diff options
| author | Matthias Springer <me@m-sp.org> | 2023-10-05 14:33:04 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-10-05 14:33:04 +0200 |
| commit | ea71d2d0fe8f172fde5b39c702cfa4dc5c3df88e (patch) | |
| tree | 2e0fb892cf821e916e7945fb81d05b4eaf9ee66c /llvm/lib/Bitcode/Reader/BitcodeReader.cpp | |
| parent | 932dc9d8c4979f4f0aec33e2b9725a110ded8b61 (diff) | |
| download | llvm-ea71d2d0fe8f172fde5b39c702cfa4dc5c3df88e.zip llvm-ea71d2d0fe8f172fde5b39c702cfa4dc5c3df88e.tar.gz llvm-ea71d2d0fe8f172fde5b39c702cfa4dc5c3df88e.tar.bz2 | |
[mlir][tensor][bufferize] Reshapes: Fix memory side effects and memory space (#68195)
* `tensor.collapse_shape` may bufferize to a memory read because the op
may have to reallocate the source buffer.
* `tensor.reshape` should not use `bufferization.clone` for
reallocation. This op has requirements wrt. the order of buffer
writes/reads. Use `memref.alloc` and `memref.copy` instead. Also fix a
bug where the memory space of the source buffer was not propagated to
the reallocated buffer.
Diffstat (limited to 'llvm/lib/Bitcode/Reader/BitcodeReader.cpp')
0 files changed, 0 insertions, 0 deletions