage images are supported in compute shaders for image views whose format features contain VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT.

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 behavior 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 views whose format features contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT.

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′C_BC_R conversion or samples a subsampled image, 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. Inline Uniform Block

An inline uniform block (VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT) is almost identical to a uniform buffer, and differs only in taking its storage directly from the encompassing descriptor set instead of being backed by buffer memory. It is typically used to access a small set of constant data that does not require the additional flexibility provided by the indirection enabled when using a uniform buffer where the descriptor and the referenced buffer memory are decoupled. Compared to push constants, they allow reusing the same set of constant data across multiple disjoint sets of draw and dispatch commands.

Inline uniform block descriptors cannot be aggregated into arrays. Instead, the array size specified for an inline uniform block descriptor binding specifies the binding’s capacity in bytes.

13.1.12. 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.

13.1.13. Acceleration Structure

An acceleration structure (VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR) is a descriptor type that is used to retrieve scene geometry from within shaders bound to ray tracing pipelines. Shaders have read-only access to the memory.

13.2. Descriptor Sets

Descriptors are grouped together into descriptor set objects. A descriptor set object is an opaque object containing 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorSetLayout)

To create descriptor set layout objects, call:

// Provided by VK_VERSION_1_0
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 a VkDescriptorSetLayoutCreateInfo structure specifying the state of the descriptor set layout object.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pSetLayout is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

Information about the descriptor set layout is passed in a VkDescriptorSetLayoutCreateInfo structure:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 all elements of pBindings must not have a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT
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 bit set, flags must include VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT
If any binding has the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT 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)

sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO
pNext must be NULL or a pointer to a valid instance of VkDescriptorSetLayoutBindingFlagsCreateInfo
The sType value of each struct in the pNext chain must be unique
flags must be a valid combination of VkDescriptorSetLayoutCreateFlagBits values
If bindingCount is not 0, pBindings must be a valid pointer to an array of bindingCount valid VkDescriptorSetLayoutBinding structures

Bits which can be set in VkDescriptorSetLayoutCreateInfo::flags to specify options for descriptor set layout are:

// Provided by VK_VERSION_1_0
typedef enum VkDescriptorSetLayoutCreateFlagBits {
  // Provided by VK_VERSION_1_2
    VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT = 0x00000002,
  // Provided by VK_KHR_push_descriptor
    VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR = 0x00000001,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT_EXT = VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT,
} 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 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 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.

// Provided by VK_VERSION_1_0
typedef VkFlags VkDescriptorSetLayoutCreateFlags;

VkDescriptorSetLayoutCreateFlags is a bitmask type for setting a mask of zero or more VkDescriptorSetLayoutCreateFlagBits.

The VkDescriptorSetLayoutBinding structure is defined as:

// Provided by VK_VERSION_1_0
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 , except if descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT in which case descriptorCount is the size in bytes of the inline uniform block . 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 and must not be changed; updating a VK_DESCRIPTOR_TYPE_SAMPLER descriptor with immutable samplers is not allowed and updates to a VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER descriptor with immutable samplers does not modify the samplers (the image views are updated, but the sampler updates are ignored). If pImmutableSamplers is not NULL, then it points to an array of sampler handles that will be copied into the set layout and used for the corresponding binding. Only the sampler handles are copied; the sampler objects must not be destroyed before the final use of the set layout and any descriptor pools and sets created using it. 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 value of descriptorType is 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 descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount must be a multiple of 4
If descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount must be less than or equal to VkPhysicalDeviceInlineUniformBlockPropertiesEXT::maxInlineUniformBlockSize
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
The sampler objects indicated by pImmutableSamplers must not have a borderColor with one of the values VK_BORDER_COLOR_FLOAT_CUSTOM_EXT or VK_BORDER_COLOR_INT_CUSTOM_EXT

Valid Usage (Implicit)

descriptorType must be a valid VkDescriptorType value

If the pNext chain of a VkDescriptorSetLayoutCreateInfo structure includes a VkDescriptorSetLayoutBindingFlagsCreateInfo structure, then that structure includes an array of flags, one for each descriptor set layout binding.

The VkDescriptorSetLayoutBindingFlagsCreateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkDescriptorSetLayoutBindingFlagsCreateInfo {
    VkStructureType                    sType;
    const void*                        pNext;
    uint32_t                           bindingCount;
    const VkDescriptorBindingFlags*    pBindingFlags;
} VkDescriptorSetLayoutBindingFlagsCreateInfo;

or the equivalent

// Provided by VK_EXT_descriptor_indexing
typedef VkDescriptorSetLayoutBindingFlagsCreateInfo VkDescriptorSetLayoutBindingFlagsCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
bindingCount is zero or the number of elements in pBindingFlags.
pBindingFlags is a pointer to an array of VkDescriptorBindingFlags bitfields, one for each descriptor set layout binding.

If bindingCount is zero or if this structure is not included in the pNext chain, the VkDescriptorBindingFlags 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, VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT, or VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT
If an element of pBindingFlags includes VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT, then all other elements of VkDescriptorSetLayoutCreateInfo::pBindings must have a smaller value of binding
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingUniformBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT
If VkPhysicalDeviceDescriptorIndexingFeatures::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
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingStorageImageUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_STORAGE_IMAGE must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingStorageBufferUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT
If VkPhysicalDeviceDescriptorIndexingFeatures::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
If VkPhysicalDeviceDescriptorIndexingFeatures::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
If VkPhysicalDeviceInlineUniformBlockFeaturesEXT::descriptorBindingInlineUniformBlockUpdateAfterBind is not enabled, all bindings with descriptor type VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT must not use VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT
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
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingUpdateUnusedWhilePending is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingPartiallyBound is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT
If VkPhysicalDeviceDescriptorIndexingFeatures::descriptorBindingVariableDescriptorCount is not enabled, all elements of pBindingFlags must not include VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT
If an element of pBindingFlags includes VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT, 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
If bindingCount is not 0, pBindingFlags must be a valid pointer to an array of bindingCount valid combinations of VkDescriptorBindingFlagBits values

Bits which can be set in each element of VkDescriptorSetLayoutBindingFlagsCreateInfo::pBindingFlags to specify options for the corresponding descriptor set layout binding are:

// Provided by VK_VERSION_1_2
typedef enum VkDescriptorBindingFlagBits {
    VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT = 0x00000001,
    VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT = 0x00000002,
    VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT = 0x00000004,
    VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT = 0x00000008,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT_EXT = VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT_EXT = VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT_EXT = VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT_EXT = VK_DESCRIPTOR_BINDING_VARIABLE_DESCRIPTOR_COUNT_BIT,
} VkDescriptorBindingFlagBits;

or the equivalent

// Provided by VK_EXT_descriptor_indexing
typedef VkDescriptorBindingFlagBits VkDescriptorBindingFlagBitsEXT;

VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT 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. Multiple descriptors with this flag set can be updated concurrently in different threads, though the same descriptor must not be updated concurrently by two threads. Descriptors with this flag set 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 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 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 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 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 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 , except for descriptor bindings with a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT where descriptorCount specifies the upper bound on the byte size of the binding, thus it counts against the maxInlineUniformBlockSize limit instead. .

Note

Note that while VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT and VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT both involve updates to descriptor sets after they are bound, VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT is a weaker requirement since it is only about descriptors that are not used, whereas VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT requires the implementation to observe updates to descriptors that are used.

// Provided by VK_VERSION_1_2
typedef VkFlags VkDescriptorBindingFlags;

or the equivalent

// Provided by VK_EXT_descriptor_indexing
typedef VkDescriptorBindingFlags VkDescriptorBindingFlagsEXT;

VkDescriptorBindingFlags is a bitmask type for setting a mask of zero or more VkDescriptorBindingFlagBits.

To query information about whether a descriptor set layout can be created, call:

// Provided by VK_VERSION_1_1
void vkGetDescriptorSetLayoutSupport(
    VkDevice                                    device,
    const VkDescriptorSetLayoutCreateInfo*      pCreateInfo,
    VkDescriptorSetLayoutSupport*               pSupport);

or the equivalent command

// Provided by VK_KHR_maintenance3
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 a VkDescriptorSetLayoutCreateInfo structure specifying the state of the descriptor set layout object.
pSupport is a pointer 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 a VkDescriptorSetLayoutSupport structure:

// Provided by VK_VERSION_1_1
typedef struct VkDescriptorSetLayoutSupport {
    VkStructureType    sType;
    void*              pNext;
    VkBool32           supported;
} VkDescriptorSetLayoutSupport;

or the equivalent

// Provided by VK_KHR_maintenance3
typedef VkDescriptorSetLayoutSupport VkDescriptorSetLayoutSupportKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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)

sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_SUPPORT
pNext must be NULL or a pointer to a valid instance of VkDescriptorSetVariableDescriptorCountLayoutSupport
The sType value of each struct in the pNext chain must be unique

If the pNext chain of a VkDescriptorSetLayoutSupport structure includes a VkDescriptorSetVariableDescriptorCountLayoutSupport structure, then that structure returns additional information about whether the descriptor set layout is supported.

// Provided by VK_VERSION_1_2
typedef struct VkDescriptorSetVariableDescriptorCountLayoutSupport {
    VkStructureType    sType;
    void*              pNext;
    uint32_t           maxVariableDescriptorCount;
} VkDescriptorSetVariableDescriptorCountLayoutSupport;

or the equivalent

// Provided by VK_EXT_descriptor_indexing
typedef VkDescriptorSetVariableDescriptorCountLayoutSupport VkDescriptorSetVariableDescriptorCountLayoutSupportEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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 highest numbered binding of the layout has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then maxVariableDescriptorCount indicates the maximum byte size supported for the binding, 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 VkPhysicalDeviceDescriptorIndexingFeatures::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

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:

// Provided by VK_VERSION_1_0
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 describing 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPipelineLayout)

To create a pipeline layout, call:

// Provided by VK_VERSION_1_0
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 a VkPipelineLayoutCreateInfo structure specifying the state of the pipeline layout object.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pPipelineLayout is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkPipelineLayoutCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 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 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 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 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 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 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 bindings in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceInlineUniformBlockPropertiesEXT::maxPerStageDescriptorInlineUniformBlocks
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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::maxPerStageDescriptorUpdateAfterBindInputAttachments
The total number of bindings with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT accessible to any given shader stage across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceInlineUniformBlockPropertiesEXT::maxPerStageDescriptorUpdateAfterBindInlineUniformBlocks
The total number of descriptors in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT 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 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 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 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 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 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 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 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 bindings in descriptor set layouts created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT bit set with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceInlineUniformBlockPropertiesEXT::maxDescriptorSetInlineUniformBlocks
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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::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 VkPhysicalDeviceDescriptorIndexingProperties::maxDescriptorSetUpdateAfterBindInputAttachments
The total number of bindings with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceInlineUniformBlockPropertiesEXT::maxDescriptorSetUpdateAfterBindInlineUniformBlocks
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
The total number of bindings with a descriptorType of VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR accessible across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceRayTracingPropertiesKHR::maxDescriptorSetAccelerationStructures
The total number of pImmutableSamplers created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT or VK_SAMPLER_CREATE_SUBSAMPLED_COARSE_RECONSTRUCTION_BIT_EXT across all shader stages and across all elements of pSetLayouts must be less than or equal to VkPhysicalDeviceFragmentDensityMap2PropertiesEXT::maxDescriptorSetSubsampledSamplers

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

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 return undefined values.
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)

stageFlags must be a valid combination of VkShaderStageFlagBits values
stageFlags must not be 0

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
`maxDescriptorSetInlineUniformBlocks` or `maxDescriptorSetUpdateAfterBindInlineUniformBlocks`	inline uniform block
`maxDescriptorSetAccelerationStructures`	acceleration structure

To destroy a pipeline layout, call:

// Provided by VK_VERSION_1_0
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
pipelineLayout must not have been passed to any vkCmd* command for any command buffers that are still in the recording state when vkDestroyPipelineLayout is called

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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorPool)

To create a descriptor pool object, call:

// Provided by VK_VERSION_1_0
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 a VkDescriptorPoolCreateInfo structure specifying the state of the descriptor pool object.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pDescriptorPool is a pointer 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY
VK_ERROR_FRAGMENTATION_EXT

Additional information about the pool is passed in a VkDescriptorPoolCreateInfo structure:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 bit set, descriptor pool creation may fail with the error VK_ERROR_FRAGMENTATION 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 or a pointer to a valid instance of VkDescriptorPoolInlineUniformBlockCreateInfoEXT
The sType value of each struct in the pNext chain must be unique
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

In order to be able to allocate descriptor sets having inline uniform block bindings the descriptor pool must be created with specifying the inline uniform block binding capacity of the descriptor pool, in addition to the total inline uniform data capacity in bytes which is specified through a VkDescriptorPoolSize structure with a descriptorType value of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT. This can be done by adding a VkDescriptorPoolInlineUniformBlockCreateInfoEXT structure to the pNext chain of VkDescriptorPoolCreateInfo.

The VkDescriptorPoolInlineUniformBlockCreateInfoEXT structure is defined as:

// Provided by VK_EXT_inline_uniform_block
typedef struct VkDescriptorPoolInlineUniformBlockCreateInfoEXT {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           maxInlineUniformBlockBindings;
} VkDescriptorPoolInlineUniformBlockCreateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
maxInlineUniformBlockBindings is the number of inline uniform block bindings to allocate.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_INLINE_UNIFORM_BLOCK_CREATE_INFO_EXT

Bits which can be set in VkDescriptorPoolCreateInfo::flags to enable operations on a descriptor pool are:

// Provided by VK_VERSION_1_0
typedef enum VkDescriptorPoolCreateFlagBits {
    VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT = 0x00000001,
  // Provided by VK_VERSION_1_2
    VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT = 0x00000002,
  // Provided by VK_EXT_descriptor_indexing
    VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT_EXT = VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT,
} 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 specifies that descriptor sets allocated from this pool can include bindings with the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT bit set. It is valid to allocate descriptor sets that have bindings that do not set the VK_DESCRIPTOR_BINDING_UPDATE_AFTER_BIND_BIT bit from a pool that has VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT set.

// Provided by VK_VERSION_1_0
typedef VkFlags VkDescriptorPoolCreateFlags;

VkDescriptorPoolCreateFlags is a bitmask type for setting a mask of zero or more VkDescriptorPoolCreateFlagBits.

The VkDescriptorPoolSize structure is defined as:

// Provided by VK_VERSION_1_0
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. If type is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount is the number of bytes to allocate for descriptors of this type.

Note

When creating a descriptor pool that will contain descriptors for combined image samplers of multi-planar formats, an application needs to account for non-trivial descriptor consumption when choosing the descriptorCount value, as indicated by VkSamplerYcbcrConversionImageFormatProperties::combinedImageSamplerDescriptorCount.

Valid Usage

descriptorCount must be greater than 0
If type is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount must be a multiple of 4

Valid Usage (Implicit)

type must be a valid VkDescriptorType value

To destroy a descriptor pool, call:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorSet)

To allocate descriptor sets from a descriptor pool, call:

// Provided by VK_VERSION_1_0
VkResult vkAllocateDescriptorSets(
    VkDevice                                    device,
    const VkDescriptorSetAllocateInfo*          pAllocateInfo,
    VkDescriptorSet*                            pDescriptorSets);

device is the logical device that owns the descriptor pool.
pAllocateInfo is a pointer to a 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. Descriptors also become undefined if the underlying resource is destroyed. Descriptor sets containing undefined descriptors can still be bound and used, subject to the following conditions:

For descriptor set bindings created with the VK_DESCRIPTOR_BINDING_PARTIALLY_BOUND_BIT 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 bit set, all descriptors in that binding that are statically used must have been populated before the descriptor set is consumed.
Descriptor bindings with descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT can be undefined when the descriptor set is consumed; though values in that block will be undefined.
Entries that are not used by a pipeline can have undefined 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 pAllocateInfo->descriptorPool, 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.

Additionally, the allocation may also fail if a call to vkAllocateDescriptorSets would cause the total number of inline uniform block bindings allocated from the pool to exceed the value of VkDescriptorPoolInlineUniformBlockCreateInfoEXT::maxInlineUniformBlockBindings used to create the descriptor pool.

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
pAllocateInfo->descriptorSetCount must be greater than 0

Host Synchronization

Host access to pAllocateInfo->descriptorPool must be externally synchronized

Return Codes

VK_SUCCESS

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:

// Provided by VK_VERSION_1_0
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 a structure extending this 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 a pointer to 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 bit set, descriptorPool must have been created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT 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 VkDescriptorSetVariableDescriptorCountAllocateInfo
The sType value of each struct in the pNext chain must be unique
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 VkDescriptorSetVariableDescriptorCountAllocateInfo structure, then that structure includes an array of descriptor counts for variable descriptor count bindings, one for each descriptor set being allocated.

The VkDescriptorSetVariableDescriptorCountAllocateInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkDescriptorSetVariableDescriptorCountAllocateInfo {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           descriptorSetCount;
    const uint32_t*    pDescriptorCounts;
} VkDescriptorSetVariableDescriptorCountAllocateInfo;

or the equivalent

// Provided by VK_EXT_descriptor_indexing
typedef VkDescriptorSetVariableDescriptorCountAllocateInfo VkDescriptorSetVariableDescriptorCountAllocateInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
descriptorSetCount is zero or the number of elements in pDescriptorCounts.
pDescriptorCounts is a pointer to 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 the variable count descriptor binding in the corresponding descriptor set layout has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then pDescriptorCounts[i] specifies the binding’s capacity in bytes. 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
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:

// Provided by VK_VERSION_1_0
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 a pointer to an array of handles to VkDescriptorSet objects.

After calling 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

VK_SUCCESS

To return all descriptor sets allocated from a given pool to the pool, rather than freeing individual descriptor sets, call:

// Provided by VK_VERSION_1_0
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

VK_SUCCESS

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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 or VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT 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 or VK_DESCRIPTOR_BINDING_UPDATE_UNUSED_WHILE_PENDING_BIT 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:

// Provided by VK_VERSION_1_0
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 a structure extending this 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. If the descriptor binding identified by dstSet and dstBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then dstArrayElement specifies the starting byte offset within the binding.
descriptorCount is the number of descriptors to update (the number of elements in pImageInfo, pBufferInfo, or pTexelBufferView , or a value matching the dataSize member of a VkWriteDescriptorSetInlineUniformBlockEXT structure in the pNext chain , or a value matching the accelerationStructureCount of a VkWriteDescriptorSetAccelerationStructureKHR structure in the pNext chain ). If the descriptor binding identified by dstSet and dstBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount specifies the number of bytes to update.
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 is a pointer to an array of VkDescriptorImageInfo structures or is ignored, as described below.
pBufferInfo is a pointer to an array of VkDescriptorBufferInfo structures or is ignored, as described below.
pTexelBufferView is a pointer 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, or none of them in case descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, in which case the source data for the descriptor writes is taken from the VkWriteDescriptorSetInlineUniformBlockEXT structure included in the pNext chain of VkWriteDescriptorSet, or if descriptorType is VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR, in which case the source data for the descriptor writes is taken from the VkWriteDescriptorSetAccelerationStructureKHR structure in the pNext chain of VkWriteDescriptorSet, as specified below.

If the nullDescriptor feature is enabled, the buffer, imageView, or bufferView can be VK_NULL_HANDLE. Loads from a null descriptor return zero values and stores and atomics to a null descriptor are discarded.

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.

Note

The same behavior applies to bindings with a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT where descriptorCount specifies the number of bytes to update while dstArrayElement specifies the starting byte offset, thus in this case if the dstBinding has a smaller byte size than the sum of dstArrayElement and descriptorCount, then the remainder will be used to update the subsequent binding - dstBinding+1 starting at offset zero. This falls out as a special case of the above rule.

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_INLINE_UNIFORM_BLOCK_EXT, dstArrayElement must be an integer multiple of 4
If descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, descriptorCount must be an integer multiple of 4
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, each element of pTexelBufferView must be either a valid VkBufferView handle or VK_NULL_HANDLE
If descriptorType is VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER or VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER and the nullDescriptor feature is not enabled, each element of pTexelBufferView must not be VK_NULL_HANDLE
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 member of each element of pImageInfo must be either a valid VkImageView handle or VK_NULL_HANDLE
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 and the nullDescriptor feature is not enabled, the imageView member of each element of pImageInfo must not be VK_NULL_HANDLE
If descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, the pNext chain must include a VkWriteDescriptorSetInlineUniformBlockEXT structure whose dataSize member equals descriptorCount
If descriptorType is VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR, the pNext chain must include a VkWriteDescriptorSetAccelerationStructureKHR structure whose accelerationStructureCount member equals descriptorCount
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, and the corresponding immutable sampler must have been created with an identically defined VkSamplerYcbcrConversionInfo object
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 an immutable sampler that enables sampler Y′C_BC_R 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 a member of the list given in Sampled Image or Combined Image Sampler, corresponding to its type
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 VkDescriptorBindingFlagBits
If descriptorType is VK_DESCRIPTOR_TYPE_SAMPLER, then dstSet must not have been allocated with a layout that included immutable samplers for dstBinding

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET
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 VkWriteDescriptorSetAccelerationStructureKHR or VkWriteDescriptorSetInlineUniformBlockEXT
The sType value of each struct in the pNext chain must be unique
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 of non-ignored parameters 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:

// Provided by VK_VERSION_1_0
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,
  // Provided by VK_EXT_inline_uniform_block
    VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT = 1000138000,
  // Provided by VK_KHR_ray_tracing
    VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR = 1000165000,
  // Provided by VK_NV_ray_tracing
    VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_NV = VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR,
} VkDescriptorType;

VK_DESCRIPTOR_TYPE_SAMPLER specifies a sampler descriptor.
VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER specifies a combined image sampler descriptor.
VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE specifies a sampled image descriptor.
VK_DESCRIPTOR_TYPE_STORAGE_IMAGE specifies a storage image descriptor.
VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER specifies a uniform texel buffer descriptor.
VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER specifies a storage texel buffer descriptor.
VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER specifies a uniform buffer descriptor.
VK_DESCRIPTOR_TYPE_STORAGE_BUFFER specifies a storage buffer descriptor.
VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC specifies a dynamic uniform buffer descriptor.
VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC specifies a dynamic storage buffer descriptor.
VK_DESCRIPTOR_TYPE_INPUT_ATTACHMENT specifies an input attachment descriptor.
VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT specifies an inline uniform block.

