vkGetExecutionGraphPipelineScratchSizeAMDX - Query scratch space required to dispatch an execution graph

Description

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

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_HOST_MEMORY

See Also

VK_AMDX_shader_enqueue, Device, ExecutionGraphPipelineScratchSizeAMDX, Pipeline

vkGetExecutionGraphPipelineNodeIndexAMDX - Query internal id of a node in an execution graph

Description

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

Valid Usage

• pNodeInfo->pName must not be NULL

• pNodeInfo->index must not be SHADER_INDEX_UNUSED_AMDX
• There must be a node in executionGraph with a shader name and index equal to pNodeInfo->pName and pNodeInfo->index

Valid Usage (Implicit)

• device must be a valid Device handle

• executionGraph must be a valid Pipeline handle
• pNodeInfo must be a valid pointer to a valid PipelineShaderStageNodeCreateInfoAMDX structure
• pNodeIndex must be a valid pointer to a uint32_t value
• executionGraph must have been created, allocated, or retrieved from device

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_HOST_MEMORY

See Also

VK_AMDX_shader_enqueue, Device, Pipeline, PipelineShaderStageNodeCreateInfoAMDX

vkCreateExecutionGraphPipelinesAMDX - Creates a new execution graph pipeline object

Description

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

Valid Usage

• The shaderEnqueue feature must be enabled

• If the flags member of any element of pCreateInfos contains the PIPELINE_CREATE_DERIVATIVE_BIT flag, and the basePipelineIndex member of that same element is not -1, basePipelineIndex must be less than the index into pCreateInfos that corresponds to that element
• If the flags member of any element of pCreateInfos contains the PIPELINE_CREATE_DERIVATIVE_BIT flag, the base pipeline must have been created with the PIPELINE_CREATE_ALLOW_DERIVATIVES_BIT flag set
• If pipelineCache was created with PIPELINE_CACHE_CREATE_EXTERNALLY_SYNCHRONIZED_BIT, host access to pipelineCache must be externally synchronized
• If PipelineBinaryInfoKHR::binaryCount is not 0 for any element of pCreateInfos, pipelineCache must be NULL_HANDLE
• If a PipelineCreateFlags2CreateInfoKHR structure with the PIPELINE_CREATE_2_CAPTURE_DATA_BIT_KHR flag set is included in the pNext chain of any element of pCreateInfos, pipelineCache must be NULL_HANDLE
• If PipelineBinaryInfoKHR::binaryCount is not 0 for any element of pCreateInfos, PIPELINE_CREATION_FEEDBACK_APPLICATION_PIPELINE_CACHE_HIT_BIT must not be set in the flags of that element
• If PipelineBinaryInfoKHR::binaryCount is not 0 for any element of pCreateInfos, PIPELINE_CREATION_FEEDBACK_BASE_PIPELINE_ACCELERATION_BIT must not be set in the flags of that element
• If PipelineBinaryInfoKHR::binaryCount is not 0 for any element of pCreateInfos, PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_EXT must not be set in the flags of that element

Valid Usage (Implicit)

• device must be a valid Device handle

• If pipelineCache is not NULL_HANDLE, pipelineCache must be a valid PipelineCache handle
• pCreateInfos must be a valid pointer to an array of createInfoCount valid ExecutionGraphPipelineCreateInfoAMDX structures
• If pAllocator is not NULL, pAllocator must be a valid pointer to a valid AllocationCallbacks structure
• pPipelines must be a valid pointer to an array of createInfoCount Pipeline handles
• createInfoCount must be greater than 0
• If pipelineCache is a valid handle, it must have been created, allocated, or retrieved from device

Return Codes

Success

• SUCCESS
• PIPELINE_COMPILE_REQUIRED_EXT

Failure

• ERROR_OUT_OF_HOST_MEMORY
• ERROR_OUT_OF_DEVICE_MEMORY

See Also

VK_AMDX_shader_enqueue, AllocationCallbacks, Device, ExecutionGraphPipelineCreateInfoAMDX, Pipeline, PipelineCache

vkCmdInitializeGraphScratchMemoryAMDX - Initialize scratch memory for an execution graph

Description

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

Execution of this command may modify any memory locations in the range [scratch,scratch + scratchSize). Accesses to this memory range are performed in the PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT and ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

If any portion of scratch is modified by any command other than cmdDispatchGraphAMDX, cmdDispatchGraphIndirectAMDX, cmdDispatchGraphIndirectCountAMDX, or cmdInitializeGraphScratchMemoryAMDX with the same execution graph, it must be reinitialized for the execution graph again before dispatching against it.

Valid Usage

• scratch must be the device address of an allocated memory range at least as large as scratchSize

• scratchSize must be greater than or equal to ExecutionGraphPipelineScratchSizeAMDX::minSize returned by getExecutionGraphPipelineScratchSizeAMDX for the bound execution graph pipeline
• scratch must be a multiple of 64

Valid Usage (Implicit)

• commandBuffer must be a valid CommandBuffer handle

• executionGraph must be a valid Pipeline handle
• commandBuffer must be in the recording state
• The CommandPool that commandBuffer was allocated from must support graphics, or compute operations
• This command must only be called outside of a video coding scope
• commandBuffer must be a primary CommandBuffer
• Both of commandBuffer, and executionGraph must have been created, allocated, or retrieved from the same Device

Host Synchronization

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

Command Properties

'

See Also

VK_AMDX_shader_enqueue, CommandBuffer, DeviceAddress, DeviceSize, Pipeline

vkCmdDispatchGraphAMDX - Dispatch an execution graph

Description

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. There are no rasterization order guarantees between separately dispatched graphics nodes, though individual primitives within a single dispatch do adhere to rasterization order. Draw calls executed before or after the execution graph also execute relative to each graphics node with respect to rasterization order.

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 + scratchSize). Accesses to this memory range are performed in the PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT and ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

This command captures command buffer state for mesh nodes similarly to draw commands.

