Tensor inputs, if not used in the body of TiledLoopOp, can be removed.
memref::CastOp can be folded into TiledLoopOp as well.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Paths
| Differential D101445
[mlir] Add folding for tensor inputs and memref.cast in linalg.tiled_loop. ClosedPublic Authored by pifon2a on Apr 28 2021, 5:17 AM.
Details Summary Tensor inputs, if not used in the body of TiledLoopOp, can be removed.
Diff Detail
Event TimelineHerald added subscribers: dcaballe, cota, teijeong and 16 others. · View Herald TranscriptApr 28 2021, 5:17 AM This revision is now accepted and ready to land.Apr 28 2021, 5:31 AM This revision was landed with ongoing or failed builds.Apr 28 2021, 5:36 AM Closed by commit rG29dbac0ae295: [mlir] Add folding for tensor inputs and memref.cast in linalg.tiled_loop. (authored by pifon2a). · Explain Why This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 341156 mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
mlir/test/Dialect/Linalg/canonicalize.mlir
|
nit: I think auto should be avoided in these two cases. Exceptions are dyn_cast/cast and long types such as iterators.