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 sampler 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.

When updating descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, none of the pImageInfo, pBufferInfo, or pTexelBufferView members are accessed, instead the source data of the descriptor update operation is taken from the VkWriteDescriptorSetInlineUniformBlockEXT structure in the pNext chain of VkWriteDescriptorSet. When updating descriptors with a descriptorType of VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR, none of the pImageInfo, pBufferInfo, or pTexelBufferView members are accessed, instead the source data of the descriptor update operation is taken from the VkWriteDescriptorSetAccelerationStructureKHR structure in the pNext chain of VkWriteDescriptorSet.

The VkDescriptorBufferInfo structure is defined as:

// Provided by VK_VERSION_1_0
typedef struct VkDescriptorBufferInfo {
    VkBuffer        buffer;
    VkDeviceSize    offset;
    VkDeviceSize    range;
} VkDescriptorBufferInfo;

buffer is VK_NULL_HANDLE or 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
If the nullDescriptor feature is not enabled, buffer must not be VK_NULL_HANDLE
If buffer is VK_NULL_HANDLE, offset must be zero and range must be VK_WHOLE_SIZE

Valid Usage (Implicit)

If buffer is not VK_NULL_HANDLE, buffer must be a valid VkBuffer handle

The VkDescriptorImageInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 VK_NULL_HANDLE or 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
If imageView is created from a depth/stencil image, the aspectMask used to create the imageView must include either VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT but not both
imageLayout must match the actual VkImageLayout of each subresource accessible from imageView at the time this descriptor is accessed as defined by the image layout matching rules
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 of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

If the descriptorType member of VkWriteDescriptorSet is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then the data to write to the descriptor set is specified through a VkWriteDescriptorSetInlineUniformBlockEXT structure included in the pNext chain of VkWriteDescriptorSet.

The VkWriteDescriptorSetInlineUniformBlockEXT structure is defined as:

// Provided by VK_EXT_inline_uniform_block
typedef struct VkWriteDescriptorSetInlineUniformBlockEXT {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           dataSize;
    const void*        pData;
} VkWriteDescriptorSetInlineUniformBlockEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
dataSize is the number of bytes of inline uniform block data pointed to by pData.
pData is a pointer to dataSize number of bytes of data to write to the inline uniform block.

Valid Usage

dataSize must be an integer multiple of 4

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET_INLINE_UNIFORM_BLOCK_EXT
pData must be a valid pointer to an array of dataSize bytes
dataSize must be greater than 0

The VkWriteDescriptorSetAccelerationStructureKHR structure is defined as:

// Provided by VK_KHR_ray_tracing
typedef struct VkWriteDescriptorSetAccelerationStructureKHR {
    VkStructureType                      sType;
    const void*                          pNext;
    uint32_t                             accelerationStructureCount;
    const VkAccelerationStructureKHR*    pAccelerationStructures;
} VkWriteDescriptorSetAccelerationStructureKHR;

or the equivalent

// Provided by VK_NV_ray_tracing
typedef VkWriteDescriptorSetAccelerationStructureKHR VkWriteDescriptorSetAccelerationStructureNV;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
accelerationStructureCount is the number of elements in pAccelerationStructures.
pAccelerationStructures are the acceleration structures to update.

Valid Usage

accelerationStructureCount must be equal to descriptorCount in the extended structure
Each acceleration structure in pAccelerationStructures must have been created with VK_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL_KHR

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET_ACCELERATION_STRUCTURE_KHR
pAccelerationStructures must be a valid pointer to an array of accelerationStructureCount valid VkAccelerationStructureKHR handles
accelerationStructureCount must be greater than 0

The VkCopyDescriptorSet structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this structure.
srcSet, srcBinding, and srcArrayElement are the source set, binding, and array element, respectively. If the descriptor binding identified by srcSet and srcBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then srcArrayElement specifies the starting byte offset within the binding to copy from.
dstSet, dstBinding, and dstArrayElement are the destination set, binding, and array element, respectively. If the descriptor binding identified by dstSet and dstBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then dstArrayElement specifies the starting byte offset within the binding to copy to.
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. If the descriptor binding identified by srcSet and srcBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount specifies the number of bytes to copy and the remaining array elements in the source or destination binding refer to the remaining number of bytes in those.

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
The type of dstBinding within dstSet must be equal to the type of srcBinding within srcSet
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 the descriptor type of the descriptor set binding specified by srcBinding is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, srcArrayElement must be an integer multiple of 4
If the descriptor type of the descriptor set binding specified by dstBinding is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, dstArrayElement must be an integer multiple of 4
If the descriptor type of the descriptor set binding specified by either srcBinding or dstBinding is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, descriptorCount must be an integer multiple of 4
If srcSet’s layout was created with the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT flag set, then dstSet’s layout must also have been created with the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT flag set
If srcSet’s layout was created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT flag set, then dstSet’s layout must also have been created without the VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT flag set
If the descriptor pool from which srcSet was allocated was created with the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT 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 flag set
If the descriptor pool from which srcSet was allocated was created without the VK_DESCRIPTOR_POOL_CREATE_UPDATE_AFTER_BIND_BIT 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 flag set
If the descriptor type of the descriptor set binding specified by dstBinding is VK_DESCRIPTOR_TYPE_SAMPLER, then dstSet must not have been allocated with a layout that included immutable samplers for dstBinding

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:

// Provided by VK_VERSION_1_1
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkDescriptorUpdateTemplate)

or the equivalent

// Provided by VK_KHR_descriptor_update_template
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:

// Provided by VK_VERSION_1_1
VkResult vkCreateDescriptorUpdateTemplate(
    VkDevice                                    device,
    const VkDescriptorUpdateTemplateCreateInfo* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorUpdateTemplate*                 pDescriptorUpdateTemplate);

or the equivalent command

// Provided by VK_KHR_descriptor_update_template
VkResult vkCreateDescriptorUpdateTemplateKHR(
    VkDevice                                    device,
    const VkDescriptorUpdateTemplateCreateInfo* pCreateInfo,
    const VkAllocationCallbacks*                pAllocator,
    VkDescriptorUpdateTemplate*                 pDescriptorUpdateTemplate);

device is the logical device that creates the descriptor update template.
pCreateInfo is a pointer to a VkDescriptorUpdateTemplateCreateInfo structure specifying the set of descriptors to update with a single call to vkCmdPushDescriptorSetWithTemplateKHR or vkUpdateDescriptorSetWithTemplate.
pAllocator controls host memory allocation as described in the Memory Allocation chapter.
pDescriptorUpdateTemplate is a pointer to a VkDescriptorUpdateTemplate handle in which the resulting descriptor update template object is returned.

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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkDescriptorUpdateTemplateCreateInfo structure is defined as:

// Provided by VK_VERSION_1_1
typedef struct VkDescriptorUpdateTemplateCreateInfo {
    VkStructureType                           sType;
    const 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

// Provided by VK_KHR_descriptor_update_template
typedef VkDescriptorUpdateTemplateCreateInfo VkDescriptorUpdateTemplateCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this 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
descriptorUpdateEntryCount must be greater than 0
Both of descriptorSetLayout, and pipelineLayout that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

// Provided by VK_VERSION_1_1
typedef VkFlags VkDescriptorUpdateTemplateCreateFlags;

or the equivalent

// Provided by VK_KHR_descriptor_update_template
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:

// Provided by VK_VERSION_1_1
typedef enum VkDescriptorUpdateTemplateType {
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET = 0,
  // Provided by VK_KHR_push_descriptor with VK_VERSION_1_1, VK_KHR_push_descriptor with VK_KHR_descriptor_update_template, VK_KHR_descriptor_update_template with VK_KHR_push_descriptor
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_PUSH_DESCRIPTORS_KHR = 1,
  // Provided by VK_KHR_descriptor_update_template
    VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET_KHR = VK_DESCRIPTOR_UPDATE_TEMPLATE_TYPE_DESCRIPTOR_SET,
} VkDescriptorUpdateTemplateType;

or the equivalent

// Provided by VK_KHR_descriptor_update_template
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:

// Provided by VK_VERSION_1_1
typedef struct VkDescriptorUpdateTemplateEntry {
    uint32_t            dstBinding;
    uint32_t            dstArrayElement;
    uint32_t            descriptorCount;
    VkDescriptorType    descriptorType;
    size_t              offset;
    size_t              stride;
} VkDescriptorUpdateTemplateEntry;

or the equivalent

// Provided by VK_KHR_descriptor_update_template
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. If the descriptor binding identified by srcBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then dstArrayElement specifies the starting byte offset to update.
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. If the descriptor binding identified by dstBinding has a descriptor type of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then descriptorCount specifies the number of bytes to update and the remaining array elements in the destination binding refer to the remaining number of bytes in it.
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. If descriptorType is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT then the value of stride is ignored and the stride is assumed to be 1, i.e. the descriptor update information for them is always specified as a contiguous range.

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
If descriptor type is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, dstArrayElement must be an integer multiple of 4
If descriptor type is VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT, descriptorCount must be an integer multiple of 4

Valid Usage (Implicit)

descriptorType must be a valid VkDescriptorType value

To destroy a descriptor update template, call:

// Provided by VK_VERSION_1_1
void vkDestroyDescriptorUpdateTemplate(
    VkDevice                                    device,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const VkAllocationCallbacks*                pAllocator);

or the equivalent command

// Provided by VK_KHR_descriptor_update_template
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:

// Provided by VK_VERSION_1_1
void vkUpdateDescriptorSetWithTemplate(
    VkDevice                                    device,
    VkDescriptorSet                             descriptorSet,
    VkDescriptorUpdateTemplate                  descriptorUpdateTemplate,
    const void*                                 pData);

or the equivalent command

// Provided by VK_KHR_descriptor_update_template
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 a VkDescriptorUpdateTemplate object specifying the update mapping between pData and the descriptor set to update.
pData is a pointer to memory containing one or more VkDescriptorImageInfo, VkDescriptorBufferInfo, or VkBufferView structures or VkAccelerationStructureKHR or VkAccelerationStructureNV handles used to write the descriptors.

Valid Usage

pData must be a valid pointer to a memory containing one or more valid instances of VkDescriptorImageInfo, VkDescriptorBufferInfo, or VkBufferView in a layout defined by descriptorUpdateTemplate when it was created with vkCreateDescriptorUpdateTemplate

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
    {
        1,                                           // 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
    {
        2,                                           // binding
        0,                                           // dstArrayElement
        2,                                           // descriptorCount
        VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER,     // descriptorType
        offsetof(AppDataStructure, bufferView) +
          offsetof(AppBufferView, bufferView),       // offset
        sizeof(AppBufferView)                        // stride, bufferViews do not have to be compact
    },
};

// create a 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:

// Provided by VK_VERSION_1_0
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 a pointer to 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 range 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 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 which corresponds to a descriptor binding with type VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC must be a multiple of VkPhysicalDeviceLimits::minUniformBufferOffsetAlignment
Each element of pDynamicOffsets which corresponds to a descriptor binding with type VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC must be a multiple of VkPhysicalDeviceLimits::minStorageBufferOffsetAlignment
For each dynamic uniform or storage buffer binding in pDescriptorSets, the sum of the effective offset, as defined above, and the range of the binding must be less than or equal to the size of the buffer

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

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:

// Provided by VK_KHR_push_descriptor
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 are undefined. 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 are undefined.

Push descriptors that are statically used by a pipeline must not be undefined 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 (the immutable samplers are pushed, rather than the samplers in pDescriptorWrites). Push descriptors that are not statically used can remain undefined.

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

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:

// Provided by VK_KHR_push_descriptor with VK_VERSION_1_1, VK_KHR_push_descriptor with VK_KHR_descriptor_update_template, VK_KHR_descriptor_update_template with VK_KHR_push_descriptor
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 is a descriptor update template defining 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 is a pointer to memory containing descriptors for the templated update.

Valid Usage

The pipelineBindPoint specified during the creation of the descriptor update template must be supported by the commandBuffer’s parent VkCommandPool’s queue family
pData must be a valid pointer to a memory containing one or more valid instances of VkDescriptorImageInfo, VkDescriptorBufferInfo, or VkBufferView in a layout defined by descriptorUpdateTemplate when it was created with vkCreateDescriptorUpdateTemplateKHR

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

Primary
Secondary

Both

Graphics
Compute

API example

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 a 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:

// Provided by VK_VERSION_1_0
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 a pointer to an array of size bytes containing the new push constant values.

Note

As stageFlags needs to include all flags the relevant push constant ranges were created with, any flags that are not supported by the queue family that the VkCommandPool used to allocate commandBuffer was created on are ignored.

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

Primary
Secondary

Both

Graphics
Compute

13.3. Physical Storage Buffer Access

To query a 64-bit buffer device address value through which buffer memory can be accessed in a shader, call:

// Provided by VK_VERSION_1_2
VkDeviceAddress vkGetBufferDeviceAddress(
    VkDevice                                    device,
    const VkBufferDeviceAddressInfo*            pInfo);

or the equivalent command

// Provided by VK_KHR_buffer_device_address
VkDeviceAddress vkGetBufferDeviceAddressKHR(
    VkDevice                                    device,
    const VkBufferDeviceAddressInfo*            pInfo);

or the equivalent command

// Provided by VK_EXT_buffer_device_address
VkDeviceAddress vkGetBufferDeviceAddressEXT(
    VkDevice                                    device,
    const VkBufferDeviceAddressInfo*            pInfo);

device is the logical device that the buffer was created on.
pInfo is a pointer to a VkBufferDeviceAddressInfo structure specifying the buffer to retrieve an address for.

The 64-bit return value is an address of the start of pInfo->buffer. The address range starting at this value and whose size is the size of the buffer can be used in a shader to access the memory bound to that buffer, using the SPV_KHR_physical_storage_buffer extension or the equivalent SPV_EXT_physical_storage_buffer extension and the PhysicalStorageBuffer storage class. For example, this value can be stored in a uniform buffer, and the shader can read the value from the uniform buffer and use it to do a dependent read/write to this buffer. A value of zero is reserved as a “null” pointer and must not be returned as a valid buffer device address. All loads, stores, and atomics in a shader through PhysicalStorageBuffer pointers must access addresses in the address range of some buffer.

If the buffer was created with a non-zero value of VkBufferOpaqueCaptureAddressCreateInfo::opaqueCaptureAddress or VkBufferDeviceAddressCreateInfoEXT::deviceAddress the return value will be the same address that was returned at capture time.

Valid Usage

The bufferDeviceAddress or VkPhysicalDeviceBufferDeviceAddressFeaturesEXT::bufferDeviceAddress feature must be enabled
If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice or VkPhysicalDeviceBufferDeviceAddressFeaturesEXT::bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkBufferDeviceAddressInfo structure

The VkBufferDeviceAddressInfo structure is defined as:

// Provided by VK_VERSION_1_2
typedef struct VkBufferDeviceAddressInfo {
    VkStructureType    sType;
    const void*        pNext;
    VkBuffer           buffer;
} VkBufferDeviceAddressInfo;

or the equivalent

// Provided by VK_KHR_buffer_device_address
typedef VkBufferDeviceAddressInfo VkBufferDeviceAddressInfoKHR;

or the equivalent

// Provided by VK_EXT_buffer_device_address
typedef VkBufferDeviceAddressInfo VkBufferDeviceAddressInfoEXT;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
buffer specifies the buffer whose address is being queried.

Valid Usage

If buffer is non-sparse and was not created with the VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT flag, then it must be bound completely and contiguously to a single VkDeviceMemory object
buffer must have been created with VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_BUFFER_DEVICE_ADDRESS_INFO
pNext must be NULL
buffer must be a valid VkBuffer handle

To query a 64-bit buffer opaque capture address, call:

// Provided by VK_VERSION_1_2
uint64_t vkGetBufferOpaqueCaptureAddress(
    VkDevice                                    device,
    const VkBufferDeviceAddressInfo*            pInfo);

or the equivalent command

// Provided by VK_KHR_buffer_device_address
uint64_t vkGetBufferOpaqueCaptureAddressKHR(
    VkDevice                                    device,
    const VkBufferDeviceAddressInfo*            pInfo);

device is the logical device that the buffer was created on.
pInfo is a pointer to a VkBufferDeviceAddressInfo structure specifying the buffer to retrieve an address for.

The 64-bit return value is an opaque capture address of the start of pInfo->buffer.

If the buffer was created with a non-zero value of VkBufferOpaqueCaptureAddressCreateInfo::opaqueCaptureAddress the return value must be the same address.

Valid Usage

The bufferDeviceAddress feature must be enabled
If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkBufferDeviceAddressInfo structure

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.

Shader Input and Output Interface
Vertex Input Interface
Fragment Output Interface
Fragment Input Attachment Interface
Ray Tracing Pipeline Interface
Shader Resource Interface

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.

All the variables forming the shader input and 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. These generally form the interfaces between consecutive shader stages, regardless of any non-shader stages between the consecutive shader stages.

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.

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.

There must be no more than one built-in interface block per shader per interface.

Built-ins must not have any Location or Component decorations.

14.1.2. User-defined Variable Interface

The non-built-in variables listed by OpEntryPoint with the Input or Output storage class form the user-defined variable interface. These must have SPIR-V numerical types or, recursively, composite types of such types. By default, the components of such types have a width of 32 or 64 bits. If an implementation supports storageInputOutput16, components can also have a width of 16 bits. These variables must be identified with a Location decoration and can also be identified with a Component decoration.

14.1.3. Interface Matching

Interface matching rules only apply to built-ins when they are declared as members of the built-in interface block.

Tessellation control and mesh 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.

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. XfbBuffer, XfbStride, Offset, and Stream are also not required to match for the purposes of interface matching. 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.

At an interface between two non-fragment shader stages, the built-in interface block must match exactly, as described above, except for per-view outputs as described in Mesh Shader Per-View Outputs. 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 location assignments for user-defined variables and how many locations are consumed by a given user-variable type. As mentioned above, some inputs and outputs 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.

An OpVariable with a structure type that is not a block must be decorated with a Location.

When an OpVariable with a structure type (either block or non-block) is decorated with a Location, the members in the structure type must not be decorated with a Location. The OpVariable’s members are assigned consecutive locations in declaration order, starting from the first member, which is assigned the location decoration from the OpVariable.

When a block-type OpVariable is declared without a Location decoration, each member in its structure type must be decorated with a Location. Types nested deeper than the top-level members must not have Location decorations.

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. All variables in both the built-in interface block and the user-defined variable interface count against these limits. Each effective Location must have a value less than the number of locations available for the given interface, as specified in the "Locations Available" column 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.

See Attribute Location and Component Assignment for details.

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 describing 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 values to blending or color attachment writes are 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 resulting values are 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 describing 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. Ray Tracing Pipeline Interface

Ray tracing pipelines may have more stages than other pipelines with multiple instances of each stage and more dynamic interactions between the stages, but still has interface structures that obey the same generally rules as interfaces between shader stages in other pipelines. The three types of inter-stage interface variables for ray tracing pipelines are:

Ray payloads which contain data tracked for the entire lifetime of the ray.
Hit attributes which contain data about a specific hit for the duration of its processing.
Callable data for passing data into and out of a callable shader.

Ray payloads and callable data are used in explicit shader call instructions, so they have an incoming variant to distinguish the parameter passed to the invocation from any other payloads or data being used by subsequent shader call instructions.

An interface structure used between stages must match between the stages using it. Specifically:

The hit attribute structure read in an any-hit or closest-hit shader must be the same structure as the hit attribute structure written in the corresponding intersection shader in the same hit group.
The incoming callable data for a callable shader must be the same structure as the callable data referenced by the execute callable instruction in the calling shader.
The ray payload for a shader invoked by a trace ray command must be the same structure for all shader stages using the payload for that ray.

Any shader with an incoming ray payload, incoming callable data, or hit attribute must only declare one variable of that type.

