Re: [PATCH] powerpc/kernel: Avoid redundancies on giveup_all

2017-06-23 Thread Breno Leitao
Hi Cyril, On Fri, Jun 23, 2017 at 04:03:12PM +1000, Cyril Bur wrote: > On Thu, 2017-06-22 at 17:27 -0300, Breno Leitao wrote: > > Currently giveup_all() calls __giveup_fpu(), __giveup_altivec(), and > > __giveup_vsx(). But __giveup_vsx() also calls __giveup_fpu() and > > __giveup_altivec() again,

Re: [PATCH] powerpc/kernel: Avoid redundancies on giveup_all

2017-06-22 Thread Cyril Bur
On Thu, 2017-06-22 at 17:27 -0300, Breno Leitao wrote: > Currently giveup_all() calls __giveup_fpu(), __giveup_altivec(), and > __giveup_vsx(). But __giveup_vsx() also calls __giveup_fpu() and > __giveup_altivec() again, in a redudant manner. > > Other than giving up FP and Altivec, __giveup_vsx()

[PATCH] powerpc/kernel: Avoid redundancies on giveup_all

2017-06-22 Thread Breno Leitao
Currently giveup_all() calls __giveup_fpu(), __giveup_altivec(), and __giveup_vsx(). But __giveup_vsx() also calls __giveup_fpu() and __giveup_altivec() again, in a redudant manner. Other than giving up FP and Altivec, __giveup_vsx() also disables MSR_VSX on MSR, but this is already done by __give