MLIR 23.0.0git
Classes | Functions
mlir::impl Namespace Reference

Attribute collections provide a dictionary-like interface. More...

Classes

class  ArithToAMDGPUConversionPassBase
class  ArithToAPFloatConversionPassBase
class  ArithToArmSMEConversionPassBase
class  ArithToLLVMConversionPassBase
class  AsyncFuncToAsyncRuntimePassBase
class  AsyncParallelForPassBase
class  AsyncRuntimePolicyBasedRefCountingPassBase
class  AsyncRuntimeRefCountingOptPassBase
class  AsyncRuntimeRefCountingPassBase
class  AsyncToAsyncRuntimePassBase
class  BubbleDownMemorySpaceCastsBase
class  CanonicalizerBase
class  CompositeFixedPointPassBase
class  ControlFlowSinkBase
class  ConvertAffineForToGPUPassBase
class  ConvertAMDGPUToROCDLPassBase
class  ConvertArithToEmitCBase
class  ConvertArithToSPIRVPassBase
class  ConvertArmNeon2dToIntrPassBase
class  ConvertArmSMEToLLVMBase
class  ConvertArmSMEToSCFPassBase
class  ConvertAsyncToLLVMPassBase
class  ConvertBufferizationToMemRefPassBase
class  ConvertComplexToLibmBase
class  ConvertComplexToLLVMPassBase
class  ConvertComplexToROCDLLibraryCallsBase
class  ConvertComplexToSPIRVPassBase
class  ConvertComplexToStandardPassBase
class  ConvertControlFlowToLLVMPassBase
class  ConvertControlFlowToSPIRVPassBase
class  ConvertElementwiseToLinalgPassBase
class  ConvertFuncToEmitCBase
class  ConvertFuncToLLVMPassBase
class  ConvertFuncToSPIRVPassBase
class  ConvertGpuOpsToLLVMSPVOpsBase
class  ConvertGpuOpsToNVVMOpsBase
class  ConvertGpuOpsToROCDLOpsBase
class  ConvertGPUToSPIRVBase
class  ConvertIndexToLLVMPassBase
class  ConvertIndexToSPIRVPassBase
class  ConvertLinalgToAffineLoopsPassBase
class  ConvertLinalgToLoopsPassBase
class  ConvertLinalgToParallelLoopsPassBase
class  ConvertLinalgToStandardPassBase
class  ConvertMathToEmitCBase
class  ConvertMathToFuncsBase
class  ConvertMathToLibmPassBase
class  ConvertMathToLLVMPassBase
class  ConvertMathToROCDLBase
class  ConvertMathToSPIRVPassBase
class  ConvertMathToXeVMBase
class  ConvertMemRefToEmitCBase
class  ConvertMemRefToSPIRVPassBase
class  ConvertNVGPUToNVVMPassBase
class  ConvertNVVMToLLVMPassBase
class  ConvertOpenACCToSCFPassBase
class  ConvertOpenMPToLLVMPassBase
class  ConvertParallelLoopToGpuPassBase
class  ConvertPDLToPDLInterpPassBase
class  ConvertSCFToOpenMPPassBase
class  ConvertShapeConstraintsPassBase
class  ConvertShapeToStandardPassBase
class  ConvertShardToMPIPassBase
class  ConvertSPIRVToLLVMPassBase
class  ConvertTensorToLinalgPassBase
class  ConvertTensorToSPIRVPassBase
class  ConvertToEmitCBase
class  ConvertToLLVMPassBase
class  ConvertVectorToAMXBase
class  ConvertVectorToArmSMEPassBase
class  ConvertVectorToGPUBase
class  ConvertVectorToLLVMPassBase
class  ConvertVectorToSCFBase
class  ConvertVectorToSPIRVPassBase
class  ConvertVectorToXeGPUBase
class  ConvertXeGPUToXeVMPassBase
class  ConvertXeVMToLLVMPassBase
class  CSEBase
class  FinalizeMemRefToLLVMConversionPassBase
class  GenerateRuntimeVerificationBase
class  GpuAsyncRegionPassBase
class  GpuDecomposeMemrefsPassBase
class  GpuEliminateBarriersBase
class  GpuKernelOutliningPassBase
class  GpuLaunchSinkIndexComputationsPassBase
class  GpuMapParallelLoopsPassBase
class  GpuModuleToBinaryPassBase
class  GpuNVVMAttachTargetBase
class  GpuROCDLAttachTargetBase
class  GpuSPIRVAttachTargetBase
class  GpuToLLVMConversionPassBase
class  GpuXeVMAttachTargetBase
class  InlinerBase
class  LiftControlFlowToSCFPassBase
class  LinalgBlockPackMatmulBase
class  LinalgDetensorizePassBase
class  LinalgElementwiseOpFusionPassBase
class  LinalgFoldIntoElementwisePassBase
class  LinalgFoldUnitExtentDimsPassBase
class  LinalgGeneralizeNamedOpsPassBase
class  LinalgInlineScalarOperandsPassBase
class  LinalgMorphOpsPassBase
class  LinalgSpecializeGenericOpsPassBase
class  LocationSnapshotBase
class  LoopInvariantCodeMotionBase
class  LoopInvariantSubsetHoistingBase
class  LowerAffinePassBase
class  LowerForeachToSCFBase
class  LowerHostCodeToLLVMPassBase
class  LowerSparseIterationToSCFBase
class  LowerSparseOpsToForeachBase
class  MapMemRefStorageClassBase
class  MathToAPFloatConversionPassBase
class  Mem2RegBase
class  OptReductionPassBase
class  OutlineShapeComputationPassBase
class  PreSparsificationRewriteBase
class  PrintOpStatsBase
class  ReconcileUnrealizedCastsPassBase
class  ReductionTreePassBase
class  RemoveDeadValuesBase
class  RemoveShapeConstraintsPassBase
class  SCCPBase
class  SCFForallToForLoopBase
class  SCFForallToParallelLoopBase
class  SCFForLoopCanonicalizationBase
class  SCFForLoopPeelingBase
class  SCFForLoopRangeFoldingBase
class  SCFForLoopSpecializationBase
class  SCFForToWhileLoopBase
class  SCFParallelForToNestedForsBase
class  SCFParallelLoopFusionBase
class  SCFParallelLoopSpecializationBase
class  SCFParallelLoopTilingBase
class  SCFToControlFlowPassBase
class  SCFToEmitCBase
class  SCFToSPIRVBase
class  SetLLVMModuleDataLayoutPassBase
class  ShapeToShapeLoweringPassBase
class  SimplifyDepthwiseConvPassBase
class  SparseAssemblerBase
class  SparseBufferRewriteBase
class  SparseGPUCodegenBase
class  SparseReinterpretMapBase
class  SparseSpaceCollapseBase
class  SparseTensorCodegenBase
class  SparseTensorConversionPassBase
class  SparseVectorizationBase
class  SparsificationAndBufferizationBase
class  SparsificationPassBase
class  SROABase
class  StageSparseOperationsBase
class  StorageSpecifierToLLVMBase
class  StripDebugInfoBase
class  SymbolDCEBase
class  SymbolPrivatizeBase
class  TestSCFParallelLoopCollapsingBase
class  TopologicalSortBase
class  TosaToArithPassBase
class  TosaToLinalgBase
class  TosaToLinalgNamedBase
class  TosaToMLProgramBase
class  TosaToSCFPassBase
class  TosaToTensorPassBase
class  UBToLLVMConversionPassBase
class  UBToSPIRVConversionPassBase
class  ViewOpGraphBase

