Home
last modified time | relevance | path

Searched +full:depth +full:- +full:wise (Results 1 – 25 of 30) sorted by relevance

12

/llvm-project/mlir/test/lib/Analysis/
H A DTestMemRefDependenceCheck.cpp1 //===- TestMemRefDependenceCheck.cpp - Test dep analysis ------------------===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
9 // This file implements a pass to run pair-wise memref access dependence checks.
11 //===----------------------------------------------------------------------===//
21 #define DEBUG_TYPE "test-memref-dependence-check"
28 // TODO: Add common surrounding loop depth-wise dependence checks.
34 StringRef getArgument() const final { return "test-memref-dependence-check"; } in MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID()
55 std::string lbStr = "-inf"; in getDirectionVectorStr()
90 srcOpInst->emitError("dependence check failed"); in checkDependences()
[all …]
/llvm-project/mlir/python/mlir/dialects/linalg/opdsl/ops/
H A Dcore_named_ops.py398 includes zero-point adjustments for the left and right operands of the
402 C[D.m, D.n] += (TypeFn.cast_signed(U, A[D.m, D.k]) - TypeFn.cast_signed(U, AZp)) * (
403 TypeFn.cast_signed(U, B[D.k, D.n]) - TypeFn.cast_signed(U, BZp)
449 """Performs a matrix-matrix-transpose multiplication of two 4D inputs.
472 """Performs a batched matrix-matrix-transpose multiplication of two
473 batched-4D (5D) inputs.
477 non-batch dimensions are the same as linalg.mmt4d vs. linalg.matmul. See the
512 has its non-batc
[all...]
/llvm-project/mlir/include/mlir/Dialect/Affine/Analysis/
H A DUtils.h1 //===- Utils.h - General analysis utilities ---------------------*- C++ -*-===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
10 // memref's and non-loop IR structures. These are not passes by themselves but
14 //===----------------------------------------------------------------------===//
50 // top-level operations in a `Block` which contain load/store ops, and edges
61 // The top-level statement which is (or contains) a load/store.
100 // non-memref value, then the dependence is between a graph node which
142 // is for 'value' if non-null, or for any value otherwise. Returns false
161 // Returns the output edge count for node 'id' and 'memref' (if non-null),
[all …]
/llvm-project/mlir/include/mlir/Dialect/Affine/
H A DLoopUtils.h1 //===- LoopUtils.h - Loop transformation utilities --------
[all...]
H A DPasses.td1 //===-- Passes.td - Affine pass definition file -------
[all...]
/llvm-project/mlir/lib/Dialect/Affine/Transforms/
H A DAffineDataCopyGeneration.cpp1 //===- AffineDataCopyGeneration.cpp - Explicit memref copying pass ------*-===//
5 // SPDX-Licens
[all...]
H A DLoopFusion.cpp1 //===- LoopFusion.cpp - Code to perform loop fusion --------
[all...]
/llvm-project/mlir/include/mlir/Dialect/Linalg/IR/
H A DLinalgNamedStructuredOps.yaml3 --- !LinalgOpConfig
7 doc: |-
13 - hasCanonicalizer
16 - !LinalgOperandDefConfig
20 shape_map: affine_map<() -> ()>
21 - !LinalgOperandDefConfig
25 shape_map: affine_map<() -> ()>
26 - !LinalgOperandDefConfi
[all...]
/llvm-project/llvm/include/llvm/CodeGen/GlobalISel/
H A DUtils.h1 //==-- llvm/CodeGen/GlobalISel/Utils.h --------
[all...]
/llvm-project/polly/lib/External/isl/
H A Disl_ast_codegen.c2 * Copyright 2012-2014 Ecole Normale Superieure
9 * and Inria Paris - Rocquencourt, Domaine de Voluceau - Rocquencourt,
10 * B.P. 105 - 78153 Le Chesnay, France
77 * S -> D
81 * [S -> D] -> D
95 build = isl_ast_build_copy(data->build); in generate_non_single_valued()
103 data->list = isl_ast_graft_list_concat(data->list, list); in generate_non_single_valued()
116 if (!build->at_each_domain) in at_each_domain()
125 graft->node = build->at_each_domain(graft->node, in at_each_domain()
126 build, build->at_each_domain_user); in at_each_domain()
[all …]
/llvm-project/clang/include/clang/Analysis/FlowSensitive/
H A DDataflowEnvironment.h1 //===-- DataflowEnvironment.h -----------------------------------*- C++ -*-===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
10 // that run over Control-Flow Graphs (CFGs) to keep track of the state of the
13 //===----------------------------------------------------------------------===//
52 /// Non-null pointer to a potentially widened version of the input value.
