From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932661Ab2IRDD2 (ORCPT ); Mon, 17 Sep 2012 23:03:28 -0400 Received: from e9.ny.us.ibm.com ([32.97.182.139]:58320 "EHLO e9.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932461Ab2IRDDZ (ORCPT ); Mon, 17 Sep 2012 23:03:25 -0400 Subject: Re: [RFC][PATCH] Improving directed yield scalability for PLE handler From: Andrew Theurer Reply-To: habanero@linux.vnet.ibm.com To: Avi Kivity Cc: Raghavendra K T , Peter Zijlstra , Srikar Dronamraju , Marcelo Tosatti , Ingo Molnar , Rik van Riel , KVM , chegu vinod , LKML , X86 , Gleb Natapov , Srivatsa Vaddagiri In-Reply-To: <50559400.8030203@redhat.com> References: <504A37B0.7020605@linux.vnet.ibm.com> <1347046931.7332.51.camel@oc2024037011.ibm.com> <20120908084345.GU30238@linux.vnet.ibm.com> <1347283005.10325.55.camel@oc6622382223.ibm.com> <1347293035.2124.22.camel@twins> <20120910165653.GA28033@linux.vnet.ibm.com> <1347297124.2124.42.camel@twins> <1347307972.7332.78.camel@oc2024037011.ibm.com> <504ED54E.6040608@linux.vnet.ibm.com> <1347388061.19098.20.camel@oc2024037011.ibm.com> <20120913114813.GA11797@linux.vnet.ibm.com> <1347571858.5586.44.camel@oc2024037011.ibm.com> <50559400.8030203@redhat.com> Content-Type: text/plain; charset="UTF-8" Organization: IBM Date: Mon, 17 Sep 2012 22:03:18 -0500 Message-ID: <1347937398.10325.190.camel@oc6622382223.ibm.com> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 (2.28.3-24.el6) Content-Transfer-Encoding: 7bit x-cbid: 12091803-7182-0000-0000-000002A5BD8D Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2012-09-16 at 11:55 +0300, Avi Kivity wrote: > On 09/14/2012 12:30 AM, Andrew Theurer wrote: > > > The concern I have is that even though we have gone through changes to > > help reduce the candidate vcpus we yield to, we still have a very poor > > idea of which vcpu really needs to run. The result is high cpu usage in > > the get_pid_task and still some contention in the double runqueue lock. > > To make this scalable, we either need to significantly reduce the > > occurrence of the lock-holder preemption, or do a much better job of > > knowing which vcpu needs to run (and not unnecessarily yielding to vcpus > > which do not need to run). > > > > On reducing the occurrence: The worst case for lock-holder preemption > > is having vcpus of same VM on the same runqueue. This guarantees the > > situation of 1 vcpu running while another [of the same VM] is not. To > > prove the point, I ran the same test, but with vcpus restricted to a > > range of host cpus, such that any single VM's vcpus can never be on the > > same runqueue. In this case, all 10 VMs' vcpu-0's are on host cpus 0-4, > > vcpu-1's are on host cpus 5-9, and so on. Here is the result: > > > > kvm_cpu_spin, and all > > yield_to changes, plus > > restricted vcpu placement: 8823 +/- 3.20% much, much better > > > > On picking a better vcpu to yield to: I really hesitate to rely on > > paravirt hint [telling us which vcpu is holding a lock], but I am not > > sure how else to reduce the candidate vcpus to yield to. I suspect we > > are yielding to way more vcpus than are prempted lock-holders, and that > > IMO is just work accomplishing nothing. Trying to think of way to > > further reduce candidate vcpus.... > > I wouldn't say that yielding to the "wrong" vcpu accomplishes nothing. > That other vcpu gets work done (unless it is in pause loop itself) and > the yielding vcpu gets put to sleep for a while, so it doesn't spend > cycles spinning. While we haven't fixed the problem at least the guest > is accomplishing work, and meanwhile the real lock holder may get > naturally scheduled and clear the lock. OK, yes, if the other thread gets useful work done, then it is not wasteful. I was thinking of the worst case scenario, where any other vcpu would likely spin as well, and the host side cpu-time for switching vcpu threads was not all that productive. Well, I suppose it does help eliminate potential lock holding vcpus; it just seems to be not that efficient or fast enough. > The main problem with this theory is that the experiments don't seem to > bear it out. Granted, my test case is quite brutal. It's nothing but over-committed VMs which always have some spin lock activity. However, we really should try to fix the worst case scenario. > So maybe one of the assumptions is wrong - the yielding > vcpu gets scheduled early. That could be the case if the two vcpus are > on different runqueues - you could be changing the relative priority of > vcpus on the target runqueue, but still remain on top yourself. Is this > possible with the current code? > > Maybe we should prefer vcpus on the same runqueue as yield_to targets, > and only fall back to remote vcpus when we see it didn't help. > > Let's examine a few cases: > > 1. spinner on cpu 0, lock holder on cpu 0 > > win! > > 2. spinner on cpu 0, random vcpu(s) (or normal processes) on cpu 0 > > Spinner gets put to sleep, random vcpus get to work, low lock contention > (no double_rq_lock), by the time spinner gets scheduled we might have won > > 3. spinner on cpu 0, another spinner on cpu 0 > > Worst case, we'll just spin some more. Need to detect this case and > migrate something in. Well, we can certainly experiment and see what we get. IMO, the key to getting this working really well on the large VMs is finding the lock-holding cpu -quickly-. What I think is happening is that we go through a relatively long process to get to that one right vcpu. I guess I need to find a faster way to get there. > 4. spinner on cpu 0, alone > > Similar > > > It seems we need to tie in to the load balancer. > > Would changing the priority of the task while it is spinning help the > load balancer? Not sure. -Andrew