| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
[MLIR][MemRefToLLVM] Remove typed pointer support (#70909) This commit removes the support for lowering MemRefToLLVM to LLVM dialect with typed pointers. Typed pointers have been deprecated for a while now and it's planned to soon remove them from the LLVM dialect. Related PSA: https://discourse.llvm.org/t/psa-removal-of-typed-pointers-from-the-llvm-dialect/74502 | 2 年前 | |
Emit nuw and nsw for mul and add when lowering to llvm.getelementptr (#140966) Now that the GEP no wrap flags are known when lowering to llvm.getelementptr, we can also emit nuw and nsw for the generated llvm.mul and llvm.add when no unsigned wrap and no signed wrap are used respectively. fixes: iree-org/iree#20483 Signed-off-by: Lin, Peiyong <linpyong@gmail.com> | 1 年前 | |
Emit nuw and nsw for mul and add when lowering to llvm.getelementptr (#140966) Now that the GEP no wrap flags are known when lowering to llvm.getelementptr, we can also emit nuw and nsw for the generated llvm.mul and llvm.add when no unsigned wrap and no signed wrap are used respectively. fixes: iree-org/iree#20483 Signed-off-by: Lin, Peiyong <linpyong@gmail.com> | 1 年前 | |
[MLIR] Fix FileCheck annotations and add 32-bit coverage (#167149) Addresses #93154 by ensuring the test RUN/FILECHECK prefixes across mlir/test/Conversion/FuncToLLVM/ and related files match their expected outputs, and adds an index-bitwidth=32 variant of expand-then-convert-to-llvm.mlir that exercises @collapse_shape_dynamic_with_non_identity_layout through the same MemRef-to-LLVM pipeline so the 32-bit layout descriptors are verified. | 9 个月前 | |
[MLIR][MemRefToLLVM] Remove typed pointer support (#70909) This commit removes the support for lowering MemRefToLLVM to LLVM dialect with typed pointers. Typed pointers have been deprecated for a while now and it's planned to soon remove them from the LLVM dialect. Related PSA: https://discourse.llvm.org/t/psa-removal-of-typed-pointers-from-the-llvm-dialect/74502 | 2 年前 | |
[mlir][memref] Check memory space before lowering alloc ops (#134427) Check the memory space before lowering allocation ops, instead of starting the lowering and then rolling back the pattern when the memory space was found to be incompatible with LLVM. Note: This is in preparation of the One-Shot Dialect Conversion refactoring. Note: isConvertibleAndHasIdentityMaps now also checks the memory space. | 1 年前 | |
Reapply "Reland "[mlir][Affine] Handle null parent op in getAffineParallelInductionVarOwner" (#142785)" This reverts commit 178b64e75bb39debc235e8008686b7c4af444013. The author misread the report of the failure, and thought that it broke the CI again. Reland the fix. | 1 年前 | |
[mlir][memref] Introduce memref.distinct_objects op (#156913) The distinct_objects operation takes a list of memrefs and returns a list of memrefs of the same types, with the additional assumption that accesses to these memrefs will never alias with each other. This means that loads and stores to different memrefs in the list can be safely reordered. The discussion https://discourse.llvm.org/t/rfc-introducing-memref-aliasing-attributes/88049 | 10 个月前 | |
[mlir][LLVM] ControlFlowToLLVM: Add 1:N type conversion support (#153937) Add support for 1:N type conversions to the ControlFlowToLLVM lowering patterns. Not applicable to cf.switch and cf.assert. --------- Co-authored-by: Tobias Gysi <tobias.gysi@nextsilicon.com> | 11 个月前 |