AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-09-20 08:30:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

x86: Add MathVec_Prefer_No_AVX512 to cpu-features [BZ #21967]

Browse Source 
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.
This commit is contained in:
H.J. Lu 2017-09-12 07:46:11 -07:00
parent a166531fdf
commit ef8adeb041
4 changed files with 28 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
ChangeLog
View File

@ -1,3 +1,15 @@
2017-09-12 H.J. Lu <hongjiu.lu@intel.com>
[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.
2017-09-12 Samuel Thibault <samuel.thibault@ens-lyon.org>
* posix/sched_primax.c (__sched_get_priority_max): Add

2
sysdeps/x86/cpu-features.h
View File

@ -40,6 +40,7 @@
#define bit_arch_Use_dl_runtime_resolve_opt (1 << 20)
#define bit_arch_Use_dl_runtime_resolve_slow (1 << 21)
#define bit_arch_Prefer_No_AVX512 (1 << 22)
#define bit_arch_MathVec_Prefer_No_AVX512 (1 << 23)
/* CPUID Feature flags. */
@ -239,6 +240,7 @@ extern const struct cpu_features *__get_cpu_features (void)
# define index_arch_Use_dl_runtime_resolve_opt FEATURE_INDEX_1
# define index_arch_Use_dl_runtime_resolve_slow FEATURE_INDEX_1
# define index_arch_Prefer_No_AVX512 FEATURE_INDEX_1
# define index_arch_MathVec_Prefer_No_AVX512 FEATURE_INDEX_1
#endif /* !__ASSEMBLER__ */

7
sysdeps/x86/cpu-tunables.c
View File

@ -303,6 +303,13 @@ TUNABLE_CALLBACK (set_hwcaps) (tunable_val_t *valp)
disable, 23);
}
break;
case 24:
{
CHECK_GLIBC_IFUNC_ARCH_NEED_ARCH_BOTH
(n, cpu_features, MathVec_Prefer_No_AVX512,
AVX512F_Usable, disable, 24);
}
break;
case 26:
{
CHECK_GLIBC_IFUNC_ARCH_NEED_CPU_BOTH

11
sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h
View File

@ -32,11 +32,14 @@ IFUNC_SELECTOR (void)
{
const struct cpu_features* cpu_features = __get_cpu_features ();
if (CPU_FEATURES_ARCH_P (cpu_features, AVX512DQ_Usable))
return OPTIMIZE (skx);
if (!CPU_FEATURES_ARCH_P (cpu_features, MathVec_Prefer_No_AVX512))
{
if (CPU_FEATURES_ARCH_P (cpu_features, AVX512DQ_Usable))
return OPTIMIZE (skx);
if (CPU_FEATURES_ARCH_P (cpu_features, AVX512F_Usable))
return OPTIMIZE (knl);
if (CPU_FEATURES_ARCH_P (cpu_features, AVX512F_Usable))
return OPTIMIZE (knl);
}
return OPTIMIZE (avx2_wrapper);
}