MLIR
18.0.0git
|
Builder for warp ids used to map scf.forall to reindexed threads. More...
#include "mlir/Dialect/GPU/TransformOps/Utils.h"
Public Member Functions | |
GpuThreadIdBuilder (MLIRContext *ctx, bool useLinearMapping=false) | |
![]() | |
GpuIdBuilder ()=default | |
GpuIdBuilder (MLIRContext *ctx, bool useLinearMapping, MappingIdBuilderFnType builder) | |
Additional Inherited Members | |
![]() | |
using | MappingIdBuilderFnType = std::function< DeviceMappingAttrInterface(MLIRContext *, mlir::gpu::MappingId)> |
![]() | |
SmallVector< DeviceMappingAttrInterface > | mappingAttributes |
The mapping attributes targeted by this generator. More... | |
GpuIdBuilderFnType | idBuilder |
The constructor that builds the concrete IR for mapping ids. More... | |
Builder for warp ids used to map scf.forall to reindexed threads.
If useLinearMapping
is false, the idBuilder
method returns 3D values used for indexing rewrites as well as 3D sizes for predicate generation. If useLinearMapping
is true, the idBuilder
method returns nD values used for indexing rewrites as well as 1D sizes for predicate generation.
mlir::transform::gpu::GpuThreadIdBuilder::GpuThreadIdBuilder | ( | MLIRContext * | ctx, |
bool | useLinearMapping = false |
||
) |
Definition at line 223 of file Utils.cpp.
References mlir::get().