[all-commits] [llvm/llvm-project] 51fdd8: [mlir][OpDSL] Add type function attributes.
Tobias Gysi via All-commits
all-commits at lists.llvm.org
Fri Feb 25 00:26:23 PST 2022
Branch: refs/heads/main
Home: https://github.com/llvm/llvm-project
Commit: 51fdd802c794faf6e5b57cccd6ea181c7f8a83e9
https://github.com/llvm/llvm-project/commit/51fdd802c794faf6e5b57cccd6ea181c7f8a83e9
Author: gysit <gysit at google.com>
Date: 2022-02-25 (Fri, 25 Feb 2022)
Changed paths:
M mlir/include/mlir/Dialect/Linalg/IR/CMakeLists.txt
M mlir/include/mlir/Dialect/Linalg/IR/Linalg.h
M mlir/include/mlir/Dialect/Linalg/IR/LinalgBase.td
M mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
M mlir/include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td
M mlir/lib/Dialect/Linalg/IR/CMakeLists.txt
M mlir/lib/Dialect/Linalg/IR/LinalgDialect.cpp
M mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
M mlir/python/mlir/dialects/linalg/opdsl/lang/comprehension.py
M mlir/python/mlir/dialects/linalg/opdsl/lang/config.py
M mlir/python/mlir/dialects/linalg/opdsl/lang/dsl.py
M mlir/python/mlir/dialects/linalg/opdsl/lang/emitter.py
M mlir/python/mlir/dialects/linalg/opdsl/lang/scalar_expr.py
M mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
M mlir/test/Dialect/Linalg/generalize-named-polymorphic-ops.mlir
M mlir/test/Integration/Dialect/SparseTensor/taco/tools/mlir_pytaco.py
M mlir/test/mlir-linalg-ods-gen/test-linalg-ods-yaml-gen.yaml
M mlir/test/python/dialects/linalg/opdsl/arguments.py
M mlir/test/python/dialects/linalg/opdsl/assignments.py
M mlir/test/python/dialects/linalg/opdsl/emit_matmul.py
M mlir/test/python/dialects/linalg/ops.py
M mlir/test/python/integration/dialects/linalg/opsrun.py
M mlir/tools/mlir-linalg-ods-gen/mlir-linalg-ods-yaml-gen.cpp
M utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
Log Message:
-----------
[mlir][OpDSL] Add type function attributes.
Previously, OpDSL operation used hardcoded type conversion operations (cast or cast_unsigned). Supporting signed and unsigned casts thus meant implementing two different operations. Type function attributes allow us to define a single operation that has a cast type function attribute which at operation instantiation time may be set to cast or cast_unsigned. We may for example, defina a matmul operation with a cast argument:
```
@linalg_structured_op
def matmul(A=TensorDef(T1, S.M, S.K), B=TensorDef(T2, S.K, S.N), C=TensorDef(U, S.M, S.N, output=True),
cast=TypeFnAttrDef(default=TypeFn.cast)):
C[D.m, D.n] += cast(U, A[D.m, D.k]) * cast(U, B[D.k, D.n])
```
When instantiating the operation the attribute may be set to the desired cast function:
```
linalg.matmul(lhs, rhs, outs=[out], cast=TypeFn.cast_unsigned)
```
The revsion introduces a enum in the Linalg dialect that maps one-by-one to the type functions defined by OpDSL.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D119718
More information about the All-commits
mailing list