Table 19. Ray Pipeline Shader Interface
Shader Stage	Ray Payload	Incoming Ray Payload	Hit Attribute	Callable Data	Incoming Callable Data
Ray Generation	r/w	r/w		r/w
Intersection			r/w
Any Hit			r
Closest Hit	r/w	r/w	r	r/w
Miss	r/w	r/w		r/w
Callable				r/w	r/w

14.6. 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.

14.6.1. 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 statically used member of a push constant block must be placed at an Offset such that the entire member 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 member of a push constant block must not cause the space required for that member to extend outside the range [0, maxPushConstantsSize).

Any member of a push constant block that is declared as an array must only be accessed with dynamically uniform indices.

14.6.2. Descriptor Set Interface

The descriptor set interface is comprised of the shader variables with the storage class of StorageBuffer, 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.

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.
For OpImageRead, if Dim is SubpassData (indicating a read from an input attachment).

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 the Uniform storage class can also be used to access transparent descriptor set backed resources when the variable is assigned to a descriptor set layout binding with a descriptorType of VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT. In this case the variable must be typed as OpTypeStruct and cannot be aggregated into arrays of that type. Further, the Offset decoration for any member of such a variable must not cause the space required for that variable to extend outside the range [0,maxInlineUniformBlockSize).

Variables identified with a storage class of UniformConstant and a decoration of InputAttachmentIndex must be declared as described in Fragment Input Attachment Interface.

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 StorageImageArrayNonUniformIndexing
Storage texel buffers:
- Dynamically uniform: shaderStorageTexelBufferArrayDynamicIndexing and StorageTexelBufferArrayDynamicIndexing
- Non-uniform: shaderStorageTexelBufferArrayNonUniformIndexing and StorageTexelBufferArrayNonUniformIndexing
Input attachments:
- Dynamically uniform: shaderInputAttachmentArrayDynamicIndexing and InputAttachmentArrayDynamicIndexing
- Non-uniform: shaderInputAttachmentArrayNonUniformIndexing and InputAttachmentArrayNonUniformIndexing
Sampled images (except uniform texel buffers), samplers and combined image samplers:
- Dynamically uniform: shaderSampledImageArrayDynamicIndexing and SampledImageArrayDynamicIndexing
- Non-uniform: shaderSampledImageArrayNonUniformIndexing and SampledImageArrayNonUniformIndexing
Uniform texel buffers:
- Dynamically uniform: shaderUniformTexelBufferArrayDynamicIndexing and UniformTexelBufferArrayDynamicIndexing
- Non-uniform: shaderUniformTexelBufferArrayNonUniformIndexing and UniformTexelBufferArrayNonUniformIndexing
Uniform buffers:
- Dynamically uniform: shaderUniformBufferArrayDynamicIndexing and UniformBufferArrayDynamicIndexing
- Non-uniform: shaderUniformBufferArrayNonUniformIndexing and UniformBufferArrayNonUniformIndexing
Storage buffers:
- Dynamically uniform: shaderStorageBufferArrayDynamicIndexing and StorageBufferArrayDynamicIndexing
- Non-uniform: shaderStorageBufferArrayNonUniformIndexing and StorageBufferArrayNonUniformIndexing
Acceleration structures:
- No additional capabilities needed.

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 the combined image sampler enables sampler Y′C_BC_R conversion or samples a subsampled image, it must be indexed only by constant integral expressions when aggregated into arrays in shader code, irrespective of the shaderSampledImageArrayDynamicIndexing feature.

Table 20. 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`
inline uniform block	`VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT`
acceleration structure	`VK_DESCRIPTOR_TYPE_ACCELERATION_STRUCTURE_KHR`

Table 21. 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`)
storage buffer	`StorageBuffer`	`OpTypeStruct`	`Block`, `Offset`, (`ArrayStride`), (`MatrixStride`)
input attachment	`UniformConstant`	`OpTypeImage` (`Dim`=`SubpassData`, `Sampled`=2)	`InputAttachmentIndex`
inline uniform block	`Uniform`	`OpTypeStruct`	`Block`, `Offset`, (`ArrayStride`), (`MatrixStride`)
acceleration structure	`UniformConstant`	`OpTypeAccelerationStructureKHR`

1: in addition to DescriptorSet and Binding

14.6.3. 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 RuntimeDescriptorArray 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 variable must be of a valid resource type determined by its SPIR-V type and storage class, as defined in Shader Resource and Storage Class Correspondence. 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 22. 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
`maxPerStageDescriptorInlineUniformBlocks` or `maxPerStageDescriptorUpdateAfterBindInlineUniformBlocks`	inline uniform block
`maxDescriptorSetAccelerationStructures`	acceleration structure

1: Input attachments can only be used in the fragment shader stage

14.6.4. Offset and Stride Assignment

Certain objects must be explicitly laid out using the Offset, ArrayStride, and MatrixStride, as described in SPIR-V explicit layout validation rules. All such layouts also must conform to the following requirements.

Note

The numeric order of Offset decorations does not need to follow member declaration order.

Alignment Requirements

There are different alignment requirements depending on the specific resources and on the features enabled on the device.

The scalar alignment of the type of an OpTypeStruct member is defined recursively as follows:

A scalar of size N has a scalar alignment of N.
A vector or matrix type has a scalar alignment equal to that of its component type.
An array type has a scalar alignment equal to that of its element type.
A structure has a scalar alignment equal to the largest scalar alignment of any of its members.

The base alignment of the type of an OpTypeStruct member is defined recursively as follows:

A scalar has a base alignment equal to its scalar alignment.
A two-component vector has a base alignment equal to twice its scalar alignment.
A three- or four-component vector has a base alignment equal to four times its scalar alignment.
An array has a base alignment equal to the base alignment of its element type.
A structure has a base alignment equal to the largest base alignment of any of its members.
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.

The extended alignment of the type of an OpTypeStruct member is similarly defined as follows:

A scalar, vector or matrix type has an extended alignment equal to its base alignment.
An array or structure type has an extended alignment equal to the largest extended alignment of any of its members, rounded up to a multiple of 16.

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.

Standard Buffer Layout

Every member of an OpTypeStruct that is required to be explicitly laid out must be aligned according to the first matching rule as follows. If the struct is contained in pointer types of multiple storage classes, it must satisfy the requirements for every storage class used to reference it.

If the scalarBlockLayout feature is enabled on the device then every member must be aligned according to its scalar alignment.
All vectors must be aligned according to their scalar alignment.
If the uniformBufferStandardLayout feature is not enabled on the device, then any member of an OpTypeStruct with a storage class of Uniform and a decoration of Block must be aligned according to its extended alignment.
Every other member must be aligned according to its base alignment.

Note

Even if scalar alignment is supported, it is generally more performant to use the base alignment.

The memory layout must obey the following rules:

The Offset decoration of any member must be a multiple of its alignment.
Any ArrayStride or MatrixStride decoration must be a multiple of the alignment of the array or matrix as defined above.

Unless the scalarBlockLayout feature is enabled on the device:

Vectors must not improperly straddle, as defined 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 alignment of that structure or array.

Note

The std430 layout in GLSL satisfies these rules for types using the base alignment. The std140 layout satisfies the rules for types using the extended alignment.

14.7. Built-In Variables

Built-in variables are accessed in shaders by declaring a variable decorated with a BuiltIn SPIR-V 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.

As mentioned above, some inputs and outputs have an additional level of arrayness relative to other shader inputs and outputs. This level of arrayness is not included in the type descriptions below, but must be included when declaring the built-in.

BaryCoordNV: The BaryCoordNV decoration can be used to decorate a fragment shader input variable. This variable will contain a three-component floating-point vector with barycentric weights that indicate the location of the fragment relative to the screen-space locations of vertices of its primitive, obtained using perspective interpolation.

The BaryCoordNV decoration must be used only within fragment shaders.

The variable decorated with BaryCoordNV must be declared using the Input storage class.

The variable decorated with BaryCoordNV must be declared as three-component vector of 32-bit floating-point values.
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 fragment’s center. The K coordinate of the barycentric coordinates can be derived given the identity I + J + K = 1.0.

BaryCoordNoPerspNV: The BaryCoordNoPerspNV decoration can be used to decorate a fragment shader input variable. This variable will contain a three-component floating-point vector with barycentric weights that indicate the location of the fragment relative to the screen-space locations of vertices of its primitive, obtained using linear interpolation.

The BaryCoordNoPerspNV decoration must be used only within fragment shaders.

The variable decorated with BaryCoordNoPerspNV must be declared using the Input storage class.

The variable decorated with BaryCoordNoPerspNV must be declared as three-component vector of 32-bit floating-point values.
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 fragment 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 fragment’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 BaseVertex 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 i^th 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 mesh, vertex, fragment, tessellation control, tessellation evaluation, and geometry shaders.

In mesh or 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.

ClipDistancePerViewNV: Decorating a variable with the ClipDistancePerViewNV built-in decoration will make that variable contain the per-view clip distances. The per-view clip distances have the same semantics as ClipDistance.

The ClipDistancePerViewNV must be used only within mesh shaders.

Any variable decorated with ClipDistancePerViewNV must be declared using the Output storage class, and must also be decorated with the PerViewNV decoration.

Any variable decorated with ClipDistancePerViewNV must be declared as a two-dimensional array of 32-bit floating-point 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 mesh, vertex, fragment, tessellation control, tessellation evaluation, and geometry shaders.

In mesh or 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.

CullDistancePerViewNV: Decorating a variable with the CullDistancePerViewNV built-in decoration will make that variable contain the per-view cull distances. The per-view clip distances have the same semantics as CullDistance.

The CullDistancePerViewNV must be used only within mesh shaders.

Any variable decorated with CullDistancePerViewNV must be declared using the Output storage class, and must also be decorated with the PerViewNV decoration.

Any variable decorated with CullDistancePerViewNV must be declared as a two-dimensional array of 32-bit floating-point values.

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 $[0, m a x (1, p h y s i c a l D e v i c e C o u n t))$ , 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 task, mesh, or 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 task, mesh or 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.

When task or mesh shaders are used, only the first active stage will have proper access to the variable. The value read by other stages is undefined.
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.

Otherwise, the x and y components of FragCoord reflect the location of the center of the fragment.

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: To have a shader supply a fragment-depth value, the shader must declare the DepthReplacing execution mode. Such a shader’s fragment-depth value will come from the variable decorated with the FragDepth built-in decoration.

This value will be used for any subsequent depth testing performed by the implementation or writes to the depth attachment.

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.

FragInvocationCountEXT: Decorating a variable with the FragInvocationCountEXT built-in decoration will make that variable contain the maximum number of fragment shader invocations for the fragment, as determined by minSampleShading.

The FragInvocationCountEXT decoration must be used only within fragment shaders and the FragmentDensityEXT capability must be declared.

If Sample Shading is not enabled, FragInvocationCountEXT will be filled with a value of 1.

The variable decorated with FragInvocationCountEXT must be declared using the Input storage class.

The variable decorated with FragInvocationCountEXT must be declared as a scalar 32-bit integer.

FragSizeEXT: Decorating a variable with the FragSizeEXT built-in decoration will make that variable contain the dimensions in pixels of the area that the fragment covers for that invocation.

The FragSizeEXT decoration must be used only within fragment shaders and the FragmentDensityEXT capability must be declared.

If fragment density map is not enabled, FragSizeEXT will be filled with a value of (1,1).

The variable decorated with FragSizeEXT must be declared using the Input storage class.

The variable decorated with FragSizeEXT must be declared as a two-component vector of 32-bit integers.
FragStencilRefEXT: Decorating a variable with the FragStencilRefEXT built-in decoration will make that variable contain the new 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.
FragmentSizeNV: Decorating a variable with the FragmentSizeNV built-in decoration will make that variable contain the width and height of the fragment.

The FragmentSizeNV decoration must be used only within fragment shaders.

The variable decorated with FragmentSizeNV must be declared using the Input storage class.

The variable decorated with FragmentSizeNV must be declared as a two-component vector of 32-bit integers.
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 area is fully covered by the generating primitive. This variable is non-zero if conservative rasterization is enabled and the current fragment area 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 task, mesh, or 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.

HitKindKHR: A variable decorated with the HitKindKHR decoration will describe the intersection that triggered the execution of the current shader. The values are determined by the intersection shader. For user-defined intersection shaders this is the value that was passed to the “Hit Kind” operand of OpReportIntersectionKHR. For triangle intersection candidates, this will be one of HitKindFrontFacingTriangleKHR or HitKindBackFacingTriangleKHR.

The HitKindKHR decoration must only be used in any-hit and closest hit shaders.

Any variable decorated with HitKindKHR must be declared using the Input storage class.

Any variable decorated with HitKindKHR must be declared as a scalar 32-bit integer.

HitTNV: A variable decorated with the HitTNV decoration is equivalent to a variable decorated with the RayTmaxKHR decoration.

The HitTNV decoration must only be used in any-hit and closest hit shaders.

Any variable decorated with HitTNV must be declared using the Input storage class.

Any variable decorated with HitTNV must be declared as a scalar 32-bit floating-point value.

IncomingRayFlagsKHR: A variable with the IncomingRayFlagsKHR decoration will contain the ray flags passed in to the trace call that invoked this particular shader.

The IncomingRayFlagsKHR decoration must only be used in the intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with IncomingRayFlagsKHR must be declared using the Input storage class.

Any variable decorated with IncomingRayFlagsKHR must be declared as a scalar 32-bit integer.

InstanceCustomIndexKHR: A variable decorated with the InstanceCustomIndexKHR decoration will contain the application-defined value of the instance that intersects the current ray. Only the lower 24 bits are valid, the upper 8 bits will be ignored.

The InstanceCustomIndexKHR decoration must only be used in the intersection, any-hit, and closest hit shaders.

Any variable decorated with InstanceCustomIndexKHR must be declared using the Input storage class.

Any variable decorated with InstanceCustomIndexKHR must be declared as a scalar 32-bit integer.

InstanceId: Decorating a variable in an intersection, any-hit, or closest hit shader with the InstanceId decoration will make that variable contain the index of the instance that intersects the current ray.

The InstanceId decoration must be used only within intersection, any-hit, or closest hit shaders.

The variable decorated with InstanceId must be declared using the Input storage class.

The variable decorated with InstanceId must be declared as a scalar 32-bit integer.
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.
InvocationsPerPixelNV: Decorating a variable with the InvocationsPerPixelNV built-in decoration will make that variable contain the maximum number of fragment shader invocations per pixel, as derived from the effective shading rate for the fragment. If a primitive does not fully cover a pixel, the number of fragment shader invocations for that pixel may be less than the value of InvocationsPerPixelNV. If the shading rate indicates a fragment covering multiple pixels, then InvocationsPerPixelNV will be one.

The InvocationsPerPixelNV decoration must be used only within fragment shaders.

The variable decorated with InvocationsPerPixelNV must be declared using the Input storage class.

The variable decorated with InvocationsPerPixelNV must be declared as a scalar 32-bit integer.
InstanceIndex: Decorating a variable in a vertex shader 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.

LaunchIDKHR: A variable decorated with the LaunchIDKHR decoration will specify the index of the work item being process. One work item is generated for each of the width × height × depth items dispatched by a vkCmdTraceRaysKHR command. All shader invocations inherit the same value for variables decorated with LaunchIDKHR.

The LaunchIDKHR decoration must only be used within the ray generation, intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with LaunchIDKHR must be declared using the Input storage class.

Any variable decorated with LaunchIDKHR must be declared as a three-component vector of 32-bit integer values.

LaunchSizeKHR: A variable decorated with the LaunchSizeKHR decoration will contain the width, height, and depth dimensions passed to the vkCmdTraceRaysKHR command that initiated this shader execution. The width is in the first component, the height is in the second component, and the depth is in the third component.

The LaunchSizeKHR decoration must only be used within ray generation, intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with LaunchSizeKHR must be declared using the Input storage class.

Any variable decorated with LaunchSizeKHR must be declared as a three-component vector of 32-bit integer values.

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 mesh, 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 mesh, vertex, tessellation evaluation, geometry, and fragment shaders. If the shaderOutputLayer feature is not enabled then the Layer decoration must be used only with geometry, and fragment shaders.

In a mesh, 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.

LayerPerViewNV: Decorating a variable with the LayerPerViewNV built-in decoration will make that variable contain the per-view layer information. The per-view layer has the same semantics as Layer, for each view.

The LayerPerViewNV must only be used within mesh shaders.

Any variable decorated with LayerPerViewNV must be declared using the Output storage class, and must also be decorated with the PerViewNV decoration.

Any variable decorated with LayerPerViewNV must be declared as an array of scalar 32-bit integer values.
LocalInvocationId: Decorating a variable with the LocalInvocationId built-in decoration will make that variable contain the location of the current task, mesh, or 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 task, mesh, or 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 task, mesh, or 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.

MeshViewCountNV: Decorating a variable with the MeshViewCountNV built-in decoration will make that variable contain the number of views processed by the current mesh or task shader invocations.

The MeshViewCountNV decoration must only be used in task and mesh shaders.

Any variable decorated with MeshViewCountNV must be declared using the Input storage class.

Any variable decorated with MeshViewCountNV must be declared as a scalar 32-bit integer.

MeshViewIndicesNV: Decorating a variable with the MeshViewIndicesNV built-in decoration will make that variable contain the mesh view indices. The mesh view indices is an array of values where each element holds the view number of one of the views being processed by the current mesh or task shader invocations. The values of array elements with indices great than or equal to MeshViewCountNV are undefined. If the value of MeshViewIndicesNV[i] is j, then any outputs decorated with PerViewNV will take on the value of array element i when processing primitives for view index j.

The MeshViewIndicesNV decoration must only be used in task and mesh shaders.

Any variable decorated with MeshViewIndicesNV must be declared using the Input storage class.

Any variable decorated with MeshViewIndicesNV must be declared as an array of scalar 32-bit integers.
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 task, mesh, or 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.

ObjectRayDirectionKHR: A variable decorated with the ObjectRayDirectionKHR decoration will specify the direction of the ray being processed, in object space.

The ObjectRayDirectionKHR decoration must only be used within intersection, any-hit, and closest hit shaders.

Any variable decorated with ObjectRayDirectionKHR must be declared using the Input storage class.

Any variable decorated with ObjectRayDirectionKHR must be declared as a three-component vector of 32-bit floating-point values.

ObjectRayOriginKHR: A variable decorated with the ObjectRayOriginKHR decoration will specify the origin of the ray being processed, in object space.

The ObjectRayOriginKHR decoration must only be used within intersection, any-hit, and closest hit shaders.

Any variable decorated with ObjectRayOriginKHR must be declared using the Input storage class.

Any variable decorated with ObjectRayOriginKHR must be declared as a three-component vector of 32-bit floating-point values.

ObjectToWorldKHR: A variable decorated with the ObjectToWorldKHR decoration will contain the current object-to-world transformation matrix, which is determined by the instance of the current intersection.

The ObjectToWorldKHR decoration must only be used within intersection, any-hit, and closest hit shaders.

Any variable decorated with ObjectToWorldKHR must be declared using the Input storage class.

Any variable decorated with ObjectToWorldKHR must be declared as a matrix with four columns of three-component vectors of 32-bit floating-point values.
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 mesh, vertex, tessellation control, tessellation evaluation, and geometry shaders.

In a mesh or 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 mesh, vertex, tessellation control, tessellation evaluation, and geometry shaders.

In a mesh or 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 mesh, 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.

PrimitiveCountNV: Decorating a variable with the PrimitiveCountNV decoration will make that variable contain the primitive count. The primitive count specifies the number of primitives in the output mesh produced by the mesh shader that will be processed by subsequent pipeline stages.

The PrimitiveCountNV decoration must only be used in mesh shaders.

Any variable decorated with PrimitiveCountNV must be declared using the Output storage class.

Any variable decorated with PrimitiveCountNV must be declared as a scalar 32-bit integer.
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.

In an intersection, any-hit, or closest hit shader, it will contain the index within the geometry of the triangle or bounding box being processed.

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.

If a mesh shader is present and the fragment shader reads from an input variable decorated with PrimitiveId, then the mesh shader must write to the output variables decorated with PrimitiveId in all execution paths.

The PrimitiveId decoration must be used only within mesh, intersection, any-hit, closest hit, fragment, tessellation control, tessellation evaluation, and geometry shaders.

In an intersection, any-hit, closest hit, 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 mesh shader, any variable decorated with PrimitiveId must be declared using the 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 mesh shader or geometry shader, the resulting value is seen through the PrimitiveId decorated input variable in the fragment shader.

PrimitiveIndicesNV

Decorating a variable with the PrimitiveIndicesNV decoration will make that variable contain the output array of vertex index values. Depending on the output primitive type declared using the execution mode, the indices are split into groups of one (OutputPoints), two (OutputLinesNV), or three (OutputTriangles) indices and each group generates a primitive.

All index values must be in the range [0, N-1], where N is the value specified by the OutputVertices execution mode.

The PrimitiveIndicesNV decoration must only be used in mesh shaders.

Any variable decorated with PrimitiveIndicesNV must be declared using the Output storage class.

Any variable decorated with PrimitiveIndicesNV must be declared as an array of scalar 32-bit integers. The array must be sized according to the primitive type and OutputPrimitivesNV execution modes, where the size is:

the value specified by OutputPrimitivesNV if the execution mode is OutputPoints,
two times the value specified by OutputPrimitivesNV if the execution mode is OutputLinesNV, or
three times the value specified by OutputPrimitivesNV if the execution mode is OutputTrianglesNV.