Functions

LogicalResult verifyHasDefaultDLTIDataLayoutTrait (Operation *op)
DataLayoutSpecInterface getDataLayoutSpec (Operation *op)
TargetSystemSpecInterface getTargetSystemSpec (Operation *op)
LogicalResult foldCastInterfaceOp (Operation *op, ArrayRef< Attribute > attrOperands, SmallVectorImpl< OpFoldResult > &foldResults)
 Attempt to fold the given cast operation.
LogicalResult verifyCastInterfaceOp (Operation *op)
 Attempt to verify the given cast operation.
void ensureRegionTerminator (Region &region, OpBuilder &builder, Location loc, function_ref< Operation *(OpBuilder &, Location)> buildTerminatorOp)
 Insert an operation, generated by buildTerminatorOp, at the end of the region's only block if it does not have a terminator already.
void ensureRegionTerminator (Region &region, Builder &builder, Location loc, function_ref< Operation *(OpBuilder &, Location)> buildTerminatorOp)
 Create a simple OpBuilder and forward to the OpBuilder version of this function.
template<typename IteratorT, typename NameT>
std::pair< IteratorT, bool > findAttrUnsorted (IteratorT first, IteratorT last, NameT name)
 Unsorted string search or identifier lookups are linear scans.
template<typename IteratorT>
std::pair< IteratorT, bool > findAttrSorted (IteratorT first, IteratorT last, StringRef name)
 Using llvm::lower_bound requires an extra string comparison to check whether the returned iterator points to the found element or whether it indicates the lower bound.
template<typename IteratorT>
std::pair< IteratorT, bool > findAttrSorted (IteratorT first, IteratorT last, StringAttr name)
 StringAttr lookups on large attribute lists will switch to string binary search.
template<typename IteratorT, typename NameT>
Attribute getAttrFromSortedRange (IteratorT first, IteratorT last, NameT name)
 Get an attribute from a sorted range of named attributes.
template<typename IteratorT, typename NameT>
std::optional< NamedAttribute > getNamedAttrFromSortedRange (IteratorT first, IteratorT last, NameT name)
 Get an attribute from a sorted range of named attributes.
ParseResult parseOptionalVisibilityKeyword (OpAsmParser &parser, NamedAttrList &attrs)
 Parse an optional visibility attribute keyword (i.e., public, private, or nested) without quotes in a string attribute named 'attrName'.
std::unique_ptr<::mlir::Pass > createLowerAffinePass ()
std::unique_ptr<::mlir::Pass > createConvertAMDGPUToROCDLPass ()
std::unique_ptr<::mlir::Pass > createConvertAMDGPUToROCDLPass (ConvertAMDGPUToROCDLPassOptions options)
std::unique_ptr<::mlir::Pass > createArithToAPFloatConversionPass ()
std::unique_ptr<::mlir::Pass > createMathToAPFloatConversionPass ()
std::unique_ptr<::mlir::Pass > createArithToAMDGPUConversionPass ()
std::unique_ptr<::mlir::Pass > createArithToAMDGPUConversionPass (ArithToAMDGPUConversionPassOptions options)
std::unique_ptr<::mlir::Pass > createArithToArmSMEConversionPass ()
std::unique_ptr<::mlir::Pass > createConvertArithToEmitC ()
std::unique_ptr<::mlir::Pass > createArithToLLVMConversionPass ()
std::unique_ptr<::mlir::Pass > createArithToLLVMConversionPass (ArithToLLVMConversionPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertArithToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertArithToSPIRVPass (ConvertArithToSPIRVPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertArmNeon2dToIntrPass ()
std::unique_ptr<::mlir::Pass > createConvertArmSMEToSCFPass ()
std::unique_ptr<::mlir::Pass > createConvertAsyncToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertBufferizationToMemRefPass ()
std::unique_ptr<::mlir::Pass > createConvertComplexToLibm ()
std::unique_ptr<::mlir::Pass > createConvertComplexToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertComplexToLLVMPass (ConvertComplexToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertComplexToROCDLLibraryCalls ()
std::unique_ptr<::mlir::Pass > createConvertComplexToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertComplexToStandardPass ()
std::unique_ptr<::mlir::Pass > createConvertComplexToStandardPass (ConvertComplexToStandardPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertControlFlowToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertControlFlowToLLVMPass (ConvertControlFlowToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createLiftControlFlowToSCFPass ()
std::unique_ptr<::mlir::Pass > createConvertControlFlowToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertControlFlowToSPIRVPass (ConvertControlFlowToSPIRVPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertToEmitC ()
std::unique_ptr<::mlir::Pass > createConvertToEmitC (ConvertToEmitCOptions options)
std::unique_ptr<::mlir::Pass > createConvertToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertToLLVMPass (ConvertToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertFuncToEmitC ()
std::unique_ptr<::mlir::Pass > createConvertFuncToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertFuncToLLVMPass (ConvertFuncToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createSetLLVMModuleDataLayoutPass ()
std::unique_ptr<::mlir::Pass > createSetLLVMModuleDataLayoutPass (SetLLVMModuleDataLayoutPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertFuncToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertFuncToSPIRVPass (ConvertFuncToSPIRVPassOptions options)
LogicalResult scalarizeVectorOp (Operation *op, ValueRange operands, ConversionPatternRewriter &rewriter, const LLVMTypeConverter &converter)
 Unrolls op to array/vector elements.
std::unique_ptr<::mlir::Pass > createGpuToLLVMConversionPass ()
std::unique_ptr<::mlir::Pass > createGpuToLLVMConversionPass (GpuToLLVMConversionPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToLLVMSPVOps ()
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToLLVMSPVOps (ConvertGpuOpsToLLVMSPVOpsOptions options)
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToNVVMOps ()
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToNVVMOps (ConvertGpuOpsToNVVMOpsOptions options)
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToROCDLOps ()
std::unique_ptr<::mlir::Pass > createConvertGpuOpsToROCDLOps (ConvertGpuOpsToROCDLOpsOptions options)
std::unique_ptr<::mlir::Pass > createConvertIndexToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertIndexToLLVMPass (ConvertIndexToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertIndexToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertIndexToSPIRVPass (ConvertIndexToSPIRVPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertLinalgToStandardPass ()
std::unique_ptr<::mlir::Pass > createConvertMathToEmitC ()
std::unique_ptr<::mlir::Pass > createConvertMathToEmitC (ConvertMathToEmitCOptions options)
std::unique_ptr<::mlir::Pass > createConvertMathToFuncs ()
std::unique_ptr<::mlir::Pass > createConvertMathToFuncs (ConvertMathToFuncsOptions options)
std::unique_ptr<::mlir::Pass > createConvertMathToLibmPass ()
std::unique_ptr<::mlir::Pass > createConvertMathToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertMathToLLVMPass (ConvertMathToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertMathToROCDL ()
std::unique_ptr<::mlir::Pass > createConvertMathToROCDL (ConvertMathToROCDLOptions options)
std::unique_ptr<::mlir::Pass > createConvertMathToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertMathToXeVM ()
std::unique_ptr<::mlir::Pass > createConvertMathToXeVM (ConvertMathToXeVMOptions options)
std::unique_ptr<::mlir::Pass > createConvertMemRefToEmitC ()
std::unique_ptr<::mlir::Pass > createConvertMemRefToEmitC (ConvertMemRefToEmitCOptions options)
