diff options
| author | gysit <gysit@google.com> | 2022-03-24 10:22:28 +0000 |
|---|---|---|
| committer | gysit <gysit@google.com> | 2022-03-24 10:22:41 +0000 |
| commit | 53f7fb0a8703ffcc5ccdb0d8133ca0df494ef68f (patch) | |
| tree | 523ccc2be26c1b4b313d6a34c32d27aa3210edf5 /clang/lib/CodeGen/CodeGenFunction.cpp | |
| parent | ec93b28909749619dbe58b092a13da9d1ff1eb1e (diff) | |
| download | llvm-53f7fb0a8703ffcc5ccdb0d8133ca0df494ef68f.zip llvm-53f7fb0a8703ffcc5ccdb0d8133ca0df494ef68f.tar.gz llvm-53f7fb0a8703ffcc5ccdb0d8133ca0df494ef68f.tar.bz2 | |
[mlir][linalg] Do not fuse shape-only producers.
This revision introduces a heuristic to stop fusion for shape-only tensors. A shape-only tensor only defines the shape of the consumer computation while the data is not used. Pure producer consumer fusion thus shall not fuse the producer of a shape-only tensor. In particular, since the shape-only tensor will have other uses that actually consume the data.
The revision enables fusion for consumers that have two uses of the same tensor. One as input operand and one as shape-only output operand. In these cases, we want to fuse only the input operand and avoid output fusion via iteration argument.
Reviewed By: hanchung
Differential Revision: https://reviews.llvm.org/D120981
Diffstat (limited to 'clang/lib/CodeGen/CodeGenFunction.cpp')
0 files changed, 0 insertions, 0 deletions