67 /// Supplements `Environment` with non-standard comparison and join
113 /// This function may widen the current value -- replace it with an
124 /// * A non-null `Value *` that points either to `Current` or a widened
184 // Copy-constructor is private, Environments should not be copied. See fork().
[all …]
/llvm-project/llvm/docs/
H A DSPIRVUsage.rst2 User Guide for SPIR-V Target
14 The SPIR-V target provides code generation for the SPIR-V binary format described
15 in `the official SPIR-V specification <https://www.khronos.org/registry/SPIR-V/>`_.
20 The SPIR-V backend can be invoked either from LLVM's Static Compiler (llc) or Clang,
22 sources directly to SPIR-V. This section outlines the usage of various commands to
23 leverage the SPIR-V backend for different purposes.
26 --
[all...]
H A DLangRef.rst7 :depth: 3
14 type safety, low-level operations, flexibility, and the capability of
15 representing 'all' high-level languages cleanly. It is the common code
23 forms: as an in-memory compiler IR, as an on-disk bitcode representation
24 (suitable for fast loading by a Just-In-Time compiler), and as a human
32 The LLVM representation aims to be light-weight and low-level while
35 high-leve
[all...]
/llvm-project/polly/lib/Analysis/
H A DScopDetection.cpp1 //===- ScopDetection.cpp - Detect Scops --------
[all...]
/llvm-project/llvm/lib/Analysis/
H A DScalarEvolution.cpp1 //===- ScalarEvolution.cpp - Scalar Evolution Analysis --------
603 CompareValueComplexity(EquivalenceClasses<const Value * > & EqCacheValue,const LoopInfo * const LI,Value * LV,Value * RV,unsigned Depth) CompareValueComplexity() argument
684 CompareSCEVComplexity(EquivalenceClasses<const SCEV * > & EqCacheSCEV,EquivalenceClasses<const Value * > & EqCacheValue,const LoopInfo * const LI,const SCEV * LHS,const SCEV * RHS,DominatorTree & DT,unsigned Depth=0) CompareSCEVComplexity() argument
1002 getLosslessPtrToIntExpr(const SCEV * Op,unsigned Depth) getLosslessPtrToIntExpr() argument
1136 getTruncateExpr(const SCEV * Op,Type * Ty,unsigned Depth) getTruncateExpr() argument
1325 getPreStartForExtend(const SCEVAddRecExpr * AR,Type * Ty,ScalarEvolution * SE,unsigned Depth) getPreStartForExtend() argument
1403 getExtendAddRecStart(const SCEVAddRecExpr * AR,Type * Ty,ScalarEvolution * SE,unsigned Depth) getExtendAddRecStart() argument
1551 getZeroExtendExpr(const SCEV * Op,Type * Ty,unsigned Depth) getZeroExtendExpr() argument
1571 getZeroExtendExprImpl(const SCEV * Op,Type * Ty,unsigned Depth) getZeroExtendExprImpl() argument
1886 getSignExtendExpr(const SCEV * Op,Type * Ty,unsigned Depth) getSignExtendExpr() argument
1906 getSignExtendExprImpl(const SCEV * Op,Type * Ty,unsigned Depth) getSignExtendExprImpl() argument
2514 getAddExpr(SmallVectorImpl<const SCEV * > & Ops,SCEV::NoWrapFlags OrigFlags,unsigned Depth) getAddExpr() argument
3109 getMulExpr(SmallVectorImpl<const SCEV * > & Ops,SCEV::NoWrapFlags OrigFlags,unsigned Depth) getMulExpr() argument
4679 getMinusSCEV(const SCEV * LHS,const SCEV * RHS,SCEV::NoWrapFlags Flags,unsigned Depth) getMinusSCEV() argument
4728 getTruncateOrZeroExtend(const SCEV * V,Type * Ty,unsigned Depth) getTruncateOrZeroExtend() argument
4740 getTruncateOrSignExtend(const SCEV * V,Type * Ty,unsigned Depth) getTruncateOrSignExtend() argument
6616 getRangeRef(const SCEV * S,ScalarEvolution::RangeSignHint SignHint,unsigned Depth) getRangeRef() argument
9502 getConstantEvolvingPHIOperands(Instruction * UseInst,const Loop * L,DenseMap<Instruction *,PHINode * > & PHIMap,unsigned Depth) getConstantEvolvingPHIOperands() argument
10663 SimplifyICmpOperands(ICmpInst::Predicate & Pred,const SCEV * & LHS,const SCEV * & RHS,unsigned Depth) SimplifyICmpOperands() argument
12126 isImpliedViaMerge(ICmpInst::Predicate Pred,const SCEV * LHS,const SCEV * RHS,const SCEV * FoundLHS,const SCEV * FoundRHS,unsigned Depth) isImpliedViaMerge() argument
12382 isImpliedViaOperations(ICmpInst::Predicate Pred,const SCEV * LHS,const SCEV * RHS,const SCEV * FoundLHS,const SCEV * FoundRHS,unsigned Depth) isImpliedViaOperations() argument
[all...]
