llvm-project.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Collapse)	Author
2025-07-28	[mlir][memref] Support test-compose-subview dynamic size (#146881)	lonely eagle
	Supports the case where the sizes of the subview op is dynamic.When there are more for loops in the tile algorithm, multiple subviews are performed and test-compose-subview does not work when the size operand of the subview ops is dynamic value.
2025-07-24	[mlir][NFC] update `mlir/Dialect` create APIs (18/n) (#149925)	Maksim Levental
	See https://github.com/llvm/llvm-project/pull/147168 for more info.
2025-07-12	[llvm] Remove unused includes (NFC) (#148342)	Kazu Hirata
	These are identified by misc-include-cleaner. I've filtered out those that break builds. Also, I'm staying away from llvm-config.h, config.h, and Compiler.h, which likely cause platform- or compiler-specific build failures.
2024-12-17	[MemRef] Migrate away from PointerUnion::{is,get} (NFC) (#120202)	Kazu Hirata
	Note that PointerUnion::{is,get} have been soft deprecated in PointerUnion.h: // FIXME: Replace the uses of is(), get() and dyn_cast() with // isa<T>, cast<T> and the llvm::dyn_cast<T> I'm not touching PointerUnion::dyn_cast for now because it's a bit complicated; we could blindly migrate it to dyn_cast_if_present, but we should probably use dyn_cast when the operand is known to be non-null.
2024-02-07	[mlir]Fix compose subview (#80551)	lonely eagle
	I found a bug in `test-compose-subview`,You can see the example I gave. ``` #map = affine_map<() -> ()> module { func.func private @fun(%arg0: memref<10x10xf32>, %arg1: memref<5x5xf32>) -> memref<5x5xf32> { %c0 = arith.constant 0 : index %c5 = arith.constant 5 : index %c1 = arith.constant 1 : index %subview = memref.subview %arg0[0, 0] [5, 5] [1, 1] : memref<10x10xf32> to memref<5x5xf32, strided<[10, 1]>> %alloc = memref.alloc() : memref<5x5xf32> scf.for %arg2 = %c0 to %c5 step %c1 { scf.for %arg3 = %c0 to %c5 step %c1 { %subview_0 = memref.subview %subview[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32, strided<[10, 1]>> to memref<f32, strided<[], offset: ?>> %subview_1 = memref.subview %arg1[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> %alloc_2 = memref.alloc() : memref<f32> linalg.generic {indexing_maps = [#map, #map, #map], iterator_types = []} ins(%subview_0, %subview_1 : memref<f32, strided<[], offset: ?>>, memref<f32, strided<[], offset: ?>>) outs(%alloc_2 : memref<f32>) { ^bb0(%in: f32, %in_4: f32, %out: f32): %0 = arith.addf %in, %in_4 : f32 linalg.yield %0 : f32 } %subview_3 = memref.subview %alloc[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> memref.copy %alloc_2, %subview_3 : memref<f32> to memref<f32, strided<[], offset: ?>> } } return %alloc : memref<5x5xf32> } func.func @test(%arg0: memref<10x10xf32>, %arg1: memref<5x5xf32>) -> memref<5x5xf32> { %0 = call @fun(%arg0, %arg1) : (memref<10x10xf32>, memref<5x5xf32>) -> memref<5x5xf32> return %0 : memref<5x5xf32> } } ``` When I run `mlir-opt test.mlir ---test-compose-subview`. ``` test.mlir:14:9: error: 'linalg.generic' op expected operand rank (2) to match the result rank of indexing_map #0 (0) linalg.generic {indexing_maps = [#map, #map, #map], iterator_types = []} ins(%subview_0, %subview_1 : memref<f32, strided<[], offset: ?>>, memref<f32, strided<[], offset: ?>>) outs(%alloc_2 : memref<f32>) { ^ test1.mlir:14:9: note: see current operation: "linalg.generic"(%4, %5, %6) <{indexing_maps = [affine_map<() -> ()>, affine_map<() -> ()>, affine_map<() -> ()>], iterator_types = [], operandSegmentSizes = array<i32: 2, 1>}> ({ ^bb0(%arg4: f32, %arg5: f32, %arg6: f32): %8 = "arith.addf"(%arg4, %arg5) <{fastmath = #arith.fastmath<none>}> : (f32, f32) -> f32 "linalg.yield"(%8) : (f32) -> () }) : (memref<1x1xf32, strided<[10, 1], offset: ?>>, memref<f32, strided<[], offset: ?