diff options
| author | Nicolas Vasilache <nicolas.vasilache@gmail.com> | 2022-06-21 05:11:29 -0700 |
|---|---|---|
| committer | Nicolas Vasilache <nicolas.vasilache@gmail.com> | 2022-06-22 12:06:58 -0700 |
| commit | d5716395792696f2b56a0d4debadd040ee385143 (patch) | |
| tree | 5cc332cd857e7520b157ca7f1afca9a0d699f671 /lldb/source/Commands/CommandObjectProcess.cpp | |
| parent | d0a4450ecdaf124f9d422dbcba8cbdbeb50c836a (diff) | |
| download | llvm-d5716395792696f2b56a0d4debadd040ee385143.zip llvm-d5716395792696f2b56a0d4debadd040ee385143.tar.gz llvm-d5716395792696f2b56a0d4debadd040ee385143.tar.bz2 | |
[mlir][Linalg] SplitReduction implementation without tensor::ExpandShapeOp
This revision proposes a different implementation of the SplitReductoin transformation that does
not rely on tensor::ExpandShapeOp.
Previously, a dimension `[k]` would be split into `[k][kk]` via an ExpandShapeOp.
Instead, this revision proposes to rewrite `[k]` into `[factor * k + kk]`.
There are different tradeoffs involved but the proposed implementation is more general because
the affine rewrite is well-defined. In particular, it works naturally with `?` parallel dimensions and
non-trivial indexing maps.
A further rewrite of `[factor * k + kk]` + ExpandShapeOp is possible as a followup.
Differential Revision: https://reviews.llvm.org/D128266
Diffstat (limited to 'lldb/source/Commands/CommandObjectProcess.cpp')
0 files changed, 0 insertions, 0 deletions