vkCreateTensorARM - Create a new tensor object

Valid Usage

• The tensors feature must be enabled

Valid Usage (Implicit)

• device must be a valid Device handle

• pCreateInfo must be a valid pointer to a valid TensorCreateInfoARM structure
• If pAllocator is not NULL, pAllocator must be a valid pointer to a valid AllocationCallbacks structure
• pTensor must be a valid pointer to a TensorARM handle
• The device must have been created with at least 1 queue

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_DEVICE_MEMORY
• ERROR_OUT_OF_HOST_MEMORY
• ERROR_UNKNOWN
• ERROR_VALIDATION_FAILED

See Also

VK_ARM_tensors, AllocationCallbacks, Device, TensorARM, TensorCreateInfoARM

A convenience wrapper to make a compatible pair of calls to createTensorARM and destroyTensorARM

To ensure that destroyTensorARM is always called: pass bracket (or the allocate function from your favourite resource management library) as the last argument. To just extract the pair pass (,) as the last argument.

vkDestroyTensorARM - Destroy a tensor object

Valid Usage

• All submitted commands that refer to tensor, either directly or via a TensorViewARM, must have completed execution

• If AllocationCallbacks were provided when tensor was created, a compatible set of callbacks must be provided here
• If no AllocationCallbacks were provided when tensor was created, pAllocator must be NULL

Valid Usage (Implicit)

• device must be a valid Device handle

• If tensor is not NULL_HANDLE, tensor must be a valid TensorARM handle
• If pAllocator is not NULL, pAllocator must be a valid pointer to a valid AllocationCallbacks structure
• If tensor is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization

• Host access to tensor must be externally synchronized

See Also

VK_ARM_tensors, AllocationCallbacks, Device, TensorARM

vkCreateTensorViewARM - Create an tensor view from an existing tensor

Description

Some of the tensor creation parameters are inherited by the view. In particular, other than format, the tensor view creation inherits all other parameters from the tensor.

The remaining parameters are contained in pCreateInfo.

Valid Usage (Implicit)

• device must be a valid Device handle

• pCreateInfo must be a valid pointer to a valid TensorViewCreateInfoARM structure
• If pAllocator is not NULL, pAllocator must be a valid pointer to a valid AllocationCallbacks structure
• pView must be a valid pointer to a TensorViewARM handle
• The device must have been created with at least 1 queue

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_DEVICE_MEMORY
• ERROR_OUT_OF_HOST_MEMORY
• ERROR_UNKNOWN
• ERROR_VALIDATION_FAILED

See Also

VK_ARM_tensors, AllocationCallbacks, Device, TensorViewARM, TensorViewCreateInfoARM

A convenience wrapper to make a compatible pair of calls to createTensorViewARM and destroyTensorViewARM

To ensure that destroyTensorViewARM is always called: pass bracket (or the allocate function from your favourite resource management library) as the last argument. To just extract the pair pass (,) as the last argument.

vkDestroyTensorViewARM - Destroy a tensor view object

Valid Usage

• All submitted commands that refer to tensorView must have completed execution

• If AllocationCallbacks were provided when tensorView was created, a compatible set of callbacks must be provided here
• If no AllocationCallbacks were provided when tensorView was created, pAllocator must be NULL

Valid Usage (Implicit)

• device must be a valid Device handle

• If tensorView is not NULL_HANDLE, tensorView must be a valid TensorViewARM handle
• If pAllocator is not NULL, pAllocator must be a valid pointer to a valid AllocationCallbacks structure
• If tensorView is a valid handle, it must have been created, allocated, or retrieved from device

Host Synchronization

• Host access to tensorView must be externally synchronized

See Also

VK_ARM_tensors, AllocationCallbacks, Device, TensorViewARM

vkGetTensorMemoryRequirementsARM - Returns the memory requirements for specified Vulkan object

Valid Usage (Implicit)

See Also

VK_ARM_tensors, Device, MemoryRequirements2, TensorMemoryRequirementsInfoARM

vkBindTensorMemoryARM - Bind device memory to tensor objects

Description

On some implementations, it may be more efficient to batch memory bindings into a single command.

