Execution Graphs

Execution graphs provide a way for applications to dispatch multiple operations dynamically from a single initial command on the host. To achieve this, a new execution graph pipeline is provided, that links together multiple shaders or pipelines which each describe one or more operations that can be dispatched within the execution graph. Each linked pipeline or shader describes an execution node within the graph, which can be dispatched dynamically from another shader within the same graph. This allows applications to describe much richer execution topologies at a finer granularity than would typically be possible with API commands alone.

Pipeline Creation

To create execution graph pipelines, call:

// Provided by VK_AMDX_shader_enqueue
VkResult vkCreateExecutionGraphPipelinesAMDX(
    VkDevice                                    device,
    VkPipelineCache                             pipelineCache,
    uint32_t                                    createInfoCount,
    const VkExecutionGraphPipelineCreateInfoAMDX* pCreateInfos,
    const VkAllocationCallbacks*                pAllocator,
    VkPipeline*                                 pPipelines);
  • device is the logical device that creates the execution graph 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 VkExecutionGraphPipelineCreateInfoAMDX 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 execution graph pipeline objects are returned.

Pipelines are created and returned as described for Multiple Pipeline Creation.

Valid Usage
  • VUID-vkCreateExecutionGraphPipelinesAMDX-shaderEnqueue-09124
    The shaderEnqueue feature must be enabled

  • VUID-vkCreateExecutionGraphPipelinesAMDX-flags-09125
    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

  • VUID-vkCreateExecutionGraphPipelinesAMDX-flags-09126
    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

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pipelineCache-09127
    If pipelineCache was created with VK_PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT, host access to pipelineCache must be externally synchronized

Valid Usage (Implicit)
  • VUID-vkCreateExecutionGraphPipelinesAMDX-device-parameter
    device must be a valid VkDevice handle

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pipelineCache-parameter
    If pipelineCache is not VK_NULL_HANDLE, pipelineCache must be a valid VkPipelineCache handle

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pCreateInfos-parameter
    pCreateInfos must be a valid pointer to an array of createInfoCount valid VkExecutionGraphPipelineCreateInfoAMDX structures

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pAllocator-parameter
    If pAllocator is not NULL, pAllocator must be a valid pointer to a valid VkAllocationCallbacks structure

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pPipelines-parameter
    pPipelines must be a valid pointer to an array of createInfoCount VkPipeline handles

  • VUID-vkCreateExecutionGraphPipelinesAMDX-createInfoCount-arraylength
    createInfoCount must be greater than 0

  • VUID-vkCreateExecutionGraphPipelinesAMDX-pipelineCache-parent
    If pipelineCache is a valid handle, it must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

  • VK_PIPELINE_COMPILE_REQUIRED_EXT

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

  • VK_ERROR_OUT_OF_DEVICE_MEMORY

The VkExecutionGraphPipelineCreateInfoAMDX structure is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef struct VkExecutionGraphPipelineCreateInfoAMDX {
    VkStructureType                           sType;
    const void*                               pNext;
    VkPipelineCreateFlags                     flags;
    uint32_t                                  stageCount;
    const VkPipelineShaderStageCreateInfo*    pStages;
    const VkPipelineLibraryCreateInfoKHR*     pLibraryInfo;
    VkPipelineLayout                          layout;
    VkPipeline                                basePipelineHandle;
    int32_t                                   basePipelineIndex;
} VkExecutionGraphPipelineCreateInfoAMDX;
  • sType is a VkStructureType value identifying 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 execution graph pipeline.

  • pLibraryInfo is a pointer to a VkPipelineLibraryCreateInfoKHR structure defining pipeline libraries to include.

  • 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.

Each shader stage provided when creating an execution graph pipeline (including those in libraries) is associated with a name and an index, determined by the inclusion or omission of a VkPipelineShaderStageNodeCreateInfoAMDX structure in its pNext chain.

In addition to the shader name and index, an internal "node index" is also generated for each node, which can be queried with vkGetExecutionGraphPipelineNodeIndexAMDX, and is used exclusively for initial dispatch of an execution graph.

Valid Usage
  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-None-09497
    If the pNext chain does not include a VkPipelineCreateFlags2CreateInfoKHR structure, flags must be a valid combination of VkPipelineCreateFlagBits values

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-07984
    If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid execution graph VkPipeline handle

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-07985
    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

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-07986
    If flags contains the VK_PIPELINE_CREATE_DERIVATIVE_BIT flag, basePipelineIndex must be -1 or basePipelineHandle must be VK_NULL_HANDLE

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-07987
    If a push constant block is declared in a shader, a push constant range in layout must match both the shader stage and range

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-07988
    If a resource variables is declared in a shader, a descriptor slot in layout must match the shader stage

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-07990
    If a resource variables is declared in a shader, and the descriptor type is not VK_DESCRIPTOR_TYPE_MUTABLE_EXT, a descriptor slot in layout must match the descriptor type

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-07991
    If a resource variables is declared in a shader as an array, a descriptor slot in layout must match the descriptor count

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03365
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03366
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03367
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03368
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03369
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03370
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-03576
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_SHADER_GROUP_HANDLE_CAPTURE_REPLAY_BIT_KHR

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-04945
    flags must not include VK_PIPELINE_CREATE_RAY_TRACING_ALLOW_MOTION_BIT_NV

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-09007
    If the VkPhysicalDeviceDeviceGeneratedCommandsComputeFeaturesNV::deviceGeneratedComputePipelines is not enabled, flags must not include VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-flags-09008
    If flags includes VK_PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV, then the pNext chain must include a pointer to a valid instance of VkComputePipelineIndirectBufferInfoNV specifying the address where the pipeline’s metadata will be saved

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pipelineCreationCacheControl-02875
    If the pipelineCreationCacheControl feature is not enabled, flags must not include VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT or VK_PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-stage-09128
    The stage member of any element of pStages must be VK_SHADER_STAGE_COMPUTE_BIT

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pStages-09129
    The shader code for the entry point identified by each element of pStages and the rest of the state identified by this structure must adhere to the pipeline linking rules described in the Shader Interfaces chapter

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-09130
    layout must be consistent with the layout of the shaders specified in pStages

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pLibraryInfo-09131
    If pLibraryInfo is not NULL, each element of its pLibraries member must have been created with a layout that is compatible with the layout in this pipeline

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-09132
    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

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pLibraryInfo-09133
    If pLibraryInfo is not NULL, each element of pLibraryInfo->libraries must be either a compute pipeline or an execution graph pipeline

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-None-09134
    There must be no two nodes in the pipeline that share both the same shader name and index, as specified by VkPipelineShaderStageNodeCreateInfoAMDX

