From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755048Ab2DPPua (ORCPT ); Mon, 16 Apr 2012 11:50:30 -0400 Received: from acsinet15.oracle.com ([141.146.126.227]:46797 "EHLO acsinet15.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753882Ab2DPPu2 (ORCPT ); Mon, 16 Apr 2012 11:50:28 -0400 Date: Mon, 16 Apr 2012 11:44:29 -0400 From: Konrad Rzeszutek Wilk To: Srivatsa Vaddagiri , Jeremy Fitzhardinge Cc: Thomas Gleixner , "H. Peter Anvin" , Raghavendra K T , Ingo Molnar , Linus Torvalds , Peter Zijlstra , the arch/x86 maintainers , LKML , Avi Kivity , Marcelo Tosatti , KVM , Andi Kleen , Xen Devel , Virtualization , Jeremy Fitzhardinge , Stephan Diestelhorst , Stefano Stabellini , Attilio Rao Subject: Re: [PATCH RFC V6 0/11] Paravirtualized ticketlocks Message-ID: <20120416154429.GB4654@phenom.dumpdata.com> References: <20120321102041.473.61069.sendpatchset@codeblue.in.ibm.com> <4F7616F5.4070000@zytor.com> <20120331040745.GC14030@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120331040745.GC14030@linux.vnet.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-Source-IP: ucsinet21.oracle.com [156.151.31.93] X-CT-RefId: str=0001.0A090208.4F8C3F95.00E5,ss=1,re=0.000,fgs=0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Mar 31, 2012 at 09:37:45AM +0530, Srivatsa Vaddagiri wrote: > * Thomas Gleixner [2012-03-31 00:07:58]: > > > I know that Peter is going to go berserk on me, but if we are running > > a paravirt guest then it's simple to provide a mechanism which allows > > the host (aka hypervisor) to check that in the guest just by looking > > at some global state. > > > > So if a guest exits due to an external event it's easy to inspect the > > state of that guest and avoid to schedule away when it was interrupted > > in a spinlock held section. That guest/host shared state needs to be > > modified to indicate the guest to invoke an exit when the last nested > > lock has been released. > > I had attempted something like that long back: > > http://lkml.org/lkml/2010/6/3/4 > > The issue is with ticketlocks though. VCPUs could go into a spin w/o > a lock being held by anybody. Say VCPUs 1-99 try to grab a lock in > that order (on a host with one cpu). VCPU1 wins (after VCPU0 releases it) > and releases the lock. VCPU1 is next eligible to take the lock. If > that is not scheduled early enough by host, then remaining vcpus would keep > spinning (even though lock is technically not held by anybody) w/o making > forward progress. > > In that situation, what we really need is for the guest to hint to host > scheduler to schedule VCPU1 early (via yield_to or something similar). > > The current pv-spinlock patches however does not track which vcpu is > spinning at what head of the ticketlock. I suppose we can consider > that optimization in future and see how much benefit it provides (over > plain yield/sleep the way its done now). Right. I think Jeremy played around with this some time? > > Do you see any issues if we take in what we have today and address the > finer-grained optimization as next step? I think that is the proper course - these patches show that on baremetal we don't incur performance regressions and in virtualization case we benefit greatly. Since these are the basic building blocks of a kernel - taking it slow and just adding this set of patches for v3.5 is a good idea - and then building on top of that for further refinement. > > - vatsa