Re: [PATCH -rt] ARM TLB flush fix: don't forget to re-enable preemption

[Kevin Hilman <khilman@mvista.com>]

On Tue, 2007-05-22 at 16:25 -0700, Daniel Walker wrote:
> On Tue, 2007-05-22 at 16:01 -0700, Kevin Hilman wrote:
> > Add a preempt_enable() to flush_tlb_kernel_page() since -rt4 patch
> > adds a preempt_disable but no preempt_enable().
> > 
> > Signed-off-by: Kevin Hilman <khilman@mvista.com>
> > 
> > 
> > ---
> >  include/asm-arm/tlbflush.h |    1 +
> >  1 file changed, 1 insertion(+)
> > 
> > Index: linux-2.6.21/include/asm-arm/tlbflush.h
> > ===================================================================
> > --- linux-2.6.21.orig/include/asm-arm/tlbflush.h
> > +++ linux-2.6.21/include/asm-arm/tlbflush.h
> > @@ -378,6 +378,7 @@ static inline void local_flush_tlb_kerne
> >  		asm("mcr p15, 0, %0, c8, c6, 1" : : "r" (kaddr) : "cc");
> >  	if (tlb_flag(TLB_V6_I_PAGE))
> >  		asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (kaddr) : "cc");
> 
> Aren't these mcr operations atomic?
> 

Individually, yes.  But the point of the preempt_disable/enable is to
make the whole sequence atomic.

Kevin