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_DEVICE_MEMORY
• ERROR_OUT_OF_HOST_MEMORY
• ERROR_UNKNOWN
• ERROR_VALIDATION_FAILED

See Also

VK_ARM_tensors, BindTensorMemoryInfoARM, Device

vkGetDeviceTensorMemoryRequirementsARM - Returns the memory requirements for specified tensor creation infos

Valid Usage

• The tensors feature must be enabled

Valid Usage (Implicit)

• device must be a valid Device handle

• pInfo must be a valid pointer to a valid DeviceTensorMemoryRequirementsARM structure
• pMemoryRequirements must be a valid pointer to a MemoryRequirements2 structure

See Also

VK_ARM_tensors, Device, DeviceTensorMemoryRequirementsARM, MemoryRequirements2

vkCmdCopyTensorARM - Copy data between tensors

Valid Usage (Implicit)

• commandBuffer must be a valid CommandBuffer handle

• pCopyTensorInfo must be a valid pointer to a valid CopyTensorInfoARM structure
• commandBuffer must be in the recording state
• The CommandPool that commandBuffer was allocated from must support QUEUE_COMPUTE_BIT, QUEUE_GRAPHICS_BIT, or QUEUE_TRANSFER_BIT operations
• This command must only be called outside of a render pass instance
• This command must not be called between suspended render pass instances
• This command must only be called outside of a video coding scope

Host Synchronization

• Host access to commandBuffer must be externally synchronized

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

Command Properties

'

Conditional Rendering

vkCmdCopyTensorARM is not affected by conditional rendering

See Also

VK_ARM_tensors, CommandBuffer, CopyTensorInfoARM

vkGetTensorOpaqueCaptureDescriptorDataARM - Get tensor opaque capture descriptor data

Valid Usage

• The descriptorBufferCaptureReplay and descriptorBufferTensorDescriptors features must be enabled

• pData must point to a buffer that is at least PhysicalDeviceDescriptorBufferTensorPropertiesARM::tensorCaptureReplayDescriptorDataSize bytes in size
• If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

• device must be a valid Device handle

• pInfo must be a valid pointer to a valid TensorCaptureDescriptorDataInfoARM structure
• pData must be a pointer value

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_DEVICE_MEMORY
• ERROR_OUT_OF_HOST_MEMORY
• ERROR_UNKNOWN
• ERROR_VALIDATION_FAILED

See Also

VK_ARM_tensors, VK_EXT_descriptor_buffer, Device, TensorCaptureDescriptorDataInfoARM

vkGetTensorViewOpaqueCaptureDescriptorDataARM - Get tensor view opaque capture descriptor data

Valid Usage

• The descriptorBufferCaptureReplay and descriptorBufferTensorDescriptors features must be enabled

• pData must point to a buffer that is at least PhysicalDeviceDescriptorBufferTensorPropertiesARM::tensorViewCaptureReplayDescriptorDataSize bytes in size
• If device was created with multiple physical devices, then the bufferDeviceAddressMultiDevice feature must be enabled

Valid Usage (Implicit)

• device must be a valid Device handle

• pInfo must be a valid pointer to a valid TensorViewCaptureDescriptorDataInfoARM structure
• pData must be a pointer value

Return Codes

Success

• SUCCESS

Failure

• ERROR_OUT_OF_DEVICE_MEMORY
• ERROR_OUT_OF_HOST_MEMORY
• ERROR_UNKNOWN
• ERROR_VALIDATION_FAILED

See Also

VK_ARM_tensors, VK_EXT_descriptor_buffer, Device, TensorViewCaptureDescriptorDataInfoARM

vkGetPhysicalDeviceExternalTensorPropertiesARM - Function for querying external tensor handle capabilities.

Valid Usage (Implicit)

See Also

VK_ARM_tensors, ExternalTensorPropertiesARM, PhysicalDevice, PhysicalDeviceExternalTensorInfoARM

VkTensorDescriptionARM - Structure describing a tensor

Description

When describing a tensor created with TENSOR_TILING_OPTIMAL_ARM, pStrides must be equal to NULL. When describing a tensor created with TENSOR_TILING_LINEAR_ARM, pStrides is either an array of size dimensionCount or NULL.