Valid Usage

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s levelCount and layerCount must be 1
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s viewType must be IMAGE_VIEW_TYPE_1D or IMAGE_VIEW_TYPE_2D
• If a ImageView is sampled with depth comparison, the image view’s format features must contain FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT
• If a ImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT
• If a 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 FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT
• If a ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT
• If the VK_EXT_filter_cubic extension is not enabled and any ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, it must not have a ImageViewType of IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, or IMAGE_VIEW_TYPE_CUBE_ARRAY
• Any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have a ImageViewType and format that supports cubic filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by getPhysicalDeviceImageFormatProperties2
• Any ImageView being sampled with FILTER_CUBIC_EXT with a reduction mode of either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a ImageViewType and format that supports cubic filtering together with minmax filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by getPhysicalDeviceImageFormatProperties2
• If the cubicRangeClamp feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must not have a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM
• Any ImageView being sampled with a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with FILTER_CUBIC_EXT
• If the selectableCubicWeights feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have SamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM
• Any Image created with a ImageCreateInfo::flags containing IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a SamplerAddressMode of SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE
• For any ImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• For any ImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• For any BufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• Any BufferView 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 FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• 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 PipelineLayout that is compatible for set n, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• For each array of resources that is used by a bound shader, the indices used to access members of the array must be less than the descriptor count for the identified binding in the descriptor sets used by this command
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout and PushConstantRange arrays used to create the current ShaderEXT , as described in ???
• Descriptors in each bound descriptor set, specified via cmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was not created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdBindDescriptorSets, the bound Pipeline must have been created without PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdSetDescriptorBufferOffsetsEXT, the bound Pipeline must have been created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If a descriptor is dynamically used with a Pipeline created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If a descriptor is dynamically used with a ShaderEXT created with a DescriptorSetLayout that was created with DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command
• 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 specified statically in the Pipeline object bound to the pipeline bind point used by this command, since that pipeline was bound
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler object that uses unnormalized coordinates, that sampler must not be used to sample from any Image with a ImageView of the type IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, IMAGE_VIEW_TYPE_1D_ARRAY, IMAGE_VIEW_TYPE_2D_ARRAY or IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the shaderObject feature 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 NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands
• If a ImageView 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 ???
• If a ImageView 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
• If a ImageView created with a format other than 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
• If a ImageView created with the format 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
• If a BufferView 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
• If a ImageView with a Format 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
• If a ImageView with a Format 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
• If a BufferView with a Format 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
• If a BufferView with a Format 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
• If the sparseImageInt64Atomics feature is not enabled, Image objects created with the 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
• If the sparseImageInt64Atomics feature is not enabled, Buffer objects created with the 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
• If OpImageWeightedSampleQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM
• If OpImageWeightedSampleQCOM uses a ImageView as a sample weight image as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM
• If OpImageBoxFilterQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM
• If OpImageBlockMatchSSDQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If OpImageBlockMatchSADQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• 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
• If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must not have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format must be a single-component format
• 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
• Any shader invocation executed by this command must terminate
• If a descriptor with type equal to any of DESCRIPTOR_TYPE_SAMPLE_WEIGHT_IMAGE_QCOM, DESCRIPTOR_TYPE_BLOCK_MATCH_IMAGE_QCOM, DESCRIPTOR_TYPE_SAMPLED_IMAGE, DESCRIPTOR_TYPE_STORAGE_IMAGE, or DESCRIPTOR_TYPE_INPUT_ATTACHMENT is accessed as a result of this command, the image subresource identified by that descriptor must be in the image layout identified when the descriptor was written
• commandBuffer must not be a protected command buffer
• commandBuffer must be a primary command buffer
• scratch must be the device address of an allocated memory range at least as large as scratchSize
• scratchSize must be greater than or equal to ExecutionGraphPipelineScratchSizeAMDX::minSize returned by getExecutionGraphPipelineScratchSizeAMDX for the bound execution graph pipeline
• scratch must be a device address within a Buffer created with the BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag
• The device memory range [scratch,scratch scratchSize] must have been initialized with cmdInitializeGraphScratchMemoryAMDX using the bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command
• 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
• Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads
• 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
• If the bound execution graph pipeline includes draw nodes, this command must be called within a render pass instance that is compatible with the graphics pipeline used to create each of those nodes
• pCountInfo->infos must be a host pointer to a memory allocation at least as large as the product of count and stride
• Host memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid DispatchGraphInfoAMDX structures in the first 24 bytes
• For each DispatchGraphInfoAMDX 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
• For each DispatchGraphInfoAMDX structure in pCountInfo->infos, nodeIndex must be a valid node index in the bound execution graph pipeline, as returned by getExecutionGraphPipelineNodeIndexAMDX
• For each DispatchGraphInfoAMDX 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)

• commandBuffer must be a valid CommandBuffer handle

• pCountInfo must be a valid pointer to a valid DispatchGraphCountInfoAMDX structure
• commandBuffer must be in the recording state
• The CommandPool that commandBuffer was allocated from must support graphics, or compute operations
• This command must only be called outside of a video coding scope
• commandBuffer must be a primary CommandBuffer

Host Synchronization

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

Command Properties

'

See Also

VK_AMDX_shader_enqueue, CommandBuffer, DeviceAddress, DeviceSize, DispatchGraphCountInfoAMDX

vkCmdDispatchGraphIndirectAMDX - Dispatch an execution graph with node and payload parameters read on the device

Description

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. There are no rasterization order guarantees between separately dispatched graphics nodes, though individual primitives within a single dispatch do adhere to rasterization order. Draw calls executed before or after the execution graph also execute relative to each graphics node with respect to rasterization order.

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 PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT access flag.

Execution of this command may modify any memory locations in the range [scratch,scratch + scratchSize). Accesses to this memory range are performed in the PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT and ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

This command captures command buffer state for mesh nodes similarly to draw commands.

