diff options
| author | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-06-09 09:58:35 -0700 |
|---|---|---|
| committer | Noah Goldstein <goldstein.w.n@gmail.com> | 2022-06-09 12:51:25 -0700 |
| commit | cffb9414c55b2e169ed8af1cefd1e3f2ea97e750 (patch) | |
| tree | 18f8ef507ae394488b620d49cfcbf8d2c5aa93c7 /sysdeps/or1k/Makefile | |
| parent | bcc41f66a48bf764ee85fea56b8e32719e230a0a (diff) | |
| download | glibc-cffb9414c55b2e169ed8af1cefd1e3f2ea97e750.tar glibc-cffb9414c55b2e169ed8af1cefd1e3f2ea97e750.tar.gz glibc-cffb9414c55b2e169ed8af1cefd1e3f2ea97e750.tar.bz2 glibc-cffb9414c55b2e169ed8af1cefd1e3f2ea97e750.zip | |
x86: Optimize svml_s_tanhf4_core_sse4.S
Optimizations are:
1. Reduce code size (-112 bytes).
2. Remove redundant move instructions.
3. Slightly improve instruction selection/scheduling where
possible.
4. Prefer registers which get short instruction encoding.
5. Reduce rodata size (-4k+ rodata is shared with avx2).
Result is roughly a 15-16% speedup:
Function, New Time, Old Time, New / Old
_ZGVbN4v_tanhf, 3.158, 3.749, 0.842
Diffstat (limited to 'sysdeps/or1k/Makefile')
0 files changed, 0 insertions, 0 deletions