mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-23 13:30:06 +00:00
ef8adeb041
AVX512 functions in mathvec are used on machines with AVX512. An AVX2 wrapper is also provided and it can be used when the AVX512 version isn't profitable. MathVec_Prefer_No_AVX512 is addded to cpu-features. If glibc.tune.hwcaps=MathVec_Prefer_No_AVX512 is set in GLIBC_TUNABLES environment variable, the AVX2 wrapper will be used. Tested on x86-64 machines with and without AVX512. Also verified glibc.tune.hwcaps=MathVec_Prefer_No_AVX512 on AVX512 machine. [BZ #21967] * sysdeps/x86/cpu-features.h (bit_arch_MathVec_Prefer_No_AVX512): New. (index_arch_MathVec_Prefer_No_AVX512): Likewise. * sysdeps/x86/cpu-tunables.c (TUNABLE_CALLBACK (set_hwcaps)): Handle MathVec_Prefer_No_AVX512. * sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h (IFUNC_SELECTOR): Return AVX2 version if MathVec_Prefer_No_AVX512 is set. |
||
---|---|---|
.. | ||
bits | ||
fpu | ||
nptl/bits | ||
cacheinfo.c | ||
cpu-features-offsets.sym | ||
cpu-features.c | ||
cpu-features.h | ||
cpu-tunables.c | ||
dl-get-cpu-features.c | ||
dl-hwcap.h | ||
dl-procinfo.c | ||
dl-procinfo.h | ||
dl-tunables.list | ||
elide.h | ||
float128-abi.h | ||
fpu_control.h | ||
init-arch.h | ||
libc-start.c | ||
linkmap.h | ||
Makefile | ||
math-tests.h | ||
string_private.h | ||
tininess.h | ||
tst-get-cpu-features-static.c | ||
tst-get-cpu-features.c | ||
Versions |