Valid Usage

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s levelCount and layerCount must be 1
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s viewType must be IMAGE_VIEW_TYPE_1D or IMAGE_VIEW_TYPE_2D
• If a ImageView is sampled with depth comparison, the image view’s format features must contain FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT
• If a ImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT
• If a 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 FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT
• If a ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT
• If the VK_EXT_filter_cubic extension is not enabled and any ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, it must not have a ImageViewType of IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, or IMAGE_VIEW_TYPE_CUBE_ARRAY
• Any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have a ImageViewType and format that supports cubic filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by getPhysicalDeviceImageFormatProperties2
• Any ImageView being sampled with FILTER_CUBIC_EXT with a reduction mode of either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a ImageViewType and format that supports cubic filtering together with minmax filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by getPhysicalDeviceImageFormatProperties2
• If the cubicRangeClamp feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must not have a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM
• Any ImageView being sampled with a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with FILTER_CUBIC_EXT
• If the selectableCubicWeights feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have SamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM
• Any Image created with a ImageCreateInfo::flags containing IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a SamplerAddressMode of SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE
• For any ImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• For any ImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• For any BufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• Any BufferView 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 FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• 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 PipelineLayout that is compatible for set n, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• For each array of resources that is used by a bound shader, the indices used to access members of the array must be less than the descriptor count for the identified binding in the descriptor sets used by this command
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout and PushConstantRange arrays used to create the current ShaderEXT , as described in ???
• Descriptors in each bound descriptor set, specified via cmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was not created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdBindDescriptorSets, the bound Pipeline must have been created without PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdSetDescriptorBufferOffsetsEXT, the bound Pipeline must have been created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If a descriptor is dynamically used with a Pipeline created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If a descriptor is dynamically used with a ShaderEXT created with a DescriptorSetLayout that was created with DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command
• 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 specified statically in the Pipeline object bound to the pipeline bind point used by this command, since that pipeline was bound
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler object that uses unnormalized coordinates, that sampler must not be used to sample from any Image with a ImageView of the type IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, IMAGE_VIEW_TYPE_1D_ARRAY, IMAGE_VIEW_TYPE_2D_ARRAY or IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the shaderObject feature 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 NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands
• If a ImageView 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 ???
• If a ImageView 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
• If a ImageView created with a format other than 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
• If a ImageView created with the format 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
• If a BufferView 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
• If a ImageView with a Format 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
• If a ImageView with a Format 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
• If a BufferView with a Format 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
• If a BufferView with a Format 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
• If the sparseImageInt64Atomics feature is not enabled, Image objects created with the 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
• If the sparseImageInt64Atomics feature is not enabled, Buffer objects created with the 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
• If OpImageWeightedSampleQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM
• If OpImageWeightedSampleQCOM uses a ImageView as a sample weight image as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM
• If OpImageBoxFilterQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM
• If OpImageBlockMatchSSDQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If OpImageBlockMatchSADQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• 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
• If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must not have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format must be a single-component format
• 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
• Any shader invocation executed by this command must terminate
• If a descriptor with type equal to any of DESCRIPTOR_TYPE_SAMPLE_WEIGHT_IMAGE_QCOM, DESCRIPTOR_TYPE_BLOCK_MATCH_IMAGE_QCOM, DESCRIPTOR_TYPE_SAMPLED_IMAGE, DESCRIPTOR_TYPE_STORAGE_IMAGE, or DESCRIPTOR_TYPE_INPUT_ATTACHMENT is accessed as a result of this command, the image subresource identified by that descriptor must be in the image layout identified when the descriptor was written
• commandBuffer must not be a protected command buffer
• commandBuffer must be a primary command buffer
• scratch must be the device address of an allocated memory range at least as large as scratchSize
• scratchSize must be greater than or equal to ExecutionGraphPipelineScratchSizeAMDX::minSize returned by getExecutionGraphPipelineScratchSizeAMDX for the bound execution graph pipeline
• scratch must be a device address within a Buffer created with the BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag
• The device memory range [scratch,scratch scratchSize] must have been initialized with cmdInitializeGraphScratchMemoryAMDX using the bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command
• 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
• Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads
• 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
• If the bound execution graph pipeline includes draw nodes, this command must be called within a render pass instance that is compatible with the graphics pipeline used to create each of those nodes
• 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
• pCountInfo->infos must be a device address within a Buffer created with the BUFFER_USAGE_INDIRECT_BUFFER_BIT flag
• pCountInfo->infos must be a multiple of executionGraphDispatchAddressAlignment
• Device memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid DispatchGraphInfoAMDX structures in the first 24 bytes when this command is executed on the device
• For each DispatchGraphInfoAMDX 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
• For each DispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a device address within a Buffer created with the BUFFER_USAGE_INDIRECT_BUFFER_BIT flag
• For each DispatchGraphInfoAMDX structure in pCountInfo->infos, payloads must be a multiple of executionGraphDispatchAddressAlignment
• For each DispatchGraphInfoAMDX structure in pCountInfo->infos, nodeIndex must be a valid node index in the bound execution graph pipeline, as returned by getExecutionGraphPipelineNodeIndexAMDX when this command is executed on the device
• For each DispatchGraphInfoAMDX 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)

• commandBuffer must be a valid CommandBuffer handle

• pCountInfo must be a valid pointer to a valid DispatchGraphCountInfoAMDX structure
• commandBuffer must be in the recording state
• The CommandPool that commandBuffer was allocated from must support graphics, or compute operations
• This command must only be called outside of a video coding scope
• commandBuffer must be a primary CommandBuffer

Host Synchronization

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

Command Properties

'

See Also

VK_AMDX_shader_enqueue, CommandBuffer, DeviceAddress, DeviceSize, DispatchGraphCountInfoAMDX

vkCmdDispatchGraphIndirectCountAMDX - Dispatch an execution graph with all parameters read on the device

Description

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 PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT access flag.

Execution of this command may modify any memory locations in the range [scratch,scratch + scratchSize). Accesses to this memory range are performed in the PIPELINE_STAGE_2_COMPUTE_SHADER_BIT pipeline stage with the ACCESS_2_SHADER_STORAGE_READ_BIT and ACCESS_2_SHADER_STORAGE_WRITE_BIT access flags.