>>, memref<f32>) -> () ``` This PR fixes that.In the meantime I've extended this PR to handle cases where stride is greater than 1. ``` func.func private @Unknown0(%arg0: memref<10x10xf32>, %arg1: memref<5x5xf32>) -> memref<5x5xf32> { %c0 = arith.constant 0 : index %c5 = arith.constant 5 : index %c1 = arith.constant 1 : index %subview = memref.subview %arg0[0, 0] [5, 5] [2, 2] : memref<10x10xf32> to memref<5x5xf32, strided<[20, 2]>> %alloc = memref.alloc() : memref<5x5xf32> scf.for %arg2 = %c0 to %c5 step %c1 { scf.for %arg3 = %c0 to %c5 step %c1 { %subview_0 = memref.subview %subview[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32, strided<[20, 2]>> to memref<f32, strided<[], offset: ?>> %subview_1 = memref.subview %arg1[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> %alloc_2 = memref.alloc() : memref<f32> linalg.generic {indexing_maps = [affine_map<() -> ()>, affine_map<() -> ()>, affine_map<() -> ()>], iterator_types = []} ins(%subview_0, %subview_1 : memref<f32, strided<[], offset: ?>>, memref<f32, strided<[], offset: ?>>) outs(%alloc_2 : memref<f32>) { ^bb0(%in: f32, %in_4: f32, %out: f32): %0 = arith.addf %in, %in_4 : f32 linalg.yield %0 : f32 } %subview_3 = memref.subview %alloc[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> memref.copy %alloc_2, %subview_3 : memref<f32> to memref<f32, strided<[], offset: ?>> } } return %alloc : memref<5x5xf32> } $ mlir-opt test.mlir -test-compose-subview #map = affine_map<()[s0] -> (s0 * 2)> #map1 = affine_map<() -> ()> module { func.func private @Unknown0(%arg0: memref<10x10xf32>, %arg1: memref<5x5xf32>) -> memref<5x5xf32> { %c0 = arith.constant 0 : index %c5 = arith.constant 5 : index %c1 = arith.constant 1 : index %alloc = memref.alloc() : memref<5x5xf32> scf.for %arg2 = %c0 to %c5 step %c1 { scf.for %arg3 = %c0 to %c5 step %c1 { %0 = affine.apply #map()[%arg2] %1 = affine.apply #map()[%arg3] %subview = memref.subview %arg0[%0, %1] [1, 1] [2, 2] : memref<10x10xf32> to memref<f32, strided<[], offset: ?>> %subview_0 = memref.subview %arg1[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> %alloc_1 = memref.alloc() : memref<f32> linalg.generic {indexing_maps = [#map1, #map1, #map1], iterator_types = []} ins(%subview, %subview_0 : memref<f32, strided<[], offset: ?>>, memref<f32, strided<[], offset: ?>>) outs(%alloc_1 : memref<f32>) { ^bb0(%in: f32, %in_3: f32, %out: f32): %2 = arith.addf %in, %in_3 : f32 linalg.yield %2 : f32 } %subview_2 = memref.subview %alloc[%arg2, %arg3] [1, 1] [1, 1] : memref<5x5xf32> to memref<f32, strided<[], offset: ?>> memref.copy %alloc_1, %subview_2 : memref<f32> to memref<f32, strided<[], offset: ?>> } } return %alloc : memref<5x5xf32> } } ```
2023-05-26	[mlir] Move casting calls from methods to function calls	Tres Popp
	The MLIR classes Type/Attribute/Operation/Op/Value support cast/dyn_cast/isa/dyn_cast_or_null functionality through llvm's doCast functionality in addition to defining methods with the same name. This change begins the migration of uses of the method to the corresponding function call as has been decided as more consistent. Note that there still exist classes that only define methods directly, such as AffineExpr, and this does not include work currently to support a functional cast/isa call. Context: - https://mlir.llvm.org/deprecation/ at "Use the free function variants for dyn_cast/cast/isa/…" - Original discussion at https://discourse.llvm.org/t/preferred-casting-style-going-forward/68443 Implementation: This patch updates all remaining uses of the deprecated functionality in mlir/. This was done with clang-tidy as described below and further modifications to GPUBase.td and OpenMPOpsInterfaces.td. Steps are described per line, as comments are removed by git: 0. Retrieve the change from the following to build clang-tidy with an additional check: main...tpopp:llvm-project:tidy-cast-check 1. Build clang-tidy 2. Run clang-tidy over your entire codebase while disabling all checks and enabling the one relevant one. Run on all header files also. 3. Delete .inc files that were also modified, so the next build rebuilds them to a pure state. ``` ninja -C $BUILD_DIR clang-tidy run-clang-tidy -clang-tidy-binary=$BUILD_DIR/bin/clang-tidy -checks='-,misc-cast-functions'\ -header-filter=mlir/ mlir/ -fix rm -rf $BUILD_DIR/tools/mlir/*/.inc ``` Differential Revision: https://reviews.llvm.org/D151542