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-None-09135
    There must be no two nodes in the pipeline that share the same shader name and have input payload declarations with different sizes

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-None-09136
    There must be no two nodes in the pipeline that share the same name but have different execution models

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-CoalescedInputCountAMDX-09137
    There must be no two nodes in the pipeline that share the same name where one includes CoalescedInputCountAMDX and the other does not

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-StaticNumWorkgroupsAMDX-09138
    There must be no two nodes in the pipeline that share the same name where one includes StaticNumWorkgroupsAMDX and the other does not

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-PayloadNodeNameAMDX-09139
    If an output payload declared in any shader in the pipeline has a PayloadNodeNameAMDX decoration with a Node Name that matches the shader name of any other node in the graph, the size of the output payload must match the size of the input payload in the matching node

Valid Usage (Implicit)
  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-sType-sType
    sType must be VK_STRUCTURE_TYPE_EXECUTION_GRAPH_PIPELINE_CREATE_INFO_AMDX

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pNext-pNext
    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 VkPipelineCreationFeedbackCreateInfo

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-sType-unique
    The sType value of each struct in the pNext chain must be unique

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pStages-parameter
    If stageCount is not 0, and pStages is not NULL, pStages must be a valid pointer to an array of stageCount valid VkPipelineShaderStageCreateInfo structures

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-pLibraryInfo-parameter
    If pLibraryInfo is not NULL, pLibraryInfo must be a valid pointer to a valid VkPipelineLibraryCreateInfoKHR structure

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-layout-parameter
    layout must be a valid VkPipelineLayout handle

  • VUID-VkExecutionGraphPipelineCreateInfoAMDX-commonparent
    Both of basePipelineHandle, and layout that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same VkDevice

VK_SHADER_INDEX_UNUSED_AMDX is a special shader index used to indicate that the created node does not override the index. In this case, the shader index is determined through other means. It is defined as:

#define VK_SHADER_INDEX_UNUSED_AMDX       (~0U)

The VkPipelineShaderStageNodeCreateInfoAMDX structure is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef struct VkPipelineShaderStageNodeCreateInfoAMDX {
      VkStructureType    sType;
    const void*          pNext;
    const char*          pName;
    uint32_t             index;
} VkPipelineShaderStageNodeCreateInfoAMDX;
  • sType is a VkStructureType value identifying this structure.

  • pNext is NULL or a pointer to a structure extending this structure.

  • pName is the shader name to use when creating a node in an execution graph. If pName is NULL, the name of the entry point specified in SPIR-V is used as the shader name.

  • index is the shader index to use when creating a node in an execution graph. If index is VK_SHADER_INDEX_UNUSED_AMDX then the original index is used, either as specified by the ShaderIndexAMDX execution mode, or 0 if that too is not specified.

When included in the pNext chain of a VkPipelineShaderStageCreateInfo structure, this structure specifies the shader name and shader index of a node when creating an execution graph pipeline. If this structure is omitted, the shader name is set to the name of the entry point in SPIR-V and the shader index is set to 0.

When dispatching a node from another shader, the name is fixed at pipeline creation, but the index can be set dynamically. By associating multiple shaders with the same name but different indexes, applications can dynamically select different nodes to execute. Applications must ensure each node has a unique name and index.

Valid Usage (Implicit)
  • VUID-VkPipelineShaderStageNodeCreateInfoAMDX-sType-sType
    sType must be VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_NODE_CREATE_INFO_AMDX

  • VUID-VkPipelineShaderStageNodeCreateInfoAMDX-pName-parameter
    If pName is not NULL, pName must be a null-terminated UTF-8 string

To query the internal node index for a particular node in an execution graph, call:

// Provided by VK_AMDX_shader_enqueue
VkResult vkGetExecutionGraphPipelineNodeIndexAMDX(
    VkDevice                                    device,
    VkPipeline                                  executionGraph,
    const VkPipelineShaderStageNodeCreateInfoAMDX* pNodeInfo,
    uint32_t*                                   pNodeIndex);
  • device is the that executionGraph was created on.

  • executionGraph is the execution graph pipeline to query the internal node index for.

  • pNodeInfo is a pointer to a VkPipelineShaderStageNodeCreateInfoAMDX structure identifying the name and index of the node to query.

  • pNodeIndex is the returned internal node index of the identified node.

Once this function returns, the contents of pNodeIndex contain the internal node index of the identified node.

Valid Usage
  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-pNodeInfo-09140
    pNodeInfo->pName must not be NULL

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-pNodeInfo-09141
    pNodeInfo->index must not be VK_SHADER_INDEX_UNUSED_AMDX

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-executionGraph-09142
    There must be a node in executionGraph with a shader name and index equal to pNodeInfo->pName and pNodeInfo->index

Valid Usage (Implicit)
  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-device-parameter
    device must be a valid VkDevice handle

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-executionGraph-parameter
    executionGraph must be a valid VkPipeline handle

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-pNodeInfo-parameter
    pNodeInfo must be a valid pointer to a valid VkPipelineShaderStageNodeCreateInfoAMDX structure

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-pNodeIndex-parameter
    pNodeIndex must be a valid pointer to a uint32_t value

  • VUID-vkGetExecutionGraphPipelineNodeIndexAMDX-executionGraph-parent
    executionGraph must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

Initializing Scratch Memory

Implementations may need scratch memory to manage dispatch queues or similar when executing a pipeline graph, and this is explicitly managed by the application.

To query the scratch space required to dispatch an execution graph, call:

// Provided by VK_AMDX_shader_enqueue
VkResult vkGetExecutionGraphPipelineScratchSizeAMDX(
    VkDevice                                    device,
    VkPipeline                                  executionGraph,
    VkExecutionGraphPipelineScratchSizeAMDX*    pSizeInfo);
  • device is the that executionGraph was created on.

  • executionGraph is the execution graph pipeline to query the scratch space for.

  • pSizeInfo is a pointer to a VkExecutionGraphPipelineScratchSizeAMDX structure that will contain the required scratch size.

After this function returns, information about the scratch space required will be returned in pSizeInfo.

Valid Usage (Implicit)
  • VUID-vkGetExecutionGraphPipelineScratchSizeAMDX-device-parameter
    device must be a valid VkDevice handle

  • VUID-vkGetExecutionGraphPipelineScratchSizeAMDX-executionGraph-parameter
    executionGraph must be a valid VkPipeline handle

  • VUID-vkGetExecutionGraphPipelineScratchSizeAMDX-pSizeInfo-parameter
    pSizeInfo must be a valid pointer to a VkExecutionGraphPipelineScratchSizeAMDX structure

  • VUID-vkGetExecutionGraphPipelineScratchSizeAMDX-executionGraph-parent
    executionGraph must have been created, allocated, or retrieved from device

Return Codes
Success
  • VK_SUCCESS

Failure
  • VK_ERROR_OUT_OF_HOST_MEMORY

The VkExecutionGraphPipelineScratchSizeAMDX structure is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef struct VkExecutionGraphPipelineScratchSizeAMDX {
    VkStructureType    sType;
    void*              pNext;
    VkDeviceSize       size;
} VkExecutionGraphPipelineScratchSizeAMDX;
  • sType is a VkStructureType value identifying this structure.

