about summary refs log tree commit diff
path: root/sysdeps/x86_64/multiarch/memmove_chk.c
diff options
context:
space:
mode:
authorAndrew Senkevich <andrew.senkevich@intel.com>2016-01-16 00:49:45 +0300
committerAndrew Senkevich <andrew.senkevich@intel.com>2016-01-16 00:49:45 +0300
commit72276d6e8843db6df5971b06787f0a5e39bda138 (patch)
treead7ed01db58285d38559773305d5d8b16eca39d3 /sysdeps/x86_64/multiarch/memmove_chk.c
parentb02840bacdefde318d2ad2f920e50785b9b25d69 (diff)
downloadglibc-72276d6e8843db6df5971b06787f0a5e39bda138.tar.gz
glibc-72276d6e8843db6df5971b06787f0a5e39bda138.tar.xz
glibc-72276d6e8843db6df5971b06787f0a5e39bda138.zip
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.
Diffstat (limited to 'sysdeps/x86_64/multiarch/memmove_chk.c')
-rw-r--r--sysdeps/x86_64/multiarch/memmove_chk.c9
1 files changed, 9 insertions, 0 deletions
diff --git a/sysdeps/x86_64/multiarch/memmove_chk.c b/sysdeps/x86_64/multiarch/memmove_chk.c
index 60ed98fd40..f64da63180 100644
--- a/sysdeps/x86_64/multiarch/memmove_chk.c
+++ b/sysdeps/x86_64/multiarch/memmove_chk.c
@@ -26,10 +26,19 @@ extern __typeof (__memmove_chk) __memmove_chk_sse2 attribute_hidden;
 extern __typeof (__memmove_chk) __memmove_chk_ssse3 attribute_hidden;
 extern __typeof (__memmove_chk) __memmove_chk_ssse3_back attribute_hidden;
 extern __typeof (__memmove_chk) __memmove_chk_avx_unaligned attribute_hidden;
+# ifdef HAVE_AVX512_ASM_SUPPORT
+  extern __typeof (__memmove_chk) __memmove_chk_avx512_no_vzeroupper attribute_hidden;
+# endif
 
 #include "debug/memmove_chk.c"
 
 libc_ifunc (__memmove_chk,
+#ifdef HAVE_AVX512_ASM_SUPPORT
+	    HAS_ARCH_FEATURE (AVX512F_Usable)
+	      && HAS_ARCH_FEATURE (Prefer_No_VZEROUPPER)
+	    ? __memmove_chk_avx512_no_vzeroupper
+	    :
+#endif
 	    HAS_ARCH_FEATURE (AVX_Fast_Unaligned_Load) ? __memmove_chk_avx_unaligned :
 	    (HAS_CPU_FEATURE (SSSE3)
 	    ? (HAS_ARCH_FEATURE (Fast_Copy_Backward)