The formats that must be supported for format are documented in https://registry.khronos.org/vulkan/specs/latest/html/vkspec.html#features-formats-mandatory-features-tensor.

Each element in the pStrides array describes the offset in bytes between increments of the given dimension. For example, pStrides[0] describes the offset between element [x0,x1,x2,x3] and element [x0+1,x1,x2,x3]. The pStrides array can be used to determine whether a tensor is packed or not. If pStrides[dimensionCount-1] is equal to the size of a tensor element and for each dimension n greater than 0 and less than dimensionCount, pStrides[n-1] is equal to pStrides[n] * pDimensions[n], then the tensor is a packed tensor. If the tensorNonPacked feature is not enabled, the tensor must be a packed tensor.

When a tensor is created with TENSOR_TILING_LINEAR_ARM and pStrides equal to NULL the tensor strides are calculated by the vulkan implementation such that the resulting tensor is a packed tensor.

Expressed as an addressing formula, the starting byte of an element in a 4-dimensional, for example, linear tensor has address:

// Assume (x0,x1,x2,x3) are in units of elements.

address(x0,x1,x2,x3) = x0*pStrides[0] + x1*pStrides[1] + x2*pStrides[2] + x3*pStrides[3]

Valid Usage

• dimensionCount must be less than or equal to PhysicalDeviceTensorPropertiesARM::maxTensorDimensionCount

• For each i where i ≤ dimensionCount-1, pDimensions[i] must be greater than 0
• For each i where i ≤ dimensionCount-1, pDimensions[i] must be less than or equal to ::maxPerDimensionTensorElements
• format must not be FORMAT_UNDEFINED and must be a one-component Format
• pStrides[dimensionCount-1] must equal the size in bytes of a tensor element
• For each i, pStrides[i] must be a multiple of the element size
• For each i, pStrides[i] must be greater than 0 and less than or equal to PhysicalDeviceTensorPropertiesARM::maxTensorStride
• pStrides[0] × pDimensions[0] must be less than or equal to ::maxTensorSize
• For each i greater than 0, pStrides[i-1] must be greater than or equal to pStrides[i] × pDimensions[i] so that no two elements of the tensor reference the same memory address
• If the tensorNonPacked feature is not enabled, then the members of TensorDescriptionARM must describe a packed tensor
• If tiling is TENSOR_TILING_OPTIMAL_ARM and usage is TENSOR_USAGE_IMAGE_ALIASING_BIT_ARM then the size of the tensor along its innermost dimension, i.e. pDimensions[dimensionCount - 1], must be less than or equal to 4
• If tiling is TENSOR_TILING_LINEAR_ARM then TENSOR_USAGE_IMAGE_ALIASING_BIT_ARM must not be set in usage

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_TENSOR_DESCRIPTION_ARM

• tiling must be a valid TensorTilingARM value
• format must be a valid Format value
• pDimensions must be a valid pointer to an array of dimensionCount int64_t values
• If pStrides is not NULL, pStrides must be a valid pointer to an array of dimensionCount int64_t values
• usage must be a valid combination of TensorUsageFlagBitsARM values
• usage must not be 0
• dimensionCount must be greater than 0

See Also

VK_ARM_tensors, Format, PhysicalDeviceExternalTensorInfoARM, StructureType, TensorCreateInfoARM, TensorTilingARM, TensorUsageFlagsARM

VkTensorViewCreateInfoARM - Structure specifying parameters of a newly created tensor view

Description

If tensor was created with the TENSOR_CREATE_MUTABLE_FORMAT_BIT_ARM flag, format can be different from the tensor’s format, but if they are not equal they must be compatible. Tensor format compatibility is defined in the Format Compatibility Classes section. Views of compatible formats will have the same mapping between element locations irrespective of the format, with only the interpretation of the bit pattern changing.

