MLIR 22.0.0git
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
[detail level 123456]
 Nimpl
 NllvmThe OpAsmOpInterface, see OpAsmInterface.td for more details
 NmlirInclude the generated interface declarations
 Nnanobind
 Npybind11
 Nstd
 CAllocaScopeHoisterMove allocations into an allocation scope, if it is legal to move them (e.g
 CAllocaScopeInlinerInline an AllocaScopeOp if either the direct parent is an allocation scope or it contains no allocation
 CAmdgpuMaskedloadToLoadPass
 CAncestorKeyStores the "key" associated with an ancestor
 CApplyConversionActionThis is the type of Action that is dispatched when a conversion is applied
 CAssertOpLowering
 CBaseTrait
 Cbasic_parser
 CBatchMatmulOp
 Cbf16
 CBinaryArithOpToAPFloatConversionRewrite a binary arithmetic operation to an APFloat function call
 CBreakDownVectorReductionExample:
 CBufferizationOptions
 CCanonicalizeAffineMinMaxOpExprAndTermOrderCanonicalize the affine map result expression order of an affine min/max operation
 CCanonicalizeContractAddReturn a fused vector::ContractionOp which represents a patterns such as:
 CCanonicalizeContractMatmulToMMTCanonicalization of a vector.contraction a, b, c with row-major matmul semantics to a contraction suitable for MMT (matrix matrix multiplication with the RHS transposed) lowering
 CCanonicalizeSingleResultAffineMinMaxOp
 CCastInfo
 CChainedReductionPattern to fold chained reduction to a series of vector additions and a final reduction. This form should require fewer subgroup operations
 CClampClampOptimization
 CClampFOpConversion
 CClampIsNoOp
 CCmpFIntToFPConst
 CCmpFOpToAPFloatConversion
 CCollapseShapeOpMemRefCastFolder
 CCommutativeOperandStores a commutative operand along with its BFS traversal information
 CConcatOptimization
 CConcatSliceOptimization
 CConsolidateTransposeOptimization
 CConstStrippingForwardingCast
 CContractOpToElementwiseLower vector.contract with all size one reduction dimensions to elementwise ops when possible
 CConversionPattern
 CConversionTarget
 CConvert1DExtractStridedSliceIntoExtractInsertChainFor a 1-D ExtractStridedSlice, breaks it down into a chain of Extract ops to extract each element from the source, and then a chain of Insert ops to insert to the target vector
 CConvert1DExtractStridedSliceIntoShuffleRewritePattern for ExtractStridedSliceOp where source and destination vectors are 1-D
 CConvertAccessChain
 CConvertAddressOf
 CConvertAliasResource
 CConvertLoad
 CConvertMathToROCDLPass
 CConvertNativeFuncPatternConvert math ops marked with fast (afn) to native OpenCL intrinsics
 CConvertSameRankInsertStridedSliceIntoShuffleRewritePattern for InsertStridedSliceOp where source and destination vectors have the same rank
 CConvertStore
 CConvertVariable
 CCopyBuilderHelper to create the tma operations corresponding to linalg::CopyOp
 CDecomposeDifferentRankInsertStridedSliceRewritePattern for InsertStridedSliceOp where source and destination vectors have different ranks
 CDecomposeNDExtractStridedSliceRewritePattern for ExtractStridedSliceOp where the source vector is n-D
 CDeduplicateAffineMinMaxExpressionsRemove duplicated expressions in affine min/max ops
 CDefaultDoCastIfPossible
 CDenseIntOrFPElementsAttr
 CDerivedTy
 CDescriptorSetInfoStruct containing information regarding a descriptor set
 CDropInnerMostUnitDimsTransferReadDrop inner most contiguous unit dimensions from transfer_read operand
 CDropInnerMostUnitDimsTransferWriteDrop inner most contiguous unit dimensions from transfer_write operand. E.g., vector.transfer_write arg1, arg0[c0, arg2, c0, c0, c0] {in_bounds = [true, true, true, true, true]} : vector<1x16x16x1x1xf32>, memref<1x512x16x1x1xf32>