RayGeometryIndexKHR: A variable decorated with the RayGeometryIndexKHR decoration will contain the generated index for the acceleration structure geometry currently being shaded.

The RayGeometryIndexKHR decoration must only be used within intersection, any-hit, and closest hit shaders.

Any variable decorated with RayGeometryIndexKHR must be declared using the Input storage class.

Any variable decorated with RayGeometryIndexKHR must be declared as a scalar 32-bit integer value.

RayTmaxKHR: A variable decorated with the RayTmaxKHR decoration will contain the parametric t_max values of the ray being processed. The values are independent of the space in which the ray origin and direction exist.

The t_max value changes throughout the lifetime of the ray query that produced the intersection. In the closest hit shader, the value reflects the closest distance to the intersected primitive. In the any-hit shader, it reflects the distance to the primitive currently being intersected. In the intersection shader, it reflects the distance to the closest primitive intersected so far. The value can change in the intersection shader after calling OpReportIntersectionKHR if the corresponding any-hit shader does not ignore the intersection. In a miss shader, the value is identical to the parameter passed into OpTraceRayKHR.

The RayTmaxKHR decoration must only be used with the intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with RayTmaxKHR must be declared with the Input storage class.

Any variable decorated with RayTmaxKHR must be declared as a scalar 32-bit floating-point value.

RayTminKHR: A variable decorated with the RayTminKHR decoration will contain the parametric t_min values of the ray being processed. The values are independent of the space in which the ray origin and direction exist.

The t_min value remains constant for the duration of the ray query.

The RayTminKHR decoration must only be used with the intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with RayTminKHR must be declared with the Input storage class.

Any variable decorated with RayTminKHR must be declared as a scalar 32-bit floating-point value.
SampleId: Decorating a variable with the SampleId built-in decoration will make that variable contain the coverage index for the current fragment shader invocation. 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 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 is still covered after depth testing. 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 the render pass has a fragment density map attachment, the variable will instead contain the sub-fragment position of the sample being shaded. The top left of the fragment is considered to be at coordinate (0,0) and the bottom right of the fragment is considered to be at coordinate (1,1) for any fragment area.

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.

SMCountNV: Decorating a variable with the SMCountNV built-in decoration will make that variable contain the number of SMs on the device.

The variable decorated with SMCountNV must be declared using the Input storage class.

The variable decorated with SMCountNV must be declared as a scalar 32-bit integer value.

SMIDNV: Decorating a variable with the SMIDNV built-in decoration will make that variable contain the ID of the SM on which the current shader invocation is running. This variable is in the range [0, SMCountNV-1].

The variable decorated with SMIDNV must be declared using the Input storage class.

The variable decorated with SMIDNV must be declared as a scalar 32-bit integer value.
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 task, mesh or, 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.

Note

There is no direct relationship between SubgroupLocalInvocationId and LocalInvocationId or LocalInvocationIndex. If the pipeline was created with VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT_EXT, applications can compute their own local invocation index to serve the same purpose:

index = SubgroupLocalInvocationId + SubgroupId * SubgroupSize

If full subgroups are not enabled, some subgroups may be dispatched with inactive invocations that don’t correspond to a local workgroup invocation, making the value of index unreliable.

SubgroupSize: Decorating a variable with the SubgroupSize builtin decoration will make that variable contain the implementation-dependent number of invocations in a subgroup. This value must be a power-of-two integer.

If the pipeline was created with the VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flag set, the SubgroupSize decorated variable will contain the subgroup size for each subgroup that gets dispatched. This value must be between minSubgroupSize and maxSubgroupSize and must be uniform with subgroup scope. The value may vary across a single draw or dispatch call, and for fragment shaders may vary across a single primitive.

If the pipeline was created with a chained VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure, the SubgroupSize decorated variable will match requiredSubgroupSize.

If the pipeline was not created with the VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT flag set and no VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT structure was chained, the variable decorated with SubgroupSize will match subgroupSize.

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.

TaskCountNV: Decorating a variable with the TaskCountNV decoration will make that variable contain the task count. The task count specifies the number of subsequent mesh shader workgroups that get generated upon completion of the task shader.

The TaskCountNV decoration must only be used in task shaders.

Any variable decorated with TaskCountNV must be declared using the Output storage class.

Any variable decorated with TaskCountNV 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 mesh, 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, scissor rectangle, and exclusive 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 mesh, vertex, tessellation evaluation, geometry, and fragment shaders. If the shaderOutputViewportIndex feature is not enabled then the ViewportIndex decoration must be used only with geometry, and fragment shaders.

In a mesh, 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 mesh, 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, scissor rectangle, and exclusive 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 mesh, 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 mesh, 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, scissor rectangle, and exclusive 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.

WarpsPerSMNV: Decorating a variable with the WarpsPerSMNV built-in decoration will make that variable contain the maximum number of warps executing on a SM.

The variable decorated with WarpsPerSMNV must be declared using the Input storage class.

The variable decorated with WarpsPerSMNV must be declared as a scalar 32-bit integer value.

WarpIDNV: Decorating a variable with the WarpIDNV built-in decoration will make that variable contain the ID of the warp on a SM on which the current shader invocation is running. This variable is in the range [0, WarpsPerSMNV-1].

The variable decorated with WarpIDNV must be declared using the Input storage class.

The variable decorated with WarpIDNV must be declared as a scalar 32-bit integer value.
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 task, mesh, or 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 task, mesh, or 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.

WorldRayDirectionKHR: A variable decorated with the WorldRayDirectionKHR decoration will specify the direction of the ray being processed, in world space.

The WorldRayDirectionKHR decoration must only be used within intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with WorldRayDirectionKHR must be declared using the Input storage class.

Any variable decorated with WorldRayDirectionKHR must be declared as a three-component vector of 32-bit floating-point values.

WorldRayOriginKHR: A variable decorated with the WorldRayOriginKHR decoration will specify the origin of the ray being processed, in world space.

The WorldRayOriginKHR decoration must only be used within intersection, any-hit, closest hit, and miss shaders.

Any variable decorated with WorldRayOriginKHR must be declared using the Input storage class.

Any variable decorated with WorldRayOriginKHR must be declared as a three-component vector of 32-bit floating-point values.

WorldToObjectKHR: A variable decorated with the WorldToObjectKHR decoration will contain the current world-to-object transformation matrix, which is determined by the instance of the current intersection.

The WorldToObjectKHR decoration must only be used within intersection, any-hit, and closest hit shaders.

Any variable decorated with WorldToObjectKHR must be declared using the Input storage class.

Any variable decorated with WorldToObjectKHR must be declared as a matrix with four columns of three-component vectors of 32-bit floating-point values.

15. Image Operations

15.1. Image Operations Overview

Vulkan Image Operations are operations performed by those SPIR-V Image 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.

Image Operations include the functionality of the following SPIR-V Image Instructions:

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.
OpImageSampleFootprintNV identifies and returns information about the set of texels in the image that would be accessed by an equivalent OpImageSample* instruction.
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), with the coordinates having the following meanings:

i: Coordinate in the first dimension of an image.
j: Coordinate in the second dimension of an image.
k: Coordinate in the third dimension of an image.
l: Coordinate for array layer.
n: Index of the sample within the texel.

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 conditionally present and is taken from the Sample image operand.

For all coordinate types, unused coordinates are assigned a value of zero.

Figure 3. Texel Coordinate Systems, Linear Filtering

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.
- The t coordinate goes from 0.0 to 1.0.
Unnormalized texel coordinates:
- The u coordinate within the range 0.0 to 8.0 is within the image, otherwise it is outside the image.
- The v coordinate within the range 0.0 to 4.0 is within the image, otherwise it is outside the image.
Integer texel coordinates:
- The i coordinate within the range 0 to 7 addresses texels within the image, otherwise it is outside the image.
- The j coordinate within the range 0 to 3 addresses texels within the image, otherwise it outside the image.
Also shown for linear filtering:
- Given the unnormalized coordinates (u,v), the four texels selected are i₀j₀, i₁j₀, i₀j₁, and i₁j₁.
- The fractions α and β.
- Given the offset Δ_i and Δ_j, the four texels selected by the offset are i₀j'₀, i₁j'₀, i₀j'₁, and i₁j'₁.

Note

For formats with reduced-resolution channels, Δ_i and Δ_j 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.

Figure 4. Texel Coordinate Systems, Nearest Filtering

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 Δ_i and Δ_j, the texel selected by the offset is ij'.

For corner-sampled images, the texel samples are located at the grid intersections instead of the texel centers.

Figure 5. Texel Coordinate Systems, Corner Sampling

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 (red_shared, green_shared, blue_shared, exp_shared) as follows:

First, the components (red, green, blue) are clamped to (red_clamped, green_clamped, blue_clamped) as:

: red_clamped = max(0, min(sharedexp_max, red))
: green_clamped = max(0, min(sharedexp_max, green))
: blue_clamped = max(0, min(sharedexp_max, blue))

where:

N B E_{m a x} s h a r e d e x p_{m a x} = 9 = 15 = 31 = \frac{( 2 ^{N} - 1 )}{2 ^{N}} \times 2^{(E_{m a x} - B)} number of mantissa bits per component exponent bias maximum possible biased exponent value

Note

NaN, if supported, is handled as in IEEE 754-2008 minNum() and maxNum(). This results in any NaN being mapped to zero.

The largest clamped component, max_clamped is determined:

: max_clamped = max(red_clamped, green_clamped, blue_clamped)

A preliminary shared exponent exp' is computed:

e x p^{'} = {⌊ lo g_{2} (m a x_{c l a m p e d}) ⌋ + (B + 1) 0 for m a x_{c l a m p e d} > 2^{- (B + 1)} for m a x_{c l a m p e d} \leq 2^{- (B + 1)}

The shared exponent exp_shared is computed:

m a x_{s h a r e d} = ⌊ \frac{m a x _{c l a m p e d}}{2 ^{(e x p^{'} - B - N)}} + \frac{1}{2} ⌋

e x p_{s h a r e d} = {e x p^{'} e x p^{'} + 1 for 0 \leq m a x_{s h a r e d} < 2^{N} for m a x_{s h a r e d} = 2^{N}

Finally, three integer values in the range 0 to 2^N are computed:

r e d_{s h a r e d} g r e e n_{s h a r e d} b l u e_{s h a r e d} = ⌊ \frac{r e d _{c l a m p e d}}{2 ^{(e x p_{s h a r e d} - B - N)}} + \frac{1}{2} ⌋ = ⌊ \frac{g r e e n _{c l a m p e d}}{2 ^{(e x p_{s h a r e d} - B - N)}} + \frac{1}{2} ⌋ = ⌊ \frac{b l u e _{c l a m p e d}}{2 ^{(e x p_{s h a r e d} - B - N)}} + \frac{1}{2} ⌋

15.2.2. Shared Exponent to RGB

A shared exponent color (red_shared, green_shared, blue_shared, exp_shared) is transformed to an RGB color (red, green, blue) as follows:

: $r e d = r e d_{s h a r e d} \times 2^{(e x p_{s h a r e d} - B - N)}$
: $g r e e n = g r e e n_{s h a r e d} \times 2^{(e x p_{s h a r e d} - B - N)}$
: $b l u e = b l u e_{s h a r e d} \times 2^{(e x p_{s h a r e d} - B - N)}$

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:

Validation operations
Format conversion
Texel replacement
Depth comparison
Conversion to RGBA
Component swizzle
Chroma reconstruction
Y′C_BC_R conversion

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′C_BC_R 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 the identity swizzle.
The VkImageLayout of any subresource in the image view does not match that specified in VkDescriptorImageInfo::imageLayout used to write the image descriptor.
The SPIR-V Image Format is not compatible with the image view’s format.
The sampler unnormalizedCoordinates is VK_TRUE and any of the limitations of unnormalized coordinates are violated.
The sampler was created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT and the image was not created with flags containing VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT.
The sampler was not created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT and the image was created with flags containing VK_IMAGE_CREATE_SUBSAMPLED_BIT_EXT.
The sampler was created with flags containing VK_SAMPLER_CREATE_SUBSAMPLED_BIT_EXT and is used with a function that is not OpImageSampleImplicitLod or OpImageSampleExplicitLod, or is used with operands Offset or ConstOffsets.
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.
- If the Sampled Type of the OpTypeImage does not match the numeric format of the image, as shown in the SPIR-V Sampled Type column of the Interpretation of Numeric Format table.
- If the signedness of any read or sample operation does not match the signedness of the image’s format.
If the image was created with VkImageCreateInfo::flags containing VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV, the sampler addressing modes must only use a VkSamplerAddressMode of VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE.
The SPIR-V instruction is OpImageSampleFootprintNV with Dim = 2D and addressModeU or addressModeV in the sampler is not VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE.
The SPIR-V instruction is OpImageSampleFootprintNV with Dim = 3D and addressModeU, addressModeV, or addressModeW in the sampler is not VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE.
The sampler was created with a specified VkSamplerCustomBorderColorCreateInfoEXT::format which does not match the VkFormat of the image view(s) it is sampling.
The sampler is sampling an image view of VK_FORMAT_B4G4R4A4_UNORM_PACK16, VK_FORMAT_B5G6R5_UNORM_PACK16, or VK_FORMAT_B5G5R5A1_UNORM_PACK16 format without a specified VkSamplerCustomBorderColorCreateInfoEXT::format.

Only OpImageSample* and OpImageSparseSample* can be used with a sampler that enables sampler Y′C_BC_R conversion.

OpImageFetch, OpImageSparseFetch, OpImage*Gather, and OpImageSparse*Gather must not be used with a sampler that enables sampler Y′C_BC_R conversion.

The ConstOffset and Offset operands must not be used with a sampler that enables sampler Y′C_BC_R 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 < w_s
: 0 ≤ j < h_s
: 0 ≤ k < d_s
: 0 ≤ l < layers
: 0 ≤ n < samples

where:

: w_s = width of the image level
: h_s = height of the image level
: d_s = 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:

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.
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.
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.
Cube Map Edge or Corner

Otherwise the texel coordinates lie beyond the edges or corners of the selected cube map face, and Cube map edge handling is performed.

Cube Map Edge Handling

If the texel coordinates lie beyond the edges or corners of the selected cube map face, the following steps are performed. Note that this does not occur when using VK_FILTER_NEAREST 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 beyond the selected cube map face in either only i or only j, then 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 beyond the selected cube map face in both i and j, then 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. The methods are subject to the constraint that for linear filtering if the three available texels have the same value, the resulting filtered texel must have that value, and for cubic filtering if the twelve available samples have the same value, the resulting filtered texel must have 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, the image must not be sampled with sampler Y′C_BC_R conversion enabled.

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 view 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 23. Border Color B, Custom Border Color VkSamplerCustomBorderColorCreateInfoEXT::`customBorderColor` U
Sampler `borderColor`	Corresponding Border Color
`VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK`	[B_r, B_g, B_b, B_a] = [0.0, 0.0, 0.0, 0.0]
`VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK`	[B_r, B_g, B_b, B_a] = [0.0, 0.0, 0.0, 1.0]
`VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE`	[B_r, B_g, B_b, B_a] = [1.0, 1.0, 1.0, 1.0]
`VK_BORDER_COLOR_INT_TRANSPARENT_BLACK`	[B_r, B_g, B_b, B_a] = [0, 0, 0, 0]
`VK_BORDER_COLOR_INT_OPAQUE_BLACK`	[B_r, B_g, B_b, B_a] = [0, 0, 0, 1]
`VK_BORDER_COLOR_INT_OPAQUE_WHITE`	[B_r, B_g, B_b, B_a] = [1, 1, 1, 1]
`VK_BORDER_COLOR_FLOAT_CUSTOM_EXT`	[B_r, B_g, B_b, B_a] = [U_r, U_g, U_b, U_a]
`VK_BORDER_COLOR_INT_CUSTOM_EXT`	[B_r, B_g, B_b, B_a] = [U_r, U_g, U_b, U_a]

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 24. Border Texel Components After Replacement
Texel Aspect or Format	Component Assignment
Depth aspect	D = B_r
Stencil aspect	S = B_r
One component color format	Color_r = B_r
Two component color format	[Color_r,Color_g] = [B_r,B_g]
Three component color format	[Color_r,Color_g,Color_b] = [B_r,B_g,B_b]
Four component color format	[Color_r,Color_g,Color_b,Color_a] = [B_r,B_g,B_b,B_a]

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 access is to an image resource and the x, y, z, or layer coordinate validation fails and robustImageAccess is enabled then zero must be returned for the R, G, and B channels, if present. Either zero or one must be returned for the A channel, if present. If robustImageAccess2 is enabled, zero values must be returned. If only the sample index was invalid, the values returned are undefined.

Additionally, if robustImageAccess is enabled, but robustImageAccess2 is not, any invalid texels may be expanded to four components prior to texel replacement. This means that components not present in the image format may be replaced with 0 or may undergo conversion to RGBA as normal.

Loads from a null descriptor return a four component color value of all zeros. However, for storage images and storage texel buffers using an explicit SPIR-V Image Format, loads from a null descriptor may return an alpha value of 1 (float or integer, depending on format) if the format doesn’t include alpha.

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.

D D = 1.0 = 0.0 ⎩ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎧ D_{ref} \leq D D_{ref} \geq D D_{ref} < D D_{ref} > D D_{ref} = D D_{ref} \neq = D true false for LEQUAL for GEQUAL for LESS for GREATER for EQUAL for NOTEQUAL for ALWAYS for NEVER otherwise

where, in the depth comparison:

: D_ref = shaderOp.D_ref (from optional SPIR-V operand)
: D (texel depth value)

15.3.5. Conversion to RGBA

The texel is expanded from one, two, or three components to four components based on the image base color:

Table 25. Texel Color After Conversion To RGBA
Texel Aspect or Format	RGBA Color
Depth aspect	[Color_r,Color_g,Color_b, Color_a] = [D,0,0,one]
Stencil aspect	[Color_r,Color_g,Color_b, Color_a] = [S,0,0,one]
One component color format	[Color_r,Color_g,Color_b, Color_a] = [Color_r,0,0,one]
Two component color format	[Color_r,Color_g,Color_b, Color_a] = [Color_r,Color_g,0,one]
Three component color format	[Color_r,Color_g,Color_b, Color_a] = [Color_r,Color_g,Color_b,one]
Four component color format	[Color_r,Color_g,Color_b, Color_a] = [Color_r,Color_g,Color_b,Color_a]

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 VkComponentSwizzle enums in the components member of the VkImageViewCreateInfo structure for the image being read if sampler Y′C_BC_R conversion is not enabled, and
the VkComponentSwizzle enums in the components member of the VkSamplerYcbcrConversionCreateInfo structure for the sampler Y′C_BC_R conversion if sampler Y′C_BC_R conversion is enabled.

The swizzle can rearrange the components of the texel, or substitute zero or one for any components. It is defined as follows for each color component:

C o l o r_{c o m p o n e n t}^{'} = ⎩ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎧ C o l o r_{r} C o l o r_{g} C o l o r_{b} C o l o r_{a} 0 o n e i d e n t i t y for RED swizzle for GREEN swizzle for BLUE swizzle for ALPHA swizzle for ZERO swizzle for ONE swizzle for IDENTITY swizzle

where:

o n e i d e n t i t y = {1.0 f 1 for floating point components for integer components = ⎩ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎧ C o l o r_{r} C o l o r_{g} C o l o r_{b} C o l o r_{a} for c o m p o n e n t = r for c o m p o n e n t = g for c o m p o n e n t = b for c o m p o n e n t = a

If the border color is one of the VK_BORDER_COLOR_*_OPAQUE_BLACK enums and the VkComponentSwizzle is not the identity swizzle for all components, 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. This process is also commonly known as “chroma subsampling”. There is one luminance sample in each texture texel, but each chrominance sample may be shared among several texels in one or both texture dimensions.

“_444” formats do not spatially downsample chroma values compared with luma: there are unique chroma samples for each texel.
“_422” formats have downsampling in the x dimension (corresponding to u or s coordinates): they are sampled at half the resolution of luma in that dimension.
“_420” formats have downsampling in the x dimension (corresponding to u or s coordinates) and the y dimension (corresponding to v or t coordinates): they are sampled at half the resolution of luma in both dimensions.

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′C_BC_R 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.