Values intended to be used with one view format may not be exactly preserved when written or read through a different format. For example, an integer value that happens to have the bit pattern of a floating-point denorm or NaN may be flushed or canonicalized when written or read through a view with a floating-point format. Similarly, a value written through a signed normalized format that has a bit pattern exactly equal to -2b may be changed to -2b + 1 as described in Conversion from Normalized Fixed-Point to Floating-Point.

Valid Usage

• If tensor was not created with TENSOR_CREATE_MUTABLE_FORMAT_BIT_ARM flag, format must be identical to the format used to create tensor

• If tensor was created with TENSOR_CREATE_MUTABLE_FORMAT_BIT_ARM flag, format must be compatible with the format used to create tensor, as defined in Format Compatibility Classes
• If flags includes TENSOR_VIEW_CREATE_DESCRIPTOR_BUFFER_CAPTURE_REPLAY_BIT_ARM, the descriptorBufferCaptureReplay feature must be enabled
• If the pNext chain includes a OpaqueCaptureDescriptorDataCreateInfoEXT structure, flags must contain TENSOR_VIEW_CREATE_DESCRIPTOR_BUFFER_CAPTURE_REPLAY_BIT_ARM

• The usage flags of tensor must have at least one of the following bits set:

  • TENSOR_USAGE_SHADER_BIT_ARM
  • TENSOR_USAGE_DATA_GRAPH_BIT_ARM
• The tensor view’s format features must contain the format feature flags required by the usage flags of tensor for format as indicated in the https://registry.khronos.org/vulkan/specs/latest/html/vkspec.html#format-feature-dependent-usage-flags section
• If tensor is non-sparse then it must be bound completely and contiguously to a single DeviceMemory object

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_TENSOR_VIEW_CREATE_INFO_ARM

• pNext must be NULL or a pointer to a valid instance of OpaqueCaptureDescriptorDataCreateInfoEXT
• The sType value of each structure in the pNext chain must be unique
• flags must be a valid combination of TensorViewCreateFlagBitsARM values
• tensor must be a valid TensorARM handle
• format must be a valid Format value

See Also

VK_ARM_tensors, VK_EXT_descriptor_heap, Format, ResourceDescriptorDataEXT, StructureType, TensorARM, TensorViewCreateFlagsARM, createTensorViewARM

VkTensorMemoryRequirementsInfoARM - Structure specifying memory requirements

Valid Usage (Implicit)