H A DLazyValueInfo.cpp1 //===- LazyValueInfo.cpp - Value constraint analysis --------
1248 getValueFromCondition(Value * Val,Value * Cond,bool IsTrueDest,bool UseBlockValue,unsigned Depth) getValueFromCondition() argument
[all...]
/llvm-project/mlir/lib/Dialect/Affine/Utils/
H A DLoopUtils.cpp1 //===- LoopUtils.cpp ---- Misc utilities for loop transformation -----
[all...]
/llvm-project/llvm/lib/Transforms/InstCombine/
H A DInstCombineShifts.cpp1 //===- InstCombineShifts.cpp ---------
[all...]
/llvm-project/llvm/lib/Target/X86/
H A DX86ISelLowering.cpp1 //===-- X86ISelLowering.cpp - X86 DAG Lowering Implementation -------
5832 getFauxShuffleMask(SDValue N,const APInt & DemandedElts,SmallVectorImpl<int> & Mask,SmallVectorImpl<SDValue> & Ops,const SelectionDAG & DAG,unsigned Depth,bool ResolveKnownElts) getFauxShuffleMask() argument
6349 getTargetShuffleInputs(SDValue Op,const APInt & DemandedElts,SmallVectorImpl<SDValue> & Inputs,SmallVectorImpl<int> & Mask,APInt & KnownUndef,APInt & KnownZero,const SelectionDAG & DAG,unsigned Depth,bool ResolveKnownElts) getTargetShuffleInputs() argument
6374 getTargetShuffleInputs(SDValue Op,const APInt & DemandedElts,SmallVectorImpl<SDValue> & Inputs,SmallVectorImpl<int> & Mask,const SelectionDAG & DAG,unsigned Depth,bool ResolveKnownElts) getTargetShuffleInputs() argument
6383 getTargetShuffleInputs(SDValue Op,SmallVectorImpl<SDValue> & Inputs,SmallVectorImpl<int> & Mask,const SelectionDAG & DAG,unsigned Depth=0,bool ResolveKnownElts=true) getTargetShuffleInputs() argument
6422 getShuffleScalarElt(SDValue Op,unsigned Index,SelectionDAG & DAG,unsigned Depth) getShuffleScalarElt() argument
27313 unsigned Depth = Op.getConstantOperandVal(0); LowerRETURNADDR() local
27365 unsigned Depth = Op.getConstantOperandVal(0); LowerFRAMEADDR() local
37125 computeKnownBitsForPSADBW(SDValue LHS,SDValue RHS,KnownBits & Known,const APInt & DemandedElts,const SelectionDAG & DAG,unsigned Depth) computeKnownBitsForPSADBW() argument
37146 computeKnownBitsForPMADDWD(SDValue LHS,SDValue RHS,KnownBits & Known,const APInt & DemandedElts,const SelectionDAG & DAG,unsigned Depth) computeKnownBitsForPMADDWD() argument
37169 computeKnownBitsForPMADDUBSW(SDValue LHS,SDValue RHS,KnownBits & Known,const APInt & DemandedElts,const SelectionDAG & DAG,unsigned Depth) computeKnownBitsForPMADDUBSW() argument
38480 combineX86ShuffleChain(ArrayRef<SDValue> Inputs,SDValue Root,ArrayRef<int> BaseMask,int Depth,bool HasVariableMask,bool AllowVariableCrossLaneMask,bool AllowVariablePerLaneMask,SelectionDAG & DAG,const X86Subtarget & Subtarget) combineX86ShuffleChain() argument
39227 combineX86ShuffleChainWithExtract(ArrayRef<SDValue> Inputs,SDValue Root,ArrayRef<int> BaseMask,int Depth,bool HasVariableMask,bool AllowVariableCrossLaneMask,bool AllowVariablePerLaneMask,SelectionDAG & DAG,const X86Subtarget & Subtarget) combineX86ShuffleChainWithExtract() argument
