projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 092c303) | patch
[mlir][linalg] Lower PadTensorOp to InitTensorOp + FillOp + SubTensorInitOp
authorMatthias Springer <springerm@google.com>
Mon, 14 Jun 2021 05:20:11 +0000 (14:20 +0900)
committerMatthias Springer <springerm@google.com>
Mon, 14 Jun 2021 05:21:08 +0000 (14:21 +0900)
commit98fff5153a81fc501a69a4c184de5236fc806b9f
tree54b9d61f450c4f952856f0354c8744386f653848tree | snapshot
parent092c303955cd18be6c0b923b1c0a1b96e2c91893commit | diff
[mlir][linalg] Lower PadTensorOp to InitTensorOp + FillOp + SubTensorInitOp

Currently limited to constant pad values. Any combination of dynamic/static tensor sizes and padding sizes is supported.

Differential Revision: https://reviews.llvm.org/D103679
mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp diff | blob | history
mlir/test/Dialect/Linalg/vectorization.mlir diff | blob | history
Domain: System / Toolchain;