Re: [PATCH RFC V4 3/3] kvm: Choose better candidate for directedyield

From: Raghavendra K T
Date: Mon Jul 16 2012 - 13:51:28 EST


On 07/16/2012 03:37 PM, Avi Kivity wrote:
On 07/16/2012 11:25 AM, Raghavendra K T wrote:
From: Raghavendra K T<raghavendra.kt@xxxxxxxxxxxxxxxxxx>

Currently, on a large vcpu guests, there is a high probability of
yielding to the same vcpu who had recently done a pause-loop exit or
cpu relax intercepted. Such a yield can lead to the vcpu spinning
again and hence degrade the performance.

The patchset keeps track of the pause loop exit/cpu relax interception
and gives chance to a vcpu which:
(a) Has not done pause loop exit or cpu relax intercepted at all
(probably he is preempted lock-holder)
(b) Was skipped in last iteration because it did pause loop exit or
cpu relax intercepted, and probably has become eligible now
(next eligible lock holder)



+#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
+/*
+ * Helper that checks whether a VCPU is eligible for directed yield.
+ * Most eligible candidate to yield is decided by following heuristics:
+ *
+ * (a) VCPU which has not done pl-exit or cpu relax intercepted recently
+ * (preempted lock holder), indicated by @cpu_relax_intercepted.
+ * Set at the beiginning and cleared at the end of interception/PLE handler.
+ *
+ * (b) VCPU which has done pl-exit/ cpu relax intercepted but did not get
+ * chance last time (mostly it has become eligible now since we have probably
+ * yielded to lockholder in last iteration. This is done by toggling
+ * @dy_eligible each time a VCPU checked for eligibility.)
+ *
+ * Yielding to a recently pl-exited/cpu relax intercepted VCPU before yielding
+ * to preempted lock-holder could result in wrong VCPU selection and CPU
+ * burning. Giving priority for a potential lock-holder increases lock
+ * progress.
+ */
+bool kvm_vcpu_check_and_update_eligible(struct kvm_vcpu *vcpu)

Predicates' names should give a hint as to what true and false returns
mean. For example vcpu_eligible_for_directed_yield().


I agree regarding the Predicate name. My confusion was it was
doing more than that (flipping eligible flag).
So, I ll go with kvm_vcpu_eligible_for_directed_yield()

+{
[...]
+ return eligible;
+}

You're accessing another vcpu's data structures without any locking.
This is probably okay since we're not basing any life or death decisions
on this, but a comment would be good to explain to readers that this has
been considered and is okay (and why).



True and agree. What we doing here is not worth of locking overhead.
will try to explain more on that.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/