projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: cb7bda2) | patch
[mlir][linalg] Return tensor::PadOp handle from transform op
authorMatthias Springer <me@m-sp.org>
Tue, 4 Jul 2023 12:18:32 +0000 (14:18 +0200)
committerMatthias Springer <me@m-sp.org>
Tue, 4 Jul 2023 12:24:47 +0000 (14:24 +0200)
commit0e06ec5961325c51ef2fa48a25c9b99038ff2f1b
tree3dde60e422ea2d649a214ce1e1306ea04549c97dtree | snapshot
parentcb7bda2ace81226c5b33165411dd0316f93fa57ecommit | diff
[mlir][linalg] Return tensor::PadOp handle from transform op

"transform.structured.pad" now returns all `tensor::PadOp` in addition to the padded ops.

Also add a test case that shows how to force an allocation for "tensor.pad" ops with a custom memory space.

Differential Revision: https://reviews.llvm.org/D153555
mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td diff | blob | history
mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h diff | blob | history
mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp diff | blob | history
mlir/lib/Dialect/Linalg/Transforms/Padding.cpp diff | blob | history
mlir/python/mlir/dialects/_structured_transform_ops_ext.py diff | blob | history
mlir/test/Dialect/Linalg/pad-to-specific-memory-space.mlir [new file with mode: 0644] blob
mlir/test/Dialect/Linalg/transform-op-hoist-pad-build-packing-loop-nest.mlir diff | blob | history
mlir/test/Dialect/Linalg/transform-op-hoist-pad.mlir diff | blob | history
mlir/test/Dialect/Linalg/transform-op-pad.mlir diff | blob | history
mlir/test/Dialect/Linalg/transform-ops-invalid.mlir diff | blob | history
mlir/test/Dialect/Linalg/transform-ops.mlir diff | blob | history
Domain: System / Toolchain;