See Also

VK_ARM_tensors, StructureType, TensorARM, getTensorMemoryRequirementsARM

VkBindTensorMemoryInfoARM - Structure specifying how to bind a tensor to memory

Valid Usage

• tensor must not already be backed by a memory object

• memoryOffset must be less than the size of memory
• memory must have been allocated using one of the memory types allowed in the memoryTypeBits member of the MemoryRequirements structure returned from a call to getTensorMemoryRequirementsARM with tensor
• memoryOffset must be an integer multiple of the alignment member of the MemoryRequirements structure returned from a call to getTensorMemoryRequirementsARM with tensor
• The size member of the MemoryRequirements structure returned from a call to getTensorMemoryRequirementsARM with tensor must be less than or equal to the size of memory minus memoryOffset
• If tensor requires a dedicated allocation (as reported by getTensorMemoryRequirementsARM in MemoryDedicatedRequirements::requiresDedicatedAllocation for tensor), memory must have been created with MemoryDedicatedAllocateInfoTensorARM::tensor equal to tensor
• If the MemoryAllocateInfo provided when memory was allocated included a MemoryDedicatedAllocateInfoTensorARM structure in its pNext chain, and MemoryDedicatedAllocateInfoTensorARM::tensor was not NULL_HANDLE, then tensor must equal MemoryDedicatedAllocateInfoTensorARM::tensor, and memoryOffset must be zero
• If the value of ExportMemoryAllocateInfo::handleTypes used to allocate memory is not 0, it must include at least one of the handles set in ExternalMemoryTensorCreateInfoARM::handleTypes when tensor was created
• If memory was allocated by a memory import operation, that is not ImportAndroidHardwareBufferInfoANDROID with a non-NULL buffer value, the external handle type of the imported memory must also have been set in ExternalMemoryTensorCreateInfoARM::handleTypes when tensor was created
• If memory was allocated with the ImportAndroidHardwareBufferInfoANDROID memory import operation with a non-NULL buffer value, EXTERNAL_MEMORY_HANDLE_TYPE_ANDROID_HARDWARE_BUFFER_BIT_ANDROID must also have been set in ExternalMemoryTensorCreateInfoARM::handleTypes when tensor was created
• If tensor was created with the TENSOR_CREATE_PROTECTED_BIT_ARM bit set, the tensor must be bound to a memory object allocated with a memory type that reports MEMORY_PROPERTY_PROTECTED_BIT
• If tensor was created with the TENSOR_CREATE_PROTECTED_BIT_ARM bit not set, the tensor must not be bound to a memory object allocated with a memory type that reports MEMORY_PROPERTY_PROTECTED_BIT
• If tensor was created with the TENSOR_CREATE_DESCRIPTOR_BUFFER_CAPTURE_REPLAY_BIT_ARM bit set, memory must have been allocated with the MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT bit set
• If tensor was created with the TENSOR_CREATE_DESCRIPTOR_BUFFER_CAPTURE_REPLAY_BIT_ARM bit set, memory must have been allocated with the MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT bit set
• If tensor was created with the TENSOR_CREATE_DESCRIPTOR_HEAP_CAPTURE_REPLAY_BIT_ARM bit set, memory must have been allocated with the MEMORY_ALLOCATE_DEVICE_ADDRESS_BIT bit set
• If tensor was created with the TENSOR_CREATE_DESCRIPTOR_HEAP_CAPTURE_REPLAY_BIT_ARM bit set, memory must have been allocated with the MEMORY_ALLOCATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT bit set

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_BIND_TENSOR_MEMORY_INFO_ARM

• pNext must be NULL
• tensor must be a valid TensorARM handle
• memory must be a valid DeviceMemory handle
• Both of memory, and tensor must have been created, allocated, or retrieved from the same Device

Host Synchronization

• Host access to tensor must be externally synchronized

See Also

VK_ARM_tensors, DeviceMemory, DeviceSize, StructureType, TensorARM, bindTensorMemoryARM

VkWriteDescriptorSetTensorARM - Structure specifying descriptor tensor info

Valid Usage

• If the nullDescriptor feature is not enabled, each element of pTensorViews must not be NULL_HANDLE

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET_TENSOR_ARM

• pTensorViews must be a valid pointer to an array of tensorViewCount valid or NULL_HANDLE TensorViewARM handles
• tensorViewCount must be greater than 0

See Also

VK_ARM_tensors, StructureType, TensorViewARM

VkTensorFormatPropertiesARM - Structure specifying properties of a format used to describe tensor elements

Valid Usage (Implicit)

See Also

VK_ARM_tensors, FormatFeatureFlags2, StructureType

VkPhysicalDeviceTensorPropertiesARM - Structure describing the tensor properties of a physical device

Description

If the PhysicalDeviceTensorPropertiesARM 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_ARM_tensors, Bool32, ShaderStageFlags, StructureType

VkTensorMemoryBarrierARM - Structure specifying a tensor memory barrier

Description

The first synchronization scope and access scope described by this structure include only operations and memory accesses specified by srcStageMask and srcAccessMask.

The second synchronization scope and access scope described by this structure include only operations and memory accesses specified by dstStageMask and dstAccessMask.

Both access scopes are limited to only memory accesses to tensor.

If tensor was created with SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex is not equal to dstQueueFamilyIndex, this memory barrier defines a queue family transfer operation. When executed on a queue in the family identified by srcQueueFamilyIndex, this barrier defines a queue family release operation for the specified tensor, and the second synchronization and access scopes do not synchronize operations on that queue. When executed on a queue in the family identified by dstQueueFamilyIndex, this barrier defines a queue family acquire operation for the specified tensor, and the first synchronization and access scopes do not synchronize operations on that queue.

