Re: [PATCH] powerpc/asm: Mark cr0 as clobbered in mftb()

2017-07-10 Thread Oliver
On Mon, Jul 10, 2017 at 9:50 PM, Michael Ellerman wrote: > Oliver O'Halloran writes: > >> The workaround for the CELL timebase bug does not correctly mark cr0 as >> being clobbered. This can result in GCC making some poor^W completely >> broken

Re: [PATCH] powerpc/asm: Mark cr0 as clobbered in mftb()

2017-07-10 Thread Michael Ellerman
Segher Boessenkool writes: > Hi! > > On Mon, Jul 10, 2017 at 09:50:06PM +1000, Michael Ellerman wrote: >> Oliver O'Halloran writes: >> >> > The workaround for the CELL timebase bug does not correctly mark cr0 as >> > being clobbered. This can

Re: [PATCH] powerpc/asm: Mark cr0 as clobbered in mftb()

2017-07-10 Thread Segher Boessenkool
Hi! On Mon, Jul 10, 2017 at 09:50:06PM +1000, Michael Ellerman wrote: > Oliver O'Halloran writes: > > > The workaround for the CELL timebase bug does not correctly mark cr0 as > > being clobbered. This can result in GCC making some poor^W completely > > broken optimisations. >

Re: [PATCH] powerpc/asm: Mark cr0 as clobbered in mftb()

2017-07-10 Thread Michael Ellerman
Oliver O'Halloran writes: > The workaround for the CELL timebase bug does not correctly mark cr0 as > being clobbered. This can result in GCC making some poor^W completely > broken optimisations. Fruit 'n oats, how did we never notice that? Luck I guess. Or subtle breakage

[PATCH] powerpc/asm: Mark cr0 as clobbered in mftb()

2017-07-06 Thread Oliver O'Halloran
The workaround for the CELL timebase bug does not correctly mark cr0 as being clobbered. This can result in GCC making some poor^W completely broken optimisations. Signed-off-by: Oliver O'Halloran --- arch/powerpc/include/asm/reg.h | 2 +- 1 file changed, 1 insertion(+), 1