projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9232ca4) | patch
[mlir] Remove old "tc" linalg ods generator.
authorStella Laurenzo <stellaraccident@gmail.com>
Wed, 29 Sep 2021 23:47:17 +0000 (23:47 +0000)
committerStella Laurenzo <stellaraccident@gmail.com>
Thu, 30 Sep 2021 16:30:06 +0000 (16:30 +0000)
commit267bb194f3cedd88db0e7a5ea86169d669948e74
tree1a2a3e1977da99107457c97067469813230b2820tree | snapshot
parent9232ca4712cf3f3a99a8cd095165241daf0ddc4ecommit | diff
[mlir] Remove old "tc" linalg ods generator.

* This could have been removed some time ago as it only had one op left in it, which is redundant with the new approach.
* `matmul_i8_i8_i32` (the remaining op) can be trivially replaced by `matmul`, which natively supports mixed precision.

Differential Revision: https://reviews.llvm.org/D110792
14 files changed:
mlir/include/mlir/Dialect/Linalg/IR/CMakeLists.txt diff | blob | history
mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOpsSpec.tc [deleted file] blob | history
mlir/include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td diff | blob | history
mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp diff | blob | history
mlir/test/CMakeLists.txt diff | blob | history
mlir/test/Dialect/Linalg/tile-and-pad-tensors.mlir diff | blob | history
mlir/test/Dialect/Linalg/vectorization.mlir diff | blob | history
mlir/test/Integration/Dialect/Linalg/CPU/benchmark_matmul_i8_i8_i32.mlir [deleted file] blob | history
mlir/test/lib/Dialect/Linalg/TestLinalgTransforms.cpp diff | blob | history
mlir/test/lit.cfg.py diff | blob | history
mlir/test/mlir-linalg-ods-gen/test-linalg-ods-gen.tc [deleted file] blob | history
mlir/tools/mlir-linalg-ods-gen/CMakeLists.txt diff | blob | history
mlir/tools/mlir-linalg-ods-gen/mlir-linalg-ods-gen.cpp [deleted file] blob | history
utils/bazel/llvm-project-overlay/mlir/BUILD.bazel diff | blob | history
Domain: System / Toolchain;