From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752135AbbGOBYT (ORCPT ); Tue, 14 Jul 2015 21:24:19 -0400 Received: from g2t2352.austin.hp.com ([15.217.128.51]:33265 "EHLO g2t2352.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751800AbbGOBYR (ORCPT ); Tue, 14 Jul 2015 21:24:17 -0400 Message-ID: <55A5B63D.1090303@hp.com> Date: Tue, 14 Jul 2015 21:24:13 -0400 From: Waiman Long User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.12) Gecko/20130109 Thunderbird/10.0.12 MIME-Version: 1.0 To: Peter Zijlstra CC: Ingo Molnar , Thomas Gleixner , "H. Peter Anvin" , x86@kernel.org, linux-kernel@vger.kernel.org, Scott J Norton , Douglas Hatch Subject: Re: [PATCH 1/7] locking/pvqspinlock: Only kick CPU at unlock time References: <1436647018-49734-1-git-send-email-Waiman.Long@hp.com> <1436647018-49734-2-git-send-email-Waiman.Long@hp.com> <20150713120202.GG3644@twins.programming.kicks-ass.net> In-Reply-To: <20150713120202.GG3644@twins.programming.kicks-ass.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 07/13/2015 08:02 AM, Peter Zijlstra wrote: > On Sat, Jul 11, 2015 at 04:36:52PM -0400, Waiman Long wrote: >> @@ -181,9 +187,9 @@ static void pv_wait_node(struct mcs_spinlock *node) >> pv_wait(&pn->state, vcpu_halted); >> >> /* >> - * Reset the vCPU state to avoid unncessary CPU kicking >> + * Reset the state except when vcpu_hashed is set. >> */ >> - WRITE_ONCE(pn->state, vcpu_running); >> + cmpxchg(&pn->state, vcpu_halted, vcpu_running); > Why? Suppose we did get advanced into the hashed state, and then get a > (spurious) wakeup, this means we'll observe our ->locked == 1 condition > and fall out of pv_wait_node(). > > We'll then enter pv_wait_head(), which with your modification: > >> @@ -229,19 +244,42 @@ static void pv_wait_head(struct qspinlock *lock, struct mcs_spinlock *node) >> { >> struct pv_node *pn = (struct pv_node *)node; >> struct __qspinlock *l = (void *)lock; >> - struct qspinlock **lp = NULL; >> + struct qspinlock **lp; >> int loop; >> >> + /* >> + * Initialize lp to a non-NULL value if it has already been in the >> + * pv_hashed state so that pv_hash() won't be called again. >> + */ >> + lp = (READ_ONCE(pn->state) == vcpu_hashed) ? (struct qspinlock **)1 >> + : NULL; >> for (;;) { >> + WRITE_ONCE(pn->state, vcpu_running); > Will instantly and unconditionally write vcpu_running. > > This code is kind of complicated. I am going to get rid of the current tri-state setup, and switch to a separate sync variable for defer kicking. Cheers, Longman