powerpc64: strchr/strchrnul optimization for power8
commit9314d3545e6641063b490918e2e8716556ba20db
authorRajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
Tue, 27 Dec 2016 19:48:37 +0000 (27 17:48 -0200)
committerTulio Magno Quites Machado Filho <tuliom@linux.vnet.ibm.com>
Wed, 28 Dec 2016 13:44:31 +0000 (28 11:44 -0200)
treecc3d6518db2c0a86db8dcbb22ee44e789a0a9920
parentffcf0f1cb77dd0d902495fd066a96784f76f2c3a
powerpc64: strchr/strchrnul optimization for power8

The P7 code is used for <=32B strings and for > 32B vectorized loops are used.
This shows as an average 25% improvement depending on the position of search
character.  The performance is same for shorter strings.
Tested on ppc64 and ppc64le.
ChangeLog
sysdeps/powerpc/powerpc64/multiarch/Makefile
sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
sysdeps/powerpc/powerpc64/multiarch/strchr-power8.S [copied from sysdeps/powerpc/powerpc64/multiarch/strchrnul.c with 54% similarity]
sysdeps/powerpc/powerpc64/multiarch/strchr.c
sysdeps/powerpc/powerpc64/multiarch/strchrnul-power8.S [copied from sysdeps/powerpc/powerpc64/multiarch/strchrnul.c with 54% similarity]
sysdeps/powerpc/powerpc64/multiarch/strchrnul.c
sysdeps/powerpc/powerpc64/power8/strchr.S [new file with mode: 0644]
sysdeps/powerpc/powerpc64/power8/strchrnul.S [copied from sysdeps/powerpc/powerpc64/multiarch/strchrnul.c with 54% similarity]