[PATCH] powerpc64: strchr/strchrnul optimization for power8

Tulio Magno Quites Machado Filho tuliom@linux.vnet.ibm.com
Tue Dec 27 19:56:00 GMT 2016


Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com> writes:

> 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.
>
> 2016-11-08  Rajalakshmi Srinivasaraghavan  <raji@linux.vnet.ibm.com>
>
> 	* sysdeps/powerpc/powerpc64/multiarch/Makefile
> 	(sysdep_routines): Add strchr-power8 and strchrnul_power8.
> 	* sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
> 	(strchr): Add __strchr_power8 to list of strchr functions.
> 	(strchrnul): Add __strchrnul_power8 to list of strchr functions.
> 	* sysdeps/powerpc/powerpc64/multiarch/strchr-power8.S: New file.
> 	* sysdeps/powerpc/powerpc64/multiarch/strchrnul-power8.S: New file.
> 	* sysdeps/powerpc/powerpc64/multiarch/strchr.c
> 	(strchr): Add __strchr_power8 to ifunc list.
> 	* sysdeps/powerpc/powerpc64/multiarch/strchrnul.c
> 	(__strchrnul): Add __strchrnul_power8 to ifunc list.
> 	* sysdeps/powerpc/powerpc64/power8/strchr.S: New file.
> 	* sysdeps/powerpc/powerpc64/power8/strchrnul.S: New file.

LGTM.

-- 
Tulio Magno



More information about the Libc-alpha mailing list