  • pNext is NULL or a pointer to a structure extending this structure.

  • size indicates the scratch space required for dispatch the queried execution graph.

Valid Usage (Implicit)
  • VUID-VkExecutionGraphPipelineScratchSizeAMDX-sType-sType
    sType must be VK_STRUCTURE_TYPE_EXECUTION_GRAPH_PIPELINE_SCRATCH_SIZE_AMDX

To initialize scratch memory for a particular execution graph, call:

// Provided by VK_AMDX_shader_enqueue
void vkCmdInitializeGraphScratchMemoryAMDX(
    VkCommandBuffer                             commandBuffer,
    VkDeviceAddress                             scratch);
  • commandBuffer is the command buffer into which the command will be recorded.

  • scratch is a pointer to the scratch memory to be initialized.

This command must be called before using scratch to dispatch the currently bound execution graph pipeline.

Execution of this command may modify any memory locations in the range [scratch,scratch + size), where size is the value returned in VkExecutionGraphPipelineScratchSizeAMDX::size by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline. Accesses to this memory range are performed in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT and VK_ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

If any portion of scratch is modified by any command other than vkCmdDispatchGraphAMDX, vkCmdDispatchGraphIndirectAMDX, vkCmdDispatchGraphIndirectCountAMDX, or vkCmdInitializeGraphScratchMemoryAMDX with the same execution graph, it must be reinitialized for the execution graph again before dispatching against it.

Valid Usage
  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-scratch-09143
    scratch must be the device address of an allocated memory range at least as large as the value of VkExecutionGraphPipelineScratchSizeAMDX::size returned by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline.

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-scratch-09144
    scratch must be a multiple of 64

Valid Usage (Implicit)
  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-commandBuffer-parameter
    commandBuffer must be a valid VkCommandBuffer handle

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-commandBuffer-recording
    commandBuffer must be in the recording state

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-commandBuffer-cmdpool
    The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-renderpass
    This command must only be called outside of a render pass instance

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-videocoding
    This command must only be called outside of a video coding scope

  • VUID-vkCmdInitializeGraphScratchMemoryAMDX-bufferlevel
    commandBuffer must be a primary VkCommandBuffer

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

Command Properties
Command Buffer Levels Render Pass Scope Video Coding Scope Supported Queue Types Command Type

Primary

Outside

Outside

Graphics
Compute

Action

Dispatching a Graph

Initial dispatch of an execution graph is done from the host in the same way as any other command, and can be used in a similar way to compute dispatch commands, with indirect variants available.

To record an execution graph dispatch, call:

// Provided by VK_AMDX_shader_enqueue
void vkCmdDispatchGraphAMDX(
    VkCommandBuffer                             commandBuffer,
    VkDeviceAddress                             scratch,
    const VkDispatchGraphCountInfoAMDX*         pCountInfo);
  • commandBuffer is the command buffer into which the command will be recorded.

  • scratch is a pointer to the scratch memory to be used.

  • pCountInfo is a host pointer to a VkDispatchGraphCountInfoAMDX structure defining the nodes which will be initially executed.

When this command is executed, the nodes specified in pCountInfo are executed. Nodes executed as part of this command are not implicitly synchronized in any way against each other once they are dispatched.

For this command, all device/host pointers in substructures are treated as host pointers and read only during host execution of this command. Once this command returns, no reference to the original pointers is retained.

Execution of this command may modify any memory locations in the range [scratch,scratch + size), where size is the value returned in VkExecutionGraphPipelineScratchSizeAMDX::size by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline Accesses to this memory range are performed in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT and VK_ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

Valid Usage
  • VUID-vkCmdDispatchGraphAMDX-magFilter-04553
    If a VkSampler created with magFilter or minFilter equal to VK_FILTER_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphAMDX-mipmapMode-04770
    If a VkSampler created with mipmapMode equal to VK_SAMPLER_MIPMAP_MODE_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphAMDX-None-06479
    If a VkImageView is sampled with depth comparison, the image view’s format features must contain VK_FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT

  • VUID-vkCmdDispatchGraphAMDX-None-02691
    If a VkImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphAMDX-None-07888
    If a VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER descriptor is accessed using atomic operations as a result of this command, then the storage texel buffer’s format features must contain VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphAMDX-None-02692
    If a VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT

  • VUID-vkCmdDispatchGraphAMDX-None-02693
    If the VK_EXT_filter_cubic extension is not enabled and any VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, it must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • VUID-vkCmdDispatchGraphAMDX-filterCubic-02694
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have a VkImageViewType and format that supports cubic filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphAMDX-filterCubicMinmax-02695
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT with a reduction mode of either VK_SAMPLER_REDUCTION_MODE_MIN or VK_SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a VkImageViewType and format that supports cubic filtering together with minmax filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphAMDX-cubicRangeClamp-09212
    If the cubicRangeClamp feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must not have a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM

  • VUID-vkCmdDispatchGraphAMDX-reductionMode-09213
    Any VkImageView being sampled with a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with VK_FILTER_CUBIC_EXT

  • VUID-vkCmdDispatchGraphAMDX-selectableCubicWeights-09214
    If the selectableCubicWeights feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have VkSamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to VK_CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM

  • VUID-vkCmdDispatchGraphAMDX-flags-02696
    Any VkImage created with a VkImageCreateInfo::flags containing VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a VkSamplerAddressMode of VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE

  • VUID-vkCmdDispatchGraphAMDX-OpTypeImage-07027
    For any VkImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphAMDX-OpTypeImage-07028
    For any VkImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphAMDX-OpTypeImage-07029
    For any VkBufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphAMDX-OpTypeImage-07030
    Any VkBufferView being read as a storage texel buffer where the image format field of the OpTypeImage is Unknown then the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphAMDX-None-08600
    For each set n that is statically used by a bound shader, a descriptor set must have been bound to n at the same pipeline bind point, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphAMDX-None-08601
    For each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphAMDX-maintenance4-08602
    If the maintenance4 feature is not enabled, then for each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout and VkPushConstantRange arrays used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphAMDX-None-08114
    Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was not created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphAMDX-None-08115
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdBindDescriptorSets, the bound VkPipeline must have been created without VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphAMDX-None-08116
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphAMDX-None-08604
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphAMDX-None-08117
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdSetDescriptorBufferOffsetsEXT, the bound VkPipeline must have been created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphAMDX-None-08119
    If a descriptor is dynamically used with a VkPipeline created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphAMDX-None-08605
    If a descriptor is dynamically used with a VkShaderEXT created with a VkDescriptorSetLayout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphAMDX-None-08606
    If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphAMDX-None-08608
    If a pipeline is bound to the pipeline bind point used by this command, there must not have been any calls to dynamic state setting commands for any state not specified as dynamic in the VkPipeline object bound to the pipeline bind point used by this command, since that pipeline was bound