std::unique_ptr<::mlir::Pass > createFinalizeMemRefToLLVMConversionPass ()
std::unique_ptr<::mlir::Pass > createFinalizeMemRefToLLVMConversionPass (FinalizeMemRefToLLVMConversionPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertMemRefToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertMemRefToSPIRVPass (ConvertMemRefToSPIRVPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertNVGPUToNVVMPass ()
std::unique_ptr<::mlir::Pass > createConvertNVVMToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertOpenACCToSCFPass ()
std::unique_ptr<::mlir::Pass > createConvertOpenMPToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertPDLToPDLInterpPass ()
std::unique_ptr<::mlir::Pass > createReconcileUnrealizedCastsPass ()
std::unique_ptr<::mlir::Pass > createSCFToControlFlowPass ()
std::unique_ptr<::mlir::Pass > createSCFToControlFlowPass (SCFToControlFlowPassOptions options)
std::unique_ptr<::mlir::Pass > createSCFToEmitC ()
std::unique_ptr<::mlir::Pass > createConvertAffineForToGPUPass ()
std::unique_ptr<::mlir::Pass > createConvertAffineForToGPUPass (ConvertAffineForToGPUPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertParallelLoopToGpuPass ()
std::unique_ptr<::mlir::Pass > createConvertSCFToOpenMPPass ()
std::unique_ptr<::mlir::Pass > createConvertSCFToOpenMPPass (ConvertSCFToOpenMPPassOptions options)
std::unique_ptr<::mlir::Pass > createSCFToSPIRV ()
std::unique_ptr<::mlir::Pass > createConvertShapeConstraintsPass ()
std::unique_ptr<::mlir::Pass > createConvertShapeToStandardPass ()
std::unique_ptr<::mlir::Pass > createConvertShardToMPIPass ()
std::unique_ptr<::mlir::Pass > createLowerHostCodeToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertSPIRVToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertSPIRVToLLVMPass (ConvertSPIRVToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertTensorToLinalgPass ()
std::unique_ptr<::mlir::Pass > createConvertTensorToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertTensorToSPIRVPass (ConvertTensorToSPIRVPassOptions options)
std::unique_ptr<::mlir::Pass > createTosaToArithPass ()
std::unique_ptr<::mlir::Pass > createTosaToArithPass (TosaToArithPassOptions options)
std::unique_ptr<::mlir::Pass > createTosaToMLProgram ()
std::unique_ptr<::mlir::Pass > createTosaToSCFPass ()
std::unique_ptr<::mlir::Pass > createTosaToTensorPass ()
std::unique_ptr<::mlir::Pass > createUBToLLVMConversionPass ()
std::unique_ptr<::mlir::Pass > createUBToLLVMConversionPass (UBToLLVMConversionPassOptions options)
std::unique_ptr<::mlir::Pass > createUBToSPIRVConversionPass ()
std::unique_ptr<::mlir::Pass > createConvertVectorToAMX ()
std::unique_ptr<::mlir::Pass > createConvertVectorToArmSMEPass ()
std::unique_ptr<::mlir::Pass > createConvertVectorToLLVMPass ()
std::unique_ptr<::mlir::Pass > createConvertVectorToLLVMPass (ConvertVectorToLLVMPassOptions options)
std::unique_ptr<::mlir::Pass > createConvertVectorToSPIRVPass ()
std::unique_ptr<::mlir::Pass > createConvertVectorToXeGPU ()
std::unique_ptr<::mlir::Pass > createConvertXeGPUToXeVMPass ()
std::unique_ptr<::mlir::Pass > createConvertXeVMToLLVMPass ()
std::unique_ptr<::mlir::Pass > createAsyncParallelForPass ()
std::unique_ptr<::mlir::Pass > createAsyncParallelForPass (AsyncParallelForPassOptions options)
std::unique_ptr<::mlir::Pass > createAsyncRuntimePolicyBasedRefCountingPass ()
std::unique_ptr<::mlir::Pass > createAsyncRuntimeRefCountingPass ()
std::unique_ptr<::mlir::Pass > createAsyncRuntimeRefCountingOptPass ()
std::unique_ptr<::mlir::Pass > createAsyncFuncToAsyncRuntimePass ()
std::unique_ptr<::mlir::Pass > createAsyncToAsyncRuntimePass ()
std::unique_ptr<::mlir::Pass > createGpuAsyncRegionPass ()
std::unique_ptr<::mlir::Pass > createGpuDecomposeMemrefsPass ()
std::unique_ptr<::mlir::Pass > createGpuEliminateBarriers ()
std::unique_ptr<::mlir::Pass > createGpuKernelOutliningPass ()
std::unique_ptr<::mlir::Pass > createGpuKernelOutliningPass (GpuKernelOutliningPassOptions options)
std::unique_ptr<::mlir::Pass > createGpuLaunchSinkIndexComputationsPass ()
std::unique_ptr<::mlir::Pass > createGpuModuleToBinaryPass ()
std::unique_ptr<::mlir::Pass > createGpuModuleToBinaryPass (GpuModuleToBinaryPassOptions options)
std::unique_ptr<::mlir::Pass > createGpuNVVMAttachTarget ()
std::unique_ptr<::mlir::Pass > createGpuNVVMAttachTarget (GpuNVVMAttachTargetOptions options)
std::unique_ptr<::mlir::Pass > createGpuMapParallelLoopsPass ()
std::unique_ptr<::mlir::Pass > createGpuMapParallelLoopsPass (GpuMapParallelLoopsPassOptions options)
std::unique_ptr<::mlir::Pass > createGpuROCDLAttachTarget ()
std::unique_ptr<::mlir::Pass > createGpuROCDLAttachTarget (GpuROCDLAttachTargetOptions options)
std::unique_ptr<::mlir::Pass > createGpuSPIRVAttachTarget ()
std::unique_ptr<::mlir::Pass > createGpuSPIRVAttachTarget (GpuSPIRVAttachTargetOptions options)
std::unique_ptr<::mlir::Pass > createGpuXeVMAttachTarget ()
std::unique_ptr<::mlir::Pass > createGpuXeVMAttachTarget (GpuXeVMAttachTargetOptions options)
std::unique_ptr<::mlir::Pass > createLinalgBlockPackMatmul ()
std::unique_ptr<::mlir::Pass > createLinalgBlockPackMatmul (LinalgBlockPackMatmulOptions options)
std::unique_ptr<::mlir::Pass > createLinalgDetensorizePass ()
std::unique_ptr<::mlir::Pass > createLinalgDetensorizePass (LinalgDetensorizePassOptions options)
std::unique_ptr<::mlir::Pass > createLinalgFoldUnitExtentDimsPass ()
std::unique_ptr<::mlir::Pass > createLinalgFoldUnitExtentDimsPass (LinalgFoldUnitExtentDimsPassOptions options)
std::unique_ptr<::mlir::Pass > createLinalgElementwiseOpFusionPass ()
std::unique_ptr<::mlir::Pass > createConvertElementwiseToLinalgPass ()
std::unique_ptr<::mlir::Pass > createLinalgFoldIntoElementwisePass ()
std::unique_ptr<::mlir::Pass > createLinalgGeneralizeNamedOpsPass ()
std::unique_ptr<::mlir::Pass > createLinalgInlineScalarOperandsPass ()
std::unique_ptr<::mlir::Pass > createConvertLinalgToAffineLoopsPass ()
std::unique_ptr<::mlir::Pass > createConvertLinalgToLoopsPass ()
std::unique_ptr<::mlir::Pass > createConvertLinalgToParallelLoopsPass ()
std::unique_ptr<::mlir::Pass > createLinalgMorphOpsPass ()
std::unique_ptr<::mlir::Pass > createLinalgMorphOpsPass (LinalgMorphOpsPassOptions options)
std::unique_ptr<::mlir::Pass > createSimplifyDepthwiseConvPass ()
std::unique_ptr<::mlir::Pass > createLinalgSpecializeGenericOpsPass ()
std::unique_ptr<::mlir::Pass > createSCFForLoopPeeling ()
std::unique_ptr<::mlir::Pass > createSCFForLoopPeeling (SCFForLoopPeelingOptions options)
std::unique_ptr<::mlir::Pass > createOutlineShapeComputationPass ()
std::unique_ptr<::mlir::Pass > createRemoveShapeConstraintsPass ()
std::unique_ptr<::mlir::Pass > createShapeToShapeLoweringPass ()
std::unique_ptr<::mlir::Pass > createOptReductionPass ()
std::unique_ptr<::mlir::Pass > createOptReductionPass (OptReductionPassOptions options)
std::unique_ptr<::mlir::Pass > createReductionTreePass ()
std::unique_ptr<::mlir::Pass > createReductionTreePass (ReductionTreePassOptions options)
std::unique_ptr<::mlir::Pass > createBubbleDownMemorySpaceCasts ()
std::unique_ptr<::mlir::Pass > createCompositeFixedPointPass ()
std::unique_ptr<::mlir::Pass > createCompositeFixedPointPass (CompositeFixedPointPassOptions options)
std::unique_ptr<::mlir::Pass > createLocationSnapshot ()
std::unique_ptr<::mlir::Pass > createLocationSnapshot (LocationSnapshotOptions options)
std::unique_ptr<::mlir::Pass > createMem2Reg ()
std::unique_ptr<::mlir::Pass > createMem2Reg (Mem2RegOptions options)
std::unique_ptr<::mlir::Pass > createSROA ()