Figure 6. 422 downsampling, xChromaOffset=COSITED_EVEN

Figure 7. 422 downsampling, xChromaOffset=MIDPOINT

Figure 8. 420 downsampling, xChromaOffset=COSITED_EVEN, yChromaOffset=COSITED_EVEN

Figure 9. 420 downsampling, xChromaOffset=MIDPOINT, yChromaOffset=COSITED_EVEN

Figure 10. 420 downsampling, xChromaOffset=COSITED_EVEN, yChromaOffset=MIDPOINT

Figure 11. 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 $τ_{i j k} [l e v e l]$ values accessed by texel filtering are reconstructed as follows:
  
  $τ_{R}^{'} (i, j) τ_{B}^{'} (i, j) = τ_{R} (⌊ i \times 0.5 ⌋, j) [l e v e l] = τ_{B} (⌊ i \times 0.5 ⌋, j) [l e v e l]$
- 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 $τ_{i j k} [l e v e l]$ values accessed by texel filtering are reconstructed as follows:
  
  $τ_{R}^{'} (i, j) τ_{B}^{'} (i, j) = τ_{R} (⌊ i \times 0.5 ⌋, ⌊ j \times 0.5 ⌋) [l e v e l] = τ_{B} (⌊ i \times 0.5 ⌋, ⌊ j \times 0.5 ⌋) [l e v e l]$
  
  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:
    
    $τ_{R B}^{'} (i, j) = ⎩ ⎪ ⎪ ⎨ ⎪ ⎪ ⎧ τ_{R B} (⌊ i \times 0.5 ⌋, j) [l e v e l], 0.5 \times τ_{R B} (⌊ i \times 0.5 ⌋, j) [l e v e l] + 0.5 \times τ_{R B} (⌊ i \times 0.5 ⌋ + 1, j) [l e v e l], 0.5 \times i = ⌊ 0.5 \times i ⌋ 0.5 \times i \neq = ⌊ 0.5 \times i ⌋$
  - If xChromaOffset is VK_CHROMA_LOCATION_MIDPOINT:
    
    $τ_{R B} (i, j)^{'} = ⎩ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎧ 0.25 \times τ_{R B} (⌊ i \times 0.5 ⌋ - 1, j) [l e v e l] + 0.75 \times τ_{R B} (⌊ i \times 0.5 ⌋, j) [l e v e l], 0.75 \times τ_{R B} (⌊ i \times 0.5 ⌋, j) [l e v e l] + 0.25 \times τ_{R B} (⌊ i \times 0.5 ⌋ + 1, j) [l e v e l], 0.5 \times i = ⌊ 0.5 \times i ⌋ 0.5 \times i \neq = ⌊ 0.5 \times i ⌋$
- 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:
  
  $i_{R B} j_{R B} i_{f l o o r} j_{f l o o r} i_{f r a c} j_{f r a c} = {0.5 \times (i) 0.5 \times (i - 0.5) If xChromaOffset = COSITED_EVEN If xChromaOffset = MIDPOINT = {0.5 \times (j) 0.5 \times (j - 0.5) If yChromaOffset = COSITED_EVEN If yChromaOffset = MIDPOINT = ⌊ i_{R B} ⌋ = ⌊ j_{R B} ⌋ = i_{R B} - i_{f l o o r} = j_{R B} - j_{f l o o r}$
  
  $τ_{R B}^{'} (i, j) = τ_{R B} (i_{f l o o r}, j_{f l o o r}) [l e v e l] τ_{R B} (1 + i_{f l o o r}, j_{f l o o r}) [l e v e l] τ_{R B} (i_{f l o o r}, 1 + j_{f l o o r}) [l e v e l] τ_{R B} (1 + i_{f l o o r}, 1 + j_{f l o o r}) [l e v e l] \times \times \times \times (1 - i_{f r a c}) (i_{f r a c}) (1 - i_{f r a c}) (i_{f r a c}) \times \times \times \times (1 - j_{f r a c}) (1 - j_{f r a c}) (j_{f r a c}) (j_{f r a c}) + + +$

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):

u_{R B}^{'} (422 / 420) v_{R B}^{'} (420) = {0.5 \times (u + 0.5), 0.5 \times u, xChromaOffset = COSITED_EVEN xChromaOffset = MIDPOINT = {0.5 \times (v + 0.5), 0.5 \times v, yChromaOffset = COSITED_EVEN yChromaOffset = MIDPOINT

15.3.9. Sampler Y′C_BC_R Conversion

Sampler Y′C_BC_R conversion performs the following operations, which an implementation may combine into a single mathematical operation:

Sampler Y′C_BC_R Range Expansion
Sampler Y′C_BC_R Model Conversion

Sampler Y′C_BC_R Range Expansion

Sampler Y′C_BC_R range expansion is applied to color channel values after all texel input operations which are not specific to sampler Y′C_BC_R conversion. For example, the input values to this stage have been converted using the normal format conversion rules.

Sampler Y′C_BC_R 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′C_BC_R 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′C_BC_R 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:

Y^{'} C_{B} C_{R} = C_{r g b a}^{'} [G] = C_{r g b a}^{'} [B] - \frac{2 ^{(n - 1)}}{( 2 ^{n} ) - 1} = C_{r g b a}^{'} [R] - \frac{2 ^{(n - 1)}}{( 2 ^{n} ) - 1}

Note

These formulae correspond to the “full range” encoding in the “Quantization schemes” chapter of 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:

$Y^{'} C_{B} C_{R} = \frac{C _{r g b a}^{'} [ G ] \times ( 2 ^{n} - 1 ) - 1 6 \times 2 ^{n - 8}}{2 1 9 \times 2 ^{n - 8}} = \frac{C _{r g b a}^{'} [ B ] \times ( 2 ^{n} - 1 ) - 1 2 8 \times 2 ^{n - 8}}{2 2 4 \times 2 ^{n - 8}} = \frac{C _{r g b a}^{'} [ R ] \times ( 2 ^{n} - 1 ) - 1 2 8 \times 2 ^{n - 8}}{2 2 4 \times 2 ^{n - 8}}$

Note

These formulae correspond to the “narrow range” encoding in the “Quantization schemes” chapter of 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 C_B and C_R fall in the range [-0.5,0.5].

Sampler Y′C_BC_R 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′C_BC_R 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′C_BC_R form both in memory and in the shader; Y′C_BC_R range expansion is applied to the channels as for other Y′C_BC_R models, with the vector (C_R,Y′,C_B,A) provided to the shader.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_709: The color channels are transformed from a Y′C_BC_R representation to an R′G′B′ representation as described in the “BT.709 Y′C_BC_R conversion” section of the Khronos Data Format Specification.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_601: The color channels are transformed from a Y′C_BC_R representation to an R′G′B′ representation as described in the “BT.601 Y′C_BC_R conversion” section of the Khronos Data Format Specification.
VK_SAMPLER_YCBCR_MODEL_CONVERSION_YCBCR_2020: The color channels are transformed from a Y′C_BC_R representation to an R′G′B′ representation as described in the “BT.2020 Y′C_BC_R 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:

Calculate the unnormalized texel coordinates corresponding to the desired sample position.
For a minFilter or magFilter of VK_FILTER_NEAREST:
1. Calculate (i,j) for the sample location as described under the “nearest filtering” formulae in (u,v,w,a) to (i,j,k,l,n) Transformation And Array Layer Selection
2. Calculate the normalized texel coordinates corresponding to these integer coordinates.
3. Sample using sampler Y′C_BC_R conversion at this location.
For a minFilter or magFilter of VK_FILTER_LINEAR:
1. Calculate (i_[0,1],j_[0,1]) for the sample location as described under the “linear filtering” formulae in (u,v,w,a) to (i,j,k,l,n) Transformation And Array Layer Selection
2. Calculate the normalized texel coordinates corresponding to these integer coordinates.
3. Sample using sampler Y′C_BC_R conversion at each of these locations.
4. Convert the non-linear A′R′G′B′ outputs of the Y′C_BC_R conversions to linear ARGB values as described in the “Transfer Functions” chapter of the Khronos Data Format Specification.
5. Interpolate the linear ARGB values using the α and β values described in the “linear filtering” section of (u,v,w,a) to (i,j,k,l,n) Transformation And Array Layer Selection and the equations in Texel Filtering.

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 or magFilter are both VK_FILTER_NEAREST, these operations are redundant and sampling using sampler Y′C_BC_R 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:

Validation operations
Texel output format conversion

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 write causes the contents of the image’s memory to become undefined.

Texel Type Validation

If the Sampled Type of the OpTypeImage does not match the type defined for the format, as specified in the SPIR-V Sampled Type column of the Interpretation of Numeric Format table, the write causes the value of the texel to become undefined. For integer types, if the Signedness of the Sampled Type of the OpTypeImage does not match the signedness of the accessed resource, the write causes the value of the texel to become undefined.

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). Floating-point outputs are converted as described in Floating-Point Format Conversions and Fixed-Point Data Conversion. Integer outputs are converted such that their value is preserved. The converted value of any integer that cannot be represented in the target format is undefined.

15.5. Normalized Texel Coordinate Operations

If the image sampler instruction provides normalized texel coordinates, some of the following operations are performed.

15.5.1. Projection Operation

For Proj image operations, the normalized texel coordinates (s,t,r,q,a) and (if present) the D_ref coordinate are transformed as follows:

s t r D_{ref} = \frac{s}{q}, = \frac{t}{q}, = \frac{r}{q}, = \frac{D _{ref}}{q}, for 1D, 2D, or 3D image for 2D or 3D image for 3D image if provided

15.5.2. 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. That is:

\partial s / \partial x \partial t / \partial x \partial u / \partial x = d P d x (s), = d P d x (t), = d P d x (u), \partial s / \partial y \partial t / \partial y \partial u / \partial y = d P d y (s), = d P d y (t), = d P d y (u), for 1D, 2D, Cube, or 3D image for 2D, Cube, or 3D image for Cube or 3D image

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 λ_base as described in Level-of-Detail Operation.

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.

15.5.3. Cube Map Face Selection and Transformations

For cube map image instructions, the (s,t,r) coordinates are treated as a direction vector (r_x,r_y,r_z). The direction vector is used to select a cube map face. The direction vector is transformed to a per-face texel coordinate system (s_face,t_face), The direction vector is also used to transform the derivatives to per-face derivatives.

15.5.4. 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 r_z wins over r_y and r_x, and the second rule that r_y wins over r_x. An implementation may choose other rules, but the rules must be deterministic and depend only on (r_x,r_y,r_z).

The layer number (corresponding to a cube map face), the coordinate selections for s_c, t_c, r_c, and the selection of derivatives, are determined by the major axis direction as specified in the following two tables.

Table 26. Cube map face and coordinate selection
Major Axis Direction	Layer Number	Cube Map Face	s_c	t_c	r_c
+r_x	0	Positive X	-r_z	-r_y	r_x
-r_x	1	Negative X	+r_z	-r_y	r_x
+r_y	2	Positive Y	+r_x	+r_z	r_y
-r_y	3	Negative Y	+r_x	-r_z	r_y
+r_z	4	Positive Z	+r_x	-r_y	r_z
-r_z	5	Negative Z	-r_x	-r_y	r_z

Table 27. Cube map derivative selection
Major Axis Direction	∂s_c / ∂x	∂s_c / ∂y	∂t_c / ∂x	∂t_c / ∂y	∂r_c / ∂x	∂r_c / ∂y
+r_x	-∂r_z / ∂x	-∂r_z / ∂y	-∂r_y / ∂x	-∂r_y / ∂y	+∂r_x / ∂x	+∂r_x / ∂y
-r_x	+∂r_z / ∂x	+∂r_z / ∂y	-∂r_y / ∂x	-∂r_y / ∂y	-∂r_x / ∂x	-∂r_x / ∂y
+r_y	+∂r_x / ∂x	+∂r_x / ∂y	+∂r_z / ∂x	+∂r_z / ∂y	+∂r_y / ∂x	+∂r_y / ∂y
-r_y	+∂r_x / ∂x	+∂r_x / ∂y	-∂r_z / ∂x	-∂r_z / ∂y	-∂r_y / ∂x	-∂r_y / ∂y
+r_z	+∂r_x / ∂x	+∂r_x / ∂y	-∂r_y / ∂x	-∂r_y / ∂y	+∂r_z / ∂x	+∂r_z / ∂y
-r_z	-∂r_x / ∂x	-∂r_x / ∂y	-∂r_y / ∂x	-∂r_y / ∂y	-∂r_z / ∂x	-∂r_z / ∂y

15.5.5. Cube Map Coordinate Transformation

s_{face} t_{face} = \frac{1}{2} \times \frac{s _{c}}{∣ r _{c} ∣} + \frac{1}{2} = \frac{1}{2} \times \frac{t _{c}}{∣ r _{c} ∣} + \frac{1}{2}

15.5.6. Cube Map Derivative Transformation

\frac{\partial s _{face}}{\partial x} \frac{\partial s _{face}}{\partial x} \frac{\partial s _{face}}{\partial x} = \frac{\partial}{\partial x} (\frac{1}{2} \times \frac{s _{c}}{∣ r _{c} ∣} + \frac{1}{2}) = \frac{1}{2} \times \frac{\partial}{\partial x} (\frac{s _{c}}{∣ r _{c} ∣}) = \frac{1}{2} \times (\frac{∣ r _{c} ∣ \times \partial s _{c} / \partial x - s _{c} \times \partial r _{c} / \partial x}{( r _{c} ) ^{2}})

\frac{\partial s _{face}}{\partial y} \frac{\partial t _{face}}{\partial x} \frac{\partial t _{face}}{\partial y} = \frac{1}{2} \times (\frac{∣ r _{c} ∣ \times \partial s _{c} / \partial y - s _{c} \times \partial r _{c} / \partial y}{( r _{c} ) ^{2}}) = \frac{1}{2} \times (\frac{∣ r _{c} ∣ \times \partial t _{c} / \partial x - t _{c} \times \partial r _{c} / \partial x}{( r _{c} ) ^{2}}) = \frac{1}{2} \times (\frac{∣ r _{c} ∣ \times \partial t _{c} / \partial y - t _{c} \times \partial r _{c} / \partial y}{( r _{c} ) ^{2}})

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 s_face and t_face are only valid for non-dependent texture fetches (pre OpenGL 2.0).

15.5.7. 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 d_l values, defined below. These values must be computed with mipmapPrecisionBits of accuracy and may be subject to implementation-specific maxima and minima for very large, out-of-range values.

Scale Factor Operation

The magnitude of the derivatives are calculated by:

: m_ux = |∂s/∂x| × w_base
: m_vx = |∂t/∂x| × h_base
: m_wx = |∂r/∂x| × d_base
: m_uy = |∂s/∂y| × w_base
: m_vy = |∂t/∂y| × h_base
: m_wy = |∂r/∂y| × d_base

where:

: ∂t/∂x = ∂t/∂y = 0 (for 1D images)
: ∂r/∂x = ∂r/∂y = 0 (for 1D, 2D or Cube images)

and:

: w_base = image.w
: h_base = image.h
: d_base = image.d

(for the baseMipLevel, from the image descriptor).

For corner-sampled images, the w_base, h_base, and d_base are instead:

: w_base = image.w - 1
: h_base = image.h - 1
: d_base = image.d - 1

A point sampled in screen space has an elliptical footprint in texture space. The minimum and maximum scale factors (ρ_min, ρ_max) should be the minor and major axes of this ellipse.

The scale factors ρ_x and ρ_y, calculated from the magnitude of the derivatives in x and y, are used to compute the minimum and maximum scale factors.

ρ_x and ρ_y may be approximated with functions f_x and f_y, subject to the following constraints:

f_{x} is continuous and monotonically increasing in each of m_{u x}, m_{v x}, and m_{w x} f_{y} is continuous and monotonically increasing in each of m_{u y}, m_{v y}, and m_{w y}

max (∣ m_{u x} ∣, ∣ m_{v x} ∣, ∣ m_{w x} ∣) \leq f_{x} \leq 2 (∣ m_{u x} ∣ + ∣ m_{v x} ∣ + ∣ m_{w x} ∣) max (∣ m_{u y} ∣, ∣ m_{v y} ∣, ∣ m_{w y} ∣) \leq f_{y} \leq 2 (∣ m_{u y} ∣ + ∣ m_{v y} ∣ + ∣ m_{w y} ∣)

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 w_base, h_base, d_base for rectangle textures either, since they do not support implicit derivatives.

The minimum and maximum scale factors (ρ_min,ρ_max) are determined by:

: ρ_max = max(ρ_x, ρ_y)
: ρ_min = min(ρ_x, ρ_y)

The ratio of anisotropy is determined by:

: η = min(ρ_max/ρ_min, max_Aniso)

where:

: sampler.max_Aniso = maxAnisotropy (from sampler descriptor)
: limits.max_Aniso = maxSamplerAnisotropy (from physical device limits)
: max_Aniso = min(sampler.max_Aniso, limits.max_Aniso)

If ρ_max = ρ_min = 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 ρ_max ≠ 0 and ρ_min = 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 max_Aniso. However, anytime the footprint is small in texel space the implementation may use a smaller value of η, even when ρ_min is zero or close to zero. If either VkPhysicalDeviceFeatures::samplerAnisotropy or VkSamplerCreateInfo::anisotropyEnable are VK_FALSE, max_Aniso 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:

λ_{b a s e} (x, y) λ^{'} (x, y) λ = {s h a d e r O p . L o d lo g_{2} (\frac{ρ _{m a x}}{η}) (from optional SPIR-V operand) otherwise = λ_{b a s e} + c l a m p (s a m p l e r . b i a s + s h a d e r O p . b i a s, - m a x S a m p l e r L o d B i a s, m a x S a m p l e r L o d B i a s) = ⎩ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎧ l o d_{m a x}, λ^{'}, l o d_{m i n}, undefined, λ^{'} > l o d_{m a x} l o d_{m i n} \leq λ^{'} \leq l o d_{m a x} λ^{'} < l o d_{m i n} l o d_{m i n} > l o d_{m a x}

where:

s a m p l e r . b i a s s h a d e r O p . b i a s s a m p l e r . l o d_{m i n} s h a d e r O p . l o d_{m i n} l o d_{m i n} l o d_{m a x} = m i p L o d B i a s = {B i a s 0 (from optional SPIR-V operand) otherwise = m i n L o d = {M i n L o d 0 (from optional SPIR-V operand) otherwise = max (s a m p l e r . l o d_{m i n}, s h a d e r O p . l o d_{m i n}) = m a x L o d (from sampler descriptor) (from sampler descriptor) (from sampler descriptor)

and maxSamplerLodBias is the value of the VkPhysicalDeviceLimits feature maxSamplerLodBias.

Image Level(s) Selection

The image level(s) d, d_hi, and d_lo which texels are read from are determined by an image-level parameter d_l, which is computed based on the LOD parameter, as follows:

d_{l} = {n e a r e s t (d^{'}), d^{'}, mipmapMode is VK_SAMPLER_MIPMAP_MODE_NEAREST otherwise

where:

d^{'} = l e v e l_{b a s e} + clamp (λ, 0, q)

n e a r e s t (d^{'}) = {⌈ d^{'} + 0.5 ⌉ - 1, ⌊ d^{'} + 0.5 ⌋, preferred alternative

and:

: level_base = 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 = d_l.

If the sampler’s mipmapMode is VK_SAMPLER_MIPMAP_MODE_LINEAR, two neighboring levels are selected:

d_{h i} d_{l o} δ = ⌊ d_{l} ⌋ = m i n (d_{h i} + 1, q) = d_{l} - d_{h i}

δ is the fractional value, quantized to the number of mipmap precision bits, used for linear filtering between levels.

15.5.8. (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 used in filtering (either d, or d_hi and d_lo).

u (x, y) v (x, y) w (x, y) a (x, y) = s (x, y) \times w i d t h_{s c a l e} + Δ_{i} = {0 t (x, y) \times h e i g h t_{s c a l e} + Δ_{j} for 1D images otherwise = {0 r (x, y) \times d e p t h_{s c a l e} + Δ_{k} for 2D or Cube images otherwise = {a (x, y) 0 for array images otherwise

where:

: width_scale = width_level
: height_scale = height_level
: depth_scale = depth_level

for conventional images, and:

: width_scale = width_level - 1
: height_scale = height_level - 1
: depth_scale = depth_level - 1

for corner-sampled images.

and where (Δ_i, Δ_j, Δ_k) are taken from the image instruction if it includes a ConstOffset or Offset operand, otherwise they are taken to be zero.

Operations then proceed to Unnormalized Texel Coordinate Operations.

15.6. Unnormalized Texel Coordinate Operations

15.6.1. (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:

R N E (a) = {r o u n d T i e s T o E v e n (a) ⌊ a + 0.5 ⌋ preferred, from IEEE Std 754-2008 Floating-Point Arithmetic alternative

The sample index n is assigned the value 0.

Nearest filtering (VK_FILTER_NEAREST) computes the integer texel coordinates that the unnormalized coordinates lie within:

i j k = ⌊ u + s h i f t ⌋ = ⌊ v + s h i f t ⌋ = ⌊ w + s h i f t ⌋

where:

: shift = 0.0

for conventional images, and:

: shift = 0.5

for corner-sampled images.

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.

i_{0} i_{1} j_{0} j_{1} k_{0} k_{1} = ⌊ u - s h i f t ⌋ = i_{0} + 1 = ⌊ v - s h i f t ⌋ = j_{0} + 1 = ⌊ w - s h i f t ⌋ = k_{0} + 1

α β γ = f r a c (u - s h i f t) = f r a c (v - s h i f t) = f r a c (w - s h i f t)

where:

: shift = 0.5

for conventional images, and:

: shift = 0.0

for corner-sampled images, and where:

f r a c (x) = x - ⌊ x ⌋

where the number of fraction bits retained is specified by VkPhysicalDeviceLimits::subTexelPrecisionBits.

