MLIR  17.0.0git
Public Member Functions | List of all members
mlir::RewritePatternSet Class Reference

#include "mlir/IR/PatternMatch.h"

Public Member Functions

 RewritePatternSet (MLIRContext *context)
 
 RewritePatternSet (MLIRContext *context, std::unique_ptr< RewritePattern > pattern)
 Construct a RewritePatternSet populated with the given pattern. More...
 
 RewritePatternSet (PDLPatternModule &&pattern)
 
MLIRContext * getContext () const
 
NativePatternListT & getNativePatterns ()
 Return the native patterns held in this list. More...
 
PDLPatternModule & getPDLPatterns ()
 Return the PDL patterns held in this list. More...
 
void clear ()
 Clear out all of the held patterns in this list. More...
 
template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet & add (ConstructorArg &&arg, ConstructorArgs &&...args)
 Add an instance of each of the pattern types 'Ts' to the pattern list with the given arguments. More...
 
template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet & addWithLabel (ArrayRef< StringRef > debugLabels, ConstructorArg &&arg, ConstructorArgs &&...args)
 An overload of the above add method that allows for attaching a set of debug labels to the attached patterns. More...
 
template<typename... Ts>
RewritePatternSet & add ()
 Add an instance of each of the pattern types 'Ts'. More...
 
RewritePatternSet & add (std::unique_ptr< RewritePattern > pattern)
 Add the given native pattern to the pattern list. More...
 
RewritePatternSet & add (PDLPatternModule &&pattern)
 Add the given PDL pattern to the pattern list. More...
 
template<typename OpType >
RewritePatternSet & add (LogicalResult(*implFn)(OpType, PatternRewriter &rewriter), PatternBenefit benefit=1, ArrayRef< StringRef > generatedNames={})
 
template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet & insert (ConstructorArg &&arg, ConstructorArgs &&...args)
 Add an instance of each of the pattern types 'Ts' to the pattern list with the given arguments. More...
 
template<typename... Ts>
RewritePatternSet & insert ()
 Add an instance of each of the pattern types 'Ts'. More...
 
RewritePatternSet & insert (std::unique_ptr< RewritePattern > pattern)
 Add the given native pattern to the pattern list. More...
 
RewritePatternSet & insert (PDLPatternModule &&pattern)
 Add the given PDL pattern to the pattern list. More...
 
template<typename OpType >
RewritePatternSet & insert (LogicalResult(*implFn)(OpType, PatternRewriter &rewriter))
 

Detailed Description

Definition at line 1563 of file PatternMatch.h.

Constructor & Destructor Documentation

◆ RewritePatternSet() [1/3]

mlir::RewritePatternSet::RewritePatternSet ( MLIRContext *  context)
inline

Definition at line 1567 of file PatternMatch.h.

◆ RewritePatternSet() [2/3]

mlir::RewritePatternSet::RewritePatternSet ( MLIRContext *  context,
std::unique_ptr< RewritePattern >  pattern 
)
inline

Construct a RewritePatternSet populated with the given pattern.

Definition at line 1570 of file PatternMatch.h.

◆ RewritePatternSet() [3/3]

mlir::RewritePatternSet::RewritePatternSet ( PDLPatternModule &&  pattern)
inline

Definition at line 1575 of file PatternMatch.h.

Member Function Documentation

◆ add() [1/5]

template<typename... Ts>
RewritePatternSet& mlir::RewritePatternSet::add ( )
inline

Add an instance of each of the pattern types 'Ts'.

Return a reference to this for chaining insertions.

Definition at line 1631 of file PatternMatch.h.

Referenced by insert().

◆ add() [2/5]

template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet& mlir::RewritePatternSet::add ( ConstructorArg &&  arg,
ConstructorArgs &&...  args 
)
inline

Add an instance of each of the pattern types 'Ts' to the pattern list with the given arguments.

Return a reference to this for chaining insertions. Note: ConstructorArg is necessary here to separate the two variadic lists.

Definition at line 1603 of file PatternMatch.h.

