| commit | ba1e150d03cac7629111a90eea1f2d727e465983 | |
| author | Kerry McLaughlin <kerry.mclaughlin@arm.com> | |
| Tue, 16 Feb 2021 10:43:42 +0000 (16 10:43 +0000) | ||
| committer | Kerry McLaughlin <kerry.mclaughlin@arm.com> | |
| Tue, 16 Feb 2021 13:50:06 +0000 (16 13:50 +0000) | ||
| tree | f46873c8017e4842f390776f88b3a9d263917a13 | treesnapshot (tar.gz zip) |
| parent | 32389346ed83e14b7a9bd3a31a96181c6a1cdb5e | commitdiff |
[SVE] Add support for scalable vectorization of loops with int/fast FP reductions
This patch enables scalable vectorization of loops with integer/fast reductions, e.g:
```
unsigned sum = 0;
for (int i = 0; i < n; ++i) {
sum += a[i];
}
```
A new TTI interface, isLegalToVectorizeReduction, has been added to prevent
reductions which are not supported for scalable types from vectorizing.
If the reduction is not supported for a given scalable VF,
computeFeasibleMaxVF will fall back to using fixed-width vectorization.
Reviewed By: david-arm, fhahn, dmgreen
Differential Revision: https://reviews.llvm.org/D95245
This patch enables scalable vectorization of loops with integer/fast reductions, e.g:
```
unsigned sum = 0;
for (int i = 0; i < n; ++i) {
sum += a[i];
}
```
A new TTI interface, isLegalToVectorizeReduction, has been added to prevent
reductions which are not supported for scalable types from vectorizing.
If the reduction is not supported for a given scalable VF,
computeFeasibleMaxVF will fall back to using fixed-width vectorization.
Reviewed By: david-arm, fhahn, dmgreen
Differential Revision: https://reviews.llvm.org/D95245