Re: [PATCH] locking/osq: Drop the overload of osq lock

[Boqun Feng <boqun.feng@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1642 bytes --]

On Sat, Jun 25, 2016 at 06:09:22PM +0200, Peter Zijlstra wrote:
> On Sat, Jun 25, 2016 at 11:21:30PM +0800, Boqun Feng wrote:
> > > 
> > > 	int vpc = vcpu_preempt_count();
> > > 
> > > 	...
> > > 
> > > 	for (;;) {
> > > 
> > > 		/* the big spin loop */
> > > 
> > > 		if (need_resched() || vpc != vcpu_preempt_count())
> > 
> > So on PPC, we have lppaca::yield_count to detect when an vcpu is
> 
> Which sounds like just the value we want.. And I suspect that on x86 KVM
> and Xen have similar numbers stashed away someplace.
> 
> > preempted, if the yield_count is even, the vcpu is running, otherwise it
> > is preempted(__spin_yield() is a user of this).
> > 
> > Therefore it makes more sense we
> > 
> > 		if (need_resched() || vcpu_is_preempted(old))
> > 
> > here, and implement vcpu_is_preempted() on PPC as
> > 
> > bool vcpu_is_preempted(int cpu)
> > {
> > 	return !!(be32_to_cpu(lppaca_of(cpu).yield_count) & 1)
> > }
> > 
> > Thoughts?
> 
> That works here, but it would not work for the need_resched() in
> mutex_spin_on_owner() and mutex_optimistic_spin() which need equal
> treatment.
> 
> Because those too we want to limit.
> 
> The count thing, while a little more cumbersome, is more widely
> applicable than just the one OSQ case where we happen to have a cpu
> number.
> 

But if we don't have a cpu number, which vcpu's preemption are we
trying to detect? I think the logic here is that if _this_ vcpu sees the
_owner_ vcpu is preempted, it should just stop spinning. Therefore, we
need to know the owner cpu number.

Am I missing something here?

Regards,
Boqun


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 473 bytes --]