/external/llvm-project/mlir/include/mlir/Pass/ |
D | AnalysisManager.h | 148 auto res = analyses.find(TypeID::get<AnalysisT>()); in getCachedAnalysis() 149 if (res == analyses.end()) in getCachedAnalysis() 158 void clear() { analyses.clear(); } in clear() 164 for (auto it = analyses.begin(), e = analyses.end(); it != e;) { in invalidate() 167 analyses.erase(curIt); in invalidate() 178 std::tie(it, wasInserted) = analyses.try_emplace(id); in getAnalysisImpl() 195 ConceptMap analyses; variable 201 NestedAnalysisMap(Operation *op) : analyses(op) {} in NestedAnalysisMap() 204 Operation *getOperation() const { return analyses.getOperation(); } in getOperation() 213 detail::AnalysisMap analyses; member [all …]
|
/external/llvm/test/Other/ |
D | new-pass-manager.ll | 139 ; Make sure no-op passes that preserve all analyses don't even try to do any 145 ; CHECK-NO-OP-INVALIDATION-NOT: Invalidating all non-preserved analyses 211 ; CHECK-INVALIDATE-ALL: Invalidating all non-preserved analyses 216 ; CHECK-INVALIDATE-ALL: Invalidating all non-preserved analyses 218 ; CHECK-INVALIDATE-ALL: Invalidating all non-preserved analyses 223 ; CHECK-INVALIDATE-ALL: Invalidating all non-preserved analyses 245 ; CHECK-INVALIDATE-ALL-CG: Invalidating all non-preserved analyses 250 ; CHECK-INVALIDATE-ALL-CG: Invalidating all non-preserved analyses 252 ; CHECK-INVALIDATE-ALL-CG: Invalidating all non-preserved analyses 257 ; CHECK-INVALIDATE-ALL-CG: Invalidating all non-preserved analyses [all …]
|
/external/llvm-project/llvm/test/Transforms/Inline/ |
D | cgscc-incremental-invalidate.ll | 1 ; Test for a subtle bug when computing analyses during inlining and mutating 2 ; the SCC structure. Without care, this can fail to invalidate analyses. 109 ; here, but because we visit this function we compute fresh analyses for it. 110 ; These analyses are then invalidated when we inline callee disrupting the
|
D | cgscc-invalidate.ll | 2 ; analyses after inlining into the function body. 87 ; This function ends up split into a separate SCC, which can cause its analyses
|
D | clear-analyses.ll | 3 ; delete, this doesn't crash or go awry despite the inliner clearing the analyses
|
/external/llvm-project/llvm/docs/CommandGuide/ |
D | opt.rst | 15 takes LLVM source files as input, runs the specified optimizations or analyses 20 When `-analyze` is specified, :program:`opt` performs various analyses 108 line options to enable various optimizations or analyses. To see the new
|
/external/llvm-project/llvm/docs/ |
D | UserGuides.rst | 109 Information on how to write LLVM transformations and analyses. 116 A list of optimizations and analyses implemented in LLVM. 127 use existing analyses.
|
D | AliasAnalysis.rst | 17 alias analyses respond to a query with a `Must, May, or No`_ alias response, 23 primary interface used by clients and implementations of alias analyses in the 29 which can provide it, allowing for powerful analyses and transformations to work 208 analyses and transformations, the answer should be fairly obvious from what type 282 benefit from both alias analyses). The alias analysis class automatically 301 which allows all alias analyses in a change to be updated. 325 use pointers). Typically alias analyses keep data structures that have entries 382 also no way of setting a chain of analyses as the default. 597 analyses have.
