Re: Question about x86/mm/gup.c's use of disabled interrupts

[Avi Kivity <avi@redhat.com>]

Jeremy Fitzhardinge wrote:
> Avi Kivity wrote:
>>> And the hypercall could result in no Xen-level IPIs at all, so it 
>>> could be very quick by comparison to an IPI-based Linux 
>>> implementation, in which case the flag polling would be particularly 
>>> harsh.
>>
>> Maybe we could bring these optimizations into Linux as well.  The 
>> only thing Xen knows that Linux doesn't is if a vcpu is not 
>> scheduled; all other information is shared.
>
> I don't think there's a guarantee that just because a vcpu isn't 
> running now, it won't need a tlb flush.  If a pcpu does runs vcpu 1 -> 
> idle -> vcpu 1, then there's no need for it to do a tlb flush, but the 
> hypercall can make force a flush when it reschedules vcpu 1 (if the 
> tlb hasn't already been flushed by some other means).

That's what I assumed you meant.  Also, if a vcpu has a different cr3 
loaded, the flush can be elided.  Looks like Linux does this 
(s/vcpu/process/).

> (I'm not sure to what extent Xen implements this now, but I wouldn't 
> want to over-constrain it.)

Well, kvm does this.


>> The nice thing about local_irq_disable() is that it scales so well.
>
> Right.  But it effectively puts the burden on the tlb-flusher to check 
> the state (implicitly, by trying to send an interrupt).  Putting an 
> explicit poll in gets the same effect, but its pure overhead just to 
> deal with the gup race.

I guess it hopes the flushes are much rarer.  Certainly for threaded 
databases doing O_DIRECT stuff, I'd expect lots of gupfs and no tlb flushes.


-- 
error compiling committee.c: too many arguments to function