Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751555AbdIEIwP (ORCPT ); Tue, 5 Sep 2017 04:52:15 -0400 Received: from mx2.suse.de ([195.135.220.15]:45539 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751507AbdIEIwL (ORCPT ); Tue, 5 Sep 2017 04:52:11 -0400 Subject: Re: [PATCH resend] x86,kvm: Add a kernel parameter to disable PV spinlock To: Peter Zijlstra Cc: Davidlohr Bueso , Oscar Salvador , Ingo Molnar , Paolo Bonzini , "H . Peter Anvin" , Thomas Gleixner , linux-kernel@vger.kernel.org, kvm@vger.kernel.org, linux-doc@vger.kernel.org, Waiman Long References: <20170904142836.15446-1-osalvador@suse.de> <20170904144011.gp7hpis6usjehbuf@hirez.programming.kicks-ass.net> <20170904222157.GD17982@linux-80c1.suse> <0869e8a5-4abd-8f7f-0135-aab3e72e2d01@suse.com> <20170905065837.rs767a4os2aumg7h@hirez.programming.kicks-ass.net> <924fec17-548a-083d-edce-7adcb662c513@suse.com> <20170905081001.hn2276qrhfyqpjdi@hirez.programming.kicks-ass.net> <83ac209b-0807-0a72-cd07-d4ccd1d1ed61@suse.com> <20170905082801.d3sr7hdjz6lpon5p@hirez.programming.kicks-ass.net> From: Juergen Gross Message-ID: <1b3afce5-c95d-5a3a-c36c-953a0654ab2d@suse.com> Date: Tue, 5 Sep 2017 10:52:06 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.3.0 MIME-Version: 1.0 In-Reply-To: <20170905082801.d3sr7hdjz6lpon5p@hirez.programming.kicks-ass.net> Content-Type: text/plain; charset=utf-8 Content-Language: de-DE Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1435 Lines: 37 On 05/09/17 10:28, Peter Zijlstra wrote: > On Tue, Sep 05, 2017 at 10:14:21AM +0200, Juergen Gross wrote: >> On 05/09/17 10:10, Peter Zijlstra wrote: >>> On Tue, Sep 05, 2017 at 09:35:40AM +0200, Juergen Gross wrote: >>>>> So the problem with qspinlock is that it will revert to a classic >>>>> test-and-set spinlock if you don't do paravirt but are running a HV. >>>> >>>> In the Xen case we just use the bare metal settings when xen_nopvspin >>>> has been specified. So paravirt, but without modifying any pv_lock_ops >>>> functions. >>> >>> See arch/x86/include/asm/qspinlock.h:virt_spin_lock(). Unless you clear >>> X86_FEATURE_HYPERVISOR you get a test-and-set spinlock. >>> >>> And as the comment there says, this is a fallback for !paravirt enabled >>> hypervisors to avoid the worst of the lock holder preemption crud. >>> >>> But this very much does not deal with the 1:1 case nicely. >>> >> >> Aah, now I've got it. >> >> So maybe we should add virt_spin_lock() to pv_lock_ops? This way e.g. >> xen_nopvspin could tweak just the virt_spin_lock() case by letting it >> return false all the time? > > Hmm, that might work. Could we somehow nop that call when > !X86_FEATURE_HYPERVISOR?, that saves native from having to do the call > and would be a win for everyone. So in fact we want a "always false" shortcut for bare metal and for any virtualization environment selecting bare metal behavior. I'll have a try. Juergen