diff options
| author | Hassnaa Hamdi <hassnaa.hamdi@arm.com> | 2023-04-20 05:34:05 +0000 |
|---|---|---|
| committer | Hassnaa Hamdi <hassnaa.hamdi@arm.com> | 2023-04-24 11:35:19 +0000 |
| commit | 23f2b865372d6f4712b360e3ac0d0d84f214c8ae (patch) | |
| tree | 00f3c3f1c07cf627491f6d0481d50a6832507189 /utils | |
| parent | 77b010474e3dab38f73f4225b54c8963a8e90744 (diff) | |
| download | llvm-23f2b865372d6f4712b360e3ac0d0d84f214c8ae.tar.gz | |
[AArch64][DAGCombiner]: combine <2xi64> mul add/sub.
64-bit vector mul is not supported in NEON,
so we use the SVE's mul.
To improve the performance, we can go one step further,
and use SVE's add/sub, so that we can use SVE's mla/mls.
That works on these patterns:
add v1, (mul v2, v3)
sub v1, (mul v2, v3)
Reviewed By: david-arm, paulwalker-arm
Differential Revision: https://reviews.llvm.org/D148118
Diffstat (limited to 'utils')
0 files changed, 0 insertions, 0 deletions