Re: [PATCH RFC V6 0/11] Paravirtualized ticketlocks

[Avi Kivity <avi@redhat.com>]

On 04/02/2012 12:51 PM, Raghavendra K T wrote:
> On 04/01/2012 07:23 PM, Avi Kivity wrote:
> > On 04/01/2012 04:48 PM, Raghavendra K T wrote:
> >>>> I have patch something like below in mind to try:
> >>>>
> >>>> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> >>>> index d3b98b1..5127668 100644
> >>>> --- a/virt/kvm/kvm_main.c
> >>>> +++ b/virt/kvm/kvm_main.c
> >>>> @@ -1608,15 +1608,18 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me)
> >>>>         * else and called schedule in __vcpu_run.  Hopefully that
> >>>>         * VCPU is holding the lock that we need and will release it.
> >>>>         * We approximate round-robin by starting at the last boosted
> >>>> VCPU.
> >>>> +     * Priority is given to vcpu that are unhalted.
> >>>>         */
> >>>> -    for (pass = 0; pass<   2&&   !yielded; pass++) {
> >>>> +    for (pass = 0; pass<   3&&   !yielded; pass++) {
> >>>>            kvm_for_each_vcpu(i, vcpu, kvm) {
> >>>>                struct task_struct *task = NULL;
> >>>>                struct pid *pid;
> >>>> -            if (!pass&&   i<   last_boosted_vcpu) {
> >>>> +            if (!pass&&   !vcpu->pv_unhalted)
> >>>> +                continue;
> >>>> +            else if (pass == 1&&   i<   last_boosted_vcpu) {
> >>>>                    i = last_boosted_vcpu;
> >>>>                    continue;
> >>>> -            } else if (pass&&   i>   last_boosted_vcpu)
> >>>> +            } else if (pass == 2&&   i>   last_boosted_vcpu)
> >>>>                    break;
> >>>>                if (vcpu == me)
> >>>>                    continue;
> >>>>
> >>>
> >>> Actually I think this is unneeded.  The loops tries to find vcpus
> that
> >>> are runnable but not running (vcpu_active(vcpu->wq)), and halted
> vcpus
> >>> don't match this condition.
> >>>
>
> Oh! I think I misinterpreted your statement. hmm I got it. you told to
> remove if (vcpu == me) condition.

No, the entire patch is unneeded.  My original comment was:

> from the PLE handler, don't wake up a vcpu that is sleeping because it
is waiting for a kick

But the PLE handler never wakes up sleeping vcpus anyway.

There's still a conflict with PLE in that it may trigger during the spin
phase and send a random yield_to() somewhere.  Maybe it's sufficient to
tune the PLE timeout to be longer than the spinlock timeout.


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