A queue family transfer operation is also defined if the values are not equal, and either is one of the special queue family values reserved for external memory ownership transfers, as described in https://registry.khronos.org/vulkan/specs/latest/html/vkspec.html#synchronization-queue-transfers. A queue family release operation is defined when dstQueueFamilyIndex is one of those values, and a queue family acquire operation is defined when srcQueueFamilyIndex is one of those values.

Valid Usage

• If tensor was created with a sharing mode of SHARING_MODE_CONCURRENT, srcQueueFamilyIndex and dstQueueFamilyIndex must both be QUEUE_FAMILY_IGNORED

• If tensor was created with a sharing mode of SHARING_MODE_EXCLUSIVE, srcQueueFamilyIndex and dstQueueFamilyIndex must both be either QUEUE_FAMILY_IGNORED, or a valid queue family (see https://registry.khronos.org/vulkan/specs/latest/html/vkspec.html#devsandqueues-queueprops)
• If tensor was created with a sharing mode of SHARING_MODE_EXCLUSIVE, and srcQueueFamilyIndex and dstQueueFamilyIndex are not QUEUE_FAMILY_IGNORED, at least one of them must be the same as the family of the queue that will execute this barrier
• If tensor is non-sparse then it must be bound completely and contiguously to a single DeviceMemory object

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_TENSOR_MEMORY_BARRIER_ARM

• srcStageMask must be a valid combination of PipelineStageFlagBits2 values
• srcAccessMask must be a valid combination of AccessFlagBits2 values
• dstStageMask must be a valid combination of PipelineStageFlagBits2 values
• dstAccessMask must be a valid combination of AccessFlagBits2 values
• tensor must be a valid TensorARM handle

See Also

VK_ARM_tensors, AccessFlags2, PipelineStageFlags2, StructureType, TensorARM, TensorDependencyInfoARM

VkTensorDependencyInfoARM - Structure specifying tensor dependency information for a synchronization command

Valid Usage (Implicit)

See Also

VK_ARM_tensors, StructureType, TensorMemoryBarrierARM

VkPhysicalDeviceTensorFeaturesARM - Structure describing tensor features that can be supported by an implementation

Members

The members of the PhysicalDeviceTensorFeaturesARM structure describe the following features:

Description

If the PhysicalDeviceTensorFeaturesARM 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. If the application wishes to use a Device with any features described by PhysicalDeviceTensorFeaturesARM, it must add an instance of the structure, with the desired feature members set to TRUE, to the pNext chain of DeviceCreateInfo when creating the Device.

Valid Usage (Implicit)

See Also

VK_ARM_tensors, Bool32, StructureType

VkDeviceTensorMemoryRequirementsARM - (None)

Valid Usage (Implicit)

See Also

VK_ARM_tensors, StructureType, TensorCreateInfoARM, getDeviceTensorMemoryRequirementsARM

VkCopyTensorInfoARM - Structure specifying an tensor copy operation

Description

Each region in pRegions describes a region to be copied from the source tensor to a corresponding region of the destination tensor. srcTensor and dstTensor can be the same tensor or alias the same memory.

The formats of srcTensor and dstTensor must be compatible. Formats are compatible if they share the same class, as shown in the Compatible Formats table.

cmdCopyTensorARM allows copying between size-compatible internal formats.

Valid Usage

• srcTensor and dstTensor must have been created with equal values for TensorDescriptionARM::dimensionCount

• For each of the elements of TensorDescriptionARM::pDimensions, srcTensor and dstTensor must be the same size
• regionCount must be equal to 1
• Each element of pRegions must be a TensorCopyARM structure whose pSrcOffset is NULL or has all its elements equal to 0
• Each element of pRegions must be a TensorCopyARM structure whose pDstOffset is NULL or has all its elements equal to 0
• Each element of pRegions must be a TensorCopyARM structure whose pExtent is NULL or equal to the TensorDescriptionARM::pDimensions array specified when srcTensor and dstTensor were created
• Each element of pRegions must be a TensorCopyARM structure whose dimensionCount, if it is not equal to 0, is equal to the largest of the TensorDescriptionARM::dimensionCount of srcTensor or dstTensor
• The format features of srcTensor must contain FORMAT_FEATURE_2_TRANSFER_SRC_BIT
• srcTensor must have been created with the TENSOR_USAGE_TRANSFER_SRC_BIT_ARM usage flag set
• The format features of dstTensor must contain FORMAT_FEATURE_2_TRANSFER_DST_BIT
• dstTensor must have been created with the TENSOR_USAGE_TRANSFER_DST_BIT_ARM usage flag set
• If srcTensor is non-sparse then it must be bound completely and contiguously to a single DeviceMemory object
• If dstTensor is non-sparse then it must be bound completely and contiguously to a single DeviceMemory object