Valid Usage

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT

• If a Sampler created with magFilter or minFilter equal to FILTER_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR, reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE, and compareEnable equal to FALSE is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_LINEAR_BIT
• If a Sampler created with mipmapMode equal to SAMPLER_MIPMAP_MODE_LINEAR and reductionMode equal to either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_MINMAX_BIT
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s levelCount and layerCount must be 1
• If a Sampler created with unnormalizedCoordinates equal to TRUE is used to sample a ImageView as a result of this command, then the image view’s viewType must be IMAGE_VIEW_TYPE_1D or IMAGE_VIEW_TYPE_2D
• If a ImageView is sampled with depth comparison, the image view’s format features must contain FORMAT_FEATURE_2_SAMPLED_IMAGE_DEPTH_COMPARISON_BIT
• If a ImageView is accessed using atomic operations as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_STORAGE_IMAGE_ATOMIC_BIT
• If a 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 FORMAT_FEATURE_STORAGE_TEXEL_BUFFER_ATOMIC_BIT
• If a ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_SAMPLED_IMAGE_FILTER_CUBIC_BIT_EXT
• If the VK_EXT_filter_cubic extension is not enabled and any ImageView is sampled with FILTER_CUBIC_EXT as a result of this command, it must not have a ImageViewType of IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, or IMAGE_VIEW_TYPE_CUBE_ARRAY
• Any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have a ImageViewType and format that supports cubic filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubic returned by getPhysicalDeviceImageFormatProperties2
• Any ImageView being sampled with FILTER_CUBIC_EXT with a reduction mode of either SAMPLER_REDUCTION_MODE_MIN or SAMPLER_REDUCTION_MODE_MAX as a result of this command must have a ImageViewType and format that supports cubic filtering together with minmax filtering, as specified by FilterCubicImageViewImageFormatPropertiesEXT::filterCubicMinmax returned by getPhysicalDeviceImageFormatProperties2
• If the cubicRangeClamp feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must not have a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM
• Any ImageView being sampled with a SamplerReductionModeCreateInfo::reductionMode equal to SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_RANGECLAMP_QCOM as a result of this command must sample with FILTER_CUBIC_EXT
• If the selectableCubicWeights feature is not enabled, then any ImageView being sampled with FILTER_CUBIC_EXT as a result of this command must have SamplerCubicWeightsCreateInfoQCOM::cubicWeights equal to CUBIC_FILTER_WEIGHTS_CATMULL_ROM_QCOM
• Any Image created with a ImageCreateInfo::flags containing IMAGE_CREATE_CORNER_SAMPLED_BIT_NV sampled as a result of this command must only be sampled using a SamplerAddressMode of SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE
• For any ImageView being written as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• For any ImageView being read as a storage image where the image format field of the OpTypeImage is Unknown, the view’s format features must contain FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• For any BufferView being written as a storage texel buffer where the image format field of the OpTypeImage is Unknown, the view’s buffer features must contain FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT
• Any BufferView 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 FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT
• 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 PipelineLayout that is compatible for set n, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout array used to create the current ShaderEXT , as described in ???
• For each array of resources that is used by a bound shader, the indices used to access members of the array must be less than the descriptor count for the identified binding in the descriptor sets used by this command
• 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 PipelineLayout that is compatible for push constants, with the PipelineLayout used to create the current Pipeline or the DescriptorSetLayout and PushConstantRange arrays used to create the current ShaderEXT , as described in ???
• Descriptors in each bound descriptor set, specified via cmdBindDescriptorSets, must be valid as described by descriptor validity if they are statically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was not created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdBindDescriptorSets, the bound Pipeline must have been created without PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by the Pipeline bound to the pipeline bind point used by this command and the bound Pipeline was created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• Descriptors in bound descriptor buffers, specified via cmdSetDescriptorBufferOffsetsEXT, must be valid if they are dynamically used by any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command
• If the descriptors used by the Pipeline bound to the pipeline bind point were specified via cmdSetDescriptorBufferOffsetsEXT, the bound Pipeline must have been created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If a descriptor is dynamically used with a Pipeline created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If a descriptor is dynamically used with a ShaderEXT created with a DescriptorSetLayout that was created with DESCRIPTOR_SET_LAYOUT_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, the descriptor memory must be resident
• If the shaderObject feature is not enabled, a valid pipeline must be bound to the pipeline bind point used by this command
• 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 specified statically in the Pipeline object bound to the pipeline bind point used by this command, since that pipeline was bound
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler object that uses unnormalized coordinates, that sampler must not be used to sample from any Image with a ImageView of the type IMAGE_VIEW_TYPE_3D, IMAGE_VIEW_TYPE_CUBE, IMAGE_VIEW_TYPE_1D_ARRAY, IMAGE_VIEW_TYPE_2D_ARRAY or IMAGE_VIEW_TYPE_CUBE_ARRAY, in any shader stage
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the Pipeline object bound to the pipeline bind point used by this command or any ShaderEXT bound to a stage corresponding to the pipeline bind point used by this command accesses a Sampler 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
• If the shaderObject feature 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 NULL_HANDLE shader objects must be bound to every supported shader stage corresponding to the pipeline bind point used by this command
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a uniform buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If any stage of the Pipeline object bound to the pipeline bind point used by this command accesses a storage buffer, and that stage was created without enabling either PIPELINE_ROBUSTNESS_BUFFER_BEHAVIOR_ROBUST_BUFFER_ACCESS_EXT or 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
• If the robustBufferAccess feature is not enabled, and any ShaderEXT 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
• If commandBuffer is an unprotected command buffer and protectedNoFault is not supported, any resource accessed by bound shaders must not be a protected resource
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must only be used with OpImageSample* or OpImageSparseSample* instructions
• If a bound shader accesses a Sampler or ImageView object that enables sampler Y′CBCR conversion, that object must not use the ConstOffset and Offset operands
• If a ImageView 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 ???
• If a ImageView 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
• If a ImageView created with a format other than 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
• If a ImageView created with the format 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
• If a BufferView 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
• If a ImageView with a Format 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
• If a ImageView with a Format 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
• If a BufferView with a Format 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
• If a BufferView with a Format 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
• If the sparseImageInt64Atomics feature is not enabled, Image objects created with the 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
• If the sparseImageInt64Atomics feature is not enabled, Buffer objects created with the 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
• If OpImageWeightedSampleQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_SAMPLED_IMAGE_BIT_QCOM
• If OpImageWeightedSampleQCOM uses a ImageView as a sample weight image as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_WEIGHT_IMAGE_BIT_QCOM
• If OpImageBoxFilterQCOM is used to sample a ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BOX_FILTER_SAMPLED_BIT_QCOM
• If OpImageBlockMatchSSDQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If OpImageBlockMatchSADQCOM is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• 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
• If OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If any command other than OpImageWeightedSampleQCOM, OpImageBoxFilterQCOM, OpImageBlockMatchWindowSSDQCOM, OpImageBlockMatchWindowSADQCOM, OpImageBlockMatchGatherSSDQCOM, OpImageBlockMatchGatherSADQCOM, OpImageBlockMatchSSDQCOM, or OpImageBlockMatchSADQCOM uses a Sampler as a result of this command, then the sampler must not have been created with SAMPLER_CREATE_IMAGE_PROCESSING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format features must contain FORMAT_FEATURE_2_BLOCK_MATCHING_BIT_QCOM
• If a OpImageBlockMatchWindow*QCOM or OpImageBlockMatchGather*QCOM instruction is used to read from an ImageView as a result of this command, then the image view’s format must be a single-component format
• 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
• Any shader invocation executed by this command must terminate
• If a descriptor with type equal to any of DESCRIPTOR_TYPE_SAMPLE_WEIGHT_IMAGE_QCOM, DESCRIPTOR_TYPE_BLOCK_MATCH_IMAGE_QCOM, DESCRIPTOR_TYPE_SAMPLED_IMAGE, DESCRIPTOR_TYPE_STORAGE_IMAGE, or DESCRIPTOR_TYPE_INPUT_ATTACHMENT is accessed as a result of this command, the image subresource identified by that descriptor must be in the image layout identified when the descriptor was written
• commandBuffer must not be a protected command buffer
• commandBuffer must be a primary command buffer
• scratch must be the device address of an allocated memory range at least as large as scratchSize
• scratchSize must be greater than or equal to ExecutionGraphPipelineScratchSizeAMDX::minSize returned by getExecutionGraphPipelineScratchSizeAMDX for the bound execution graph pipeline
• scratch must be a device address within a Buffer created with the BUFFER_USAGE_EXECUTION_GRAPH_SCRATCH_BIT_AMDX or BUFFER_USAGE_2_EXECUTION_GRAPH_SCRATCH_BIT_AMDX flag
• The device memory range [scratch,scratch scratchSize] must have been initialized with cmdInitializeGraphScratchMemoryAMDX using the bound execution graph pipeline, and not modified after that by anything other than another execution graph dispatch command
• 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
• Execution of this command must not cause any shader to initialize more than maxExecutionGraphShaderPayloadCount output payloads
• 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
• If the bound execution graph pipeline includes draw nodes, this command must be called within a render pass instance that is compatible with the graphics pipeline used to create each of those nodes
• countInfo must be a device pointer to a memory allocation containing a valid DispatchGraphCountInfoAMDX structure when this command is executed on the device
• countInfo must be a device address within a Buffer created with the BUFFER_USAGE_INDIRECT_BUFFER_BIT flag
• countInfo must be a multiple of executionGraphDispatchAddressAlignment
• 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
• countInfo->infos must be a device address within a Buffer created with the BUFFER_USAGE_INDIRECT_BUFFER_BIT flag
• countInfo->infos must be a multiple of executionGraphDispatchAddressAlignment
• Device memory locations at indexes in the range [infos, infos + (count*stride)), at a granularity of stride must contain valid DispatchGraphInfoAMDX structures in the first 24 bytes when this command is executed on the device
• For each DispatchGraphInfoAMDX 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
• For each DispatchGraphInfoAMDX structure in countInfo->infos, payloads must be a device address within a Buffer created with the BUFFER_USAGE_INDIRECT_BUFFER_BIT flag
• For each DispatchGraphInfoAMDX structure in countInfo->infos, payloads must be a multiple of executionGraphDispatchAddressAlignment
• For each DispatchGraphInfoAMDX structure in countInfo->infos, nodeIndex must be a valid node index in the bound execution graph pipeline, as returned by getExecutionGraphPipelineNodeIndexAMDX when this command is executed on the device
• For each DispatchGraphInfoAMDX 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)