 CDropUnitDimFromElementwiseOpsFor vectors with at least one unit dim, replaces: elementwise(a, b) with: sc_a = shape_cast(a) sc_b = shape_cast(b) res = elementwise(sc_a, sc_b) return shape_cast(res) The newly inserted shape_cast Ops fold (before elementwise Op) and then restore (after elementwise Op) the unit dim. Vectors a and b are required to be rank > 1
 CDropUnitDimsFromScfForOpA pattern to drop unit dims from the iter_args of an scf.for
 CDropUnitDimsFromTransposeOpA pattern to drop unit dims from vector.transpose
 CDynamicEventPool
 CDynamicMemRefIteratorIterate over all elements in a dynamic memref
 CDynamicMemRefType
 CExecuteRegionForwardingEliminator
 CExpected
 Cf16
 CFileLineColRange
 CFoldAddIntoDestReplace a linalg.add with one operand the single user of a contraction, which has a zero-filled, "identity-mapped" destination and is dominated by the other operand, by the contraction with other as its dest
 CFoldArithExtIntoContractionOpPattern to fold arithmetic extensions on floating point data types into vector contraction operations. linalg.matmul introduces arithmetic extensions on its operands. Please mlir snippets below for more details
 CFoldBroadcastsFold back-to-back broadcasts together
 CFoldConstantCase
 CFoldConvertLayoutOp
 CFoldLaunchArgumentsSimplify the gpu.launch when the range of a thread or block ID is trivially known to be one
 CFoldTensorCastProducerOpFolds a tensor.cast op into a consuming DestinationStyleOpInterface op if the tensor.cast has source that is more static than the consuming op
 CFoldTransposeWithTransposeFold transpose with transpose
 CForallRewriteResultStruct to return the result of the rewrite of a forall operation
 CFpToFpConversion
 CFpToIntConversion
 CFromElementsToShapeCastRewrite from_elements on multiple scalar extracts as a shape_cast on a single extract
 CFusedLoc
 CGenericOptionValue
 CGPUAllReduceConversionPattern to convert a gpu.all_reduce op into a SPIR-V group op
 CGPUSubgroupReduceConversionPattern to convert a gpu.subgroup_reduce op into a SPIR-V group op
 CHasAffineDimExprVisitorVisitor to check if any of the given set of positions from AffineDimExprs are used within an AffineExpr
 CHopperBuilderHelper to create the base Hopper-specific operations that are reused in various other places
 CIAddCarryFold
 Cilist_alloc_traits
 CIndexSwitchOpLowering
 CInputAndOutputIndices
 CInsertSliceOfInsertSliceFolder
 CIntToFpConversion
 CjEliminates variable at the specified position using Fourier-Motzkin variable elimination
 CL0RTContextWrapper
 CMatmulOp
 CMaxPool2dIsNoOp
 CMemoryRequirements
 CMemRefDescriptor
 CMergeAffineMinMaxOpMerge an affine min/max op to its consumers if its consumer is also an affine min/max op
 CMergeArithBitcast
 CMergeComplexBitcast
 CMLIRConvertBytecodeParamsThis class represents the parameters used when converting between MLIR's bytecode and textual format
 CMLIRConvertBytecodeResultThis class represents the result of converting between MLIR's bytecode and textual format
 CMlirDiagnosticAn opaque reference to a diagnostic, always owned by the diagnostics engine (context)
 CMlirDialectHandle
 CMlirDialectRegistrationHooksStructure of dialect registration hooks
 CMlirLinalgContractionDimensions
 CMlirLinalgConvolutionDimensions
 CMlirLogicalResultA logical result value, essentially a boolean with named states
 CMlirNamedAttributeNamed MLIR attribute
 CMlirRewritePatternCallbacksRewritePattern API
 CMlirStringRefA pointer to a sized fragment of a string, not necessarily null-terminated
 CMlirTypeFromLLVMIRTranslator
 CMlirTypeToLLVMIRTranslator
 CMmaSyncBuilderHelper struct to provide a simple mapping from matmul operations to the corresponding mma.sync operation
 CMulExtendedFold
 CMultiBlockExecuteInliner
 CNegFOpToAPFloatConversion
 CNodeAccess
 CNullableValueCastFailed
 CNumWorkGroupsStruct containing the number of local workgroups to dispatch for each dimension
 COpComplianceInfo
 COptionValueBase
