about summary refs log tree commit diff
path: root/sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
diff options
context:
space:
mode:
authorAnton Blanchard via Libc-alpha <libc-alpha@sourceware.org>2020-05-14 09:08:35 +1000
committerPaul E. Murphy <murphyp@linux.vnet.ibm.com>2020-05-18 08:26:22 -0500
commitaa70d0563256b8ea053203177f756bca33b5cf37 (patch)
tree152631b5f088b02e3a42f3e1f7defe63cefe33dd /sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
parent39037048502d52ab6422c18f2d178d6228d2c7b9 (diff)
downloadglibc-aa70d0563256b8ea053203177f756bca33b5cf37.tar.gz
glibc-aa70d0563256b8ea053203177f756bca33b5cf37.tar.xz
glibc-aa70d0563256b8ea053203177f756bca33b5cf37.zip
powerpc: Optimized stpcpy for POWER9
Add stpcpy support to the POWER9 strcpy. This is up to 40% faster on
small strings and up to 90% faster on long relatively unaligned strings,
compared to the POWER8 version. A few examples:

                                        __stpcpy_power9  __stpcpy_power8
Length   20, alignments in bytes  4/ 4:  2.58246          4.8788
Length 1024, alignments in bytes  1/ 6: 24.8186          47.8528
Diffstat (limited to 'sysdeps/powerpc/powerpc64/multiarch/stpcpy.c')
-rw-r--r--sysdeps/powerpc/powerpc64/multiarch/stpcpy.c17
1 files changed, 12 insertions, 5 deletions
diff --git a/sysdeps/powerpc/powerpc64/multiarch/stpcpy.c b/sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
index 0a63c8f51a..68e1e648c3 100644
--- a/sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
+++ b/sysdeps/powerpc/powerpc64/multiarch/stpcpy.c
@@ -26,13 +26,20 @@
 extern __typeof (__stpcpy) __stpcpy_ppc attribute_hidden;
 extern __typeof (__stpcpy) __stpcpy_power7 attribute_hidden;
 extern __typeof (__stpcpy) __stpcpy_power8 attribute_hidden;
+# ifdef __LITTLE_ENDIAN__
+extern __typeof (__stpcpy) __stpcpy_power9 attribute_hidden;
+# endif
 
 libc_ifunc_hidden (__stpcpy, __stpcpy,
-		   (hwcap2 & PPC_FEATURE2_ARCH_2_07)
-		   ? __stpcpy_power8
-		   : (hwcap & PPC_FEATURE_HAS_VSX)
-		     ? __stpcpy_power7
-		     : __stpcpy_ppc);
+# ifdef __LITTLE_ENDIAN__
+		   (hwcap2 & PPC_FEATURE2_ARCH_3_00)
+		   ? __stpcpy_power9 :
+# endif
+		     (hwcap2 & PPC_FEATURE2_ARCH_2_07)
+		     ? __stpcpy_power8
+		     : (hwcap & PPC_FEATURE_HAS_VSX)
+		       ? __stpcpy_power7
+		       : __stpcpy_ppc);
 
 weak_alias (__stpcpy, stpcpy)
 libc_hidden_def (__stpcpy)