diff options
| author | Mahesh Ravishankar <ravishankarm@google.com> | 2022-11-11 23:09:08 +0000 |
|---|---|---|
| committer | Mahesh Ravishankar <ravishankarm@google.com> | 2022-11-11 23:17:07 +0000 |
| commit | cb40d5291e6113fddcfa5fc3861b7320f78dbfe8 (patch) | |
| tree | 1ab4bc38fefd29bd09b6e7bdb8273534e0390356 /llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp | |
| parent | 57dbca20eae8db0fd17cc59fe98c7804f02d6644 (diff) | |
| download | llvm-cb40d5291e6113fddcfa5fc3861b7320f78dbfe8.zip llvm-cb40d5291e6113fddcfa5fc3861b7320f78dbfe8.tar.gz llvm-cb40d5291e6113fddcfa5fc3861b7320f78dbfe8.tar.bz2 | |
[mlir][Linalg] Avoid using `tensor.cast` by default while folding `fill` with `pad`.
This is unnecessary if the generated operation type already matches
the type of the replaced value. Also use `OpFoldResult` to reduce the
number of cases the casts are needed.
Reviewed By: springerm, hanchung, antiagainst
Differential Revision: https://reviews.llvm.org/D137479
Diffstat (limited to 'llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp')
0 files changed, 0 insertions, 0 deletions