Andrew Senkevich
72276d6e88
Added memcpy/memmove family optimized with AVX512 for KNL hardware.
...
Added AVX512 implementations of memcpy, mempcpy, memmove, memcpy_chk,
mempcpy_chk, memmove_chk.
It shows average improvement more than 30% over AVX versions on KNL
hardware (performance results in the thread
<https://sourceware.org/ml/libc-alpha/2016-01/msg00258.html >).
* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Added new files.
* sysdeps/x86_64/multiarch/ifunc-impl-list.c: Added new tests.
* sysdeps/x86_64/multiarch/memcpy-avx512-no-vzeroupper.S: New file.
* sysdeps/x86_64/multiarch/mempcpy-avx512-no-vzeroupper.S: Likewise.
* sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S: Likewise.
* sysdeps/x86_64/multiarch/memcpy.S: Added new IFUNC branch.
* sysdeps/x86_64/multiarch/memcpy_chk.S: Likewise.
* sysdeps/x86_64/multiarch/memmove.c: Likewise.
* sysdeps/x86_64/multiarch/memmove_chk.c: Likewise.
* sysdeps/x86_64/multiarch/mempcpy.S: Likewise.
* sysdeps/x86_64/multiarch/mempcpy_chk.S: Likewise.
2016-01-16 00:49:45 +03:00
Joseph Myers
f7a9f785e5
Update copyright dates with scripts/update-copyrights.
2016-01-04 16:05:18 +00:00
H.J. Lu
0b5395f052
Update x86_64 multiarch functions for <cpu-features.h>
...
This patch updates x86_64 multiarch functions to use the newly defined
HAS_CPU_FEATURE, HAS_ARCH_FEATURE and LOAD_RTLD_GLOBAL_RO_RDX from
<cpu-features.h>.
* sysdeps/x86_64/fpu/multiarch/e_asin.c: Replace HAS_XXX with
HAS_CPU_FEATURE/HAS_ARCH_FEATURE (XXX).
* sysdeps/x86_64/fpu/multiarch/e_atan2.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/e_exp.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/e_log.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/e_pow.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_atan.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_fma.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_fmaf.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_sin.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_tan.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_ceil.S: Use
LOAD_RTLD_GLOBAL_RO_RDX and HAS_CPU_FEATURE (SSE4_1).
* sysdeps/x86_64/fpu/multiarch/s_ceilf.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_floor.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_floorf.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_nearbyint.S : Likewise.
* sysdeps/x86_64/fpu/multiarch/s_nearbyintf.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_rintf.S: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_rintf.S : Likewise.
* sysdeps/x86_64/multiarch/ifunc-impl-list.c: Likewise.
* sysdeps/x86_64/multiarch/sched_cpucount.c: Likewise.
* sysdeps/x86_64/multiarch/strstr.c: Likewise.
* sysdeps/x86_64/multiarch/memmove.c: Likewise.
* sysdeps/x86_64/multiarch/memmove_chk.c: Likewise.
* sysdeps/x86_64/multiarch/test-multiarch.c: Likewise.
* sysdeps/x86_64/multiarch/memcmp.S: Remove __init_cpu_features
call. Add LOAD_RTLD_GLOBAL_RO_RDX. Replace HAS_XXX with
HAS_CPU_FEATURE/HAS_ARCH_FEATURE (XXX).
* sysdeps/x86_64/multiarch/memcpy.S: Likewise.
* sysdeps/x86_64/multiarch/memcpy_chk.S: Likewise.
* sysdeps/x86_64/multiarch/mempcpy.S: Likewise.
* sysdeps/x86_64/multiarch/mempcpy_chk.S: Likewise.
* sysdeps/x86_64/multiarch/memset.S: Likewise.
* sysdeps/x86_64/multiarch/memset_chk.S: Likewise.
* sysdeps/x86_64/multiarch/strcat.S: Likewise.
* sysdeps/x86_64/multiarch/strchr.S: Likewise.
* sysdeps/x86_64/multiarch/strcmp.S: Likewise.
* sysdeps/x86_64/multiarch/strcpy.S: Likewise.
* sysdeps/x86_64/multiarch/strcspn.S: Likewise.
* sysdeps/x86_64/multiarch/strspn.S: Likewise.
* sysdeps/x86_64/multiarch/wcscpy.S: Likewise.
* sysdeps/x86_64/multiarch/wmemcmp.S: Likewise.
2015-08-13 03:41:30 -07:00
H.J. Lu
5f3d0b78e0
Use AVX unaligned memcpy only if AVX2 is available
...
memcpy with unaligned 256-bit AVX register loads/stores are slow on older
processorsl like Sandy Bridge. This patch adds bit_AVX_Fast_Unaligned_Load
and sets it only when AVX2 is available.
[BZ #17801 ]
* sysdeps/x86_64/multiarch/init-arch.c (__init_cpu_features):
Set the bit_AVX_Fast_Unaligned_Load bit for AVX2.
* sysdeps/x86_64/multiarch/init-arch.h (bit_AVX_Fast_Unaligned_Load):
New.
(index_AVX_Fast_Unaligned_Load): Likewise.
(HAS_AVX_FAST_UNALIGNED_LOAD): Likewise.
* sysdeps/x86_64/multiarch/memcpy.S (__new_memcpy): Check the
bit_AVX_Fast_Unaligned_Load bit instead of the bit_AVX_Usable bit.
* sysdeps/x86_64/multiarch/memcpy_chk.S (__memcpy_chk): Likewise.
* sysdeps/x86_64/multiarch/mempcpy.S (__mempcpy): Likewise.
* sysdeps/x86_64/multiarch/mempcpy_chk.S (__mempcpy_chk): Likewise.
* sysdeps/x86_64/multiarch/memmove.c (__libc_memmove): Replace
HAS_AVX with HAS_AVX_FAST_UNALIGNED_LOAD.
* sysdeps/x86_64/multiarch/memmove_chk.c (__memmove_chk): Likewise.
2015-01-30 15:37:58 -08:00
Joseph Myers
b168057aaa
Update copyright dates with scripts/update-copyrights.
2015-01-02 16:29:47 +00:00
Ling Ma
05f3633da4
Improve 64bit memcpy performance for Haswell CPU with AVX instruction
...
In this patch we take advantage of HSW memory bandwidth, manage to
reduce miss branch prediction by avoiding using branch instructions and
force destination to be aligned with avx instruction.
The CPU2006 403.gcc benchmark indicates this patch improves performance
from 2% to 10%.
2014-07-30 08:02:35 -07:00
Allan McRae
d4697bc93d
Update copyright notices with scripts/update-copyrights
2014-01-01 22:00:23 +10:00
Joseph Myers
568035b787
Update copyright notices with scripts/update-copyrights.
2013-01-02 19:05:09 +00:00
H.J. Lu
ac49ecaf9d
Add x86-64 __libc_ifunc_impl_list
2012-10-11 16:41:12 -07:00
Paul Eggert
59ba27a63a
Replace FSF snail mail address with URLs.
2012-02-09 23:18:22 +00:00
Roland McGrath
8b2b771538
Clean up warnings in new x86_64/multiarch code.
2010-08-25 12:13:08 -07:00
H.J. Lu
6fb8cbcb58
Improve 64bit memcpy/memmove for Atom, Core 2 and Core i7
...
This patch includes optimized 64bit memcpy/memmove for Atom, Core 2 and
Core i7. It improves memcpy by up to 3X on Atom, up to 4X on Core 2 and
up to 1X on Core i7. It also improves memmove by up to 3X on Atom, up to
4X on Core 2 and up to 2X on Core i7.
2010-06-30 08:26:11 -07:00