Referenced by mlir::populateAffineToStdConversionPatterns(), mlir::populateAffineToVectorConversionPatterns(), mlir::populateAMDGPUToROCDLConversionPatterns(), mlir::populateAMXLegalizeForLLVMExportPatterns(), mlir::populateAnyFunctionOpInterfaceTypeConversionPattern(), mlir::arith::populateArithExpandOpsPatterns(), mlir::arith::populateArithToLLVMConversionPatterns(), mlir::arith::populateArithToSPIRVPatterns(), mlir::arith::populateArithWideIntEmulationPatterns(), mlir::populateArmSVELegalizeForLLVMExportPatterns(), mlir::cf::populateAssertToLLVMConversionPattern(), mlir::populateAsyncFuncToAsyncRuntimeConversionPatterns(), mlir::async::populateAsyncParallelForPatterns(), mlir::populateAsyncStructuralTypeConversionsAndLegality(), mlir::linalg::populateBubbleUpExtractSliceOpPatterns(), mlir::vector::populateBubbleVectorBitCastOpPatterns(), mlir::populateBufferizationToMemRefConversionPatterns(), mlir::populateBuiltinFuncToSPIRVPatterns(), mlir::populateCallOpTypeConversionPattern(), mlir::vector::populateCastAwayVectorLeadingOneDimPatterns(), mlir::arith::populateCeilFloorDivExpandOpsPatterns(), mlir::linalg::populateCollapseDimensions(), mlir::populateCommutativityUtilsPatterns(), mlir::populateComplexToLibmConversionPatterns(), mlir::populateComplexToLLVMConversionPatterns(), mlir::populateComplexToStandardConversionPatterns(), mlir::memref::populateComposeSubViewPatterns(), mlir::cf::populateControlFlowToLLVMConversionPatterns(), mlir::cf::populateControlFlowToSPIRVPatterns(), mlir::populateConvertArmNeon2dToIntrPatterns(), mlir::populateConvertShapeConstraintsConversionPatterns(), mlir::linalg::populateConvolutionVectorizationPatterns(), mlir::populateDecomposeCallGraphTypesPatterns(), mlir::linalg::populateDecomposeConvolutionPatterns(), mlir::linalg::populateElementwiseOpsFusionPatterns(), mlir::linalg::populateElementwiseToLinalgConversionPatterns(), mlir::bufferization::populateEliminateBufferizeMaterializationsPatterns(), mlir::populateExpandCtlzPattern(), mlir::memref::populateExpandOpsPatterns(), mlir::memref::populateExpandStridedMetadataPatterns(), mlir::populateExpandTanhPattern(), mlir::populateFinalizeMemRefToLLVMConversionPatterns(), mlir::vector::populateFlattenVectorTransferPatterns(), mlir::tensor::populateFoldConstantExtractSlicePatterns(), mlir::memref::populateFoldMemRefAliasOpPatterns(), mlir::linalg::populateFoldReshapeOpsByCollapsingPatterns(), mlir::linalg::populateFoldReshapeOpsByExpansionPatterns(), mlir::tensor::populateFoldTensorEmptyPatterns(), mlir::linalg::populateFoldUnitExtentDimsViaReshapesPatterns(), mlir::linalg::populateFoldUnitExtentDimsViaSlicesPatterns(), mlir::populateFunctionOpInterfaceTypeConversionPattern(), mlir::populateFuncToLLVMConversionPatterns(), mlir::populateFuncToLLVMFuncOpConversionPattern(), mlir::populateFuncToSPIRVPatterns(), mlir::linalg::populateFuseTensorPadWithProducerLinalgOpPatterns(), mlir::populateGpuAllReducePatterns(), mlir::populateGpuSubgroupReduceOpLoweringPattern(), mlir::populateGpuToLLVMConversionPatterns(), mlir::populateGpuToNVVMConversionPatterns(), mlir::populateGpuToROCDLConversionPatterns(), mlir::populateGPUToSPIRVPatterns(), mlir::populateGpuWMMAToNVVMConversionPatterns(), mlir::populateGpuWMMAToSPIRVConversionPatterns(), mlir::linalg::populateInlineConstantOperandsPatterns(), mlir::arith::populateIntRangeOptimizationsPatterns(), mlir::linalg::populateLinalgNamedOpConversionPatterns(), mlir::linalg::populateLinalgNamedOpsGeneralizationPatterns(), mlir::populateLinalgToLLVMConversionPatterns(), mlir::linalg::populateLinalgToStandardConversionPatterns(), mlir::populateMathAlgebraicSimplificationPatterns(), mlir::populateMathPolynomialApproximationPatterns(), mlir::populateMathToLibmConversionPatterns(), mlir::populateMathToLLVMConversionPatterns(), mlir::populateMathToSPIRVPatterns(), mlir::gpu::populateMemorySpaceLoweringPatterns(), mlir::spirv::populateMemorySpaceToStorageClassPatterns(), mlir::populateMemRefToSPIRVPatterns(), mlir::memref::populateMemRefWideIntEmulationPatterns(), mlir::tensor::populateMergeConsecutiveInsertExtractSlicePatterns(), mlir::nvgpu::populateMmaSyncF32ToTF32Patterns(), mlir::linalg::populateMoveInitOperandsToInputPattern(), mlir::populateNVGPUToNVVMConversionPatterns(), mlir::populateOpenACCToLLVMConversionPatterns(), mlir::populateOpenACCToSCFConversionPatterns(), mlir::populateOpenMPToLLVMConversionPatterns(), populateOpPatterns(), mlir::linalg::populatePadOpVectorizationPatterns(), mlir::linalg::populatePadTensorTilingPatterns(), mlir::populateParallelLoopToGPUPatterns(), mlir::populatePostSparsificationRewriting(), mlir::populatePrepareVectorToMMAPatterns(), mlir::populatePreSparsificationRewriting(), mlir::tensor::populateReassociativeReshapeFoldingPatterns(), mlir::populateReconcileUnrealizedCastsPatterns(), mlir::populateRemoveShapeConstraintsPatterns(), mlir::memref::populateResolveRankedShapeTypeResultDimsPatterns(), mlir::memref::populateResolveShapedTypeResultDimsPatterns(), mlir::populateReturnOpTypeConversionPattern(), mlir::vector::populateScalarVectorTransferLoweringPatterns(), mlir::scf::populateSCFForLoopCanonicalizationPatterns(), mlir::scf::populateSCFLoopPipeliningPatterns(), mlir::scf::populateSCFStructuralTypeConversionsAndLegality(), mlir::populateSCFToControlFlowConversionPatterns(), mlir::populateSCFToSPIRVPatterns(), mlir::vector::populateShapeCastFoldingPatterns(), mlir::populateShapeRewritePatterns(), mlir::populateShapeToStandardConversionPatterns(), mlir::tensor::populateSimplifyTensorPack(), mlir::populateSparseBufferRewriting(), mlir::populateSparseTensorCodegenPatterns(), mlir::populateSparseTensorConversionPatterns(), mlir::populateSparseVectorizationPatterns(), mlir::populateSparsificationPatterns(), mlir::x86vector::avx2::populateSpecializedTransposeLoweringPatterns(), mlir::spirv::populateSPIRVExpandExtendedMultiplicationPatterns(), mlir::spirv::populateSPIRVGLCanonicalizationPatterns(), populateSPIRVLayoutInfoPatterns(), mlir::populateSPIRVToLLVMConversionPatterns(), mlir::populateSPIRVToLLVMFunctionConversionPatterns(), mlir::populateSPIRVToLLVMModuleConversionPatterns(), mlir::tensor::populateSplitPaddingPatterns(), mlir::linalg::populateSplitReductionPattern(), mlir::populateStorageSpecifierToLLVMPatterns(), mlir::linalg::populateSwapExtractSliceWithFillPatterns(), mlir::populateTensorToLinalgPatterns(), mlir::populateTensorToSPIRVPatterns(), mlir::tosa::populateTosaDecomposeConv2D(), mlir::tosa::populateTosaDecomposeDepthwise(), mlir::tosa::populateTosaDecomposeTransposeConv(), mlir::tosa::populateTosaFoldConstantTransposePatterns(), mlir::tosa::populateTosaRescaleToArithConversionPatterns(), mlir::tosa::populateTosaToArithConversionPatterns(), mlir::tosa::populateTosaToLinalgConversionPatterns(), mlir::tosa::populateTosaToLinalgNamedConversionPatterns(), mlir::tosa::populateTosaToSCFConversionPatterns(), mlir::tosa::populateTosaToTensorConversionPatterns(), mlir::vector::populateVectorBroadcastLoweringPatterns(), mlir::vector::populateVectorContractLoweringPatterns(), mlir::vector::populateVectorExtractStridedSliceToExtractInsertChainPatterns(), mlir::vector::populateVectorInsertExtractStridedSliceDecompositionPatterns(), mlir::vector::populateVectorInsertExtractStridedSliceTransforms(), mlir::vector::populateVectorMaskLoweringPatternsForSideEffectingOps(), mlir::vector::populateVectorMaskMaterializationPatterns(), mlir::vector::populateVectorMaskOpLoweringPatterns(), mlir::vector::populateVectorMultiReductionLoweringPatterns(), mlir::vector::populateVectorReductionToContractPatterns(), mlir::vector::populateVectorScanLoweringPatterns(), mlir::vector::populateVectorShapeCastLoweringPatterns(), mlir::populateVectorToLLVMConversionPatterns(), mlir::populateVectorToLLVMMatrixConversionPatterns(), mlir::populateVectorToSCFConversionPatterns(), mlir::populateVectorToSPIRVPatterns(), mlir::vector::populateVectorTransferCollapseInnerMostContiguousDimsPatterns(), mlir::vector::populateVectorTransferDropUnitDimsPatterns(), mlir::vector::populateVectorTransferLoweringPatterns(), mlir::vector::populateVectorTransferPermutationMapLoweringPatterns(), mlir::vector::populateVectorTransposeLoweringPatterns(), mlir::vector::populateVectorUnrollPatterns(), mlir::vector::populateWarpExecuteOnLane0OpToScfForPattern(), and mlir::populateX86VectorLegalizeForLLVMExportPatterns().

