about summary refs log tree commit diff
path: root/sysdeps/x86_64/multiarch/init-arch.h
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 /sysdeps/x86_64/multiarch/init-arch.h
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 'sysdeps/x86_64/multiarch/init-arch.h')
-rw-r--r--sysdeps/x86_64/multiarch/init-arch.h4
1 files changed, 4 insertions, 0 deletions
diff --git a/sysdeps/x86_64/multiarch/init-arch.h b/sysdeps/x86_64/multiarch/init-arch.h
index 55f1c5b34c..e6b5ba5530 100644
--- a/sysdeps/x86_64/multiarch/init-arch.h
+++ b/sysdeps/x86_64/multiarch/init-arch.h
@@ -25,6 +25,7 @@
 #define bit_FMA4_Usable			(1 << 8)
 #define bit_Slow_SSE4_2			(1 << 9)
 #define bit_AVX2_Usable			(1 << 10)
+#define bit_AVX_Fast_Unaligned_Load	(1 << 11)
 
 /* CPUID Feature flags.  */
 
@@ -74,6 +75,7 @@
 # define index_FMA4_Usable		FEATURE_INDEX_1*FEATURE_SIZE
 # define index_Slow_SSE4_2		FEATURE_INDEX_1*FEATURE_SIZE
 # define index_AVX2_Usable		FEATURE_INDEX_1*FEATURE_SIZE
+# define index_AVX_Fast_Unaligned_Load	FEATURE_INDEX_1*FEATURE_SIZE
 
 #else	/* __ASSEMBLER__ */
 
@@ -169,6 +171,7 @@ extern const struct cpu_features *__get_cpu_features (void)
 # define index_FMA4_Usable		FEATURE_INDEX_1
 # define index_Slow_SSE4_2		FEATURE_INDEX_1
 # define index_AVX2_Usable		FEATURE_INDEX_1
+# define index_AVX_Fast_Unaligned_Load	FEATURE_INDEX_1
 
 # define HAS_ARCH_FEATURE(name) \
   ((__get_cpu_features ()->feature[index_##name] & (bit_##name)) != 0)
@@ -181,5 +184,6 @@ extern const struct cpu_features *__get_cpu_features (void)
 # define HAS_AVX2			HAS_ARCH_FEATURE (AVX2_Usable)
 # define HAS_FMA			HAS_ARCH_FEATURE (FMA_Usable)
 # define HAS_FMA4			HAS_ARCH_FEATURE (FMA4_Usable)
+# define HAS_AVX_FAST_UNALIGNED_LOAD	HAS_ARCH_FEATURE (AVX_Fast_Unaligned_Load)
 
 #endif	/* __ASSEMBLER__ */