2023-05-12	[mlir] Move casting calls from methods to function calls	Tres Popp
	The MLIR classes Type/Attribute/Operation/Op/Value support cast/dyn_cast/isa/dyn_cast_or_null functionality through llvm's doCast functionality in addition to defining methods with the same name. This change begins the migration of uses of the method to the corresponding function call as has been decided as more consistent. Note that there still exist classes that only define methods directly, such as AffineExpr, and this does not include work currently to support a functional cast/isa call. Caveats include: - This clang-tidy script probably has more problems. - This only touches C++ code, so nothing that is being generated. Context: - https://mlir.llvm.org/deprecation/ at "Use the free function variants for dyn_cast/cast/isa/…" - Original discussion at https://discourse.llvm.org/t/preferred-casting-style-going-forward/68443 Implementation: This first patch was created with the following steps. The intention is to only do automated changes at first, so I waste less time if it's reverted, and so the first mass change is more clear as an example to other teams that will need to follow similar steps. Steps are described per line, as comments are removed by git: 0. Retrieve the change from the following to build clang-tidy with an additional check: https://github.com/llvm/llvm-project/compare/main...tpopp:llvm-project:tidy-cast-check 1. Build clang-tidy 2. Run clang-tidy over your entire codebase while disabling all checks and enabling the one relevant one. Run on all header files also. 3. Delete .inc files that were also modified, so the next build rebuilds them to a pure state. 4. Some changes have been deleted for the following reasons: - Some files had a variable also named cast - Some files had not included a header file that defines the cast functions - Some files are definitions of the classes that have the casting methods, so the code still refers to the method instead of the function without adding a prefix or removing the method declaration at the same time. ``` ninja -C $BUILD_DIR clang-tidy run-clang-tidy -clang-tidy-binary=$BUILD_DIR/bin/clang-tidy -checks='-,misc-cast-functions'\ -header-filter=mlir/ mlir/ -fix rm -rf $BUILD_DIR/tools/mlir/*/.inc git restore mlir/lib/IR mlir/lib/Dialect/DLTI/DLTI.cpp\ mlir/lib/Dialect/Complex/IR/ComplexDialect.cpp\ mlir/lib/**/IR/\ mlir/lib/Dialect/SparseTensor/Transforms/SparseVectorization.cpp\ mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp\ mlir/test/lib/Dialect/Test/TestTypes.cpp\ mlir/test/lib/Dialect/Transform/TestTransformDialectExtension.cpp\ mlir/test/lib/Dialect/Test/TestAttributes.cpp\ mlir/unittests/TableGen/EnumsGenTest.cpp\ mlir/test/python/lib/PythonTestCAPI.cpp\ mlir/include/mlir/IR/ ``` Differential Revision: https://reviews.llvm.org/D150123
2023-04-20	[mlir][Affine][NFC] Wrap dialect in "affine" namespace	Matthias Springer
	This cleanup aligns the affine dialect with all the other dialects. Differential Revision: https://reviews.llvm.org/D148687
2022-07-14	[mlir] (NFC) run clang-format on all files	Jeff Niu

2022-07-10	[mlir] Flip accessors to prefixed form (NFC)	Jacques Pienaar
	Another mechanical sweep to keep diff small for flip to _Prefixed.
2022-02-18	Rename PatternRewriteSet::insert to add	Tres Popp
	insert is soft deprecated, so remove all references so it's less likely to be used and can be easily removed in the future. Differential Revision: https://reviews.llvm.org/D120021
2022-01-26	[mlir] Finish replacing OwningRewritePatternList with RewritePatternSet	River Riddle
	OwningRewritePatternList has been deprecated for ~10 months now, we can remove the leftover using directives at this point. Differential Revision: https://reviews.llvm.org/D118287
2022-01-26	[mlir] Move ComposeSubView+ExpandOps from Standard to MemRef	River Riddle
	These transformations already operate on memref operations (as part of splitting up the standard dialect). Now that the operations have moved, it's time for these transformations to move as well. Differential Revision: https://reviews.llvm.org/D118285