• commandBuffer must be a valid CommandBuffer handle

• commandBuffer must be in the recording state
• The CommandPool that commandBuffer was allocated from must support graphics, or compute operations
• This command must only be called outside of a video coding scope
• commandBuffer must be a primary CommandBuffer

Host Synchronization

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

Command Properties

'

See Also

VK_AMDX_shader_enqueue, CommandBuffer, DeviceAddress, DeviceSize

VkPhysicalDeviceShaderEnqueuePropertiesAMDX - Structure describing shader enqueue limits of an implementation

Members

The members of the PhysicalDeviceShaderEnqueuePropertiesAMDX structure describe the following limits:

Description

If the PhysicalDeviceShaderEnqueuePropertiesAMDX structure is included in the pNext chain of the PhysicalDeviceProperties2 structure passed to getPhysicalDeviceProperties2, it is filled in with each corresponding implementation-dependent property.

Valid Usage (Implicit)

See Also

VK_AMDX_shader_enqueue, StructureType

VkPhysicalDeviceShaderEnqueueFeaturesAMDX - Structure describing whether shader enqueue within execution graphs are supported by the implementation

Members

This structure describes the following feature:

Description

If the PhysicalDeviceShaderEnqueueFeaturesAMDX structure is included in the pNext chain of the PhysicalDeviceFeatures2 structure passed to getPhysicalDeviceFeatures2, it is filled in to indicate whether each corresponding feature is supported. PhysicalDeviceShaderEnqueueFeaturesAMDX can also be used in the pNext chain of DeviceCreateInfo to selectively enable these features.