Detailed Description

Attribute collections provide a dictionary-like interface.

Define common lookup functions.

Function Documentation

◆ createArithToAMDGPUConversionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToAMDGPUConversionPass ( )

Definition at line 174 of file ArithToAMDGPU.cpp.

◆ createArithToAMDGPUConversionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToAMDGPUConversionPass ( ArithToAMDGPUConversionPassOptions options)

Definition at line 178 of file ArithToAMDGPU.cpp.

◆ createArithToAPFloatConversionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToAPFloatConversionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 259 of file ArithToAPFloat.cpp.

◆ createArithToArmSMEConversionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToArmSMEConversionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 334 of file ArithToArmSME.cpp.

◆ createArithToLLVMConversionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToLLVMConversionPass ( )

Definition at line 421 of file ArithToLLVM.cpp.

◆ createArithToLLVMConversionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createArithToLLVMConversionPass ( ArithToLLVMConversionPassOptions options)

Definition at line 425 of file ArithToLLVM.cpp.

◆ createAsyncFuncToAsyncRuntimePass()

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncFuncToAsyncRuntimePass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 79 of file AsyncToAsyncRuntime.cpp.

◆ createAsyncParallelForPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncParallelForPass ( )

Definition at line 175 of file AsyncParallelFor.cpp.

◆ createAsyncParallelForPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncParallelForPass ( AsyncParallelForPassOptions options)

Definition at line 179 of file AsyncParallelFor.cpp.

◆ createAsyncRuntimePolicyBasedRefCountingPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncRuntimePolicyBasedRefCountingPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 258 of file AsyncRuntimeRefCounting.cpp.

◆ createAsyncRuntimeRefCountingOptPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncRuntimeRefCountingOptPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 333 of file AsyncRuntimeRefCountingOpt.cpp.

◆ createAsyncRuntimeRefCountingPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncRuntimeRefCountingPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 408 of file AsyncRuntimeRefCounting.cpp.

◆ createAsyncToAsyncRuntimePass()

std::unique_ptr<::mlir::Pass > mlir::impl::createAsyncToAsyncRuntimePass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 485 of file AsyncToAsyncRuntime.cpp.

◆ createBubbleDownMemorySpaceCasts()

std::unique_ptr<::mlir::Pass > mlir::impl::createBubbleDownMemorySpaceCasts ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 93 of file BubbleDownMemorySpaceCasts.cpp.

◆ createCompositeFixedPointPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createCompositeFixedPointPass ( )

Definition at line 344 of file CompositePass.cpp.

◆ createCompositeFixedPointPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createCompositeFixedPointPass ( CompositeFixedPointPassOptions options)

Definition at line 348 of file CompositePass.cpp.

◆ createConvertAffineForToGPUPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertAffineForToGPUPass ( )

Definition at line 615 of file SCFToGPUPass.cpp.

◆ createConvertAffineForToGPUPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertAffineForToGPUPass ( ConvertAffineForToGPUPassOptions options)

Definition at line 619 of file SCFToGPUPass.cpp.

◆ createConvertAMDGPUToROCDLPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertAMDGPUToROCDLPass ( )

Definition at line 517 of file AMDGPUToROCDL.cpp.

◆ createConvertAMDGPUToROCDLPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertAMDGPUToROCDLPass ( ConvertAMDGPUToROCDLPassOptions options)

Definition at line 521 of file AMDGPUToROCDL.cpp.

◆ createConvertArithToEmitC()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertArithToEmitC ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 698 of file ArithToEmitCPass.cpp.

◆ createConvertArithToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertArithToSPIRVPass ( )

Definition at line 788 of file ArithToSPIRV.cpp.

◆ createConvertArithToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertArithToSPIRVPass ( ConvertArithToSPIRVPassOptions options)

Definition at line 792 of file ArithToSPIRV.cpp.

◆ createConvertArmNeon2dToIntrPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertArmNeon2dToIntrPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 872 of file ArmNeon2dToIntr.cpp.

◆ createConvertArmSMEToSCFPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertArmSMEToSCFPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1020 of file ArmSMEToSCF.cpp.

◆ createConvertAsyncToLLVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertAsyncToLLVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1098 of file AsyncToLLVM.cpp.

◆ createConvertBufferizationToMemRefPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertBufferizationToMemRefPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1176 of file BufferizationToMemRef.cpp.