  • VUID-vkCmdDispatchGraphAMDX-None-08609
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • VUID-vkCmdDispatchGraphAMDX-None-08610
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • VUID-vkCmdDispatchGraphAMDX-None-08611
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • VUID-vkCmdDispatchGraphAMDX-None-08607
    If the shaderObject is enabled, either a valid pipeline must be bound to the pipeline bind point used by this command, or a valid combination of valid and VK_NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphAMDX-uniformBuffers-06935
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for uniformBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphAMDX-None-08612
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a uniform buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphAMDX-storageBuffers-06936
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for storageBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphAMDX-None-08613
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a storage buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-02707
    If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource

  • VUID-vkCmdDispatchGraphAMDX-None-06550
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions

  • VUID-vkCmdDispatchGraphAMDX-ConstOffset-06551
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands

  • VUID-vkCmdDispatchGraphAMDX-viewType-07752
    If a VkImageView is accessed as a result of this command, then the image view’s viewType must match the Dim operand of the OpTypeImage as described in Instruction/Sampler/Image View Validation

  • VUID-vkCmdDispatchGraphAMDX-format-07753
    If a VkImageView is accessed as a result of this command, then the numeric type of the image view’s format and the Sampled Type operand of the OpTypeImage must match

  • VUID-vkCmdDispatchGraphAMDX-OpImageWrite-08795
    If a VkImageView created with a format other than VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the image view’s format

  • VUID-vkCmdDispatchGraphAMDX-OpImageWrite-08796
    If a VkImageView created with the format VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have four components

  • VUID-vkCmdDispatchGraphAMDX-OpImageWrite-04469
    If a VkBufferView is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the buffer view’s format

  • VUID-vkCmdDispatchGraphAMDX-SampledType-04470
    If a VkImageView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphAMDX-SampledType-04471
    If a VkImageView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphAMDX-SampledType-04472
    If a VkBufferView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphAMDX-SampledType-04473
    If a VkBufferView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphAMDX-sparseImageInt64Atomics-04474
    If the sparseImageInt64Atomics feature is not enabled, VkImage objects created with the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphAMDX-sparseImageInt64Atomics-04475
    If the sparseImageInt64Atomics feature is not enabled, VkBuffer objects created with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphAMDX-OpImageWeightedSampleQCOM-06971
    If OpImageWeightedSampleQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageWeightedSampleQCOM-06972
    If OpImageWeightedSampleQCOM uses a VkImageView as a sample weight image as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBoxFilterQCOM-06973
    If OpImageBoxFilterQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchSSDQCOM-06974
    If OpImageBlockMatchSSDQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchSADQCOM-06975
    If OpImageBlockMatchSADQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchSADQCOM-06976
    If OpImageBlockMatchSADQCOM or OpImageBlockMatchSSDQCOM is used to read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphAMDX-OpImageWeightedSampleQCOM-06977
    If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageWeightedSampleQCOM-06978
    If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must not have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchWindow-09215
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchWindow-09216
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format must be a single-component format.

  • VUID-vkCmdDispatchGraphAMDX-OpImageBlockMatchWindow-09217
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphAMDX-None-07288
    Any shader invocation executed by this command must terminate

  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-09181
    commandBuffer must not be a protected command buffer

  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-09182
    commandBuffer must be a primary command buffer

  • VUID-vkCmdDispatchGraphAMDX-scratch-09183
    scratch must be the device address of an allocated memory range at least as large as the value of VkExecutionGraphPipelineScratchSizeAMDX::size returned by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline

  • VUID-vkCmdDispatchGraphAMDX-scratch-09184
    scratch must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or VK_BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag

  • VUID-vkCmdDispatchGraphAMDX-scratch-09185
    Device memory in the range [scratch,scratch
    VkExecutionGraphPipelineScratchSizeAMDX::size) must have been initialized with vkCmdInitializeGraphScratchMemoryAMDX using the currently bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command

  • VUID-vkCmdDispatchGraphAMDX-maxComputeWorkGroupCount-09186
    Execution of this command must not cause a node to be dispatched with a larger number of workgroups than that specified by either a MaxNumWorkgroupsAMDX decoration in the dispatched node or maxComputeWorkGroupCount

  • VUID-vkCmdDispatchGraphAMDX-maxExecutionGraphShaderPayloadCount-09187
    Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads

  • VUID-vkCmdDispatchGraphAMDX-NodeMaxPayloadsAMDX-09188
    Execution of this command must not cause any shader that declares NodeMaxPayloadsAMDX to initialize more output payloads than specified by the max number of payloads for that decoration. This requirement applies to each NodeMaxPayloadsAMDX decoration separately

  • VUID-vkCmdDispatchGraphAMDX-pCountInfo-09145
    pCountInfo->infos must be a host pointer to a memory allocation at least as large as the product of count and stride

  • VUID-vkCmdDispatchGraphAMDX-infos-09146
    Host memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid VkDispatchGraphInfoAMDX structures in the first 24 bytes

  • VUID-vkCmdDispatchGraphAMDX-pCountInfo-09147
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a host pointer to a memory allocation at least as large as the product of payloadCount and payloadStride

  • VUID-vkCmdDispatchGraphAMDX-pCountInfo-09148
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, nodeIndex must be a valid node index in the currently bound execution graph pipeline, as returned by vkGetExecutionGraphPipelineNodeIndexAMDX

  • VUID-vkCmdDispatchGraphAMDX-pCountInfo-09149
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, host memory locations at indexes in the range [payloads, payloads + (payloadCount * payloadStride)), at a granularity of payloadStride must contain a payload matching the size of the input payload expected by the node in nodeIndex in the first bytes

Valid Usage (Implicit)
  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-parameter
    commandBuffer must be a valid VkCommandBuffer handle

  • VUID-vkCmdDispatchGraphAMDX-pCountInfo-parameter
    pCountInfo must be a valid pointer to a valid VkDispatchGraphCountInfoAMDX structure