Cubic filtering (VK_FILTER_CUBIC_EXT) 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, k₀, k₁, k₂ or k₃, as well as weights α, β, and γ.

i_{0} j_{0} k_{0} = ⌊ u - \frac{3}{2} ⌋ = ⌊ v - \frac{3}{2} ⌋ = ⌊ w - \frac{3}{2} ⌋ i_{1} j_{1} k_{1} = i_{0} + 1 = j_{0} + 1 = k_{0} + 1 i_{2} j_{2} k_{2} = i_{1} + 1 = j_{1} + 1 = k_{1} + 1 i_{3} j_{3} k_{3} = i_{2} + 1 = j_{2} + 1 = k_{2} + 1

α β γ = f r a c (u - \frac{1}{2}) = f r a c (v - \frac{1}{2}) = f r a c (w - \frac{1}{2})

where:

f r a c (x) = x - ⌊ x ⌋

where the number of fraction bits retained is specified by VkPhysicalDeviceLimits::subTexelPrecisionBits.

15.7. 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:

d = l e v e l_{b a s e} + {L o d 0 (from optional SPIR-V operand) otherwise

If d does not lie in the range [baseMipLevel, baseMipLevel + levelCount) then any values fetched are zero if robustImageAccess2 is enabled, otherwise are undefined, and any writes (if supported) are discarded.

15.8. Image Sample Operations

15.8.1. 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.

i = ⎩ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎧ i m o d s i z e (s i z e - 1) - m i r r o r ((i m o d (2 \times s i z e)) - s i z e) c l a m p (i, 0, s i z e - 1) c l a m p (i, - 1, s i z e) c l a m p (m i r r o r (i), 0, s i z e - 1) for repeat for mirrored repeat for clamp to edge for clamp to border for mirror clamp to edge

where:

m i r r o r (n) = {n - (1 + n) for n \geq 0 otherwise

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.

15.8.2. Texel Gathering

SPIR-V instructions with Gather in the name return a vector derived from 4 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. If the operation does not use the ConstOffsets image operand then the four texels form the 2 × 2 rectangle used for texture filtering:

τ [R] τ [G] τ [B] τ [A] = τ_{i 0 j 1} [l e v e l_{b a s e}] [c o m p] = τ_{i 1 j 1} [l e v e l_{b a s e}] [c o m p] = τ_{i 1 j 0} [l e v e l_{b a s e}] [c o m p] = τ_{i 0 j 0} [l e v e l_{b a s e}] [c o m p]

If the operation does use the ConstOffsets image operand then the offsets allow a custom filter to be defined:

τ [R] τ [G] τ [B] τ [A] = τ_{i 0 j 0 + Δ_{0}} [l e v e l_{b a s e}] [c o m p] = τ_{i 0 j 0 + Δ_{1}} [l e v e l_{b a s e}] [c o m p] = τ_{i 0 j 0 + Δ_{2}} [l e v e l_{b a s e}] [c o m p] = τ_{i 0 j 0 + Δ_{3}} [l e v e l_{b a s e}] [c o m p]

where:

τ [l e v e l_{b a s e}] [c o m p] c o m p = ⎩ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎧ τ [l e v e l_{b a s e}] [R], τ [l e v e l_{b a s e}] [G], τ [l e v e l_{b a s e}] [B], τ [l e v e l_{b a s e}] [A], for c o m p = 0 for c o m p = 1 for c o m p = 2 for c o m p = 3 from SPIR-V operand Component

OpImage*Gather must not be used on a sampled image with sampler Y′C_BC_R conversion enabled.

15.8.3. Texel Filtering

Texel filtering is first performed for each level (either d or d_hi and d_lo).

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.

Texel Nearest Filtering

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.

τ [l e v e l] = ⎩ ⎪ ⎪ ⎨ ⎪ ⎪ ⎧ τ_{i j k} [l e v e l], τ_{i j} [l e v e l], τ_{i} [l e v e l], for 3D image for 2D or Cube image for 1D image

Texel Linear Filtering

Within a mip level, VK_FILTER_LINEAR filtering combines 8 (for 3D), 4 (for 2D or Cube), or 2 (for 1D) texel values, together with their linear weights. The linear weights are derived from the fractions computed earlier:

w_{i_{0}} w_{i_{1}} w_{j_{0}} w_{j_{1}} w_{k_{0}} w_{k_{1}} = (1 - α) = (α) = (1 - β) = (β) = (1 - γ) = (γ)

The values of multiple texels, together with their weights, are combined to produce a filtered value.

The VkSamplerReductionModeCreateInfo::reductionMode can control the process by which multiple texels, together with their weights, are combined to produce a filtered texture value.

When the reductionMode is set (explicitly or implicitly) to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, a weighted average is computed:

τ_{3 D} τ_{2 D} τ_{1 D} = k = k_{0} \sum k_{1} j = j_{0} \sum j_{1} i = i_{0} \sum i_{1} (w_{i}) (w_{j}) (w_{k}) τ_{i j k} = j = j_{0} \sum j_{1} i = i_{0} \sum i_{1} (w_{i}) (w_{j}) τ_{i j} = i = i_{0} \sum i_{1} (w_{i}) τ_{i}

However, if the reduction mode is VK_SAMPLER_REDUCTION_MODE_MIN or VK_SAMPLER_REDUCTION_MODE_MAX, the process operates on the above set of multiple texels, together with their weights, computing a component-wise minimum or maximum, respectively, of the components of the set of texels with non-zero weights.

Texel Cubic Filtering

Within a mip level, VK_FILTER_CUBIC_EXT, filtering computes a weighted average of 64 (for 3D), 16 (for 2D), or 4 (for 1D) texel values, together with their Catmull-Rom weights.

Catmull-Rom weights are derived from the fractions computed earlier.

[w_{i_{0}}, w_{i_{1}}, w_{i_{2}}, w_{i_{3}}] = \frac{1}{2} [1 α α^{2} α^{3}] ⎣ ⎢ ⎢ ⎢ ⎡ - 0 - 1 - 2 - 1 - 2 - 0 - 5 - 3 - 0 - 1 - 4 - 3 - 0 - 0 - 1 - 1 ⎦ ⎥ ⎥ ⎥ ⎤ [w_{j_{0}}, w_{j_{1}}, w_{j_{2}}, w_{j_{3}}] = \frac{1}{2} [1 β β^{2} β^{3}] ⎣ ⎢ ⎢ ⎢ ⎡ - 0 - 1 - 2 - 1 - 2 - 0 - 5 - 3 - 0 - 1 - 4 - 3 - 0 - 0 - 1 - 1 ⎦ ⎥ ⎥ ⎥ ⎤ [w_{k_{0}}, w_{k_{1}}, w_{k_{2}}, w_{k_{3}}] = \frac{1}{2} [1 γ γ^{2} γ^{3}] ⎣ ⎢ ⎢ ⎢ ⎡ - 0 - 1 - 2 - 1 - 2 - 0 - 5 - 3 - 0 - 1 - 4 - 3 - 0 - 0 - 1 - 1 ⎦ ⎥ ⎥ ⎥ ⎤

The values of multiple texels, together with their weights, are combined to produce a filtered value.

The VkSamplerReductionModeCreateInfo::reductionMode can control the process by which multiple texels, together with their weights, are combined to produce a filtered texture value.

When the reductionMode is set (explicitly or implicitly) to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, a weighted average is computed:

τ_{3 D} τ_{2 D} τ_{1 D} = k = j_{0} \sum k_{3} j = j_{0} \sum j_{3} i = i_{0} \sum i_{3} (w_{i}) (w_{j}) (w_{k}) τ_{i j k} = j = j_{0} \sum j_{3} i = i_{0} \sum i_{3} (w_{i}) (w_{j}) τ_{i j} = i = i_{0} \sum i_{3} (w_{i}) τ_{i}

Texel Mipmap Filtering

VK_SAMPLER_MIPMAP_MODE_NEAREST filtering returns the value of a single mipmap level,

τ = τ[d].

VK_SAMPLER_MIPMAP_MODE_LINEAR filtering combines the values of multiple mipmap levels (τ[hi] and τ[lo]), together with their linear weights.

The linear weights are derived from the fraction computed earlier:

w_{h i} w_{l o} = (1 - δ) = (δ)

The values of multiple mipmap levels, together with their weights, are combined to produce a final filtered value.

The VkSamplerReductionModeCreateInfo::reductionMode can control the process by which multiple texels, together with their weights, are combined to produce a filtered texture value.

When the reductionMode is set (explicitly or implicitly) to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, a weighted average is computed:

τ = (w_{h i}) τ [h i] + (w_{l o}) τ [l o]

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 τ_2Daniso is defined using the single isotropic τ_2D(u,v) at level d.

τ_{2 D a n i s o} τ_{2 D a n i s o} = \frac{1}{N} i = 1 \sum N τ_{2 D} (u (x - \frac{1}{2} + \frac{i}{N + 1}, y), (v (x - \frac{1}{2} + \frac{i}{N + 1}), y)), = \frac{1}{N} i = 1 \sum N τ_{2 D} (u (x, y - \frac{1}{2} + \frac{i}{N + 1}), (v (x, y - \frac{1}{2} + \frac{i}{N + 1}))), when ρ_{x} > ρ_{y} when ρ_{y} \geq ρ_{x}

When VkSamplerReductionModeCreateInfo::reductionMode is set to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, the above summation is used. However, if the reduction mode is VK_SAMPLER_REDUCTION_MODE_MIN or VK_SAMPLER_REDUCTION_MODE_MAX, the process operates on the above values, together with their weights, computing a component-wise minimum or maximum, respectively, of the components of the values with non-zero weights.

15.9. Texel Footprint Evaluation

The SPIR-V instruction OpImageSampleFootprintNV evaluates the set of texels from a single mip level that would be accessed during a texel filtering operation. In addition to the inputs that would be accepted by an equivalent OpImageSample* instruction, OpImageSampleFootprintNV accepts two additional inputs. The Granularity input is an integer identifying the size of texel groups used to evaluate the footprint. Each bit in the returned footprint mask corresponds to an aligned block of texels whose size is given by the following table:

Table 28. Texel footprint granularity values
`Granularity`	`Dim` = 2D	`Dim` = 3D
0	unsupported	unsupported
1	2x2	2x2x2
2	4x2	unsupported
3	4x4	4x4x2
4	8x4	unsupported
5	8x8	unsupported
6	16x8	unsupported
7	16x16	unsupported
8	unsupported	unsupported
9	unsupported	unsupported
10	unsupported	16x16x16
11	64x64	32x16x16
12	128x64	32x32x16
13	128x128	32x32x32
14	256x128	64x32x32
15	256x256	unsupported

The Coarse input is used to select between the two mip levels that may be accessed during texel filtering when using a mipmapMode of VK_SAMPLER_MIPMAP_MODE_LINEAR. When filtering between two mip levels, a Coarse value of true requests the footprint in the lower-resolution mip level (higher level number), while false requests the footprint in the higher-resolution mip level. If texel filtering would access only a single mip level, the footprint in that level would be returned when Coarse is set to false; an empty footprint would be returned when Coarse is set to true.

The footprint for OpImageSampleFootprintNV is returned in a structure with six members:

The first member is a boolean value that is true if the texel filtering operation would access only a single mip level.
The second member is a two- or three-component integer vector holding the footprint anchor location. For two-dimensional images, the returned components are in units of eight texel groups. For three-dimensional images, the returned components are in units of four texel groups.
The third member is a two- or three-component integer vector holding a footprint offset relative to the anchor. All returned components are in units of texel groups.
The fourth member is a two-component integer vector mask, which holds a bitfield identifying the set of texel groups in an 8x8 or 4x4x4 neighborhood relative to the anchor and offset.
The fifth member is an integer identifying the mip level containing the footprint identified by the anchor, offset, and mask.
The sixth member is an integer identifying the granularity of the returned footprint.

For footprints in two-dimensional images (Dim2D), the mask returned by OpImageSampleFootprintNV indicates whether each texel group in a 8x8 local neighborhood of texel groups would have one or more texels accessed during texel filtering. In the mask, the texel group with local group coordinates $(l g x, l g y)$ is considered covered if and only if

0 \neq = ((m a s k . x + (m a s k . y < < 32)) & (1 < < (l g y \times 8 + l g x)))

where:

$0 < = l g x < 8$ and $0 < = l g y < 8$ ; and
$m a s k$ is the returned two-component mask.

The local group with coordinates $(l g x, l g y)$ in the mask is considered covered if and only if the texel filtering operation would access one or more texels $τ_{i j}$ in the returned miplevel where:

i 0 i 1 j 0 j 1 = {g r a n . x \times (8 \times a n c h o r . x + l g x), g r a n . x \times (8 \times (a n c h o r . x - 1) + l g x), if l g x + o f f s e t . x < 8 otherwise = i 0 + g r a n . x - 1 = {g r a n . y \times (8 \times a n c h o r . y + l g y), g r a n . y \times (8 \times (a n c h o r . y - 1) + l g y), if l g y + o f f s e t . y < 8 o t h e r w i s e = j 0 + g r a n . y - 1

and

$i 0 < = i < = i 1$ and $j 0 < = j < = j 1$ ;
$g r a n$ is a two-component vector holding the width and height of the texel group identified by the granularity;
$a n c h o r$ is the returned two-component anchor vector; and
$o f f s e t$ is the returned two-component offset vector.

For footprints in three-dimensional images (Dim3D), the mask returned by OpImageSampleFootprintNV indicates whether each texel group in a 4x4x4 local neighborhood of texel groups would have one or more texels accessed during texel filtering. In the mask, the texel group with local group coordinates $(l g x, l g y, l g z)$ , is considered covered if and only if:

0 \neq = ((m a s k . x + (m a s k . y < < 32)) & (1 < < (l g z \times 16 + l g y \times 4 + l g x)))

where:

$0 < = l g x < 4$ , $0 < = l g y < 4$ , and $0 < = l g z < 4$ ; and
$m a s k$ is the returned two-component mask.

The local group with coordinates $(l g x, l g y, l g z)$ in the mask is considered covered if and only if the texel filtering operation would access one or more texels $τ_{i j k}$ in the returned miplevel where:

i 0 i 1 j 0 j 1 k 0 k 1 = {g r a n . x \times (4 \times a n c h o r . x + l g x), g r a n . x \times (4 \times (a n c h o r . x - 1) + l g x), if l g x + o f f s e t . x < 4 otherwise = i 0 + g r a n . x - 1 = {g r a n . y \times (4 \times a n c h o r . y + l g y), g r a n . y \times (4 \times (a n c h o r . y - 1) + l g y), if l g y + o f f s e t . y < 4 o t h e r w i s e = j 0 + g r a n . y - 1 = {g r a n . z \times (4 \times a n c h o r . z + l g z), g r a n . z \times (4 \times (a n c h o r . z - 1) + l g z), if l g z + o f f s e t . z < 4 o t h e r w i s e = k 0 + g r a n . z - 1

and

$i 0 < = i < = i 1$ , $j 0 < = j < = j 1$ , $k 0 < = k < = k 1$ ;
$g r a n$ is a three-component vector holding the width, height, and depth of the texel group identified by the granularity;
$a n c h o r$ is the returned three-component anchor vector; and
$o f f s e t$ is the returned three-component offset vector.

If the sampler used by OpImageSampleFootprintNV enables anisotropic texel filtering via anisotropyEnable, it is possible that the set of texel groups accessed in a mip level may be too large to be expressed using an 8x8 or 4x4x4 mask using the granularity requested in the instruction. In this case, the implementation uses a texel group larger than the requested granularity. When a larger texel group size is used, OpImageSampleFootprintNV returns an integer granularity value that can be interpreted in the same manner as the granularity value provided to the instruction to determine the texel group size used. If anisotropic texel filtering is disabled in the sampler, or if an anisotropic footprint can be represented as an 8x8 or 4x4x4 mask with the requested granularity, OpImageSampleFootprintNV will use the requested granularity as-is and return a granularity value of zero.

OpImageSampleFootprintNV supports only two- and three-dimensional image accesses (Dim2D and Dim3D), and the footprint returned is undefined if a sampler uses an addressing mode other than VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE.

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 Footprint Evaluation: Performed by OpImageSampleFootprint instructions.
Texel Filtering: Performed by all OpImageSample* and OpImageSparseSample* instructions.
Sparse Residency: Performed by all OpImageSparse* instructions.

16. Fragment Density Map Operations

16.1. Fragment Density Map Operations Overview

When a fragment is generated in a render pass that has a fragment density map attachment, its area is determined by the properties of the local framebuffer region that the fragment occupies. The framebuffer is divided into a uniform grid of these local regions, and their fragment area property is derived from the density map with the following operations:

Fetch density value
- Component swizzle
- Component mapping
Fragment area conversion
- Fragment area filter
- Fragment area clamp

16.2. Fetch Density Value

Each local framebuffer region at center coordinate (x,y) fetches a texel from the fragment density map at integer coordinates:

: $i = ⌊ \frac{x}{f r a g m e n t D e n s i t y T e x e l S i z e _{w i d t h}} ⌋$
: $j = ⌊ \frac{y}{f r a g m e n t D e n s i t y T e x e l S i z e _{h e i g h t}} ⌋$

Where the size of each region in the framebuffer is:

: $f r a g m e n t D e n s i t y T e x e l S i z e_{w i d t h}^{'} = 2^{⌈ l o g_{2} (\frac{f r a m e b u f f e r _{w i d t h}}{f r a g m e n t D e n s i t y M a p _{w i d t h}}) ⌉}$
: $f r a g m e n t D e n s i t y T e x e l S i z e_{h e i g h t}^{'} = 2^{⌈ l o g_{2} (\frac{f r a m e b u f f e r _{h e i g h t}}{f r a g m e n t D e n s i t y M a p _{h e i g h t}}) ⌉}$

This region is subject to the limits in VkPhysicalDeviceFragmentDensityMapPropertiesEXT and therefore the final region size is clamped:

: $f r a g m e n t D e n s i t y T e x e l S i z e_{w i d t h} = c l a m p (f r a g m e n t D e n s i t y T e x e l S i z e_{w i d t h}^{'}, m i n F r a g m e n t D e n s i t y T e x e l S i z e_{w i d t h}, m a x F r a g m e n t D e n s i t y T e x e l S i z e_{w i d t h})$
: $f r a g m e n t D e n s i t y T e x e l S i z e_{h e i g h t} = c l a m p (f r a g m e n t D e n s i t y T e x e l S i z e_{h e i g h t}^{'}, m i n F r a g m e n t D e n s i t y T e x e l S i z e_{h e i g h t}, m a x F r a g m e n t D e n s i t y T e x e l S i z e_{h e i g h t})$

When multiview is enabled for the render pass and the fragment density map attachment view was created with layerCount greater than 1, the density map layer that the texel is fetched from is:

: $l a y e r = b a s e A r r a y L a y e r + V i e w I n d e x$

Otherwise:

: $l a y e r = b a s e A r r a y L a y e r$

The texel fetched from the density map at (i,j,layer) is next converted to density with the following operations.

16.2.1. Component Swizzle

The components member of VkImageViewCreateInfo is applied to the fetched texel as defined in Image component swizzle.

16.2.2. Component Mapping

The swizzled texel’s components are mapped to a density value:

: $d e n s i t y V a l u e_{x y} = (C_{r}^{'}, C_{g}^{'})$

16.3. Fragment Area Conversion

Fragment area for the framebuffer region is undefined if the density fetched is not a normalized floating-point value greater than 0.0. Otherwise, the fetched fragment area for that region is derived as:

: $f r a g m e n t A r e a_{w h} = \frac{1 . 0}{d e n s i t y V a l u e _{x y}}$

16.3.1. Fragment Area Filter

Optionally, the implementation may fetch additional density map texels in an implementation defined window around (i,j). The texels follow the standard conversion steps up to and including fragment area conversion.

A single fetched fragment area for the framebuffer region is chosen by the implementation and must have an area between the min and max areas of the fetched set.

16.3.2. Fragment Area Clamp

The implementation may clamp the fetched fragment area to one that it supports. The clamped fragment area must have a size less than or equal to the original fetched value. Implementations may vary the supported set of fragment areas per framebuffer region. Fragment area (1,1) must always be in the supported set.

Note

For example, if the fetched fragment area is (1,4) but the implementation only supports areas of {(1,1),(2,2)}, it could choose to clamp the area to (2,2) since it has the same size as (1,4). While this would produce fragments that have lower quality strictly in the x-axis, the overall density is maintained.

The clamped fragment area is assigned to the corresponding framebuffer region.

17. 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. Performance Queries are also supported if the associated extension is available. Intel performance queries are also supported if the associated extension is available.