◆ createConvertComplexToLibm()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToLibm ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1350 of file ComplexToLibm.cpp.

◆ createConvertComplexToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToLLVMPass ( )

Definition at line 1267 of file ComplexToLLVM.cpp.

◆ createConvertComplexToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToLLVMPass ( ConvertComplexToLLVMPassOptions options)

Definition at line 1271 of file ComplexToLLVM.cpp.

◆ createConvertComplexToROCDLLibraryCalls()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToROCDLLibraryCalls ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1425 of file ComplexToROCDLLibraryCalls.cpp.

◆ createConvertComplexToSPIRVPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToSPIRVPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1500 of file ComplexToSPIRVPass.cpp.

◆ createConvertComplexToStandardPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToStandardPass ( )

Definition at line 1591 of file ComplexToStandard.cpp.

◆ createConvertComplexToStandardPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertComplexToStandardPass ( ConvertComplexToStandardPassOptions options)

Definition at line 1595 of file ComplexToStandard.cpp.

◆ createConvertControlFlowToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertControlFlowToLLVMPass ( )

Definition at line 1686 of file ControlFlowToLLVM.cpp.

◆ createConvertControlFlowToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertControlFlowToLLVMPass ( ConvertControlFlowToLLVMPassOptions options)

Definition at line 1690 of file ControlFlowToLLVM.cpp.

◆ createConvertControlFlowToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertControlFlowToSPIRVPass ( )

Definition at line 1784 of file ControlFlowToSPIRVPass.cpp.

◆ createConvertControlFlowToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertControlFlowToSPIRVPass ( ConvertControlFlowToSPIRVPassOptions options)

Definition at line 1788 of file ControlFlowToSPIRVPass.cpp.

◆ createConvertElementwiseToLinalgPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertElementwiseToLinalgPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 87 of file ElementwiseToLinalg.cpp.

◆ createConvertFuncToEmitC()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertFuncToEmitC ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1867 of file FuncToEmitCPass.cpp.

◆ createConvertFuncToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertFuncToLLVMPass ( )

Definition at line 1957 of file FuncToLLVM.cpp.

◆ createConvertFuncToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertFuncToLLVMPass ( ConvertFuncToLLVMPassOptions options)

Definition at line 1961 of file FuncToLLVM.cpp.

◆ createConvertFuncToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertFuncToSPIRVPass ( )

Definition at line 2055 of file FuncToSPIRVPass.cpp.

◆ createConvertFuncToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertFuncToSPIRVPass ( ConvertFuncToSPIRVPassOptions options)

Definition at line 2059 of file FuncToSPIRVPass.cpp.

◆ createConvertGpuOpsToLLVMSPVOps() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToLLVMSPVOps ( )

Definition at line 2220 of file GPUToLLVMSPV.cpp.

◆ createConvertGpuOpsToLLVMSPVOps() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToLLVMSPVOps ( ConvertGpuOpsToLLVMSPVOpsOptions options)

Definition at line 2224 of file GPUToLLVMSPV.cpp.

◆ createConvertGpuOpsToNVVMOps() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToNVVMOps ( )

Definition at line 2329 of file LowerGpuOpsToNVVMOps.cpp.

◆ createConvertGpuOpsToNVVMOps() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToNVVMOps ( ConvertGpuOpsToNVVMOpsOptions options)

Definition at line 2333 of file LowerGpuOpsToNVVMOps.cpp.

◆ createConvertGpuOpsToROCDLOps() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToROCDLOps ( )

Definition at line 2444 of file LowerGpuOpsToROCDLOps.cpp.

◆ createConvertGpuOpsToROCDLOps() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertGpuOpsToROCDLOps ( ConvertGpuOpsToROCDLOpsOptions options)

Definition at line 2448 of file LowerGpuOpsToROCDLOps.cpp.

◆ createConvertIndexToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertIndexToLLVMPass ( )

Definition at line 2539 of file IndexToLLVM.cpp.

◆ createConvertIndexToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertIndexToLLVMPass ( ConvertIndexToLLVMPassOptions options)

Definition at line 2543 of file IndexToLLVM.cpp.

◆ createConvertIndexToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertIndexToSPIRVPass ( )

Definition at line 2634 of file IndexToSPIRV.cpp.

◆ createConvertIndexToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertIndexToSPIRVPass ( ConvertIndexToSPIRVPassOptions options)

Definition at line 2638 of file IndexToSPIRV.cpp.

◆ createConvertLinalgToAffineLoopsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertLinalgToAffineLoopsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 164 of file Loops.cpp.

◆ createConvertLinalgToLoopsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertLinalgToLoopsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 241 of file Loops.cpp.

◆ createConvertLinalgToParallelLoopsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertLinalgToParallelLoopsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 319 of file Loops.cpp.

◆ createConvertLinalgToStandardPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertLinalgToStandardPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 2718 of file LinalgToStandard.cpp.

◆ createConvertMathToEmitC() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToEmitC ( )

Definition at line 2808 of file MathToEmitCPass.cpp.

◆ createConvertMathToEmitC() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToEmitC ( ConvertMathToEmitCOptions options)

Definition at line 2812 of file MathToEmitCPass.cpp.

◆ createConvertMathToFuncs() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToFuncs ( )

Definition at line 2911 of file MathToFuncs.cpp.

◆ createConvertMathToFuncs() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToFuncs ( ConvertMathToFuncsOptions options)

Definition at line 2915 of file MathToFuncs.cpp.

◆ createConvertMathToLibmPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToLibmPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3091 of file MathToLibm.cpp.

◆ createConvertMathToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToLLVMPass ( )

Definition at line 3006 of file MathToLLVM.cpp.

◆ createConvertMathToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToLLVMPass ( ConvertMathToLLVMPassOptions options)

Definition at line 3010 of file MathToLLVM.cpp.

◆ createConvertMathToROCDL() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToROCDL ( )

Definition at line 3181 of file MathToROCDL.cpp.

◆ createConvertMathToROCDL() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToROCDL ( ConvertMathToROCDLOptions options)

Definition at line 3185 of file MathToROCDL.cpp.

◆ createConvertMathToSPIRVPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToSPIRVPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3264 of file MathToSPIRVPass.cpp.

◆ createConvertMathToXeVM() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToXeVM ( )

Definition at line 3353 of file MathToXeVM.cpp.

◆ createConvertMathToXeVM() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMathToXeVM ( ConvertMathToXeVMOptions options)

Definition at line 3357 of file MathToXeVM.cpp.

◆ createConvertMemRefToEmitC() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMemRefToEmitC ( )

Definition at line 3448 of file MemRefToEmitCPass.cpp.

◆ createConvertMemRefToEmitC() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMemRefToEmitC ( ConvertMemRefToEmitCOptions options)

Definition at line 3452 of file MemRefToEmitCPass.cpp.

◆ createConvertMemRefToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMemRefToSPIRVPass ( )

Definition at line 3546 of file MemRefToSPIRVPass.cpp.

◆ createConvertMemRefToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertMemRefToSPIRVPass ( ConvertMemRefToSPIRVPassOptions options)

Definition at line 3550 of file MemRefToSPIRVPass.cpp.

◆ createConvertNVGPUToNVVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertNVGPUToNVVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3632 of file NVGPUToNVVM.cpp.

◆ createConvertNVVMToLLVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertNVVMToLLVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3707 of file NVVMToLLVM.cpp.

◆ createConvertOpenACCToSCFPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertOpenACCToSCFPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3783 of file OpenACCToSCF.cpp.

◆ createConvertOpenMPToLLVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertOpenMPToLLVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3858 of file OpenMPToLLVM.cpp.

◆ createConvertParallelLoopToGpuPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertParallelLoopToGpuPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4009 of file SCFToGPUPass.cpp.

◆ createConvertPDLToPDLInterpPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertPDLToPDLInterpPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 3933 of file PDLToPDLInterp.cpp.

◆ createConvertSCFToOpenMPPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertSCFToOpenMPPass ( )

Definition at line 4098 of file SCFToOpenMP.cpp.

◆ createConvertSCFToOpenMPPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertSCFToOpenMPPass ( ConvertSCFToOpenMPPassOptions options)

Definition at line 4102 of file SCFToOpenMP.cpp.

◆ createConvertShapeConstraintsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertShapeConstraintsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4283 of file ConvertShapeConstraints.cpp.

◆ createConvertShapeToStandardPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertShapeToStandardPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4358 of file ShapeToStandard.cpp.

◆ createConvertShardToMPIPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertShardToMPIPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4438 of file ShardToMPI.cpp.

◆ createConvertSPIRVToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertSPIRVToLLVMPass ( )

Definition at line 4199 of file SPIRVToLLVMPass.cpp.

◆ createConvertSPIRVToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertSPIRVToLLVMPass ( ConvertSPIRVToLLVMPassOptions options)

Definition at line 4203 of file SPIRVToLLVMPass.cpp.

◆ createConvertTensorToLinalgPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertTensorToLinalgPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4514 of file TensorToLinalgPass.cpp.

◆ createConvertTensorToSPIRVPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertTensorToSPIRVPass ( )

Definition at line 4604 of file TensorToSPIRVPass.cpp.

◆ createConvertTensorToSPIRVPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertTensorToSPIRVPass ( ConvertTensorToSPIRVPassOptions options)

Definition at line 4608 of file TensorToSPIRVPass.cpp.

◆ createConvertToEmitC() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertToEmitC ( )

Definition at line 4699 of file ConvertToEmitCPass.cpp.

◆ createConvertToEmitC() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertToEmitC ( ConvertToEmitCOptions options)

Definition at line 4703 of file ConvertToEmitCPass.cpp.

◆ createConvertToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertToLLVMPass ( )

Definition at line 4800 of file ConvertToLLVMPass.cpp.

◆ createConvertToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertToLLVMPass ( ConvertToLLVMPassOptions options)

Definition at line 4804 of file ConvertToLLVMPass.cpp.

◆ createConvertVectorToAMX()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToAMX ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4888 of file VectorToAMX.cpp.

◆ createConvertVectorToArmSMEPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToArmSMEPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 4964 of file VectorToArmSMEPass.cpp.

◆ createConvertVectorToLLVMPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToLLVMPass ( )

Definition at line 5172 of file ConvertVectorToLLVMPass.cpp.

◆ createConvertVectorToLLVMPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToLLVMPass ( ConvertVectorToLLVMPassOptions options)

Definition at line 5176 of file ConvertVectorToLLVMPass.cpp.

◆ createConvertVectorToSPIRVPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToSPIRVPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5337 of file VectorToSPIRVPass.cpp.

◆ createConvertVectorToXeGPU()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertVectorToXeGPU ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5415 of file VectorToXeGPU.cpp.

◆ createConvertXeGPUToXeVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertXeGPUToXeVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5497 of file XeGPUToXeVM.cpp.

◆ createConvertXeVMToLLVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createConvertXeVMToLLVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5572 of file XeVMToLLVM.cpp.

◆ createFinalizeMemRefToLLVMConversionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createFinalizeMemRefToLLVMConversionPass ( )

Definition at line 5665 of file MemRefToLLVM.cpp.

◆ createFinalizeMemRefToLLVMConversionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createFinalizeMemRefToLLVMConversionPass ( FinalizeMemRefToLLVMConversionPassOptions options)

Definition at line 5669 of file MemRefToLLVM.cpp.

◆ createGpuAsyncRegionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuAsyncRegionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 83 of file AsyncRegionRewriter.cpp.

◆ createGpuDecomposeMemrefsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuDecomposeMemrefsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 160 of file DecomposeMemRefs.cpp.

◆ createGpuEliminateBarriers()

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuEliminateBarriers ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 237 of file EliminateBarriers.cpp.

◆ createGpuKernelOutliningPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuKernelOutliningPass ( )

Definition at line 325 of file KernelOutlining.cpp.

◆ createGpuKernelOutliningPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuKernelOutliningPass ( GpuKernelOutliningPassOptions options)

Definition at line 329 of file KernelOutlining.cpp.

◆ createGpuLaunchSinkIndexComputationsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuLaunchSinkIndexComputationsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 408 of file KernelOutlining.cpp.

◆ createGpuMapParallelLoopsPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuMapParallelLoopsPass ( )

Definition at line 495 of file ParallelLoopMapper.cpp.

◆ createGpuMapParallelLoopsPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuMapParallelLoopsPass ( GpuMapParallelLoopsPassOptions options)

Definition at line 499 of file ParallelLoopMapper.cpp.

◆ createGpuModuleToBinaryPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuModuleToBinaryPass ( )

Definition at line 602 of file ModuleToBinary.cpp.

◆ createGpuModuleToBinaryPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuModuleToBinaryPass ( GpuModuleToBinaryPassOptions options)

Definition at line 606 of file ModuleToBinary.cpp.

◆ createGpuNVVMAttachTarget() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuNVVMAttachTarget ( )

Definition at line 721 of file NVVMAttachTarget.cpp.

◆ createGpuNVVMAttachTarget() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuNVVMAttachTarget ( GpuNVVMAttachTargetOptions options)

Definition at line 725 of file NVVMAttachTarget.cpp.

◆ createGpuROCDLAttachTarget() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuROCDLAttachTarget ( )

Definition at line 852 of file ROCDLAttachTarget.cpp.

◆ createGpuROCDLAttachTarget() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuROCDLAttachTarget ( GpuROCDLAttachTargetOptions options)

Definition at line 856 of file ROCDLAttachTarget.cpp.

◆ createGpuSPIRVAttachTarget() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuSPIRVAttachTarget ( )

Definition at line 968 of file SPIRVAttachTarget.cpp.

◆ createGpuSPIRVAttachTarget() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuSPIRVAttachTarget ( GpuSPIRVAttachTargetOptions options)

Definition at line 972 of file SPIRVAttachTarget.cpp.

◆ createGpuToLLVMConversionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuToLLVMConversionPass ( )

Definition at line 5767 of file GPUToLLVMConversion.cpp.

◆ createGpuToLLVMConversionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuToLLVMConversionPass ( GpuToLLVMConversionPassOptions options)

Definition at line 5771 of file GPUToLLVMConversion.cpp.