  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-recording
    commandBuffer must be in the recording state

  • VUID-vkCmdDispatchGraphAMDX-commandBuffer-cmdpool
    The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • VUID-vkCmdDispatchGraphAMDX-renderpass
    This command must only be called outside of a render pass instance

  • VUID-vkCmdDispatchGraphAMDX-videocoding
    This command must only be called outside of a video coding scope

  • VUID-vkCmdDispatchGraphAMDX-bufferlevel
    commandBuffer must be a primary VkCommandBuffer

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

Command Properties
Command Buffer Levels Render Pass Scope Video Coding Scope Supported Queue Types Command Type

Primary

Outside

Outside

Graphics
Compute

Action

To record an execution graph dispatch with node and payload parameters read on device, call:

// Provided by VK_AMDX_shader_enqueue
void vkCmdDispatchGraphIndirectAMDX(
    VkCommandBuffer                             commandBuffer,
    VkDeviceAddress                             scratch,
    const VkDispatchGraphCountInfoAMDX*         pCountInfo);
  • commandBuffer is the command buffer into which the command will be recorded.

  • scratch is a pointer to the scratch memory to be used.

  • pCountInfo is a host pointer to a VkDispatchGraphCountInfoAMDX structure defining the nodes which will be initially executed.

When this command is executed, the nodes specified in pCountInfo are executed. Nodes executed as part of this command are not implicitly synchronized in any way against each other once they are dispatched.

For this command, all device/host pointers in substructures are treated as device pointers and read during device execution of this command. The allocation and contents of these pointers only needs to be valid during device execution. All of these addresses will be read in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT access flag.

Execution of this command may modify any memory locations in the range [scratch,scratch + size), where size is the value returned in VkExecutionGraphPipelineScratchSizeAMDX::size by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline. Accesses to this memory range are performed in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT and VK_ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

Valid Usage
  • VUID-vkCmdDispatchGraphIndirectAMDX-magFilter-04553
    If a VkSampler created with magFilter or minFilter equal to VK_FILTER_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-mipmapMode-04770
    If a VkSampler created with mipmapMode equal to VK_SAMPLER_MIPMAP_MODE_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-06479
    If a VkImageView is sampled with depth comparison, the image view’s format features must contain VK_FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-02691
    If a VkImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-07888
    If a VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER descriptor is accessed using atomic operations as a result of this command, then the storage texel buffer’s format features must contain VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-02692
    If a VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-02693
    If the VK_EXT_filter_cubic extension is not enabled and any VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, it must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • VUID-vkCmdDispatchGraphIndirectAMDX-filterCubic-02694
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have a VkImageViewType and format that supports cubic filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphIndirectAMDX-filterCubicMinmax-02695
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT with a reduction mode of either VK_SAMPLER_REDUCTION_MODE_MIN or VK_SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a VkImageViewType and format that supports cubic filtering together with minmax filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphIndirectAMDX-cubicRangeClamp-09212
    If the cubicRangeClamp feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must not have a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-reductionMode-09213
    Any VkImageView being sampled with a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with VK_FILTER_CUBIC_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-selectableCubicWeights-09214
    If the selectableCubicWeights feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have VkSamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to VK_CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-flags-02696
    Any VkImage created with a VkImageCreateInfo::flags containing VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a VkSamplerAddressMode of VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpTypeImage-07027
    For any VkImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpTypeImage-07028
    For any VkImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpTypeImage-07029
    For any VkBufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpTypeImage-07030
    Any VkBufferView being read as a storage texel buffer where the image format field of the OpTypeImage is Unknown then the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08600
    For each set n that is statically used by a bound shader, a descriptor set must have been bound to n at the same pipeline bind point, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08601
    For each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectAMDX-maintenance4-08602
    If the maintenance4 feature is not enabled, then for each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout and VkPushConstantRange arrays used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08114
    Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was not created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08115
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdBindDescriptorSets, the bound VkPipeline must have been created without VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08116
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08604
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08117
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdSetDescriptorBufferOffsetsEXT, the bound VkPipeline must have been created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08119
    If a descriptor is dynamically used with a VkPipeline created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08605
    If a descriptor is dynamically used with a VkShaderEXT created with a VkDescriptorSetLayout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08606
    If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08608
    If a pipeline is bound to the pipeline bind point used by this command, there must not have been any calls to dynamic state setting commands for any state not specified as dynamic in the VkPipeline object bound to the pipeline bind point used by this command, since that pipeline was bound

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08609
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08610
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08611
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08607
    If the shaderObject is enabled, either a valid pipeline must be bound to the pipeline bind point used by this command, or a valid combination of valid and VK_NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectAMDX-uniformBuffers-06935
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for uniformBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08612
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a uniform buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectAMDX-storageBuffers-06936
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for storageBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-08613
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a storage buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-02707
    If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-06550
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions

  • VUID-vkCmdDispatchGraphIndirectAMDX-ConstOffset-06551
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands

  • VUID-vkCmdDispatchGraphIndirectAMDX-viewType-07752
    If a VkImageView is accessed as a result of this command, then the image view’s viewType must match the Dim operand of the OpTypeImage as described in Instruction/Sampler/Image View Validation

  • VUID-vkCmdDispatchGraphIndirectAMDX-format-07753
    If a VkImageView is accessed as a result of this command, then the numeric type of the image view’s format and the Sampled Type operand of the OpTypeImage must match

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWrite-08795
    If a VkImageView created with a format other than VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the image view’s format

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWrite-08796
    If a VkImageView created with the format VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have four components

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWrite-04469
    If a VkBufferView is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the buffer view’s format

  • VUID-vkCmdDispatchGraphIndirectAMDX-SampledType-04470
    If a VkImageView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphIndirectAMDX-SampledType-04471
    If a VkImageView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphIndirectAMDX-SampledType-04472
    If a VkBufferView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphIndirectAMDX-SampledType-04473
    If a VkBufferView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphIndirectAMDX-sparseImageInt64Atomics-04474
    If the sparseImageInt64Atomics feature is not enabled, VkImage objects created with the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphIndirectAMDX-sparseImageInt64Atomics-04475
    If the sparseImageInt64Atomics feature is not enabled, VkBuffer objects created with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWeightedSampleQCOM-06971
    If OpImageWeightedSampleQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWeightedSampleQCOM-06972
    If OpImageWeightedSampleQCOM uses a VkImageView as a sample weight image as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBoxFilterQCOM-06973
    If OpImageBoxFilterQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchSSDQCOM-06974
    If OpImageBlockMatchSSDQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchSADQCOM-06975
    If OpImageBlockMatchSADQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchSADQCOM-06976
    If OpImageBlockMatchSADQCOM or OpImageBlockMatchSSDQCOM is used to read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWeightedSampleQCOM-06977
    If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageWeightedSampleQCOM-06978
    If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must not have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchWindow-09215
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchWindow-09216
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format must be a single-component format.