17.1. 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:

// Provided by VK_VERSION_1_0
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkQueryPool)

To create a query pool, call:

// Provided by VK_VERSION_1_0
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 a 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

VK_SUCCESS

VK_ERROR_OUT_OF_HOST_MEMORY
VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkQueryPoolCreateInfo structure is defined as:

// Provided by VK_VERSION_1_0
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 a structure extending this 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

If the pipeline statistics queries feature is not enabled, queryType must not be VK_QUERY_TYPE_PIPELINE_STATISTICS
If queryType is VK_QUERY_TYPE_PIPELINE_STATISTICS, pipelineStatistics must be a valid combination of VkQueryPipelineStatisticFlagBits values
If queryType is VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the pNext chain must include a structure of type VkQueryPoolPerformanceCreateInfoKHR
queryCount must be greater than 0

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_QUERY_POOL_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 VkQueryPoolPerformanceCreateInfoKHR or VkQueryPoolPerformanceQueryCreateInfoINTEL
The sType value of each struct in the pNext chain must be unique
flags must be 0
queryType must be a valid VkQueryType value

// Provided by VK_VERSION_1_0
typedef VkFlags VkQueryPoolCreateFlags;

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

The VkQueryPoolPerformanceCreateInfoKHR structure is defined as:

// Provided by VK_KHR_performance_query
typedef struct VkQueryPoolPerformanceCreateInfoKHR {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           queueFamilyIndex;
    uint32_t           counterIndexCount;
    const uint32_t*    pCounterIndices;
} VkQueryPoolPerformanceCreateInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
queueFamilyIndex is the queue family index to create this performance query pool for.
counterIndexCount is size of the pCounterIndices array.
pCounterIndices is the array of indices into the vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR::pCounters to enable in this performance query pool.

Valid Usage

queueFamilyIndex must be a valid queue family index of the device
The performanceCounterQueryPools feature must be enabled
Each element of pCounterIndices must be in the range of counters reported by vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR for the queue family specified in queueFamilyIndex

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_QUERY_POOL_PERFORMANCE_CREATE_INFO_KHR
pCounterIndices must be a valid pointer to an array of counterIndexCount uint32_t values
counterIndexCount must be greater than 0

To query the number of passes required to query a performance query pool on a physical device, call:

// Provided by VK_KHR_performance_query
void vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR(
    VkPhysicalDevice                            physicalDevice,
    const VkQueryPoolPerformanceCreateInfoKHR*  pPerformanceQueryCreateInfo,
    uint32_t*                                   pNumPasses);

physicalDevice is the handle to the physical device whose queue family performance query counter properties will be queried.
pPerformanceQueryCreateInfo is a pointer to a VkQueryPoolPerformanceCreateInfoKHR of the performance query that is to be created.
pNumPasses is a pointer to an integer related to the number of passes required to query the performance query pool, as described below.

The pPerformanceQueryCreateInfo member VkQueryPoolPerformanceCreateInfoKHR::queueFamilyIndex must be a queue family of physicalDevice. The number of passes required to capture the counters specified in the pPerformanceQueryCreateInfo member VkQueryPoolPerformanceCreateInfoKHR::pCounters is returned in pNumPasses.

Valid Usage (Implicit)

physicalDevice must be a valid VkPhysicalDevice handle
pPerformanceQueryCreateInfo must be a valid pointer to a valid VkQueryPoolPerformanceCreateInfoKHR structure
pNumPasses must be a valid pointer to a uint32_t value

To destroy a query pool, call:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
typedef enum VkQueryType {
    VK_QUERY_TYPE_OCCLUSION = 0,
    VK_QUERY_TYPE_PIPELINE_STATISTICS = 1,
    VK_QUERY_TYPE_TIMESTAMP = 2,
  // Provided by VK_EXT_transform_feedback
    VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT = 1000028004,
  // Provided by VK_KHR_performance_query
    VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR = 1000116000,
  // Provided by VK_KHR_ray_tracing
    VK_QUERY_TYPE_ACCELERATION_STRUCTURE_COMPACTED_SIZE_KHR = 1000165000,
  // Provided by VK_KHR_ray_tracing
    VK_QUERY_TYPE_ACCELERATION_STRUCTURE_SERIALIZATION_SIZE_KHR = 1000150000,
  // Provided by VK_INTEL_performance_query
    VK_QUERY_TYPE_PERFORMANCE_QUERY_INTEL = 1000210000,
  // Provided by VK_NV_ray_tracing
    VK_QUERY_TYPE_ACCELERATION_STRUCTURE_COMPACTED_SIZE_NV = VK_QUERY_TYPE_ACCELERATION_STRUCTURE_COMPACTED_SIZE_KHR,
} VkQueryType;

VK_QUERY_TYPE_OCCLUSION specifies an occlusion query.
VK_QUERY_TYPE_PIPELINE_STATISTICS specifies a pipeline statistics query.
VK_QUERY_TYPE_TIMESTAMP specifies a timestamp query.
VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR specifies a performance query.
VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT specifies a transform feedback query.
VK_QUERY_TYPE_ACCELERATION_STRUCTURE_COMPACTED_SIZE_KHR specifies a ray tracing acceleration structure size query.
VK_QUERY_TYPE_ACCELERATION_STRUCTURE_SERIALIZATION_SIZE_KHR specifies a ray tracing serialization acceleration structure size query
VK_QUERY_TYPE_PERFORMANCE_QUERY_INTEL specifies a Intel performance query.

17.2. Query Operation

The operation of queries is controlled by the commands vkCmdBeginQuery, vkCmdEndQuery, vkCmdBeginQueryIndexedEXT, vkCmdEndQueryIndexedEXT, 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 or vkResetQueryPool 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 must be reset before it is used. Queries must also be reset between uses.

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 on a queue, call:

// Provided by VK_VERSION_1_0
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.

If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, this command sets the status of query indices [firstQuery, firstQuery + queryCount - 1] to unavailable for each pass of queryPool, as indicated by a call to vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR.

Note

Because vkCmdResetQueryPool resets all the passes of the indicated queries, applications must not record a vkCmdResetQueryPool command for a queryPool created with VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR in a command buffer that needs to be submitted multiple times as indicated by a call to vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR. Otherwise applications will never be able to complete the recorded queries.

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
All queries used by the command must not be active
If queryPool was created with VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, this command must not be recorded in a command buffer that, either directly or through secondary command buffers, also contains begin commands for a query from the set of queries [firstQuery, firstQuery + queryCount - 1]

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

Primary
Secondary

Outside

Graphics
Compute

To reset a range of queries in a query pool on the host, call:

// Provided by VK_VERSION_1_2
void vkResetQueryPool(
    VkDevice                                    device,
    VkQueryPool                                 queryPool,
    uint32_t                                    firstQuery,
    uint32_t                                    queryCount);

or the equivalent command

// Provided by VK_EXT_host_query_reset
void vkResetQueryPoolEXT(
    VkDevice                                    device,
    VkQueryPool                                 queryPool,
    uint32_t                                    firstQuery,
    uint32_t                                    queryCount);

device is the logical device that owns the query pool.
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.

This command sets the status of query indices [firstQuery, firstQuery + queryCount - 1] to unavailable.

If queryPool is VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR this command sets the status of query indices [firstQuery, firstQuery + queryCount - 1] to unavailable for each pass.

Valid Usage

The hostQueryReset feature must be enabled
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
Submitted commands that refer to the range specified by firstQuery and queryCount in queryPool must have completed execution
The range of queries specified by firstQuery and queryCount in queryPool must not be in use by calls to vkGetQueryPoolResults or vkResetQueryPool in other threads

Valid Usage (Implicit)

device must be a valid VkDevice handle
queryPool must be a valid VkQueryPool handle
queryPool must have been created, allocated, or retrieved from device

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 dispatch commands. Timestamp queries write timestamps to a query pool. Performance queries record performance counters 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:

// Provided by VK_VERSION_1_0
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.

Calling vkCmdBeginQuery is equivalent to calling vkCmdBeginQueryIndexedEXT with the index parameter set to zero.

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
The queryType used to create queryPool must not be VK_QUERY_TYPE_TIMESTAMP
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 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
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the VkCommandPool that commandBuffer was allocated from must support graphics operations
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT then VkPhysicalDeviceTransformFeedbackPropertiesEXT::transformFeedbackQueries must be supported

If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the profiling lock must have been held before vkBeginCommandBuffer was called on commandBuffer
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR, the query begin must be the first recorded command in commandBuffer
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR, the begin command must not be recorded within a render pass instance
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and another query pool with a queryType VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR has been used within commandBuffer, its parent primary command buffer or secondary command buffer recorded within the same parent primary command buffer as commandBuffer, the performanceCounterMultipleQueryPools feature must be enabled
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, this command must not be recorded in a command buffer that, either directly or through secondary command buffers, also contains a vkCmdResetQueryPool command affecting the same query

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

Primary
Secondary

Both

Graphics
Compute

To begin an indexed query, call:

// Provided by VK_EXT_transform_feedback
void vkCmdBeginQueryIndexedEXT(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    query,
    VkQueryControlFlags                         flags,
    uint32_t                                    index);

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.
index is the query type specific index. When the query type is VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the index represents the vertex stream.

The vkCmdBeginQueryIndexedEXT command operates the same as the vkCmdBeginQuery command, except that it also accepts a query type specific index parameter.

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
The queryType used to create queryPool must not be VK_QUERY_TYPE_TIMESTAMP
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 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
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the VkCommandPool that commandBuffer was allocated from must support graphics operations
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the index parameter must be less than VkPhysicalDeviceTransformFeedbackPropertiesEXT::maxTransformFeedbackStreams
If the queryType used to create queryPool was not VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the index must be zero
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT then VkPhysicalDeviceTransformFeedbackPropertiesEXT::transformFeedbackQueries must be supported

If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the profiling lock must have been held before vkBeginCommandBuffer was called on commandBuffer
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR, the query begin must be the first recorded command in commandBuffer
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR, the begin command must not be recorded within a render pass instance
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and another query pool with a queryType VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR has been used within commandBuffer, its parent primary command buffer or secondary command buffer recorded within the same parent primary command buffer as commandBuffer, the performanceCounterMultipleQueryPools feature must be enabled
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, this command must not be recorded in a command buffer that, either directly or through secondary command buffers, also contains a vkCmdResetQueryPool command affecting the same query

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

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:

// Provided by VK_VERSION_1_0
typedef enum VkQueryControlFlagBits {
    VK_QUERY_CONTROL_PRECISE_BIT = 0x00000001,
} VkQueryControlFlagBits;

VK_QUERY_CONTROL_PRECISE_BIT specifies the precision of occlusion queries.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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.

Calling vkCmdEndQuery is equivalent to calling vkCmdEndQueryIndexedEXT with the index parameter set to zero.

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
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one or more of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_COMMAND_BUFFER_KHR, the vkCmdEndQuery must be the last recorded command in commandBuffer
If queryPool was created with a queryType of VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR and one or more of the counters used to create queryPool was VK_PERFORMANCE_COUNTER_SCOPE_RENDER_PASS_KHR, the vkCmdEndQuery must not be recorded within a render pass instance

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

Primary
Secondary

Both

Graphics
Compute

To end an indexed query after the set of desired draw or dispatch commands is recorded, call:

// Provided by VK_EXT_transform_feedback
void vkCmdEndQueryIndexedEXT(
    VkCommandBuffer                             commandBuffer,
    VkQueryPool                                 queryPool,
    uint32_t                                    query,
    uint32_t                                    index);

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.
index is the query type specific index.

The vkCmdEndQueryIndexedEXT command operates the same as the vkCmdEndQuery command, except that it also accepts a query type specific index parameter.

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 vkCmdEndQueryIndexedEXT 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
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the index parameter must be less than VkPhysicalDeviceTransformFeedbackPropertiesEXT::maxTransformFeedbackStreams
If the queryType used to create queryPool was not VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT the index must be zero
If the queryType used to create queryPool was VK_QUERY_TYPE_TRANSFORM_FEEDBACK_STREAM_EXT index must equal the index used to begin the query

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

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.
Occlusion queries, pipeline statistics queries, transform feedback queries, and timestamp queries store results in 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.
Performance queries store results in a tightly packed array whose type is determined by the unit member of the corresponding VkPerformanceCounterKHR.
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. Timestamp queries write one integer value. Performance queries write one VkPerformanceCounterResultKHR value for each VkPerformanceCounterKHR in the query. Transform feedback queries write two integers; the first integer is the number of primitives successfully written to the corresponding transform feedback buffer and the second is the number of primitives output to the vertex stream, regardless of whether they were successfully captured or not. In other words, if the transform feedback buffer was sized too small for the number of primitives output by the vertex stream, the first integer represents the number of primitives actually written and the second is the number that would have been written if all the transform feedback buffers associated with that vertex stream were large enough.
If more than one query is retrieved and stride is not at least as large as the size of the array of values corresponding to a single query, the values written to memory are undefined.

To retrieve status and results for a set of queries, call:

// Provided by VK_VERSION_1_0
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 to read.
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.

The range of queries read is defined by [firstQuery, firstQuery + queryCount - 1]. 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.

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 vkResetQueryPool is called or 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 vkResetQueryPool is called or 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.

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 not set in flags and the queryType used to create queryPool was not VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, 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
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, then pData and stride must be multiples of the size of VkPerformanceCounterResultKHR
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
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, flags must not contain VK_QUERY_RESULT_WITH_AVAILABILITY_BIT, VK_QUERY_RESULT_PARTIAL_BIT or VK_QUERY_RESULT_64_BIT
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the queryPool must have been recorded once for each pass as retrieved via a call to vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR

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

VK_SUCCESS
VK_NOT_READY

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:

// Provided by VK_VERSION_1_0
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.

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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.

Results for all requested occlusion queries, pipeline statistics queries, transform feedback queries, and timestamp queries are written as 64-bit unsigned integer values if VK_QUERY_RESULT_64_BIT is set or 32-bit unsigned integer values otherwise. Performance queries store results in a tightly packed array whose type is determined by the unit member of the corresponding VkPerformanceCounterKHR.

If neither of VK_QUERY_RESULT_WAIT_BIT and VK_QUERY_RESULT_WITH_AVAILABILITY_BIT are set, results are only written out for queries in the available state.

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.

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
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, VkPhysicalDevicePerformanceQueryPropertiesKHR::allowCommandBufferQueryCopies must be VK_TRUE
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, flags must not contain VK_QUERY_RESULT_WITH_AVAILABILITY_BIT, VK_QUERY_RESULT_PARTIAL_BIT or VK_QUERY_RESULT_64_BIT
If the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_KHR, the queryPool must have been submitted once for each pass as retrieved via a call to vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR
vkCmdCopyQueryPoolResults must not be called if the queryType used to create queryPool was VK_QUERY_TYPE_PERFORMANCE_QUERY_INTEL

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

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.

17.3. 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.

17.4. 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. 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, the value of 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:

// Provided by VK_VERSION_1_0
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).

// Provided by VK_VERSION_1_0
typedef VkFlags VkQueryPipelineStatisticFlags;

VkQueryPipelineStatisticFlags is a bitmask type for setting a mask of zero or more VkQueryPipelineStatisticFlagBits.

17.5. 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:

// Provided by VK_VERSION_1_0
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.

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

pipelineStage must be a valid stage for the queue family that was used to create the command pool that commandBuffer was allocated from
If the geometry shaders feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT
If the tessellation shaders feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT or VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT
If the conditional rendering feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_CONDITIONAL_RENDERING_BIT_EXT
If the fragment density map feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_FRAGMENT_DENSITY_PROCESS_BIT_EXT
If the transform feedback feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT
If the mesh shaders feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_MESH_SHADER_BIT_NV or VK_PIPELINE_STAGE_TASK_SHADER_BIT_NV
If the shading rate image feature is not enabled, pipelineStage must not be VK_PIPELINE_STAGE_SHADING_RATE_IMAGE_BIT_NV
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

Primary
Secondary

Both

Transfer
Graphics
Compute

Transfer

17.6. Performance Queries

Performance queries provide applications with a mechanism for getting performance counter information about the execution of command buffers, render passes, and commands.

Each queue family advertises the performance counters that can be queried on a queue of that family via a call to vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR. Implementations may limit access to performance counters based on platform requirements or only to specialized drivers for development purposes.

Note

This may include no performance counters being enumerated, or a reduced set. Please refer to platform-specific documentation for guidance on any such restrictions.

Performance queries use the existing vkCmdBeginQuery and vkCmdEndQuery to control what command buffers, render passes, or commands to get performance information for.

Implementations may require multiple passes where the command buffer, render passes, or commands being recorded are the same and are executed on the same queue to record performance counter data. This is achieved by submitting the same batch and providing a VkPerformanceQuerySubmitInfoKHR structure containing a counter pass index. The number of passes required for a given performance query pool can be queried via a call to vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR.

Note

Command buffers created with VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT must not be re-submitted. Changing command buffer usage bits may affect performance. To avoid this, the application should re-record any command buffers with the VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT when multiple counter passes are required.

Performance counter results from a performance query pool can be obtained with the command vkGetQueryPoolResults.

Performance query results are returned in an array of VkPerformanceCounterResultKHR unions containing the data associated with each counter in the query, stored in the same order as the counters supplied in pCounterIndices when creating the performance query. The VkPerformanceCounterKHR::unit enumeration specifies how to parse the counter data.

// Provided by VK_KHR_performance_query
typedef union VkPerformanceCounterResultKHR {
    int32_t     int32;
    int64_t     int64;
    uint32_t    uint32;
    uint64_t    uint64;
    float       float32;
    double      float64;
} VkPerformanceCounterResultKHR;

17.6.1. Profiling Lock

To record and submit a command buffer that contains a performance query pool the profiling lock must be held. The profiling lock must be acquired prior to any call to vkBeginCommandBuffer that will be using a performance query pool. The profiling lock must be held while any command buffer that contains a performance query pool is in the recording, executable, or pending state. To acquire the profiling lock, call:

// Provided by VK_KHR_performance_query
VkResult vkAcquireProfilingLockKHR(
    VkDevice                                    device,
    const VkAcquireProfilingLockInfoKHR*        pInfo);

device is the logical device to profile.
pInfo is a pointer to a VkAcquireProfilingLockInfoKHR structure which contains information about how the profiling is to be acquired.

Implementations may allow multiple actors to hold the profiling lock concurrently.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInfo must be a valid pointer to a valid VkAcquireProfilingLockInfoKHR structure

Return Codes

VK_SUCCESS

VK_TIMEOUT

The VkAcquireProfilingLockInfoKHR structure is defined as:

// Provided by VK_KHR_performance_query
typedef struct VkAcquireProfilingLockInfoKHR {
    VkStructureType                   sType;
    const void*                       pNext;
    VkAcquireProfilingLockFlagsKHR    flags;
    uint64_t                          timeout;
} VkAcquireProfilingLockInfoKHR;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
flags is reserved for future use.
timeout indicates how long the function waits, in nanoseconds, if the profiling lock is not available.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_ACQUIRE_PROFILING_LOCK_INFO_KHR
pNext must be NULL
flags must be 0

If timeout is 0, vkAcquireProfilingLockKHR will not block while attempting to acquire the profling lock. If timeout is UINT64_MAX, the function will not return until the profiling lock was acquired.

// Provided by VK_KHR_performance_query
typedef enum VkAcquireProfilingLockFlagBitsKHR {
} VkAcquireProfilingLockFlagBitsKHR;

// Provided by VK_KHR_performance_query
typedef VkFlags VkAcquireProfilingLockFlagsKHR;

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

To release the profiling lock, call:

// Provided by VK_KHR_performance_query
void vkReleaseProfilingLockKHR(
    VkDevice                                    device);

device is the logical device to cease profiling on.

Valid Usage

The profiling lock of device must have been held via a previous successful call to vkAcquireProfilingLockKHR

Valid Usage (Implicit)

device must be a valid VkDevice handle

17.7. Transform Feedback Queries

Transform feedback queries track the number of primitives attempted to be written and actually written, by the vertex stream being captured, to a transform feedback buffer. This query is updated during draw commands while transform feedback is active. The number of primitives actually written will be less than the number attempted to be written if the bound transform feedback buffer size was too small for the number of primitives actually drawn. Primitives are not written beyond the bound range of the transform feedback buffer. A transform feedback query is begun and ended by calling vkCmdBeginQuery and vkCmdEndQuery, respectively to query for vertex stream zero. vkCmdBeginQueryIndexedEXT and vkCmdEndQueryIndexedEXT can be used to begin and end transform feedback queries for any supported vertex stream. When a transform feedback query begins, the count of primitives written and primitives needed starts from zero. For each drawing command, the count is incremented as vertex attribute outputs are captured to the transform feedback buffers while transform feedback is active.

When a transform feedback 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).

17.8. Intel performance queries

Intel performance queries allow an application to capture performance data for a set of commands. Performance queries are used in a similar way than other types of queries. A main difference with existing queries is that the resulting data should be handed over to a library capabable to produce human readable results rather than being read directly by an application.

Prior to creating a performance query pool, initialize the device for performance queries with the call:

// Provided by VK_INTEL_performance_query
VkResult vkInitializePerformanceApiINTEL(
    VkDevice                                    device,
    const VkInitializePerformanceApiInfoINTEL*  pInitializeInfo);

device is the logical device used for the queries.
pInitializeInfo is a pointer to a VkInitializePerformanceApiInfoINTEL structure specifying initialization parameters.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pInitializeInfo must be a valid pointer to a valid VkInitializePerformanceApiInfoINTEL structure

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkInitializePerformanceApiInfoINTEL structure is defined as :

// Provided by VK_INTEL_performance_query
typedef struct VkInitializePerformanceApiInfoINTEL {
    VkStructureType    sType;
    const void*        pNext;
    void*              pUserData;
} VkInitializePerformanceApiInfoINTEL;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
pUserData is a pointer for application data.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_INITIALIZE_PERFORMANCE_API_INFO_INTEL
pNext must be NULL

Once performance query operations have completed, uninitalize the device for performance queries with the call:

// Provided by VK_INTEL_performance_query
void vkUninitializePerformanceApiINTEL(
    VkDevice                                    device);

device is the logical device used for the queries.