◆ add() [3/5]

template<typename OpType >
RewritePatternSet& mlir::RewritePatternSet::add ( LogicalResult(*)(OpType, PatternRewriter &rewriter)  implFn,
PatternBenefit  benefit = 1,
ArrayRef< StringRef >  generatedNames = {} 
)
inline

Definition at line 1653 of file PatternMatch.h.

◆ add() [4/5]

RewritePatternSet& mlir::RewritePatternSet::add ( PDLPatternModule &&  pattern)
inline

Add the given PDL pattern to the pattern list.

Return a reference to this for chaining insertions.

Definition at line 1645 of file PatternMatch.h.

References mlir::PDLPatternModule::mergeIn().

◆ add() [5/5]

RewritePatternSet& mlir::RewritePatternSet::add ( std::unique_ptr< RewritePattern >  pattern)
inline

Add the given native pattern to the pattern list.

Return a reference to this for chaining insertions.

Definition at line 1638 of file PatternMatch.h.

◆ addWithLabel()

template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet& mlir::RewritePatternSet::addWithLabel ( ArrayRef< StringRef >  debugLabels,
ConstructorArg &&  arg,
ConstructorArgs &&...  args 
)
inline

An overload of the above add method that allows for attaching a set of debug labels to the attached patterns.

This is useful for labeling groups of patterns that may be shared between multiple different passes/users.

Definition at line 1619 of file PatternMatch.h.

◆ clear()

void mlir::RewritePatternSet::clear ( )
inline

Clear out all of the held patterns in this list.

Definition at line 1588 of file PatternMatch.h.

References mlir::PDLPatternModule::clear().

◆ getContext()

MLIRContext* mlir::RewritePatternSet::getContext ( ) const
inline

Definition at line 1579 of file PatternMatch.h.