  • VUID-vkCmdDispatchGraphIndirectAMDX-OpImageBlockMatchWindow-09217
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphIndirectAMDX-None-07288
    Any shader invocation executed by this command must terminate

  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-09181
    commandBuffer must not be a protected command buffer

  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-09182
    commandBuffer must be a primary command buffer

  • VUID-vkCmdDispatchGraphIndirectAMDX-scratch-09183
    scratch must be the device address of an allocated memory range at least as large as the value of VkExecutionGraphPipelineScratchSizeAMDX::size returned by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline

  • VUID-vkCmdDispatchGraphIndirectAMDX-scratch-09184
    scratch must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or VK_BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag

  • VUID-vkCmdDispatchGraphIndirectAMDX-scratch-09185
    Device memory in the range [scratch,scratch
    VkExecutionGraphPipelineScratchSizeAMDX::size) must have been initialized with vkCmdInitializeGraphScratchMemoryAMDX using the currently bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command

  • VUID-vkCmdDispatchGraphIndirectAMDX-maxComputeWorkGroupCount-09186
    Execution of this command must not cause a node to be dispatched with a larger number of workgroups than that specified by either a MaxNumWorkgroupsAMDX decoration in the dispatched node or maxComputeWorkGroupCount

  • VUID-vkCmdDispatchGraphIndirectAMDX-maxExecutionGraphShaderPayloadCount-09187
    Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads

  • VUID-vkCmdDispatchGraphIndirectAMDX-NodeMaxPayloadsAMDX-09188
    Execution of this command must not cause any shader that declares NodeMaxPayloadsAMDX to initialize more output payloads than specified by the max number of payloads for that decoration. This requirement applies to each NodeMaxPayloadsAMDX decoration separately

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09150
    pCountInfo->infos must be a device pointer to a memory allocation at least as large as the product of count and stride when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09151
    pCountInfo->infos must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT flag

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09152
    pCountInfo->infos must be a multiple of executionGraphDispatchAddressAlignment

  • VUID-vkCmdDispatchGraphIndirectAMDX-infos-09153
    Device memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid VkDispatchGraphInfoAMDX structures in the first 24 bytes when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09154
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a device pointer to a memory allocation at least as large as the product of payloadCount and payloadStride when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09155
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT flag

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09156
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a multiple of executionGraphDispatchAddressAlignment

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09157
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, nodeIndex must be a valid node index in the currently bound execution graph pipeline, as returned by vkGetExecutionGraphPipelineNodeIndexAMDX when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-09158
    For each VkDispatchGraphInfoAMDX structure in pCountInfo->infos, device memory locations at indexes in the range [payloads, payloads + (payloadCount * payloadStride)), at a granularity of payloadStride must contain a payload matching the size of the input payload expected by the node in nodeIndex in the first bytes when this command is executed on the device

Valid Usage (Implicit)
  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-parameter
    commandBuffer must be a valid VkCommandBuffer handle

  • VUID-vkCmdDispatchGraphIndirectAMDX-pCountInfo-parameter
    pCountInfo must be a valid pointer to a valid VkDispatchGraphCountInfoAMDX structure

  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-recording
    commandBuffer must be in the recording state

  • VUID-vkCmdDispatchGraphIndirectAMDX-commandBuffer-cmdpool
    The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • VUID-vkCmdDispatchGraphIndirectAMDX-renderpass
    This command must only be called outside of a render pass instance

  • VUID-vkCmdDispatchGraphIndirectAMDX-videocoding
    This command must only be called outside of a video coding scope

  • VUID-vkCmdDispatchGraphIndirectAMDX-bufferlevel
    commandBuffer must be a primary VkCommandBuffer

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

Command Properties
Command Buffer Levels Render Pass Scope Video Coding Scope Supported Queue Types Command Type

Primary

Outside

Outside

Graphics
Compute

Action

To record an execution graph dispatch with all parameters read on device, call:

// Provided by VK_AMDX_shader_enqueue
void vkCmdDispatchGraphIndirectCountAMDX(
    VkCommandBuffer                             commandBuffer,
    VkDeviceAddress                             scratch,
    VkDeviceAddress                             countInfo);
  • commandBuffer is the command buffer into which the command will be recorded.

  • scratch is a pointer to the scratch memory to be used.

  • countInfo is a device address of a VkDispatchGraphCountInfoAMDX structure defining the nodes which will be initially executed.

When this command is executed, the nodes specified in countInfo are executed. Nodes executed as part of this command are not implicitly synchronized in any way against each other once they are dispatched.

For this command, all pointers in substructures are treated as device pointers and read during device execution of this command. The allocation and contents of these pointers only needs to be valid during device execution. All of these addresses will be read in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT access flag.

Execution of this command may modify any memory locations in the range [scratch,scratch + size), where size is the value returned in VkExecutionGraphPipelineScratchSizeAMDX::size by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline. Accesses to this memory range are performed in the VK_PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the VK_ACCESS_2_SHADER_STORAGE_READ_BIT and VK_ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

Valid Usage
  • VUID-vkCmdDispatchGraphIndirectCountAMDX-magFilter-04553
    If a VkSampler created with magFilter or minFilter equal to VK_FILTER_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-mipmapMode-04770
    If a VkSampler created with mipmapMode equal to VK_SAMPLER_MIPMAP_MODE_LINEAR and compareEnable equal to VK_FALSE is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-06479
    If a VkImageView is sampled with depth comparison, the image view’s format features must contain VK_FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-02691
    If a VkImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-07888
    If a VK_DESCRIPTOR_TYPE_STORAGE_TEXEL_BUFFER descriptor is accessed using atomic operations as a result of this command, then the storage texel buffer’s format features must contain VK_FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-02692
    If a VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-02693
    If the VK_EXT_filter_cubic extension is not enabled and any VkImageView is sampled with VK_FILTER_CUBIC_EXT as a result of this command, it must not have a VkImageViewType of VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-filterCubic-02694
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have a VkImageViewType and format that supports cubic filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-filterCubicMinmax-02695
    Any VkImageView being sampled with VK_FILTER_CUBIC_EXT with a reduction mode of either VK_SAMPLER_REDUCTION_MODE_MIN or VK_SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a VkImageViewType and format that supports cubic filtering together with minmax filtering, as specified by VkFilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by vkGetPhysicalDeviceImageFormatProperties2