Valid Usage (Implicit)

device must be a valid VkDevice handle

Some performance query features of a device can be discovered with the call:

// Provided by VK_INTEL_performance_query
VkResult vkGetPerformanceParameterINTEL(
    VkDevice                                    device,
    VkPerformanceParameterTypeINTEL             parameter,
    VkPerformanceValueINTEL*                    pValue);

device is the logical device to query.
parameter is the parameter to query.
pValue is a pointer to a VkPerformanceValueINTEL structure in which the type and value of the parameter are returned.

Valid Usage (Implicit)

device must be a valid VkDevice handle
parameter must be a valid VkPerformanceParameterTypeINTEL value
pValue must be a valid pointer to a VkPerformanceValueINTEL structure

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

Possible values of vkGetPerformanceParameterINTEL::parameter, specifying a performance query feature, are:

// Provided by VK_INTEL_performance_query
typedef enum VkPerformanceParameterTypeINTEL {
    VK_PERFORMANCE_PARAMETER_TYPE_HW_COUNTERS_SUPPORTED_INTEL = 0,
    VK_PERFORMANCE_PARAMETER_TYPE_STREAM_MARKER_VALID_BITS_INTEL = 1,
} VkPerformanceParameterTypeINTEL;

VK_PERFORMANCE_PARAMETER_TYPE_HW_COUNTERS_SUPPORTED_INTEL has a boolean result which tells whether hardware counters can be captured.
VK_PERFORMANCE_PARAMETER_TYPE_STREAM_MARKER_VALID_BITS_INTEL has a 32 bits integer result which tells how many bits can be written into the VkPerformanceValueINTEL value.

The VkPerformanceValueINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkPerformanceValueINTEL {
    VkPerformanceValueTypeINTEL    type;
    VkPerformanceValueDataINTEL    data;
} VkPerformanceValueINTEL;

type is a VkPerformanceValueTypeINTEL value specifying the type of the returned data.
data is a VkPerformanceValueDataINTEL union specifying the value of the returned data.

Valid Usage (Implicit)

type must be a valid VkPerformanceValueTypeINTEL value
If type is VK_PERFORMANCE_VALUE_TYPE_STRING_INTEL, the valueString member of data must be a null-terminated UTF-8 string

Possible values of VkPerformanceValueINTEL::type, specifying the type of the data returned in VkPerformanceValueINTEL::data, are:

VK_PERFORMANCE_VALUE_TYPE_UINT32_INTEL specifies that unsigned 32-bit integer data is returned in data.value32.
VK_PERFORMANCE_VALUE_TYPE_UINT64_INTEL specifies that unsigned 64-bit integer data is returned in data.value64.
VK_PERFORMANCE_VALUE_TYPE_FLOAT_INTEL specifies that floating-point data is returned in data.valueFloat.
VK_PERFORMANCE_VALUE_TYPE_BOOL_INTEL specifies that Bool32 data is returned in data.valueBool.
VK_PERFORMANCE_VALUE_TYPE_STRING_INTEL specifies that a pointer to a null-terminated UTF-8 string is returned in data.valueString. The pointer is valid for the lifetime of the device parameter passed to vkGetPerformanceParameterINTEL.

// Provided by VK_INTEL_performance_query
typedef enum VkPerformanceValueTypeINTEL {
    VK_PERFORMANCE_VALUE_TYPE_UINT32_INTEL = 0,
    VK_PERFORMANCE_VALUE_TYPE_UINT64_INTEL = 1,
    VK_PERFORMANCE_VALUE_TYPE_FLOAT_INTEL = 2,
    VK_PERFORMANCE_VALUE_TYPE_BOOL_INTEL = 3,
    VK_PERFORMANCE_VALUE_TYPE_STRING_INTEL = 4,
} VkPerformanceValueTypeINTEL;

The VkPerformanceValueDataINTEL union is defined as:

// Provided by VK_INTEL_performance_query
typedef union VkPerformanceValueDataINTEL {
    uint32_t       value32;
    uint64_t       value64;
    float          valueFloat;
    VkBool32       valueBool;
    const char*    valueString;
} VkPerformanceValueDataINTEL;

data.value32 represents 32-bit integer data.
data.value64 represents 64-bit integer data.
data.valueFloat represents floating-point data.
data.valueBool represents Bool32 data.
data.valueString represents a pointer to a null-terminated UTF-8 string.

The correct member of the union is determined by the associated VkPerformanceValueTypeINTEL value.

The VkQueryPoolPerformanceQueryCreateInfoINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkQueryPoolPerformanceQueryCreateInfoINTEL {
    VkStructureType                 sType;
    const void*                     pNext;
    VkQueryPoolSamplingModeINTEL    performanceCountersSampling;
} VkQueryPoolPerformanceQueryCreateInfoINTEL;

To create a pool for Intel performance queries, set VkQueryPoolCreateInfo::queryType to VK_QUERY_TYPE_PERFORMANCE_QUERY_INTEL and add a VkQueryPoolPerformanceQueryCreateInfoINTEL structure to the pNext chain of the VkQueryPoolCreateInfo structure.

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
performanceCountersSampling describe how performance queries should be captured.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_QUERY_POOL_PERFORMANCE_QUERY_CREATE_INFO_INTEL
performanceCountersSampling must be a valid VkQueryPoolSamplingModeINTEL value

Possible values of VkQueryPoolPerformanceQueryCreateInfoINTEL::performanceCountersSampling are:

// Provided by VK_INTEL_performance_query
typedef enum VkQueryPoolSamplingModeINTEL {
    VK_QUERY_POOL_SAMPLING_MODE_MANUAL_INTEL = 0,
} VkQueryPoolSamplingModeINTEL;

VK_QUERY_POOL_SAMPLING_MODE_MANUAL_INTEL is the default mode in which the application calls vkCmdBeginQuery and vkCmdEndQuery to record performance data.

To help associate query results with a particular point at which an application emitted commands, markers can be set into the command buffers with the call:

// Provided by VK_INTEL_performance_query
VkResult vkCmdSetPerformanceMarkerINTEL(
    VkCommandBuffer                             commandBuffer,
    const VkPerformanceMarkerInfoINTEL*         pMarkerInfo);

The last marker set onto a command buffer before the end of a query will be part of the query result.

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pMarkerInfo must be a valid pointer to a valid VkPerformanceMarkerInfoINTEL structure
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

Primary
Secondary

Both

Graphics
Compute
Transfer

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkPerformanceMarkerInfoINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkPerformanceMarkerInfoINTEL {
    VkStructureType    sType;
    const void*        pNext;
    uint64_t           marker;
} VkPerformanceMarkerInfoINTEL;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
marker is the marker value that will be recorded into the opaque query results.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_MARKER_INFO_INTEL
pNext must be NULL

When monitoring the behavior of an application wihtin the dataset generated by the entire set of applications running on the system, it is useful to identify draw calls within a potentially huge amount of performance data. To do so, application can generate stream markers that will be used to trace back a particular draw call with a particular performance data item.

// Provided by VK_INTEL_performance_query
VkResult vkCmdSetPerformanceStreamMarkerINTEL(
    VkCommandBuffer                             commandBuffer,
    const VkPerformanceStreamMarkerInfoINTEL*   pMarkerInfo);

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pMarkerInfo must be a valid pointer to a valid VkPerformanceStreamMarkerInfoINTEL structure
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

Primary
Secondary

Both

Graphics
Compute
Transfer

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkPerformanceStreamMarkerInfoINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkPerformanceStreamMarkerInfoINTEL {
    VkStructureType    sType;
    const void*        pNext;
    uint32_t           marker;
} VkPerformanceStreamMarkerInfoINTEL;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
marker is the marker value that will be recorded into the reports consumed by an external application.

Valid Usage

The value written by the application into marker must only used the valid bits as reported by vkGetPerformanceParameterINTEL with the VK_PERFORMANCE_PARAMETER_TYPE_STREAM_MARKER_VALID_BITS_INTEL

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_STREAM_MARKER_INFO_INTEL
pNext must be NULL

Some applications might want measure the effect of a set of commands with a different settings. It is possible to override a particular settings using :

// Provided by VK_INTEL_performance_query
VkResult vkCmdSetPerformanceOverrideINTEL(
    VkCommandBuffer                             commandBuffer,
    const VkPerformanceOverrideInfoINTEL*       pOverrideInfo);

commandBuffer is the command buffer where the override takes place.
pOverrideInfo is a pointer to a VkPerformanceOverrideInfoINTEL structure selecting the parameter to override.

Valid Usage

pOverrideInfo must not be used with a VkPerformanceOverrideTypeINTEL that is not reported available by vkGetPerformanceParameterINTEL

Valid Usage (Implicit)

commandBuffer must be a valid VkCommandBuffer handle
pOverrideInfo must be a valid pointer to a valid VkPerformanceOverrideInfoINTEL structure
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

Primary
Secondary

Both

Graphics
Compute
Transfer

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkPerformanceOverrideInfoINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkPerformanceOverrideInfoINTEL {
    VkStructureType                   sType;
    const void*                       pNext;
    VkPerformanceOverrideTypeINTEL    type;
    VkBool32                          enable;
    uint64_t                          parameter;
} VkPerformanceOverrideInfoINTEL;

type is the particular VkPerformanceOverrideTypeINTEL to set.
enable defines whether the override is enabled.
parameter is a potential required parameter for the override.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_OVERRIDE_INFO_INTEL
pNext must be NULL
type must be a valid VkPerformanceOverrideTypeINTEL value

Possible values of VkPerformanceOverrideInfoINTEL::type, specifying performance override types, are:

// Provided by VK_INTEL_performance_query
typedef enum VkPerformanceOverrideTypeINTEL {
    VK_PERFORMANCE_OVERRIDE_TYPE_NULL_HARDWARE_INTEL = 0,
    VK_PERFORMANCE_OVERRIDE_TYPE_FLUSH_GPU_CACHES_INTEL = 1,
} VkPerformanceOverrideTypeINTEL;

VK_PERFORMANCE_OVERRIDE_TYPE_NULL_HARDWARE_INTEL turns all rendering operations into noop.
VK_PERFORMANCE_OVERRIDE_TYPE_FLUSH_GPU_CACHES_INTEL stalls the stream of commands until all previously emitted commands have completed and all caches been flushed and invalidated.

Before submitting command buffers containing performance queries commands to a device queue, the application must acquire and set a performance query configuration. The configuration can be released once all command buffers containing performance query commands are not in a pending state.

// Provided by VK_INTEL_performance_query
VK_DEFINE_NON_DISPATCHABLE_HANDLE(VkPerformanceConfigurationINTEL)

To acquire a device performance configuration, call:

// Provided by VK_INTEL_performance_query
VkResult vkAcquirePerformanceConfigurationINTEL(
    VkDevice                                    device,
    const VkPerformanceConfigurationAcquireInfoINTEL* pAcquireInfo,
    VkPerformanceConfigurationINTEL*            pConfiguration);

device is the logical device that the performance query commands will be submitted to.
pAcquireInfo is a pointer to a VkPerformanceConfigurationAcquireInfoINTEL structure, specifying the performance configuration to acquire.
pConfiguration is a pointer to a VkPerformanceConfigurationINTEL handle in which the resulting configuration object is returned.

Valid Usage (Implicit)

device must be a valid VkDevice handle
pAcquireInfo must be a valid pointer to a valid VkPerformanceConfigurationAcquireInfoINTEL structure
pConfiguration must be a valid pointer to a VkPerformanceConfigurationINTEL handle

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

The VkPerformanceConfigurationAcquireInfoINTEL structure is defined as:

// Provided by VK_INTEL_performance_query
typedef struct VkPerformanceConfigurationAcquireInfoINTEL {
    VkStructureType                        sType;
    const void*                            pNext;
    VkPerformanceConfigurationTypeINTEL    type;
} VkPerformanceConfigurationAcquireInfoINTEL;

sType is the type of this structure.
pNext is NULL or a pointer to a structure extending this structure.
type is one of the VkPerformanceConfigurationTypeINTEL type of performance configuration that will be acquired.

Valid Usage (Implicit)

sType must be VK_STRUCTURE_TYPE_PERFORMANCE_CONFIGURATION_ACQUIRE_INFO_INTEL
pNext must be NULL
type must be a valid VkPerformanceConfigurationTypeINTEL value

Possible values of VkPerformanceConfigurationAcquireInfoINTEL::type, specifying performance configuration types, are:

// Provided by VK_INTEL_performance_query
typedef enum VkPerformanceConfigurationTypeINTEL {
    VK_PERFORMANCE_CONFIGURATION_TYPE_COMMAND_QUEUE_METRICS_DISCOVERY_ACTIVATED_INTEL = 0,
} VkPerformanceConfigurationTypeINTEL;

To set a performance configuration, call:

// Provided by VK_INTEL_performance_query
VkResult vkQueueSetPerformanceConfigurationINTEL(
    VkQueue                                     queue,
    VkPerformanceConfigurationINTEL             configuration);

queue is the queue on which the configuration will be used.
configuration is the configuration to use.

Valid Usage (Implicit)

queue must be a valid VkQueue handle
configuration must be a valid VkPerformanceConfigurationINTEL handle
Both of configuration, and queue must have been created, allocated, or retrieved from the same VkDevice

Command Properties

Any

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

To release a device performance configuration, call:

// Provided by VK_INTEL_performance_query
VkResult vkReleasePerformanceConfigurationINTEL(
    VkDevice                                    device,
    VkPerformanceConfigurationINTEL             configuration);

device is the device associated to the configuration object to release.
configuration is the configuration object to release.

Valid Usage

configuration must not be released before all command buffers submitted while the configuration was set are in pending state

Valid Usage (Implicit)

device must be a valid VkDevice handle
configuration must be a valid VkPerformanceConfigurationINTEL handle
configuration must have been created, allocated, or retrieved from device

Return Codes

VK_SUCCESS

VK_ERROR_TOO_MANY_OBJECTS
VK_ERROR_OUT_OF_HOST_MEMORY

18. Clear Commands

18.1. 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:

// Provided by VK_VERSION_1_0
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 containing the values that the image subresource ranges will be cleared to (see Clear Values below).
rangeCount is the number of image subresource range structures in pRanges.
pRanges is a pointer to an array of VkImageSubresourceRange structures describing a range of mipmap levels, array layers, and aspects to be cleared, as described in Image Views.

Each specified range in pRanges is cleared to the value specified by pColor.

Valid Usage

The format features of image must contain VK_FORMAT_FEATURE_TRANSFER_DST_BIT
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′C_BC_R 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::aspectMask members of the elements of the pRanges array must each only include VK_IMAGE_ASPECT_COLOR_BIT
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

Primary
Secondary

Outside

Graphics
Compute

Transfer

To clear one or more subranges of a depth/stencil image, call:

// Provided by VK_VERSION_1_0
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 containing the values that 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 is a pointer to an array of VkImageSubresourceRange structures describing a range of mipmap levels, array layers, and aspects to be cleared, as described in Image Views.

Valid Usage

The format features of image must contain VK_FORMAT_FEATURE_TRANSFER_DST_BIT
If the aspect member of any element of pRanges includes VK_IMAGE_ASPECT_STENCIL_BIT, and image was created with separate stencil usage, VK_IMAGE_USAGE_TRANSFER_DST_BIT must have been included in the VkImageStencilUsageCreateInfo::stencilUsage used to create image
If the aspect member of any element of pRanges includes VK_IMAGE_ASPECT_STENCIL_BIT, and image was not created with separate stencil usage, VK_IMAGE_USAGE_TRANSFER_DST_BIT must have been included in the VkImageCreateInfo::usage used to create image
If the aspect member of any element of pRanges includes VK_IMAGE_ASPECT_DEPTH_BIT, VK_IMAGE_USAGE_TRANSFER_DST_BIT must have been included in the VkImageCreateInfo::usage used to create image
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::aspectMask member of each element of the pRanges array must not include bits other than VK_IMAGE_ASPECT_DEPTH_BIT or VK_IMAGE_ASPECT_STENCIL_BIT
If the image’s format does not have a stencil component, then the VkImageSubresourceRange::aspectMask member of each element of the pRanges array must not include the VK_IMAGE_ASPECT_STENCIL_BIT bit
If the image’s format does not have a depth component, then the VkImageSubresourceRange::aspectMask member of each element of the pRanges array must not include the VK_IMAGE_ASPECT_DEPTH_BIT bit
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

Primary
Secondary

Outside

Graphics

Transfer

Clears outside render pass instances are treated as transfer operations for the purposes of memory barriers.

18.2. 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:

// Provided by VK_VERSION_1_0
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. If any attachment to be cleared in the current subpass is VK_ATTACHMENT_UNUSED, then the clear has no effect on that attachment.
rectCount is the number of entries in the pRects array.
pRects is a pointer 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.

If the render pass has a fragment density map attachment, clears follow the operations of fragment density maps as if each clear region was a primitive which generates fragments. The clear color is applied to all pixels inside each fragment’s area regardless if the pixels lie outside of the clear region. Clears may have a different set of supported fragment areas than draws.

Unlike other clear commands, vkCmdClearAttachments executes as a drawing command, rather than a transfer command, with writes performed by it executing in rasterization order. Clears to color attachments are executed as color attachment writes, by the VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT stage. Clears to depth/stencil attachments are executed as depth writes and writes by the VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT and VK_PIPELINE_STAGE_LATE_FRAGMENT_TESTS_BIT stages.

Valid Usage

If the aspectMask member of any element of pAttachments contains VK_IMAGE_ASPECT_COLOR_BIT, then the colorAttachment member of that element must either refer to a color attachment which is VK_ATTACHMENT_UNUSED, or must be a valid color attachment
If the aspectMask member of any element of pAttachments contains VK_IMAGE_ASPECT_DEPTH_BIT, then the current subpass' depth/stencil attachment must either be VK_ATTACHMENT_UNUSED, or must have a depth component
If the aspectMask member of any element of pAttachments contains VK_IMAGE_ASPECT_STENCIL_BIT, then the current subpass' depth/stencil attachment must either be VK_ATTACHMENT_UNUSED, or must have a stencil component
The rect member of each element of pRects must have an extent.width greater than 0
The rect member of each element of pRects must have an extent.height greater than 0
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 commandBuffer is an unprotected command buffer, then each attachment to be cleared must not be a protected image
If commandBuffer is a protected command buffer, then each attachment to be cleared must not be an unprotected image
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

Primary
Secondary

Inside

Graphics

The VkClearRect structure is defined as:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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.
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.
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
aspectMask must not include VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT for any index i
clearValue must be a valid VkClearValue union

Valid Usage (Implicit)

aspectMask must be a valid combination of VkImageAspectFlagBits values
aspectMask must not be 0

18.3. Clear Values

The VkClearColorValue structure is defined as:

// Provided by VK_VERSION_1_0
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:

// Provided by VK_VERSION_1_0
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

Unless the VK_EXT_depth_range_unrestricted extension is enabled depth must be between 0.0 and 1.0, inclusive

The VkClearValue union is defined as:

// Provided by VK_VERSION_1_0
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.

18.4. Filling Buffers

To clear buffer data, call:

// Provided by VK_VERSION_1_0
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

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

18.5. Updating Buffers

To update buffer data inline in a command buffer, call:

// Provided by VK_VERSION_1_0
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

Primary
Secondary

Outside

Transfer
Graphics
Compute

Transfer

Note

The pData parameter was of type uint32_t* instead of void* prior to version 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.

19. 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.

19.1. 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 have VK_FORMAT_FEATURE_TRANSFER_SRC_BIT in their format features.
Destination images must have VK_FORMAT_FEATURE_TRANSFER_DST_BIT in their format features.
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.
If the stencil aspect of source image is accessed, and the source image was not created with separate stencil usage, the source image must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT set in VkImageCreateInfo::usage
If the stencil aspect of destination image is accessed, and the destination image was not created with separate stencil usage, the destination image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT set in VkImageCreateInfo::usage
If the stencil aspect of source image is accessed, and the source image was created with separate stencil usage, the source image must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT set in VkImageStencilUsageCreateInfo::stencilUsage
If the stencil aspect of destination image is accessed, and the destination image was created with separate stencil usage, the destination image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT set in VkImageStencilUsageCreateInfo::stencilUsage
If non-stencil aspects of a source image are accessed, the source image must have been created with VK_IMAGE_USAGE_TRANSFER_SRC_BIT set in VkImageCreateInfo::usage
If non-stencil aspects of a source image are accessed, the source image must have been created with VK_IMAGE_USAGE_TRANSFER_DST_BIT set in VkImageCreateInfo::usage

All copy commands are treated as “transfer” operations for the purposes of synchronization barriers.

All copy commands that have a source format with an X component in its format description read undefined values from those bits.

All copy commands that have a destination format with an X component in its format description write undefined values to those bits.

19.2. Copying Data Between Buffers

To copy data between buffer objects, call:

// Provided by VK_VERSION_1_0
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 resulting values are undefined if the copy regions overlap in memory.

Valid Usage

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

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

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 valid 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