◆ createGpuXeVMAttachTarget() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuXeVMAttachTarget ( )

Definition at line 1078 of file XeVMAttachTarget.cpp.

◆ createGpuXeVMAttachTarget() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createGpuXeVMAttachTarget ( GpuXeVMAttachTargetOptions options)

Definition at line 1082 of file XeVMAttachTarget.cpp.

◆ createLiftControlFlowToSCFPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLiftControlFlowToSCFPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5853 of file ControlFlowToSCF.cpp.

◆ createLinalgBlockPackMatmul() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgBlockPackMatmul ( )

Definition at line 428 of file BlockPackMatmul.cpp.

◆ createLinalgBlockPackMatmul() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgBlockPackMatmul ( LinalgBlockPackMatmulOptions options)

Definition at line 432 of file BlockPackMatmul.cpp.

◆ createLinalgDetensorizePass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgDetensorizePass ( )

Definition at line 523 of file Detensorize.cpp.

◆ createLinalgDetensorizePass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgDetensorizePass ( LinalgDetensorizePassOptions options)

Definition at line 527 of file Detensorize.cpp.

◆ createLinalgElementwiseOpFusionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgElementwiseOpFusionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 608 of file ElementwiseOpFusion.cpp.

◆ createLinalgFoldIntoElementwisePass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgFoldIntoElementwisePass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 683 of file FoldIntoElementwise.cpp.

◆ createLinalgFoldUnitExtentDimsPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgFoldUnitExtentDimsPass ( )

Definition at line 772 of file DropUnitDims.cpp.

◆ createLinalgFoldUnitExtentDimsPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgFoldUnitExtentDimsPass ( LinalgFoldUnitExtentDimsPassOptions options)

Definition at line 776 of file DropUnitDims.cpp.

◆ createLinalgGeneralizeNamedOpsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgGeneralizeNamedOpsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 855 of file Generalization.cpp.

◆ createLinalgInlineScalarOperandsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgInlineScalarOperandsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 930 of file InlineScalarOperands.cpp.

◆ createLinalgMorphOpsPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgMorphOpsPass ( )

Definition at line 1026 of file MorphOps.cpp.

◆ createLinalgMorphOpsPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgMorphOpsPass ( LinalgMorphOpsPassOptions options)

Definition at line 1030 of file MorphOps.cpp.

◆ createLinalgSpecializeGenericOpsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLinalgSpecializeGenericOpsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1109 of file Specialize.cpp.

◆ createLocationSnapshot() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLocationSnapshot ( )

Definition at line 664 of file LocationSnapshot.cpp.

◆ createLocationSnapshot() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createLocationSnapshot ( LocationSnapshotOptions options)

Definition at line 668 of file LocationSnapshot.cpp.

◆ createLowerAffinePass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLowerAffinePass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 5931 of file AffineToStandard.cpp.

◆ createLowerHostCodeToLLVMPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createLowerHostCodeToLLVMPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6006 of file ConvertLaunchFuncToLLVMCalls.cpp.

◆ createMathToAPFloatConversionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createMathToAPFloatConversionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6151 of file MathToAPFloat.cpp.

◆ createMem2Reg() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createMem2Reg ( )

Definition at line 885 of file Mem2Reg.cpp.

◆ createMem2Reg() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createMem2Reg ( Mem2RegOptions options)

Definition at line 889 of file Mem2Reg.cpp.

◆ createOptReductionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createOptReductionPass ( )

Definition at line 92 of file OptReductionPass.cpp.

◆ createOptReductionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createOptReductionPass ( OptReductionPassOptions options)

Definition at line 96 of file OptReductionPass.cpp.

◆ createOutlineShapeComputationPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createOutlineShapeComputationPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 75 of file OutlineShapeComputation.cpp.

◆ createReconcileUnrealizedCastsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createReconcileUnrealizedCastsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6226 of file ReconcileUnrealizedCasts.cpp.

◆ createReductionTreePass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createReductionTreePass ( )

Definition at line 193 of file ReductionTreePass.cpp.

◆ createReductionTreePass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createReductionTreePass ( ReductionTreePassOptions options)

Definition at line 197 of file ReductionTreePass.cpp.

◆ createRemoveShapeConstraintsPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createRemoveShapeConstraintsPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 150 of file RemoveShapeConstraints.cpp.

◆ createSCFForLoopPeeling() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFForLoopPeeling ( )

Definition at line 163 of file LoopSpecialization.cpp.

◆ createSCFForLoopPeeling() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFForLoopPeeling ( SCFForLoopPeelingOptions options)

Definition at line 167 of file LoopSpecialization.cpp.

◆ createSCFToControlFlowPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFToControlFlowPass ( )

Definition at line 6313 of file SCFToControlFlow.cpp.

◆ createSCFToControlFlowPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFToControlFlowPass ( SCFToControlFlowPassOptions options)

Definition at line 6317 of file SCFToControlFlow.cpp.

◆ createSCFToEmitC()

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFToEmitC ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6396 of file SCFToEmitC.cpp.

◆ createSCFToSPIRV()

std::unique_ptr<::mlir::Pass > mlir::impl::createSCFToSPIRV ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6471 of file SCFToSPIRVPass.cpp.

◆ createSetLLVMModuleDataLayoutPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSetLLVMModuleDataLayoutPass ( )

Definition at line 6558 of file FuncToLLVM.cpp.

◆ createSetLLVMModuleDataLayoutPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createSetLLVMModuleDataLayoutPass ( SetLLVMModuleDataLayoutPassOptions options)

Definition at line 6562 of file FuncToLLVM.cpp.

◆ createShapeToShapeLoweringPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createShapeToShapeLoweringPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 225 of file ShapeToShapeLowering.cpp.

◆ createSimplifyDepthwiseConvPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createSimplifyDepthwiseConvPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 1185 of file SimplifyDepthwiseConv.cpp.

◆ createSROA()

std::unique_ptr<::mlir::Pass > mlir::impl::createSROA ( )

Definition at line 1240 of file SROA.cpp.

◆ createTosaToArithPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createTosaToArithPass ( )

Definition at line 6656 of file TosaToArithPass.cpp.

◆ createTosaToArithPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createTosaToArithPass ( TosaToArithPassOptions options)

Definition at line 6660 of file TosaToArithPass.cpp.

◆ createTosaToMLProgram()

std::unique_ptr<::mlir::Pass > mlir::impl::createTosaToMLProgram ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6880 of file TosaToMLProgramPass.cpp.

◆ createTosaToSCFPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createTosaToSCFPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 6955 of file TosaToSCFPass.cpp.

◆ createTosaToTensorPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createTosaToTensorPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 7030 of file TosaToTensorPass.cpp.

◆ createUBToLLVMConversionPass() [1/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createUBToLLVMConversionPass ( )

Definition at line 7117 of file UBToLLVM.cpp.

◆ createUBToLLVMConversionPass() [2/2]

std::unique_ptr<::mlir::Pass > mlir::impl::createUBToLLVMConversionPass ( UBToLLVMConversionPassOptions options)

Definition at line 7121 of file UBToLLVM.cpp.