39684 combineX86ShufflesRecursively(ArrayRef<SDValue> SrcOps,int SrcOpIndex,SDValue Root,ArrayRef<int> RootMask,ArrayRef<const SDNode * > SrcNodes,unsigned Depth,unsigned MaxDepth,bool HasVariableMask,bool AllowVariableCrossLaneMask,bool AllowVariablePerLaneMask,SelectionDAG & DAG,const X86Subtarget & Subtarget) combineX86ShufflesRecursively() argument
43718 combineBitcastToBoolVector(EVT VT,SDValue V,const SDLoc & DL,SelectionDAG & DAG,const X86Subtarget & Subtarget,unsigned Depth=0) combineBitcastToBoolVector() argument
48968 PromoteMaskArithmetic(SDValue N,const SDLoc & DL,EVT VT,SelectionDAG & DAG,unsigned Depth) PromoteMaskArithmetic() argument
49508 getBMIMatchingOp(unsigned Opc,SelectionDAG & DAG,SDValue OpMustEq,SDValue Op,unsigned Depth) getBMIMatchingOp() argument
52552 isFNEG(SelectionDAG & DAG,SDNode * N,unsigned Depth=0) isFNEG() argument
[all...]
/llvm-project/llvm/unittests/CodeGen/GlobalISel/
H A DKnownBitsTest.cpp1 //===- KnownBitsTest.cpp -------------------------------------------===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
19 unsigned CopyReg = Copies[Copies.size() - 1]; in TEST_F()
20 MachineInstr *FinalCopy = MRI->getVRegDef(CopyReg); in TEST_F()
21 unsigned SrcReg = FinalCopy->getOperand(1).getReg(); in TEST_F()
22 unsigned DstReg = FinalCopy->getOperand(0).getReg(); in TEST_F()
39 unsigned CopyReg = Copies[Copies.size() - 1]; in TEST_F()
40 MachineInstr *FinalCopy = MRI->getVRegDef(CopyReg); in TEST_F()
41 unsigned SrcReg = FinalCopy->getOperand(1).getReg(); in TEST_F()
[all …]
/llvm-project/llvm/lib/CodeGen/SelectionDAG/
H A DStatepointLowering.cpp1 //===- StatepointLowering.cpp - SDAGBuilder's statepoint code --------
[all...]
H A DDAGCombiner.cpp1 //===- DAGCombiner.cpp - Implement a DAG node combiner --------
8691 calculateByteProvider(SDValue Op,unsigned Index,unsigned Depth,std::optional<uint64_t> VectorIndex,unsigned StartingIndex=0) calculateByteProvider() argument
27864 takeInexpensiveLog2(SelectionDAG & DAG,const SDLoc & DL,EVT VT,SDValue Op,unsigned Depth,bool AssumeNonZero) takeInexpensiveLog2() argument
28361 unsigned Depth = 0; GatherAllAliases() local
[all...]
/llvm-project/llvm/lib/IR/
H A DLegacyPassManager.cpp1 //===- LegacyPassManager.cpp - LLVM Pass Infrastructure Implementation ----===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exceptio
[all...]
/llvm-project/mlir/lib/Dialect/Linalg/Transforms/
H A DVectorization.cpp1 //===- Vectorization.cpp - Implementation of linalg Vectorization --------
[all...]
/llvm-project/clang/lib/Headers/
H A Dopencl-c.h1 //===--- opencl-c.h - OpenCL C language builtin function header -----------===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
12 #include "opencl-c-base.h"
43 // OpenCL v1.1/1.2/2.0 s6.2.3 - Explicit conversions
5464 // Convert half types to non-double types.
5976 // Convert non-double types to half types.
6346 // OpenCL v1.1 s6.11.1, v1.2 s6.12.1, v2.0 s6.13.1 - Work-item Functions
6357 * Returns the number of global work-items specified for
6361 * are 0 to get_work_dim() - 1. For other values of
[all …]

12