 COuterProductOpLoweringProgressive lowering of OuterProductOp
 CPadSliceOptimization
 CPassPipelineOptions
 CPrivateVarsInfoA util to collect info needed to convert delayed privatizers from MLIR to LLVM
 CProfileInfoDepot
 CPyAttrBuilderMap
 CPyGlobalDebugFlagWrapper for the global LLVM debugging flag
 CPyOpResultPython wrapper for MlirOpResult
 CPyOpResultListA list of operation results
 Craw_ostream
 CReduceRedundantZeroPattern to eliminate redundant zero-constants added to reduction operands. It's enough for there to be one initial zero value, so we can eliminate the extra ones that feed into vector.reduction <add>. These get created by the ChainedReduction pattern
 CReifyCollapseShapeOp
 CRemoveUnusedLvlCrds
 CRowColIndexingHelper struct to encode a pair of row/column indexings in the form of affine expressions
 CScopedContext
 CSelectToExtUI
 CShapeDimensionImplements a simple high-level fusion pass on linalg structured operations
 CShardingPropagation
 CSimplifyAffineMinMaxPassCreates a simplification pass for affine min/max/apply
 CSincosOpLowering
 CSingleBlockExecuteInliner
 CSinkVectorProducerOpsSink vector producers forward to reduce live ranges
 CSliceCanonicalizerA canonicalizer wrapper to replace ExtractSliceOps
 CSliceDynamicSizeCanonicalization
 CSliceReturnTypeCanonicalizerReturn the canonical type of the result of an extract_slice op
 CSmallDenseMap
 CSmallPtrSetImpl
 CSmallVector
 Csme_state
 CSortCommutativeOperandsSorts the operands of op in ascending order of the "key" associated with each operand iff op is commutative
 CSpecifierGetterSetterOpConverter
 CStorageSpecifierGetOpConverter
 CStorageSpecifierInitOpConverter
 CStorageSpecifierSetOpConverter
 CStreamWrapper
 CStridedMemrefIteratorIterate over all elements in a strided memref
 CStridedMemrefIterator< T, 0 >Iterate over all elements in a 0-ranked strided memref
 CStridedMemRefTypeStridedMemRef descriptor type with static rank
 CStridedMemRefType< T, 0 >StridedMemRef descriptor type specialized for rank 0
 CStridedMemRefType< T, 1 >StridedMemRef descriptor type specialized for rank 1
 CStrideInfo
 CSubViewCanonicalizerA canonicalizer wrapper to replace SubViewOps
 CSubViewReturnTypeCanonicalizerReturn the canonical type of the result of a subview
 CSwapExtractSliceOfFillSwaps: tensor.extract_slice(linalg.fill(cst, init)) with: linalg.fill(cst, tensor.extract_slice(init))
 CSwapTransposeWithBroadcastThis pattern canonicalize transpose by swapping the order of broadcast and transpose: transpose(broadcast(input)) -> broadcast(transpose(input))
 CTargetToDataLayoutPass
 CTargetToTargetFeaturesPass
 CToElementsOfBroadcastCanonicalize vector.to_elements(vector.broadcast(v)) where v is a vector
 CTosaProfileCompliance
 CTranslationOptions
 CTransposeIsReshape
 CTransposeOpLoweringRewrite AVX2-specific vector.transpose, for the supported cases and depending on the TransposeLoweringOptions
 CTuneExtension
 CTypeConverter
 CTypeInfo
 CUModSimplification
 CUMulExtendedOpXOne
 CUnitExtentReplacementInfoCompute the modified metadata for an operands of operation whose unit dims are being dropped
 CUnrankedMemRefType
 CValueEquivalenceCache
 CValueFromPointerCast
 CVector
 CVector< T, Dim >
 CVectorizationHookResultVectorizationHookResult contains the vectorized op returned from a CustomVectorizationHook
 CVectorizationStateContains the vectorization state and related methods used across the vectorization process of a given operation
 CVectorShape
 CVulkanDeviceMemoryBufferStruct containing information regarding to a device memory buffer
 CVulkanHostMemoryBufferStruct containing information regarding to a host memory buffer
 CVulkanRuntimeVulkan runtime
 CWhileLowering
 CWrapFuncInClass
 CYieldOpLowering
 CZeCommandListDeleter
 CZeContextDeleter
 CZeEventDeleter
 CZeEventPoolDeleter
Generated on for MLIR by doxygen 1.14.0