diff options
| author | Joe Ramsay <Joe.Ramsay@arm.com> | 2024-02-20 16:44:13 +0000 |
|---|---|---|
| committer | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2024-02-26 09:45:50 -0300 |
| commit | e302e1021391d13a9611ba3a910df128830bd19e (patch) | |
| tree | 6b672bb0065aa687d1e5e35c6c7fde07c5345961 /sysdeps/aarch64/fpu/exp2f_sve.c | |
| parent | 02782fd12849b6673cb5c2728cb750e8ec295aa3 (diff) | |
| download | glibc-e302e1021391d13a9611ba3a910df128830bd19e.tar.gz glibc-e302e1021391d13a9611ba3a910df128830bd19e.tar.xz glibc-e302e1021391d13a9611ba3a910df128830bd19e.zip | |
aarch64/fpu: Sync libmvec routines from 2.39 and before with AOR
This includes a fix for big-endian in AdvSIMD log, some cosmetic changes, and numerous small optimisations mainly around inlining and using indexed variants of MLA intrinsics. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Diffstat (limited to 'sysdeps/aarch64/fpu/exp2f_sve.c')
| -rw-r--r-- | sysdeps/aarch64/fpu/exp2f_sve.c | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/sysdeps/aarch64/fpu/exp2f_sve.c b/sysdeps/aarch64/fpu/exp2f_sve.c index 9a5a523a10..8a686e3e05 100644 --- a/sysdeps/aarch64/fpu/exp2f_sve.c +++ b/sysdeps/aarch64/fpu/exp2f_sve.c @@ -20,6 +20,8 @@ #include "sv_math.h" #include "poly_sve_f32.h" +#define Thres 0x1.5d5e2ap+6f + static const struct data { float poly[5]; @@ -33,7 +35,7 @@ static const struct data .shift = 0x1.903f8p17f, /* Roughly 87.3. For x < -Thres, the result is subnormal and not handled correctly by FEXPA. */ - .thres = 0x1.5d5e2ap+6f, + .thres = Thres, }; static svfloat32_t NOINLINE |