On Fri, 27 Jul 2012 12:57:50 -0400
Konrad Rzeszutek Wilk <konrad@xxxxxxxxxx> wrote:
> On Thu, Jul 26, 2012 at 05:47:13PM +0200, Martin Schwidefsky wrote:
> > To enable lazy TLB flush schemes with a scope limited to a single
> > mm_struct add the mm pointer as argument to the three lazy mmu mode
> > hooks.
> >
> > diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
> > index 0b47ddb..b097945 100644
> > --- a/arch/x86/include/asm/paravirt.h
> > +++ b/arch/x86/include/asm/paravirt.h
> > @@ -694,17 +694,17 @@ static inline void arch_end_context_switch(struct task_struct *next)
> > }
> >
> > #define __HAVE_ARCH_ENTER_LAZY_MMU_MODE
> > -static inline void arch_enter_lazy_mmu_mode(void)
> > +static inline void arch_enter_lazy_mmu_mode(struct mm_struct *mm)
> > {
> > PVOP_VCALL0(pv_mmu_ops.lazy_mode.enter);
>
> If you are doing that, you should probably also update the pvops call to
> pass in the 'struct mm_struct'?
>
Seems reasonable, if we limit the lazy mmu flushing to a single mm then
that fact should be represented on the pvops calls as well.
--
blue skies,
Martin.
"Reality continues to ruin my life." - Calvin.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>