diff options
| author | Ling Ma <ling.ml@alibaba-inc.com> | 2014-07-14 00:02:52 -0400 |
|---|---|---|
| committer | H.J. Lu <hjl.tools@gmail.com> | 2014-07-30 08:02:35 -0700 |
| commit | 05f3633da4f9df870d04dd77336e793746e57ed4 (patch) | |
| tree | bd980df4702ffd9ebfa9cb4dac8823a659f0c430 /sysdeps/x86_64/multiarch/ifunc-impl-list.c | |
| parent | a53fbd8e6cd2f69bdfa3431d616a5f332aea6664 (diff) | |
| download | glibc-05f3633da4f9df870d04dd77336e793746e57ed4.tar glibc-05f3633da4f9df870d04dd77336e793746e57ed4.tar.gz glibc-05f3633da4f9df870d04dd77336e793746e57ed4.tar.bz2 glibc-05f3633da4f9df870d04dd77336e793746e57ed4.zip | |
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%.
Diffstat (limited to 'sysdeps/x86_64/multiarch/ifunc-impl-list.c')
| -rw-r--r-- | sysdeps/x86_64/multiarch/ifunc-impl-list.c | 12 |
1 files changed, 12 insertions, 0 deletions
diff --git a/sysdeps/x86_64/multiarch/ifunc-impl-list.c b/sysdeps/x86_64/multiarch/ifunc-impl-list.c index 7e93e598db..78e9b20079 100644 --- a/sysdeps/x86_64/multiarch/ifunc-impl-list.c +++ b/sysdeps/x86_64/multiarch/ifunc-impl-list.c @@ -46,6 +46,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, /* Support sysdeps/x86_64/multiarch/memmove_chk.S. */ IFUNC_IMPL (i, name, __memmove_chk, + IFUNC_IMPL_ADD (array, i, __memmove_chk, HAS_AVX, + __memmove_chk_avx_unaligned) IFUNC_IMPL_ADD (array, i, __memmove_chk, HAS_SSSE3, __memmove_chk_ssse3_back) IFUNC_IMPL_ADD (array, i, __memmove_chk, HAS_SSSE3, @@ -55,6 +57,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, /* Support sysdeps/x86_64/multiarch/memmove.S. */ IFUNC_IMPL (i, name, memmove, + IFUNC_IMPL_ADD (array, i, memmove, HAS_AVX, + __memmove_avx_unaligned) IFUNC_IMPL_ADD (array, i, memmove, HAS_SSSE3, __memmove_ssse3_back) IFUNC_IMPL_ADD (array, i, memmove, HAS_SSSE3, @@ -214,6 +218,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, #ifdef SHARED /* Support sysdeps/x86_64/multiarch/memcpy_chk.S. */ IFUNC_IMPL (i, name, __memcpy_chk, + IFUNC_IMPL_ADD (array, i, __memcpy_chk, HAS_AVX, + __memcpy_chk_avx_unaligned) IFUNC_IMPL_ADD (array, i, __memcpy_chk, HAS_SSSE3, __memcpy_chk_ssse3_back) IFUNC_IMPL_ADD (array, i, __memcpy_chk, HAS_SSSE3, @@ -223,6 +229,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, /* Support sysdeps/x86_64/multiarch/memcpy.S. */ IFUNC_IMPL (i, name, memcpy, + IFUNC_IMPL_ADD (array, i, memcpy, HAS_AVX, + __memcpy_avx_unaligned) IFUNC_IMPL_ADD (array, i, memcpy, HAS_SSSE3, __memcpy_ssse3_back) IFUNC_IMPL_ADD (array, i, memcpy, HAS_SSSE3, __memcpy_ssse3) @@ -231,6 +239,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, /* Support sysdeps/x86_64/multiarch/mempcpy_chk.S. */ IFUNC_IMPL (i, name, __mempcpy_chk, + IFUNC_IMPL_ADD (array, i, __mempcpy_chk, HAS_AVX, + __mempcpy_chk_avx_unaligned) IFUNC_IMPL_ADD (array, i, __mempcpy_chk, HAS_SSSE3, __mempcpy_chk_ssse3_back) IFUNC_IMPL_ADD (array, i, __mempcpy_chk, HAS_SSSE3, @@ -240,6 +250,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, /* Support sysdeps/x86_64/multiarch/mempcpy.S. */ IFUNC_IMPL (i, name, mempcpy, + IFUNC_IMPL_ADD (array, i, mempcpy, HAS_AVX, + __mempcpy_avx_unaligned) IFUNC_IMPL_ADD (array, i, mempcpy, HAS_SSSE3, __mempcpy_ssse3_back) IFUNC_IMPL_ADD (array, i, mempcpy, HAS_SSSE3, |