Valid Usage (Implicit)

See Also

VK_AMDX_shader_enqueue, Bool32, StructureType

VkExecutionGraphPipelineCreateInfoAMDX - Structure specifying parameters of a newly created execution graph pipeline

Description

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 PipelineShaderStageNodeCreateInfoAMDX structure in its pNext chain. For any graphics pipeline libraries, only the name and index of the vertex or mesh shader stage is linked directly to the graph as a node - other shader stages in the pipeline will be executed after those shader stages as normal. Task shaders cannot be included in a graphics pipeline used for a draw node.

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

Valid Usage

• If the pNext chain does not include a PipelineCreateFlags2CreateInfoKHR structure, flags must be a valid combination of PipelineCreateFlagBits values

• If flags contains the PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineIndex is -1, basePipelineHandle must be a valid execution graph Pipeline handle
• If flags contains the PIPELINE_CREATE_DERIVATIVE_BIT flag, and basePipelineHandle is NULL_HANDLE, basePipelineIndex must be a valid index into the calling command’s pCreateInfos parameter
• If flags contains the PIPELINE_CREATE_DERIVATIVE_BIT flag, basePipelineIndex must be -1 or basePipelineHandle must be NULL_HANDLE
• If a push constant block is declared in a shader, a push constant range in layout must match the shader stage
• If a push constant block is declared in a shader, the block must be contained inside the push constant range in layout that matches the stage
• If a resource variables is declared in a shader, a descriptor slot in layout must match the shader stage
• If a resource variables is declared in a shader, and the descriptor type is not DESCRIPTOR_TYPE_MUTABLE_EXT, a descriptor slot in layout must match the descriptor type
• If a resource variables is declared in a shader as an array, a descriptor slot in layout must match the descriptor count
• flags must not include PIPELINE_CREATE_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_SHADER_GROUP_HANDLE_CAPTURE_REPLAY_BIT_KHR
• flags must not include PIPELINE_CREATE_RAY_TRACING_ALLOW_MOTION_BIT_NV
• If the ::deviceGeneratedComputePipelines feature is not enabled, flags must not include PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV
• If flags includes PIPELINE_CREATE_INDIRECT_BINDABLE_BIT_NV, then the pNext chain must include a pointer to a valid instance of ComputePipelineIndirectBufferInfoNV specifying the address where the pipeline’s metadata will be saved
• If flags includes PIPELINE_CREATE_2_INDIRECT_BINDABLE_BIT_EXT, then the ::deviceGeneratedCommands feature must be enabled
• If the pipelineCreationCacheControl feature is not enabled, flags must not include PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT or PIPELINE_CREATE_EARLY_RETURN_ON_FAILURE_BIT
• The stage member of any element of pStages must be SHADER_STAGE_COMPUTE_BIT
• 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
• layout must be consistent with the layout of the shaders specified in pStages
• 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
• The number of resources in layout accessible to each shader stage that is used by the pipeline must be less than or equal to PhysicalDeviceLimits::maxPerStageResources
• If pLibraryInfo is not NULL, each element of pLibraryInfo->pLibraries must be either a compute pipeline, an execution graph pipeline, or a graphics pipeline
• If pLibraryInfo is not NULL, each element of pLibraryInfo->pLibraries that is a compute pipeline or a graphics pipeline must have been created with PIPELINE_CREATE_2_EXECUTION_GRAPH_BIT_AMDX set
• If the shaderMeshEnqueue feature is not enabled, and pLibraryInfo->pLibraries is not NULL, pLibraryInfo->pLibraries must not contain any graphics pipelines
• Any element of pLibraryInfo->pLibraries identifying a graphics pipeline must have been created with all possible state subsets
• There must be no two nodes in the pipeline that share both the same shader name and index, as specified by PipelineShaderStageNodeCreateInfoAMDX
• There must be no two nodes in the pipeline that share the same shader name and have input payload declarations with different sizes
• There must be no two nodes in the pipeline that share the same name but have different execution models
• There must be no two nodes in the pipeline that share the same name where one includes CoalescedInputCountAMDX and the other does not
• There must be no two nodes in the pipeline that share the same name where one includes StaticNumWorkgroupsAMDX and the other does not
• 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
• If flags does not include PIPELINE_CREATE_LIBRARY_BIT_KHR, and an output payload declared in any shader in the pipeline does not have a PayloadNodeSparseArrayAMDX decoration, there must be a node in the graph corresponding to every index from 0 to its PayloadNodeArraySizeAMDX decoration

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_EXECUTION_GRAPH_PIPELINE_CREATE_INFO_AMDX

• 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 PipelineCompilerControlCreateInfoAMD or PipelineCreationFeedbackCreateInfo
• The sType value of each struct in the pNext chain must be unique
• If stageCount is not 0, and pStages is not NULL, pStages must be a valid pointer to an array of stageCount valid PipelineShaderStageCreateInfo structures
• If pLibraryInfo is not NULL, pLibraryInfo must be a valid pointer to a valid PipelineLibraryCreateInfoKHR structure
• layout must be a valid PipelineLayout handle
• Both of basePipelineHandle, and layout that are valid handles of non-ignored parameters must have been created, allocated, or retrieved from the same Device

See Also

VK_AMDX_shader_enqueue, Pipeline, PipelineCreateFlags, PipelineLayout, PipelineLibraryCreateInfoKHR, PipelineShaderStageCreateInfo, StructureType, createExecutionGraphPipelinesAMDX

VkPipelineShaderStageNodeCreateInfoAMDX - Structure specifying the shader name and index with an execution graph

Description

When included in the pNext chain of a PipelineShaderStageCreateInfo 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.

Shaders with the same name must be of the same type - e.g. a compute and graphics shader, or even two compute shaders where one is coalescing and the other is not, cannot share the same name.

