LLVM: llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata Struct Reference (original) (raw)
In-memory representation of kernel code properties metadata. More...
#include "[llvm/Support/AMDGPUMetadata.h](AMDGPUMetadata%5F8h%5Fsource.html)"
| Public Attributes | |
|---|---|
| uint64_t | mKernargSegmentSize = 0 |
| Size in bytes of the kernarg segment memory. | |
| uint32_t | mGroupSegmentFixedSize = 0 |
| Size in bytes of the group segment memory required by a workgroup. | |
| uint32_t | mPrivateSegmentFixedSize = 0 |
| Size in bytes of the private segment memory required by a workitem. | |
| uint32_t | mKernargSegmentAlign = 0 |
| Maximum byte alignment of variables used by the kernel in the kernarg memory segment. | |
| uint32_t | mWavefrontSize = 0 |
| Wavefront size. Required. | |
| uint16_t | mNumSGPRs = 0 |
| Total number of SGPRs used by a wavefront. Optional. | |
| uint16_t | mNumVGPRs = 0 |
| Total number of VGPRs used by a workitem. Optional. | |
| uint32_t | mMaxFlatWorkGroupSize = 0 |
| Maximum flat work-group size supported by the kernel. Optional. | |
| bool | mIsDynamicCallStack = false |
| True if the generated machine code is using a dynamically sized call stack. | |
| bool | mIsXNACKEnabled = false |
| True if the generated machine code is capable of supporting XNACK. | |
| uint16_t | mNumSpilledSGPRs = 0 |
| Number of SGPRs spilled by a wavefront. Optional. | |
| uint16_t | mNumSpilledVGPRs = 0 |
| Number of VGPRs spilled by a workitem. Optional. |
In-memory representation of kernel code properties metadata.
Definition at line 276 of file AMDGPUMetadata.h.
| llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::Metadata ( ) | default |
|---|
Default constructor.
◆ empty()
| bool llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::empty ( ) const | inline |
|---|
◆ notEmpty()
| bool llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::notEmpty ( ) const | inline |
|---|
Returns
True if kernel code properties metadata is not empty, false otherwise.
Definition at line 320 of file AMDGPUMetadata.h.
Referenced by empty().
◆ mGroupSegmentFixedSize
uint32_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mGroupSegmentFixedSize = 0
◆ mIsDynamicCallStack
bool llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mIsDynamicCallStack = false
◆ mIsXNACKEnabled
bool llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mIsXNACKEnabled = false
◆ mKernargSegmentAlign
uint32_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mKernargSegmentAlign = 0
◆ mKernargSegmentSize
uint64_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mKernargSegmentSize = 0
◆ mMaxFlatWorkGroupSize
uint32_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mMaxFlatWorkGroupSize = 0
◆ mNumSGPRs
uint16_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mNumSGPRs = 0
◆ mNumSpilledSGPRs
uint16_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mNumSpilledSGPRs = 0
◆ mNumSpilledVGPRs
uint16_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mNumSpilledVGPRs = 0
◆ mNumVGPRs
uint16_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mNumVGPRs = 0
◆ mPrivateSegmentFixedSize
uint32_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mPrivateSegmentFixedSize = 0
◆ mWavefrontSize
uint32_t llvm::AMDGPU::HSAMD::Kernel::CodeProps::Metadata::mWavefrontSize = 0
The documentation for this struct was generated from the following file:
- include/llvm/Support/AMDGPUMetadata.h