◆ createUBToSPIRVConversionPass()

std::unique_ptr<::mlir::Pass > mlir::impl::createUBToSPIRVConversionPass ( )

We declare an explicit private instantiation because Pass classes should only be visible by the current library.

Definition at line 7200 of file UBToSPIRV.cpp.

◆ ensureRegionTerminator() [1/2]

void mlir::impl::ensureRegionTerminator ( Region & region,
Builder & builder,
Location loc,
function_ref< Operation *(OpBuilder &, Location)> buildTerminatorOp )

Create a simple OpBuilder and forward to the OpBuilder version of this function.

Definition at line 1420 of file Operation.cpp.

References ensureRegionTerminator(), and mlir::Builder::getContext().

◆ ensureRegionTerminator() [2/2]

void mlir::impl::ensureRegionTerminator ( Region & region,
OpBuilder & builder,
Location loc,
function_ref< Operation *(OpBuilder &, Location)> buildTerminatorOp )

Insert an operation, generated by buildTerminatorOp, at the end of the region's only block if it does not have a terminator already.

If the region is empty, insert a new block first. buildTerminatorOp should return the terminator operation to insert.

Definition at line 1403 of file Operation.cpp.

References mlir::Block::back(), mlir::Region::back(), mlir::OpBuilder::createBlock(), mlir::Block::empty(), mlir::Region::empty(), mlir::Operation::hasTrait(), mlir::OpBuilder::insert(), and mlir::OpBuilder::setInsertionPointToEnd().

Referenced by ensureRegionTerminator(), mlir::OpTrait::SingleBlockImplicitTerminator< TerminatorOpType >::Impl< ConcreteType >::ensureTerminator(), and mlir::OpTrait::SingleBlockImplicitTerminator< TerminatorOpType >::Impl< ConcreteType >::ensureTerminator().

◆ findAttrSorted() [1/2]

template<typename IteratorT>
std::pair< IteratorT, bool > mlir::impl::findAttrSorted ( IteratorT first,
IteratorT last,
StringAttr name )

StringAttr lookups on large attribute lists will switch to string binary search.

String binary searches become significantly faster than linear scans with the identifier when the attribute list becomes very large.

Definition at line 763 of file OperationSupport.h.

References findAttrSorted(), and findAttrUnsorted().

◆ findAttrSorted() [2/2]

template<typename IteratorT>
std::pair< IteratorT, bool > mlir::impl::findAttrSorted ( IteratorT first,
IteratorT last,
StringRef name )

Using llvm::lower_bound requires an extra string comparison to check whether the returned iterator points to the found element or whether it indicates the lower bound.

Skip this redundant comparison by checking if compare == / 0 during the binary search.

Definition at line 739 of file OperationSupport.h.

Referenced by findAttrSorted(), getAttrFromSortedRange(), and getNamedAttrFromSortedRange().

◆ findAttrUnsorted()

template<typename IteratorT, typename NameT>
std::pair< IteratorT, bool > mlir::impl::findAttrUnsorted ( IteratorT first,
IteratorT last,
NameT name )

Unsorted string search or identifier lookups are linear scans.

Definition at line 726 of file OperationSupport.h.

Referenced by findAttrSorted().

◆ foldCastInterfaceOp()

LogicalResult mlir::impl::foldCastInterfaceOp ( Operation * op,
ArrayRef< Attribute > attrOperands,
SmallVectorImpl< OpFoldResult > & foldResults )

Attempt to fold the given cast operation.

Definition at line 22 of file CastInterfaces.cpp.

References foldCastInterfaceOp(), mlir::Operation::getOperands(), mlir::Operation::getResults(), mlir::OperandRange::getTypes(), mlir::ResultRange::getTypes(), and success().

Referenced by foldCastInterfaceOp().

◆ getAttrFromSortedRange()

template<typename IteratorT, typename NameT>
Attribute mlir::impl::getAttrFromSortedRange ( IteratorT first,
IteratorT last,
NameT name )

Get an attribute from a sorted range of named attributes.

Returns null if the attribute was not found.

Definition at line 774 of file OperationSupport.h.

References findAttrSorted(), and result.

◆ getDataLayoutSpec()

DataLayoutSpecInterface mlir::impl::getDataLayoutSpec ( Operation * op)

Definition at line 26 of file Traits.cpp.

References mlir::Operation::getAttrOfType().

Referenced by mlir::HasDefaultDLTIDataLayout< ConcreteOp >::getDataLayoutSpec().

◆ getNamedAttrFromSortedRange()

template<typename IteratorT, typename NameT>
std::optional< NamedAttribute > mlir::impl::getNamedAttrFromSortedRange ( IteratorT first,
IteratorT last,
NameT name )

Get an attribute from a sorted range of named attributes.

Returns std::nullopt if the attribute was not found.

Definition at line 783 of file OperationSupport.h.

References findAttrSorted(), and result.

◆ getTargetSystemSpec()

TargetSystemSpecInterface mlir::impl::getTargetSystemSpec ( Operation * op)

Definition at line 31 of file Traits.cpp.

References mlir::Operation::getAttrOfType().

Referenced by mlir::HasDefaultDLTIDataLayout< ConcreteOp >::getTargetSystemSpec().

◆ parseOptionalVisibilityKeyword()

ParseResult mlir::impl::parseOptionalVisibilityKeyword ( OpAsmParser & parser,
NamedAttrList & attrs )

Parse an optional visibility attribute keyword (i.e., public, private, or nested) without quotes in a string attribute named 'attrName'.

Definition at line 1124 of file SymbolTable.cpp.

References mlir::AsmParser::getBuilder(), mlir::Builder::getNamedAttr(), mlir::Builder::getStringAttr(), mlir::SymbolTable::getVisibilityAttrName(), mlir::AsmParser::parseOptionalKeyword(), mlir::NamedAttrList::push_back(), and success().

Referenced by mlir::function_interface_impl::parseFunctionOp().

◆ scalarizeVectorOp()

LogicalResult mlir::impl::scalarizeVectorOp ( Operation * op,
ValueRange operands,
ConversionPatternRewriter & rewriter,
const LLVMTypeConverter & converter )

Unrolls op to array/vector elements.

Definition at line 649 of file GPUOpsLowering.cpp.

References mlir::Operation::getResultTypes(), mlir::LLVM::detail::handleMultidimensionalVectors(), scalarizeVectorOpHelper(), and success().

Referenced by mlir::ScalarizeVectorOpLowering< SourceOp >::matchAndRewrite().

◆ verifyCastInterfaceOp()

LogicalResult mlir::impl::verifyCastInterfaceOp ( Operation * op)

Attempt to verify the given cast operation.

Definition at line 39 of file CastInterfaces.cpp.

References diag(), mlir::Operation::emitOpError(), mlir::Operation::getOperandTypes(), mlir::Operation::getResultTypes(), success(), and verifyCastInterfaceOp().

Referenced by verifyCastInterfaceOp().

◆ verifyHasDefaultDLTIDataLayoutTrait()

LogicalResult mlir::impl::verifyHasDefaultDLTIDataLayoutTrait ( Operation * op)

Definition at line 15 of file Traits.cpp.

References success().

Referenced by mlir::HasDefaultDLTIDataLayout< ConcreteOp >::verifyTrait().

Generated on for MLIR by doxygen 1.14.0