about summary refs log tree commit diff
path: root/ChangeLog
diff options
context:
space:
mode:
authorH.J. Lu <hjl.tools@gmail.com>2015-01-30 06:50:20 -0800
committerH.J. Lu <hjl.tools@gmail.com>2015-01-30 15:37:58 -0800
commit5f3d0b78e011d2a72f9e88b0e9ef5bc081d18f97 (patch)
tree8eabf127206283d2421bc40b6bc44e123e346598 /ChangeLog
parentb658fdd82b4524cf6a39881d092caa23f63d93ac (diff)
downloadglibc-5f3d0b78e011d2a72f9e88b0e9ef5bc081d18f97.tar.gz
glibc-5f3d0b78e011d2a72f9e88b0e9ef5bc081d18f97.tar.xz
glibc-5f3d0b78e011d2a72f9e88b0e9ef5bc081d18f97.zip
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.
Diffstat (limited to 'ChangeLog')
-rw-r--r--ChangeLog18
1 files changed, 18 insertions, 0 deletions
diff --git a/ChangeLog b/ChangeLog
index 26f7f3f3b1..a696e396b2 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,21 @@
+2015-01-30  H.J. Lu  <hongjiu.lu@intel.com>
+
+	[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.
+
 2015-01-29  Andreas Schwab  <schwab@suse.de>
 
 	* sysdeps/nptl/allocrtsig.c: Include <signal.h>.