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-cubicRangeClamp-09212
    If the cubicRangeClamp feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must not have a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-reductionMode-09213
    Any VkImageView being sampled with a VkSamplerReductionModeCreateInfo::reductionMode equal to VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with VK_FILTER_CUBIC_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-selectableCubicWeights-09214
    If the selectableCubicWeights feature is not enabled, then any VkImageView being sampled with VK_FILTER_CUBIC_EXT as a result of this command must have VkSamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to VK_CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-flags-02696
    Any VkImage created with a VkImageCreateInfo::flags containing VK_IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a VkSamplerAddressMode of VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpTypeImage-07027
    For any VkImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpTypeImage-07028
    For any VkImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpTypeImage-07029
    For any VkBufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpTypeImage-07030
    Any VkBufferView being read as a storage texel buffer where the image format field of the OpTypeImage is Unknown then the view’s buffer features must contain VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08600
    For each set n that is statically used by a bound shader, a descriptor set must have been bound to n at the same pipeline bind point, with a VkPipelineLayout that is compatible for set n, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08601
    For each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout array used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-maintenance4-08602
    If the maintenance4 feature is not enabled, then for each push constant that is statically used by a bound shader, a push constant value must have been set for the same pipeline bind point, with a VkPipelineLayout that is compatible for push constants, with the VkPipelineLayout used to create the current VkPipeline or the VkDescriptorSetLayout and VkPushConstantRange arrays used to create the current VkShaderEXT , as described in Pipeline Layout Compatibility

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08114
    Descriptors in each bound descriptor set, specified via vkCmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was not created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08115
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdBindDescriptorSets, the bound VkPipeline must have been created without VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08116
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the VkPipeline bound to the pipeline bind point used by this command and the bound VkPipeline was created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08604
    Descriptors in bound descriptor buffers, specified via vkCmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08117
    If the descriptors used by the VkPipeline bound to the pipeline bind point were specified via vkCmdSetDescriptorBufferOffsetsEXT, the bound VkPipeline must have been created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08119
    If a descriptor is dynamically used with a VkPipeline created with VK_PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08605
    If a descriptor is dynamically used with a VkShaderEXT created with a VkDescriptorSetLayout that was created with VK_DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08606
    If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08608
    If a pipeline is bound to the pipeline bind point used by this command, there must not have been any calls to dynamic state setting commands for any state not specified as dynamic in the VkPipeline object bound to the pipeline bind point used by this command, since that pipeline was bound

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08609
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used to sample from any VkImage with a VkImageView of the type VK_IMAGE_VIEW_TYPE_3D, VK_IMAGE_VIEW_TYPE_CUBE, VK_IMAGE_VIEW_TYPE_1D_ARRAY, VK_IMAGE_VIEW_TYPE_2D_ARRAY or VK_IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08610
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions with ImplicitLod, Dref or Proj in their name, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08611
    If the VkPipeline object bound to the pipeline bind point used by this command or any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a VkSampler object that uses unnormalized coordinates, that sampler must not be used with any of the SPIR-V OpImageSample* or OpImageSparseSample* instructions that includes a LOD bias or any offset values, in any shader stage

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08607
    If the shaderObject is enabled, either a valid pipeline must be bound to the pipeline bind point used by this command, or a valid combination of valid and VK_NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-uniformBuffers-06935
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for uniformBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08612
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a uniform buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-storageBuffers-06936
    If any stage of the VkPipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or VK_PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_2_EXT for storageBuffers, and the robustBufferAccess feature is not enabled, that stage must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-08613
    If the robustBufferAccess feature is not enabled, and any VkShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a storage buffer, it must not access values outside of the range of the buffer as specified in the descriptor set bound to the same pipeline bind point

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-02707
    If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-06550
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-ConstOffset-06551
    If a bound shader accesses a VkSampler or VkImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-viewType-07752
    If a VkImageView is accessed as a result of this command, then the image view’s viewType must match the Dim operand of the OpTypeImage as described in Instruction/Sampler/Image View Validation

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-format-07753
    If a VkImageView is accessed as a result of this command, then the numeric type of the image view’s format and the Sampled Type operand of the OpTypeImage must match

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWrite-08795
    If a VkImageView created with a format other than VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the image view’s format

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWrite-08796
    If a VkImageView created with the format VK_FORMAT_A8_UNORM_KHR is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have four components

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWrite-04469
    If a VkBufferView is accessed using OpImageWrite as a result of this command, then the Type of the Texel operand of that instruction must have at least as many components as the buffer view’s format

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-SampledType-04470
    If a VkImageView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-SampledType-04471
    If a VkImageView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-SampledType-04472
    If a VkBufferView with a VkFormat that has a 64-bit component width is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 64

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-SampledType-04473
    If a VkBufferView with a VkFormat that has a component width less than 64-bit is accessed as a result of this command, the SampledType of the OpTypeImage operand of that instruction must have a Width of 32

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-sparseImageInt64Atomics-04474
    If the sparseImageInt64Atomics feature is not enabled, VkImage objects created with the VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-sparseImageInt64Atomics-04475
    If the sparseImageInt64Atomics feature is not enabled, VkBuffer objects created with the VK_BUFFER_CREATE_SPARSE_RESIDENCY_BIT flag must not be accessed by atomic instructions through an OpTypeImage with a SampledType with a Width of 64 by this command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWeightedSampleQCOM-06971
    If OpImageWeightedSampleQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWeightedSampleQCOM-06972
    If OpImageWeightedSampleQCOM uses a VkImageView as a sample weight image as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBoxFilterQCOM-06973
    If OpImageBoxFilterQCOM is used to sample a VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchSSDQCOM-06974
    If OpImageBlockMatchSSDQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchSADQCOM-06975
    If OpImageBlockMatchSADQCOM is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchSADQCOM-06976
    If OpImageBlockMatchSADQCOM or OpImageBlockMatchSSDQCOM is used to read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWeightedSampleQCOM-06977
    If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageWeightedSampleQCOM-06978
    If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a VkSampler as a result of this command, then the sampler must not have been created with VK_SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchWindow-09215
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format features must contain VK_FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchWindow-09216
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an VkImageView as a result of this command, then the image view’s format must be a single-component format.