|
/external/llvm/docs/CommandGuide/ |
D | opt.rst | 13 takes LLVM source files as input, runs the specified optimizations or analyses 18 When :option:`-analyze` is specified, :program:`opt` performs various analyses 106 line options to enable various optimizations or analyses. To see the new
|
/external/llvm-project/mlir/docs/ |
D | Interfaces.md | 7 transformations and analyses need to account for the semantics of 16 able to express transformations/analyses in terms of these interfaces without 20 transformations/analyses. 24 Dialect interfaces are generally useful for transformation passes or analyses 28 transformations/analyses. In these cases, registering the interface directly on
|
D | PassManagement.md | 118 An important concept, along with transformation passes, are analyses. These are 120 information on a specific operation without modifying it. In MLIR, analyses are 135 invalidated) based upon other properties such as analyses sets. 140 analyses for the current operation being processed. 143 analyses: 188 analyses, all analyses are assumed to be invalidated by a pass. To avoid 189 invalidation, a pass must specifically mark analyses that are known to be 193 preserving analyses: 199 // Mark all analyses as preserved. This is useful if a pass can guarantee 203 // Mark specific analyses as preserved. This is used if some transformation [all …]
|
/external/llvm/test/Analysis/GlobalsModRef/ |
D | chaining-analysis.ll | 3 ; This test requires the use of previous analyses to determine that
|
/external/llvm-project/llvm/test/Analysis/GlobalsModRef/ |
D | chaining-analysis.ll | 3 ; This test requires the use of previous analyses to determine that
|
/external/llvm/test/Transforms/InstCombine/ |
D | preserve-sminmax.ll | 7 ; other analyses wouldn't recognize it.
|
/external/llvm-project/llvm/test/Transforms/InstCombine/ |
D | preserve-sminmax.ll | 7 ; other analyses wouldn't recognize it.
|
D | preserved-analyses.ll | 1 ; This is really testing that instcombine preserves analyses correctly, so we
|
/external/llvm-project/llvm/test/Transforms/LoopDeletion/ |
D | invalidation.ll | 1 ; Ensure we don't run analyses over loops after they've been deleted. We run
|
/external/llvm-project/clang/lib/StaticAnalyzer/ |
D | README.txt | 123 -cfg-add-implicit-dtors Add C++ implicit destructors to CFGs for all analyses 124 -cfg-add-initializers Add C++ initializers to CFGs for all analyses 127 -unoptimized-cfg Generate unoptimized CFGs for all analyses
|
/external/clang/lib/StaticAnalyzer/ |
D | README.txt | 123 -cfg-add-implicit-dtors Add C++ implicit destructors to CFGs for all analyses 124 -cfg-add-initializers Add C++ initializers to CFGs for all analyses 127 -unoptimized-cfg Generate unoptimized CFGs for all analyses
|
/external/llvm-project/llvm/test/Transforms/LoopVectorize/ |
D | novect-lcssa-cfg-invalidation.ll | 5 ; Checks what analyses are invalidated after Loop Vectorization when no actual
|
/external/clang/include/clang/StaticAnalyzer/Core/ |
D | Analyses.def | 10 // This file defines the set of static analyses used by AnalysisConsumer.
|
/external/llvm/bindings/ocaml/analysis/ |
D | llvm_analysis.mli | 12 This interface provides an OCaml API for LLVM IR analyses, the classes in
|
/external/llvm-project/llvm/test/Transforms/ArgumentPromotion/ |
D | invalidation.ll | 3 ; the call graph, any analyses that happened to be cached for that function are
|
/external/llvm-project/mlir/docs/Rationale/ |
D | RationaleSimplifiedPolyhedralForm.md | 54 exactly what form is best to perform the analyses and transformations we want? 104 make analyses and optimizations more efficient, and also makes the scoping of 316 The simplified form is much easier for analyses and transformations to build 319 what target code will be generated. With the traditional form, these analyses 381 This is a problem for analyses and cost models, because they each have to reason 396 instructions and other analyses we need to perform.
|
/external/llvm-project/llvm/bindings/ocaml/analysis/ |
D | llvm_analysis.mli | 11 This interface provides an OCaml API for LLVM IR analyses, the classes in
|