diff options
| author | Spenser Bauman <sbauman@mathworks.com> | 2024-06-13 09:12:18 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-06-13 09:12:18 -0400 |
| commit | 846103c7e3899a654c470fa360b22d35df888427 (patch) | |
| tree | 867b4c5f751c780078de4359b3430f91be34373c /lld/test | |
| parent | 24c6579ed0cb05b4e54520d118ab3a0b70693f8d (diff) | |
| download | llvm-846103c7e3899a654c470fa360b22d35df888427.zip llvm-846103c7e3899a654c470fa360b22d35df888427.tar.gz llvm-846103c7e3899a654c470fa360b22d35df888427.tar.bz2 | |
[mlir][memref] Unranked support for extract_aligned_pointer_as_index (#93908)
memref.extract_aligned_pointer_as_index currently does not support
unranked inputs. This lack of support interferes with the folding
operations in the expand-strided-metadata pass.
%r = memref.reinterpret_cast %arg0 to
offset: [0],
sizes: [],
strides: [] : memref<*xf32> to memref<f32>
%i = memref.extract_aligned_pointer_as_index %r : memref<f32> -> index
Patterns like this occur when bufferizing operations on unranked
tensors.
This change modifies the extract_aligned_pointer_as_index operation to
support unranked inputs with corresponding support in the MemRef->LLVM
conversion.
Co-authored-by: Spenser Bauman <sabauma@fastmail>
Diffstat (limited to 'lld/test')
0 files changed, 0 insertions, 0 deletions