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-OpImageBlockMatchWindow-09217
    If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM read from a reference image as result of this command, then the specified reference coordinates must not fail integer texel coordinate validation

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-None-07288
    Any shader invocation executed by this command must terminate

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-09181
    commandBuffer must not be a protected command buffer

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-09182
    commandBuffer must be a primary command buffer

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-scratch-09183
    scratch must be the device address of an allocated memory range at least as large as the value of VkExecutionGraphPipelineScratchSizeAMDX::size returned by VkExecutionGraphPipelineScratchSizeAMDX for the currently bound execution graph pipeline

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-scratch-09184
    scratch must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or VK_BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-scratch-09185
    Device memory in the range [scratch,scratch
    VkExecutionGraphPipelineScratchSizeAMDX::size) must have been initialized with vkCmdInitializeGraphScratchMemoryAMDX using the currently bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-maxComputeWorkGroupCount-09186
    Execution of this command must not cause a node to be dispatched with a larger number of workgroups than that specified by either a MaxNumWorkgroupsAMDX decoration in the dispatched node or maxComputeWorkGroupCount

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-maxExecutionGraphShaderPayloadCount-09187
    Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-NodeMaxPayloadsAMDX-09188
    Execution of this command must not cause any shader that declares NodeMaxPayloadsAMDX to initialize more output payloads than specified by the max number of payloads for that decoration. This requirement applies to each NodeMaxPayloadsAMDX decoration separately

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09159
    countInfo must be a device pointer to a memory allocation containing a valid VkDispatchGraphCountInfoAMDX structure when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09160
    countInfo must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT flag

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09161
    countInfo must be a multiple of executionGraphDispatchAddressAlignment

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09162
    countInfo->infos must be a device pointer to a memory allocation at least as large as the product of count and stride when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09163
    countInfo->infos must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT flag

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09164
    countInfo->infos must be a multiple of executionGraphDispatchAddressAlignment

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-infos-09165
    Device memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid VkDispatchGraphInfoAMDX structures in the first 24 bytes when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09166
    For each VkDispatchGraphInfoAMDX structure in countInfo->infos, payloads must be a device pointer to a memory allocation at least as large as the product of payloadCount and payloadStride when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09167
    For each VkDispatchGraphInfoAMDX structure in countInfo->infos, payloads must be a device address within a VkBuffer created with the VK_BUFFER_USAGE_INDIRECT_BUFFER_BIT flag

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09168
    For each VkDispatchGraphInfoAMDX structure in countInfo->infos, payloads must be a multiple of executionGraphDispatchAddressAlignment

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09169
    For each VkDispatchGraphInfoAMDX structure in countInfo->infos, nodeIndex must be a valid node index in the currently bound execution graph pipeline, as returned by vkGetExecutionGraphPipelineNodeIndexAMDX when this command is executed on the device

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-countInfo-09170
    For each VkDispatchGraphInfoAMDX structure in countInfo->infos, device memory locations at indexes in the range [payloads, payloads + (payloadCount * payloadStride)), at a granularity of payloadStride must contain a payload matching the size of the input payload expected by the node in nodeIndex in the first bytes when this command is executed on the device

Valid Usage (Implicit)
  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-parameter
    commandBuffer must be a valid VkCommandBuffer handle

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-recording
    commandBuffer must be in the recording state

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-commandBuffer-cmdpool
    The VkCommandPool that commandBuffer was allocated from must support graphics, or compute operations

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-renderpass
    This command must only be called outside of a render pass instance

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-videocoding
    This command must only be called outside of a video coding scope

  • VUID-vkCmdDispatchGraphIndirectCountAMDX-bufferlevel
    commandBuffer must be a primary VkCommandBuffer

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

Command Properties
Command Buffer Levels Render Pass Scope Video Coding Scope Supported Queue Types Command Type

Primary

Outside

Outside

Graphics
Compute

Action

The VkDeviceOrHostAddressConstAMDX union is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef union VkDeviceOrHostAddressConstAMDX {
    VkDeviceAddress    deviceAddress;
    const void*        hostAddress;
} VkDeviceOrHostAddressConstAMDX;
  • deviceAddress is a buffer device address as returned by the vkGetBufferDeviceAddressKHR command.

  • hostAddress is a const host memory address.

The VkDispatchGraphCountInfoAMDX structure is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef struct VkDispatchGraphCountInfoAMDX {
    uint32_t                          count;
    VkDeviceOrHostAddressConstAMDX    infos;
    uint64_t                          stride;
} VkDispatchGraphCountInfoAMDX;

Whether infos is consumed as a device or host pointer is defined by the command this structure is used in.

The VkDispatchGraphInfoAMDX structure is defined as:

// Provided by VK_AMDX_shader_enqueue
typedef struct VkDispatchGraphInfoAMDX {
    uint32_t                          nodeIndex;
    uint32_t                          payloadCount;
    VkDeviceOrHostAddressConstAMDX    payloads;
    uint64_t                          payloadStride;
} VkDispatchGraphInfoAMDX;
  • nodeIndex is the index of a node in an execution graph to be dispatched.

  • payloadCount is the number of payloads to dispatch for the specified node.

  • payloads is a device or host address pointer to a flat array of payloads with size equal to the product of payloadCount and payloadStride

  • payloadStride is the byte stride between successive payloads in payloads

Whether payloads is consumed as a device or host pointer is defined by the command this structure is used in.

Valid Usage

Shader Enqueue

Compute shaders in an execution graph can use the OpInitializeNodePayloadsAMDX to initialize nodes for dispatch. Any node payload initialized in this way will be enqueued for dispatch once the shader is done writing to the payload. As compilers may be conservative when making this determination, shaders can further call OpFinalizeNodePayloadsAMDX to guarantee that the payload is no longer being written.

The Node Name operand of the PayloadNodeNameAMDX decoration on a payload identifies the shader name of the node to be enqueued, and the Shader Index operand of OpInitializeNodePayloadsAMDX identifies the shader index. A node identified in this way is dispatched as described in the following sections.

Compute Nodes

Compute shaders added as nodes to an execution graph are executed differently based on the presence or absence of the StaticNumWorkgroupsAMDX or CoalescingAMDX execution modes.

Dispatching a compute shader node that does not declare either the StaticNumWorkgroupsAMDX or CoalescingAMDX execution mode will execute a number of workgroups in each dimension specified by the first 12 bytes of the payload, interpreted as a VkDispatchIndirectCommand. The same payload will be broadcast to each workgroup in the same dispatch. Additional values in the payload are have no effect on execution.

Dispatching a compute shader node with the StaticNumWorkgroupsAMDX execution mode will execute workgroups in each dimension according to the x, y, and z size operands to the StaticNumWorkgroupsAMDX execution mode. The same payload will be broadcast to each workgroup in the same dispatch. Any values in the payload have no effect on execution.

Dispatching a compute shader node with the CoalescingAMDX execution mode will enqueue a single invocation for execution. Implementations may combine multiple such dispatches into the same workgroup, up to the size of the workgroup. The number of invocations coalesced into a given workgroup in this way can be queried via the CoalescedInputCountAMDX built-in. Any values in the payload have no effect on execution.