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_NODE_CREATE_INFO_AMDX

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

See Also

VK_AMDX_shader_enqueue, StructureType, getExecutionGraphPipelineNodeIndexAMDX

VkExecutionGraphPipelineScratchSizeAMDX - Structure describing the scratch space required to dispatch an execution graph

Description

Applications can use any amount of scratch memory greater than minSize for dispatching a graph, however only the values equal to minSize + an integer multiple of sizeGranularity will be used. Greater values may result in higher performance, up to maxSize which indicates the most memory that an implementation can use effectively.

Valid Usage (Implicit)

See Also

VK_AMDX_shader_enqueue, DeviceSize, StructureType, getExecutionGraphPipelineScratchSizeAMDX

VkDispatchGraphInfoAMDX - Structure specifying node parameters for execution graph dispatch

Description

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

Valid Usage

See Also

VK_AMDX_shader_enqueue, DeviceOrHostAddressConstAMDX, DispatchGraphCountInfoAMDX

VkDispatchGraphCountInfoAMDX - Structure specifying count parameters for execution graph dispatch

Description

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

See Also

VK_AMDX_shader_enqueue, DeviceOrHostAddressConstAMDX, cmdDispatchGraphAMDX, cmdDispatchGraphIndirectAMDX, cmdDispatchGraphIndirectCountAMDX

VkPipelineLibraryCreateInfoKHR - Structure specifying pipeline libraries to use when creating a pipeline

Valid Usage

• Each element of pLibraries must have been created with PIPELINE_CREATE_LIBRARY_BIT_KHR

• If any library in pLibraries was created with a shader stage with PipelineShaderStageModuleIdentifierCreateInfoEXT and identifierSize not equal to 0, the pipeline must be created with the PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT flag set
• If any element of pLibraries was created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT, all elements must have been created with PIPELINE_CREATE_DESCRIPTOR_BUFFER_BIT_EXT
• If pipeline is being created with PIPELINE_CREATE_NO_PROTECTED_ACCESS_BIT_EXT, every element of pLibraries must have been created with PIPELINE_CREATE_NO_PROTECTED_ACCESS_BIT_EXT
• If pipeline is being created without PIPELINE_CREATE_NO_PROTECTED_ACCESS_BIT_EXT, every element of pLibraries must have been created without PIPELINE_CREATE_NO_PROTECTED_ACCESS_BIT_EXT
• If pipeline is being created with PIPELINE_CREATE_PROTECTED_ACCESS_ONLY_BIT_EXT, every element of pLibraries must have been created with PIPELINE_CREATE_PROTECTED_ACCESS_ONLY_BIT_EXT
• If pipeline is being created without PIPELINE_CREATE_PROTECTED_ACCESS_ONLY_BIT_EXT, every element of pLibraries must have been created without PIPELINE_CREATE_PROTECTED_ACCESS_ONLY_BIT_EXT

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_PIPELINE_LIBRARY_CREATE_INFO_KHR

• If libraryCount is not 0, pLibraries must be a valid pointer to an array of libraryCount valid Pipeline handles

See Also

VK_KHR_pipeline_library, ExecutionGraphPipelineCreateInfoAMDX, Pipeline, RayTracingPipelineCreateInfoKHR, StructureType

VkBufferUsageFlagBits2KHR - Bitmask controlling how a pipeline is created

See Also

VK_KHR_maintenance5, BufferUsageFlags2KHR

VkPipelineCreateFlagBits2KHR - Bitmask controlling how a pipeline is created

Description

• PIPELINE_CREATE_2_DISABLE_OPTIMIZATION_BIT_KHR specifies that the created pipeline will not be optimized. Using this flag may reduce the time taken to create the pipeline.

