On Wednesday 25 April 2007 20:13:34 Jeremy Fitzhardinge wrote:
> Andi Kleen wrote:
> >> +
> >> + do {
> >> + state_time = state->state_entry_time;
> >> + barrier();
> >>
> >
> > Should be likely rmb
>
> We discussed this, and decided that it wasn't necessary. The state is
> always updated by the current CPU, so if it changes under our feet it
> will be because we were preempted, and so that should shoot down any
> speculated reads.
Then the barrier shouldn't be needed at all?
Anyways needs comments
-Andi
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/virtualization