Referenced by mlir::affineForOpBodySkew(), mlir::hoistAffineIfOp(), mlir::linalg::hoistRedundantVectorTransfersOnTensor(), insert(), mlir::populateAffineExpandIndexOpsPatterns(), mlir::populateAffineToStdConversionPatterns(), mlir::populateAffineToVectorConversionPatterns(), mlir::populateAnyFunctionOpInterfaceTypeConversionPattern(), mlir::arith::populateArithExpandOpsPatterns(), mlir::arith::populateArithToSPIRVPatterns(), mlir::arith::populateArithWideIntEmulationPatterns(), mlir::populateAsyncFuncToAsyncRuntimeConversionPatterns(), mlir::async::populateAsyncParallelForPatterns(), mlir::populateAsyncStructuralTypeConversionsAndLegality(), mlir::linalg::populateBubbleUpExtractSliceOpPatterns(), mlir::vector::populateBubbleVectorBitCastOpPatterns(), mlir::populateBufferizationToMemRefConversionPatterns(), mlir::populateBuiltinFuncToSPIRVPatterns(), mlir::populateCallOpTypeConversionPattern(), mlir::vector::populateCastAwayVectorLeadingOneDimPatterns(), mlir::arith::populateCeilFloorDivExpandOpsPatterns(), mlir::linalg::populateCollapseDimensions(), mlir::populateCommutativityUtilsPatterns(), mlir::populateComplexToLibmConversionPatterns(), mlir::populateComplexToStandardConversionPatterns(), mlir::linalg::populateConstantFoldLinalgOperations(), mlir::cf::populateControlFlowToSPIRVPatterns(), mlir::populateConvertArmNeon2dToIntrPatterns(), mlir::populateConvertShapeConstraintsConversionPatterns(), mlir::linalg::populateConvertToDestinationStylePatterns(), mlir::linalg::populateConvolutionVectorizationPatterns(), mlir::linalg::populateDataLayoutPropagationPatterns(), mlir::linalg::populateDecomposeConvolutionPatterns(), mlir::linalg::populateDecomposeLinalgOpsPattern(), mlir::linalg::populateElementwiseOpsFusionPatterns(), mlir::linalg::populateElementwiseToLinalgConversionPatterns(), mlir::bufferization::populateEliminateBufferizeMaterializationsPatterns(), mlir::bufferization::populateEmptyTensorToAllocTensorPattern(), mlir::linalg::populateEraseUnnecessaryInputsPatterns(), mlir::linalg::populateEraseUnusedOperandsAndResultsPatterns(), mlir::memref::populateExpandOpsPatterns(), mlir::memref::populateExpandStridedMetadataPatterns(), mlir::vector::populateFlattenVectorTransferPatterns(), mlir::tensor::populateFoldConstantExtractSlicePatterns(), mlir::tensor::populateFoldIntoPackAndUnpackPatterns(), mlir::memref::populateFoldMemRefAliasOpPatterns(), mlir::linalg::populateFoldReshapeOpsByCollapsingPatterns(), mlir::linalg::populateFoldReshapeOpsByExpansionPatterns(), mlir::tensor::populateFoldTensorEmptyPatterns(), mlir::linalg::populateFoldUnitExtentDimsViaReshapesPatterns(), mlir::linalg::populateFoldUnitExtentDimsViaSlicesPatterns(), mlir::populateFunctionOpInterfaceTypeConversionPattern(), mlir::populateFuncToSPIRVPatterns(), mlir::linalg::populateFuseTensorPadWithProducerLinalgOpPatterns(), mlir::populateGpuAllReducePatterns(), mlir::populateGPUToSPIRVPatterns(), mlir::populateGpuWMMAToSPIRVConversionPatterns(), mlir::linalg::populateInlineConstantOperandsPatterns(), mlir::arith::populateIntRangeOptimizationsPatterns(), mlir::linalg::populateLinalgNamedOpConversionPatterns(), mlir::linalg::populateLinalgNamedOpsGeneralizationPatterns(), mlir::linalg::populateLinalgTilingCanonicalizationPatterns(), mlir::linalg::populateLinalgToStandardConversionPatterns(), mlir::populateMathAlgebraicSimplificationPatterns(), mlir::populateMathPolynomialApproximationPatterns(), mlir::populateMathToLibmConversionPatterns(), mlir::populateMathToSPIRVPatterns(), mlir::gpu::populateMemorySpaceLoweringPatterns(), mlir::spirv::populateMemorySpaceToStorageClassPatterns(), mlir::populateMemRefToSPIRVPatterns(), mlir::memref::populateMemRefWideIntEmulationPatterns(), mlir::tensor::populateMergeConsecutiveInsertExtractSlicePatterns(), mlir::nvgpu::populateMmaSyncF32ToTF32Patterns(), mlir::linalg::populateMoveInitOperandsToInputPattern(), mlir::populateOpenACCToSCFConversionPatterns(), mlir::linalg::populatePadOpVectorizationPatterns(), mlir::linalg::populatePadTensorTilingPatterns(), mlir::populateParallelLoopToGPUPatterns(), mlir::populatePostSparsificationRewriting(), mlir::populatePrepareVectorToMMAPatterns(), mlir::populatePreSparsificationRewriting(), mlir::tensor::populateReassociativeReshapeFoldingPatterns(), mlir::populateReconcileUnrealizedCastsPatterns(), mlir::populateRemoveShapeConstraintsPatterns(), mlir::memref::populateResolveRankedShapeTypeResultDimsPatterns(), mlir::memref::populateResolveShapedTypeResultDimsPatterns(), mlir::populateReturnOpTypeConversionPattern(), mlir::vector::populateScalarVectorTransferLoweringPatterns(), mlir::scf::populateSCFForLoopCanonicalizationPatterns(), mlir::scf::populateSCFLoopPipeliningPatterns(), mlir::scf::populateSCFStructuralTypeConversionsAndLegality(), mlir::populateSCFToControlFlowConversionPatterns(), mlir::populateSCFToSPIRVPatterns(), mlir::vector::populateShapeCastFoldingPatterns(), mlir::populateShapeRewritePatterns(), mlir::populateShapeToStandardConversionPatterns(), mlir::tensor::populateSimplifyTensorPack(), mlir::populateSparseBufferRewriting(), mlir::populateSparseTensorCodegenPatterns(), mlir::populateSparseTensorConversionPatterns(), mlir::populateSparseVectorizationPatterns(), mlir::populateSparsificationPatterns(), mlir::x86vector::avx2::populateSpecializedTransposeLoweringPatterns(), mlir::spirv::populateSPIRVExpandExtendedMultiplicationPatterns(), mlir::spirv::populateSPIRVGLCanonicalizationPatterns(), populateSPIRVLayoutInfoPatterns(), mlir::populateSPIRVToLLVMConversionPatterns(), mlir::populateSPIRVToLLVMFunctionConversionPatterns(), mlir::populateSPIRVToLLVMModuleConversionPatterns(), mlir::tensor::populateSplitPaddingPatterns(), mlir::linalg::populateSplitReductionPattern(), mlir::populateStorageSpecifierToLLVMPatterns(), mlir::linalg::populateSwapExtractSliceWithFillPatterns(), mlir::populateTensorToLinalgPatterns(), mlir::populateTensorToSPIRVPatterns(), mlir::tosa::populateTosaRescaleToArithConversionPatterns(), mlir::tosa::populateTosaToArithConversionPatterns(), mlir::tosa::populateTosaToLinalgConversionPatterns(), mlir::tosa::populateTosaToLinalgNamedConversionPatterns(), mlir::tosa::populateTosaToSCFConversionPatterns(), mlir::tosa::populateTosaToTensorConversionPatterns(), mlir::vector::populateVectorBroadcastLoweringPatterns(), mlir::vector::populateVectorContractLoweringPatterns(), mlir::vector::populateVectorExtractStridedSliceToExtractInsertChainPatterns(), mlir::vector::populateVectorInsertExtractStridedSliceDecompositionPatterns(), mlir::vector::populateVectorInsertExtractStridedSliceTransforms(), mlir::vector::populateVectorMaskLoweringPatternsForSideEffectingOps(), mlir::vector::populateVectorMaskMaterializationPatterns(), mlir::vector::populateVectorMaskOpLoweringPatterns(), mlir::vector::populateVectorMultiReductionLoweringPatterns(), mlir::vector::populateVectorReductionToContractPatterns(), mlir::vector::populateVectorScanLoweringPatterns(), mlir::vector::populateVectorShapeCastLoweringPatterns(), mlir::populateVectorToSCFConversionPatterns(), mlir::populateVectorToSPIRVPatterns(), mlir::vector::populateVectorTransferCollapseInnerMostContiguousDimsPatterns(), mlir::vector::populateVectorTransferDropUnitDimsPatterns(), mlir::vector::populateVectorTransferLoweringPatterns(), mlir::vector::populateVectorTransferPermutationMapLoweringPatterns(), mlir::vector::populateVectorTransposeLoweringPatterns(), mlir::vector::populateVectorUnrollPatterns(), and mlir::vector::populateWarpExecuteOnLane0OpToScfForPattern().