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_COPY_TENSOR_INFO_ARM

• pNext must be NULL
• srcTensor must be a valid TensorARM handle
• dstTensor must be a valid TensorARM handle
• pRegions must be a valid pointer to an array of regionCount valid TensorCopyARM structures
• regionCount must be greater than 0
• Both of dstTensor, and srcTensor must have been created, allocated, or retrieved from the same Device

See Also

VK_ARM_tensors, StructureType, TensorARM, TensorCopyARM, cmdCopyTensorARM

VkTensorCopyARM - Structure specifying an tensor copy region

Valid Usage

• dimensionCount must be greater than 0 if pSrcOffset, pDstOffset, or pExtent is not NULL

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_TENSOR_COPY_ARM

• pNext must be NULL
• If dimensionCount is not 0, and pSrcOffset is not NULL, pSrcOffset must be a valid pointer to an array of dimensionCount uint64_t values
• If dimensionCount is not 0, and pDstOffset is not NULL, pDstOffset must be a valid pointer to an array of dimensionCount uint64_t values
• If dimensionCount is not 0, and pExtent is not NULL, pExtent must be a valid pointer to an array of dimensionCount uint64_t values

See Also

VK_ARM_tensors, CopyTensorInfoARM, StructureType

VkMemoryDedicatedAllocateInfoTensorARM - Specify a dedicated memory allocation tensor resource

Valid Usage

• MemoryAllocateInfo::allocationSize must equal the MemoryRequirements::size of the tensor

• If MemoryAllocateInfo defines a memory import operation with handle type EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT, the memory being imported must also be a dedicated tensor allocation and tensor must be identical to the tensor associated with the imported memory

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_MEMORY_DEDICATED_ALLOCATE_INFO_TENSOR_ARM

• tensor must be a valid TensorARM handle

See Also

VK_ARM_tensors, StructureType, TensorARM

VkPhysicalDeviceDescriptorBufferTensorPropertiesARM - Structure describing descriptor buffer tensor properties supported by an implementation

Description

If the PhysicalDeviceDescriptorBufferTensorPropertiesARM 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_ARM_tensors, VK_EXT_descriptor_buffer, StructureType

VkPhysicalDeviceDescriptorBufferTensorFeaturesARM - Structure describing the descriptor buffer tensor features that can be supported by an implementation

Members

This structure describes the following features:

Description

If the PhysicalDeviceDescriptorBufferTensorFeaturesARM 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. If the application wishes to use a Device with any features described by PhysicalDeviceDescriptorBufferTensorFeaturesARM, it must add an instance of the structure, with the desired feature members set to TRUE, to the pNext chain of DeviceCreateInfo when creating the Device.

Valid Usage (Implicit)

See Also

VK_ARM_tensors, VK_EXT_descriptor_buffer, Bool32, StructureType

VkTensorCaptureDescriptorDataInfoARM - Structure specifying a tensor for descriptor capture

Valid Usage

• If tensor is not NULL_HANDLE then tensor must have been created with TENSOR_CREATE_DESCRIPTOR_BUFFER_CAPTURE_REPLAY_BIT_ARM set in TensorCreateInfoARM::flags

Valid Usage (Implicit)

• sType must be STRUCTURE_TYPE_TENSOR_CAPTURE_DESCRIPTOR_DATA_INFO_ARM

• pNext must be NULL
• tensor must be a valid TensorARM handle