On Thu, Oct 07, 2010 at 07:18:03PM +0200, Avi Kivity wrote:
quoted text
>  On 10/07/2010 07:14 PM, Gleb Natapov wrote:
> >On Thu, Oct 07, 2010 at 03:10:27PM +0200, Avi Kivity wrote:
> >>   On 10/04/2010 05:56 PM, Gleb Natapov wrote:
> >>  >When async PF capability is detected hook up special page fault handler
> >>  >that will handle async page fault events and bypass other page faults to
> >>  >regular page fault handler. Also add async PF handling to nested SVM
> >>  >emulation. Async PF always generates exit to L1 where vcpu thread will
> >>  >be scheduled out until page is available.
> >>  >
> >>
> >>  Please separate guest and host changes.
> >>
> >>  >+void kvm_async_pf_task_wait(u32 token)
> >>  >+{
> >>  >+	u32 key = hash_32(token, KVM_TASK_SLEEP_HASHBITS);
> >>  >+	struct kvm_task_sleep_head *b =&async_pf_sleepers[key];
> >>  >+	struct kvm_task_sleep_node n, *e;
> >>  >+	DEFINE_WAIT(wait);
> >>  >+
> >>  >+	spin_lock(&b->lock);
> >>  >+	e = _find_apf_task(b, token);
> >>  >+	if (e) {
> >>  >+		/* dummy entry exist ->   wake up was delivered ahead of PF */
> >>  >+		hlist_del(&e->link);
> >>  >+		kfree(e);
> >>  >+		spin_unlock(&b->lock);
> >>  >+		return;
> >>  >+	}
> >>  >+
> >>  >+	n.token = token;
> >>  >+	n.cpu = smp_processor_id();
> >>  >+	init_waitqueue_head(&n.wq);
> >>  >+	hlist_add_head(&n.link,&b->list);
> >>  >+	spin_unlock(&b->lock);
> >>  >+
> >>  >+	for (;;) {
> >>  >+		prepare_to_wait(&n.wq,&wait, TASK_UNINTERRUPTIBLE);
> >>  >+		if (hlist_unhashed(&n.link))
> >>  >+			break;
> >>  >+		local_irq_enable();
> >>
> >>  Suppose we take another apf here.  And another, and another (for
> >>  different pages, while executing schedule()).  What's to prevent
> >>  kernel stack overflow?
> >>
> >Host side keeps track of outstanding apfs and will not send apf for the
> >same phys address twice. It will halt vcpu instead.
> 
> What about different pages, running the scheduler code?
> 
We can get couple of nested apfs, just like we can get nested
interrupts. Since scheduler disables preemption second apf will halt.

quoted text
> Oh, and we'll run the scheduler recursively.
> 
As rick said scheduler disables preemption.  And this is actually first
thing it does. Otherwise any interrupt may cause recursive scheduler
invocation.
 
--
			Gleb.
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/