◆ getNativePatterns()

NativePatternListT& mlir::RewritePatternSet::getNativePatterns ( )
inline

Return the native patterns held in this list.

Definition at line 1582 of file PatternMatch.h.

◆ getPDLPatterns()

PDLPatternModule& mlir::RewritePatternSet::getPDLPatterns ( )
inline

Return the PDL patterns held in this list.

Definition at line 1585 of file PatternMatch.h.

Referenced by mlir::registerConversionPDLFunctions().

◆ insert() [1/5]

template<typename... Ts>
RewritePatternSet& mlir::RewritePatternSet::insert ( )
inline

Add an instance of each of the pattern types 'Ts'.

Return a reference to this for chaining insertions.

Definition at line 1697 of file PatternMatch.h.

◆ insert() [2/5]

template<typename... Ts, typename ConstructorArg , typename... ConstructorArgs, typename = std::enable_if_t<sizeof...(Ts) != 0>>
RewritePatternSet& mlir::RewritePatternSet::insert ( ConstructorArg &&  arg,
ConstructorArgs &&...  args 
)
inline

Add an instance of each of the pattern types 'Ts' to the pattern list with the given arguments.

Return a reference to this for chaining insertions. Note: ConstructorArg is necessary here to separate the two variadic lists.

Definition at line 1687 of file PatternMatch.h.

Referenced by mlir::populateAffineExpandIndexOpsPatterns(), mlir::linalg::populateConstantFoldLinalgOperations(), mlir::linalg::populateConvertToDestinationStylePatterns(), mlir::linalg::populateDataLayoutPropagationPatterns(), mlir::linalg::populateDecomposeLinalgOpsPattern(), mlir::bufferization::populateEmptyTensorToAllocTensorPattern(), mlir::linalg::populateEraseUnnecessaryInputsPatterns(), mlir::linalg::populateEraseUnusedOperandsAndResultsPatterns(), mlir::tensor::populateFoldIntoPackAndUnpackPatterns(), and mlir::index::populateIndexToLLVMConversionPatterns().

◆ insert() [3/5]

template<typename OpType >
RewritePatternSet& mlir::RewritePatternSet::insert ( LogicalResult(*)(OpType, PatternRewriter &rewriter)  implFn)
inline

Definition at line 1719 of file PatternMatch.h.

References add(), and getContext().

◆ insert() [4/5]

RewritePatternSet& mlir::RewritePatternSet::insert ( PDLPatternModule &&  pattern)
inline

Add the given PDL pattern to the pattern list.

Return a reference to this for chaining insertions.

Definition at line 1711 of file PatternMatch.h.

References mlir::PDLPatternModule::mergeIn().

◆ insert() [5/5]

RewritePatternSet& mlir::RewritePatternSet::insert ( std::unique_ptr< RewritePattern >  pattern)
inline

Add the given native pattern to the pattern list.

Return a reference to this for chaining insertions.

Definition at line 1704 of file PatternMatch.h.

The documentation for this class was generated from the following file:
Generated on Fri Jan 27 2023 12:38:05 for MLIR by doxygen 1.9.1