diff options
| author | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2022-05-31 18:13:54 -0300 |
|---|---|---|
| committer | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2022-06-01 10:47:44 -0300 |
| commit | 5a6f2cabb62b2d5e4c356718283b6cb7e440ac9c (patch) | |
| tree | 42acdf13f5f471a297a87d2f18769ff7bd9ca794 /sysdeps/ieee754 | |
| parent | dc208f4a53e71560432ad6a4aa42da1becc7bd3f (diff) | |
| download | glibc-5a6f2cabb62b2d5e4c356718283b6cb7e440ac9c.tar glibc-5a6f2cabb62b2d5e4c356718283b6cb7e440ac9c.tar.gz glibc-5a6f2cabb62b2d5e4c356718283b6cb7e440ac9c.tar.bz2 glibc-5a6f2cabb62b2d5e4c356718283b6cb7e440ac9c.zip | |
i686: Use generic sincosf implementation for SSE2 version
The generic implementation shows slight better performance
(gcc 11.2.1 on a Ryzen 9 5900X):
* s_sincosf-sse2.S:
"sincosf": {
"workload-random": {
"duration": 3.89961e+09,
"iterations": 9.5472e+07,
"reciprocal-throughput": 40.8429,
"latency": 40.8483,
"max-throughput": 2.4484e+07,
"min-throughput": 2.44808e+07
}
}
* generic s_cossinf.c:
"sincosf": {
"workload-random": {
"duration": 3.71953e+09,
"iterations": 1.48512e+08,
"reciprocal-throughput": 25.0515,
"latency": 25.0391,
"max-throughput": 3.99177e+07,
"min-throughput": 3.99375e+07
}
}
Checked on i686-linux-gnu.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Diffstat (limited to 'sysdeps/ieee754')
| -rw-r--r-- | sysdeps/ieee754/flt-32/s_sincosf.c | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/sysdeps/ieee754/flt-32/s_sincosf.c b/sysdeps/ieee754/flt-32/s_sincosf.c index 918f6a655a..4fab42d974 100644 --- a/sysdeps/ieee754/flt-32/s_sincosf.c +++ b/sysdeps/ieee754/flt-32/s_sincosf.c @@ -24,6 +24,10 @@ #include "math_config.h" #include "s_sincosf.h" +#ifndef SECTION +# define SECTION +#endif + #ifndef SINCOSF # define SINCOSF_FUNC __sincosf #else @@ -35,6 +39,7 @@ small values. Large inputs have their range reduced using fast integer arithmetic. */ void +SECTION SINCOSF_FUNC (float y, float *sinp, float *cosp) { double x = y; |