|
| std::unique_ptr<::mlir::Pass > | mlir::createGpuAsyncRegionPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuDecomposeMemrefsPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuEliminateBarriers () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuKernelOutliningPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuKernelOutliningPass (GpuKernelOutliningPassOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuLaunchSinkIndexComputationsPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuMapParallelLoopsPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuMapParallelLoopsPass (GpuMapParallelLoopsPassOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuModuleToBinaryPass () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuModuleToBinaryPass (GpuModuleToBinaryPassOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuNVVMAttachTarget () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuNVVMAttachTarget (GpuNVVMAttachTargetOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuROCDLAttachTarget () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuROCDLAttachTarget (GpuROCDLAttachTargetOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuSPIRVAttachTarget () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuSPIRVAttachTarget (GpuSPIRVAttachTargetOptions options) |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuXeVMAttachTarget () |
| std::unique_ptr<::mlir::Pass > | mlir::createGpuXeVMAttachTarget (GpuXeVMAttachTargetOptions options) |
| void | mlir::populateGpuGlobalIdPatterns (RewritePatternSet &patterns) |
| | Collect a set of patterns to rewrite GlobalIdOp op within the GPU dialect.
|
| void | mlir::populateGpuSubgroupIdPatterns (RewritePatternSet &patterns) |
| | Collect a set of patterns to rewrite SubgroupIdOp op within the GPU dialect.
|
| void | mlir::populateGpuShufflePatterns (RewritePatternSet &patterns) |
| | Collect a set of patterns to rewrite shuffle ops within the GPU dialect.
|
| void | mlir::populateGpuAllReducePatterns (RewritePatternSet &patterns) |
| | Collect a set of patterns to rewrite all-reduce ops within the GPU dialect.
|
| void | mlir::populateGpuBreakDownSubgroupReducePatterns (RewritePatternSet &patterns, unsigned maxShuffleBitwidth=32, PatternBenefit benefit=1) |
| | Collect a set of patterns to break down subgroup_reduce ops into smaller ones supported by the target of size <= maxShuffleBitwidth, where size is the subgroup_reduce value bitwidth.
|
| void | mlir::populateGpuLowerSubgroupReduceToShufflePatterns (RewritePatternSet &patterns, unsigned subgroupSize, unsigned shuffleBitwidth=32, PatternBenefit benefit=1) |
| | Collect a set of patterns to lower gpu.subgroup_reduce into gpu.shuffle ops over shuffleBitwidth scalar types.
|
| void | mlir::populateGpuLowerClusteredSubgroupReduceToShufflePatterns (RewritePatternSet &patterns, unsigned subgroupSize, unsigned shuffleBitwidth=32, PatternBenefit benefit=1) |
| | Disjoint counterpart of populateGpuLowerSubgroupReduceToShufflePatterns that only matches gpu.subgroup_reduce ops with a cluster_size.
|
| void | mlir::populateGpuLowerSubgroupReduceToDPPPatterns (RewritePatternSet &patterns, unsigned subgroupSize, amdgpu::Chipset chipset, PatternBenefit benefit=1) |
| | Collect a set of patterns to lower gpu.subgroup_reduce into amdgpu.dpp ops over scalar types.
|
| void | mlir::populateGpuLowerClusteredSubgroupReduceToDPPPatterns (RewritePatternSet &patterns, unsigned subgroupSize, amdgpu::Chipset chipset, PatternBenefit benefit=1) |
| | Disjoint counterpart of populateGpuLowerSubgroupReduceToDPPPatterns that only matches gpu.subgroup_reduce ops with a cluster_size.
|
| void | mlir::populateGpuRewritePatterns (RewritePatternSet &patterns) |
| | Collect all patterns to rewrite ops within the GPU dialect.
|
| LogicalResult | mlir::gpu::transformGpuModulesToBinaries (Operation *op, OffloadingLLVMTranslationAttrInterface handler=nullptr, const gpu::TargetOptions &options={}) |
| | Searches for all GPU modules in op and transforms them into GPU binary operations.
|
| void | mlir::populateGpuDecomposeMemrefsPatterns (RewritePatternSet &patterns) |
| | Collect a set of patterns to decompose memrefs ops.
|
| void | mlir::populateGpuEliminateBarriersPatterns (RewritePatternSet &patterns) |
| | Erase barriers that do not enforce conflicting memory side effects.
|
| void | mlir::populateGpuPromoteShuffleToAMDGPUPatterns (RewritePatternSet &patterns, std::optional< amdgpu::Chipset > maybeChipset) |
| | Tries to promote gpu.shuffles to specialized AMDGPU intrinsics.
|
| void | mlir::registerGpuAsyncRegionPass () |
| void | mlir::registerGpuAsyncRegionPassPass () |
| void | mlir::registerGpuDecomposeMemrefsPass () |
| void | mlir::registerGpuDecomposeMemrefsPassPass () |
| void | mlir::registerGpuEliminateBarriers () |
| void | mlir::registerGpuEliminateBarriersPass () |
| void | mlir::registerGpuKernelOutliningPass () |
| void | mlir::registerGpuKernelOutliningPassPass () |
| void | mlir::registerGpuLaunchSinkIndexComputationsPass () |
| void | mlir::registerGpuLaunchSinkIndexComputationsPassPass () |
| void | mlir::registerGpuMapParallelLoopsPass () |
| void | mlir::registerGpuMapParallelLoopsPassPass () |
| void | mlir::registerGpuModuleToBinaryPass () |
| void | mlir::registerGpuModuleToBinaryPassPass () |
| void | mlir::registerGpuNVVMAttachTarget () |
| void | mlir::registerGpuNVVMAttachTargetPass () |
| void | mlir::registerGpuROCDLAttachTarget () |
| void | mlir::registerGpuROCDLAttachTargetPass () |
| void | mlir::registerGpuSPIRVAttachTarget () |
| void | mlir::registerGpuSPIRVAttachTargetPass () |
| void | mlir::registerGpuXeVMAttachTarget () |
| void | mlir::registerGpuXeVMAttachTargetPass () |
| void | mlir::registerGPUPasses () |
1.14.0