From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [PATCH] use unfair spinlock when running on hypervisor. Date: Tue, 1 Jun 2010 21:36:59 +0200 Message-ID: <20100601193659.GC11880@basil.fritz.box> References: <20100601093515.GH24302@redhat.com> <87sk56ycka.fsf@basil.nowhere.org> <20100601162414.GA6191@redhat.com> <20100601163807.GA11880@basil.fritz.box> <4C053ACC.5020708@redhat.com> <4C05493F.1040107@third-harmonic.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Avi Kivity , Andi Kleen , Gleb Natapov , linux-kernel@vger.kernel.org, kvm@vger.kernel.org, hpa@zytor.com, mingo@elte.hu, npiggin@suse.de, tglx@linutronix.de, mtosatti@redhat.com, john cooper To: john cooper Return-path: Content-Disposition: inline In-Reply-To: <4C05493F.1040107@third-harmonic.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org > Collecting the contention/usage statistics on a per spinlock > basis seems complex. I believe a practical approximation > to this are adaptive mutexes where upon hitting a spin > time threshold, punt and let the scheduler reconcile fairness. That would probably work, except: how do you get the adaptive spinlock into a paravirt op without slowing down a standard kernel? -Andi -- ak@linux.intel.com -- Speaking for myself only.