]> sourceware.org Git - glibc.git/commit
x86: Optimize svml_s_tanhf8_core_avx2.S
authorNoah Goldstein <goldstein.w.n@gmail.com>
Thu, 9 Jun 2022 18:16:36 +0000 (11:16 -0700)
committerNoah Goldstein <goldstein.w.n@gmail.com>
Thu, 9 Jun 2022 19:51:22 +0000 (12:51 -0700)
commitbcc41f66a48bf764ee85fea56b8e32719e230a0a
tree227a84a2baf472eac75532ef4c10ec4bf884dae7
parent3a49ce8799e7233946f8154ba05c185277f33f11
x86: Optimize svml_s_tanhf8_core_avx2.S

Optimizations are:
    1. Reduce code size (-81 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 (-32 bytes).

Result is roughly a 17-18% speedup:

       Function, New Time, Old Time, New / Old
_ZGVdN8v_tanhf,     1.977,    2.402,     0.823
sysdeps/x86_64/fpu/multiarch/svml_s_tanhf8_core_avx2.S
This page took 0.040944 seconds and 5 git commands to generate.