diff options
| author | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-06-09 09:58:32 -0700 |
|---|---|---|
| committer | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-06-09 12:51:12 -0700 |
| commit | e560b3c2d2b1aeae68bc2da901f773b9a253e594 (patch) | |
| tree | 032c0d20e0cd42058ded58fc51a8f6727f88ee8f /benchtests/ilogbf128-inputs | |
| parent | fe1915d4f6dd5db92ce9d0be47c9d4e04f95946c (diff) | |
| download | glibc-e560b3c2d2b1aeae68bc2da901f773b9a253e594.tar.gz glibc-e560b3c2d2b1aeae68bc2da901f773b9a253e594.tar.xz glibc-e560b3c2d2b1aeae68bc2da901f773b9a253e594.zip | |
x86: Optimize svml_s_tanhf16_core_avx512.S
Optimizations are:
1. Reduce code size (-67 bytes).
2. Remove redundant move instructions.
3. Slightly improve instruction selection/scheduling where
possible.
4. Reduce rodata usage (-448 bytes).
Result is roughly a 14% speedup:
Function, New Time, Old Time, New / Old
_ZGVeN16v_tanhf, 0.649, 0.752, 0.863
Diffstat (limited to 'benchtests/ilogbf128-inputs')
0 files changed, 0 insertions, 0 deletions