Re: [PATCH 4/4] powerpc/32: remove a NOP from memset()

2017-08-24 Thread Michael Ellerman
Christophe LEROY writes: > Le 24/08/2017 à 12:51, Michael Ellerman a écrit : >> Christophe Leroy writes: >> >>> memset() is patched after initialisation to activate the >>> optimised part which uses cache instructions. >>> >>> Today we have a 'b 2f' to skip the optimised patch, which then gets

Re: [PATCH 4/4] powerpc/32: remove a NOP from memset()

2017-08-24 Thread Christophe LEROY
Le 24/08/2017 à 12:51, Michael Ellerman a écrit : Christophe Leroy writes: memset() is patched after initialisation to activate the optimised part which uses cache instructions. Today we have a 'b 2f' to skip the optimised patch, which then gets replaced by a NOP, implying a useless cycle c

Re: [PATCH 4/4] powerpc/32: remove a NOP from memset()

2017-08-24 Thread Michael Ellerman
Christophe Leroy writes: > memset() is patched after initialisation to activate the > optimised part which uses cache instructions. > > Today we have a 'b 2f' to skip the optimised patch, which then gets > replaced by a NOP, implying a useless cycle consumption. > As we have a 'bne 2f' just befor

[PATCH 4/4] powerpc/32: remove a NOP from memset()

2017-08-23 Thread Christophe Leroy
memset() is patched after initialisation to activate the optimised part which uses cache instructions. Today we have a 'b 2f' to skip the optimised patch, which then gets replaced by a NOP, implying a useless cycle consumption. As we have a 'bne 2f' just before, we could use that instruction for t