Hello,
On 11/30/2009 07:02 PM, Thomas Gleixner wrote:
> No, it _CANNOT_ be preempted at that point:
>
> schedule()
> {
> preempt_disable();
>
> switch_to();
>
> preempt_enable();
> }
Yes, you're right.
>> For the time being, maybe it's best to back out the fix given that the
>> only architecture which may be affected by the original bug is ia64
>> which is the only one with both kvm and the unlocked context switch.
>
> Do you have a pointer to the original bug report ?
Nope, I was referring to the imaginary race condition, so there's no
bug to worry about. The only problem is the asymmetry between in and
out callbacks. Then again, it's not really possible to match them on
unlocked ctxsw archs anyway, so I guess the only thing to do is to
document the context difference between in and out.
Sorry about the fuss. I'll send out patch to revert it and document
the difference.
Thanks.
--
tejun
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
