[PATCH 05/05] Math optimizations for POWER8

Adhemerval Zanella azanella@linux.vnet.ibm.com
Wed Feb 19 10:36:00 GMT 2014


On 19-02-2014 04:51, Rajalakshmi Srinivasaraghavan wrote:
>> This patch add a optimized llround/llroundf implementation for POWER8
>> using the new Move From VSR Doubleword instruction to gains some
>> cycles from FP to GRP register move.
>
> In sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround.c
>
> in libc_ifunc() should it end as  : __llround_ppc64); instead of   : __llround_power8);  for default.
>
Thanks for spotting it, fixed.



More information about the Libc-alpha mailing list