• PIPELINE_CREATE_2_ALLOW_DERIVATIVES_BIT_KHR specifies that the pipeline to be created is allowed to be the parent of a pipeline that will be created in a subsequent pipeline creation call.
• PIPELINE_CREATE_2_DERIVATIVE_BIT_KHR specifies that the pipeline to be created will be a child of a previously created parent pipeline.
• PIPELINE_CREATE_2_VIEW_INDEX_FROM_DEVICE_INDEX_BIT_KHR specifies that any shader input variables decorated as ViewIndex will be assigned values as if they were decorated as DeviceIndex.
• PIPELINE_CREATE_2_DISPATCH_BASE_BIT_KHR specifies that a compute pipeline can be used with cmdDispatchBase with a non-zero base workgroup.
• PIPELINE_CREATE_2_DEFER_COMPILE_BIT_NV specifies that a pipeline is created with all shaders in the deferred state. Before using the pipeline the application must call compileDeferredNV exactly once on each shader in the pipeline before using the pipeline.
• PIPELINE_CREATE_2_CAPTURE_STATISTICS_BIT_KHR specifies that the shader compiler should capture statistics for the pipeline executables produced by the compile process which can later be retrieved by calling getPipelineExecutableStatisticsKHR. Enabling this flag must not affect the final compiled pipeline but may disable pipeline caching or otherwise affect pipeline creation time.
• PIPELINE_CREATE_2_CAPTURE_INTERNAL_REPRESENTATIONS_BIT_KHR specifies that the shader compiler should capture the internal representations of pipeline executables produced by the compile process which can later be retrieved by calling getPipelineExecutableInternalRepresentationsKHR. Enabling this flag must not affect the final compiled pipeline but may disable pipeline caching or otherwise affect pipeline creation time. When capturing IR from pipelines created with pipeline libraries, there is no guarantee that IR from libraries can be retrieved from the linked pipeline. Applications should retrieve IR from each library, and any linked pipelines, separately.
• PIPELINE_CREATE_2_LIBRARY_BIT_KHR specifies that the pipeline cannot be used directly, and instead defines a pipeline library that can be combined with other pipelines using the PipelineLibraryCreateInfoKHR structure. This is available in ray tracing and graphics pipelines.
• PIPELINE_CREATE_2_RAY_TRACING_NO_NULL_ANY_HIT_SHADERS_BIT_KHR specifies that an any-hit shader will always be present when an any-hit shader would be executed. A NULL any-hit shader is an any-hit shader which is effectively SHADER_UNUSED_KHR, such as from a shader group consisting entirely of zeros.
• PIPELINE_CREATE_2_RAY_TRACING_NO_NULL_CLOSEST_HIT_SHADERS_BIT_KHR specifies that a closest hit shader will always be present when a closest hit shader would be executed. A NULL closest hit shader is a closest hit shader which is effectively SHADER_UNUSED_KHR, such as from a shader group consisting entirely of zeros.
• PIPELINE_CREATE_2_RAY_TRACING_NO_NULL_MISS_SHADERS_BIT_KHR specifies that a miss shader will always be present when a miss shader would be executed. A NULL miss shader is a miss shader which is effectively SHADER_UNUSED_KHR, such as from a shader group consisting entirely of zeros.
• PIPELINE_CREATE_2_RAY_TRACING_NO_NULL_INTERSECTION_SHADERS_BIT_KHR specifies that an intersection shader will always be present when an intersection shader would be executed. A NULL intersection shader is an intersection shader which is effectively SHADER_UNUSED_KHR, such as from a shader group consisting entirely of zeros.
• PIPELINE_CREATE_2_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR specifies that triangle primitives will be skipped during traversal using pipeline trace ray instructions.
• PIPELINE_CREATE_2_RAY_TRACING_SKIP_AABBS_BIT_KHR specifies that AABB primitives will be skipped during traversal using pipeline trace ray instructions.
• PIPELINE_CREATE_2_RAY_TRACING_SHADER_GROUP_HANDLE_CAPTURE_REPLAY_BIT_KHR specifies that the shader group handles can be saved and reused on a subsequent run (e.g. for trace capture and replay).
• PIPELINE_CREATE_2_INDIRECT_BINDABLE_BIT_NV specifies that the pipeline can be used in combination with https://registry.khronos.org/vulkan/specs/1.3-extensions/html/vkspec.html#device-generated-commands.
• PIPELINE_CREATE_2_INDIRECT_BINDABLE_BIT_EXT specifies that the pipeline can be used in a IndirectExecutionSetEXT.
• PIPELINE_CREATE_2_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT_KHR specifies that pipeline creation will fail if a compile is required for creation of a valid Pipeline object; PIPELINE_COMPILE_REQUIRED will be returned by pipeline creation, and the Pipeline will be NULL_HANDLE.
• When creating multiple pipelines, PIPELINE_CREATE_2_EARLY_RETURN_ON_FAILURE_BIT_KHR specifies that control will be returned to the application if any individual pipeline returns a result which is not SUCCESS rather than continuing to create additional pipelines.
• PIPELINE_CREATE_2_RAY_TRACING_ALLOW_MOTION_BIT_NV specifies that the pipeline is allowed to use OpTraceRayMotionNV.
• PIPELINE_CREATE_2_RENDERING_FRAGMENT_SHADING_RATE_ATTACHMENT_BIT_KHR specifies that the pipeline will be used with a fragment shading rate attachment.
• PIPELINE_CREATE_2_RENDERING_FRAGMENT_DENSITY_MAP_ATTACHMENT_BIT_EXT specifies that the pipeline will be used with a fragment density map attachment.
• PIPELINE_CREATE_2_LINK_TIME_OPTIMIZATION_BIT_EXT specifies that pipeline libraries being linked into this library should have link time optimizations applied. If this bit is omitted, implementations should instead perform linking as rapidly as possible.
• PIPELINE_CREATE_2_RETAIN_LINK_TIME_OPTIMIZATION_INFO_BIT_EXT specifies that pipeline libraries should retain any information necessary to later perform an optimal link with PIPELINE_CREATE_2_LINK_TIME_OPTIMIZATION_BIT_EXT.
• PIPELINE_CREATE_2_DESCRIPTOR_BUFFER_BIT_EXT specifies that a pipeline will be used with descriptor buffers, rather than descriptor sets.
• PIPELINE_CREATE_2_COLOR_ATTACHMENT_FEEDBACK_LOOP_BIT_EXT specifies that the pipeline may be used with an attachment feedback loop including color attachments.
• PIPELINE_CREATE_2_DEPTH_STENCIL_ATTACHMENT_FEEDBACK_LOOP_BIT_EXT specifies that the pipeline may be used with an attachment feedback loop including depth-stencil attachments.
• PIPELINE_CREATE_2_RAY_TRACING_OPACITY_MICROMAP_BIT_EXT specifies that the ray tracing pipeline can be used with acceleration structures which reference an opacity micromap array.
• PIPELINE_CREATE_2_RAY_TRACING_DISPLACEMENT_MICROMAP_BIT_NV specifies that the ray tracing pipeline can be used with acceleration structures which reference a displacement micromap array.
• PIPELINE_CREATE_2_NO_PROTECTED_ACCESS_BIT_EXT specifies that the pipeline must not be bound to a protected command buffer.
• PIPELINE_CREATE_2_PROTECTED_ACCESS_ONLY_BIT_EXT specifies that the pipeline must not be bound to an unprotected command buffer.
• PIPELINE_CREATE_2_CAPTURE_DATA_BIT_KHR specifies that PipelineBinaryKHR objects can be created from the pipeline. If PIPELINE_CREATE_2_CAPTURE_DATA_BIT_KHR is used, implementations should not store pipeline data to an internal cache, if such a cache exists as stated by pipelineBinaryInternalCache. If pipelineBinaryPrefersInternalCache is TRUE, applications should not use PIPELINE_CREATE_2_CAPTURE_DATA_BIT_KHR.
• PIPELINE_CREATE_2_ENABLE_LEGACY_DITHERING_BIT_EXT specifies that the pipeline will be used in a render pass that is begun with RENDERING_ENABLE_LEGACY_DITHERING_BIT_EXT.
• PIPELINE_CREATE_2_EXECUTION_GRAPH_BIT_AMDX specifies that the pipeline will be used in an execution graph