From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.3 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS,URG_BIZ,USER_AGENT_MUTT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 86ADDC5CFEB for ; Mon, 9 Jul 2018 20:32:33 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 322F9208DB for ; Mon, 9 Jul 2018 20:32:33 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 322F9208DB Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.vnet.ibm.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933135AbeGIUca (ORCPT ); Mon, 9 Jul 2018 16:32:30 -0400 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:54928 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932856AbeGIUc2 (ORCPT ); Mon, 9 Jul 2018 16:32:28 -0400 Received: from pps.filterd (m0098399.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.22/8.16.0.22) with SMTP id w69KSmGJ125204 for ; Mon, 9 Jul 2018 16:32:28 -0400 Received: from e14.ny.us.ibm.com (e14.ny.us.ibm.com [129.33.205.204]) by mx0a-001b2d01.pphosted.com with ESMTP id 2k4dp9c9d1-1 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=NOT) for ; Mon, 09 Jul 2018 16:32:28 -0400 Received: from localhost by e14.ny.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 9 Jul 2018 16:32:27 -0400 Received: from b01cxnp23034.gho.pok.ibm.com (9.57.198.29) by e14.ny.us.ibm.com (146.89.104.201) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; (version=TLSv1/SSLv3 cipher=AES256-GCM-SHA384 bits=256/256) Mon, 9 Jul 2018 16:32:24 -0400 Received: from b01ledav003.gho.pok.ibm.com (b01ledav003.gho.pok.ibm.com [9.57.199.108]) by b01cxnp23034.gho.pok.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id w69KWN009699830 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL); Mon, 9 Jul 2018 20:32:23 GMT Received: from b01ledav003.gho.pok.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 5E0C3B2066; Mon, 9 Jul 2018 16:31:58 -0400 (EDT) Received: from b01ledav003.gho.pok.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 2E62DB205F; Mon, 9 Jul 2018 16:31:58 -0400 (EDT) Received: from paulmck-ThinkPad-W541 (unknown [9.70.82.159]) by b01ledav003.gho.pok.ibm.com (Postfix) with ESMTP; Mon, 9 Jul 2018 16:31:58 -0400 (EDT) Received: by paulmck-ThinkPad-W541 (Postfix, from userid 1000) id F3B0616C1972; Mon, 9 Jul 2018 13:34:41 -0700 (PDT) Date: Mon, 9 Jul 2018 13:34:41 -0700 From: "Paul E. McKenney" To: David Woodhouse Cc: Peter Zijlstra , mhillenb@amazon.de, linux-kernel Subject: Re: [RFC] Make need_resched() return true when rcu_urgent_qs requested Reply-To: paulmck@linux.vnet.ibm.com References: <20180709104429.GI2476@hirez.programming.kicks-ass.net> <1531133801.18697.73.camel@infradead.org> <20180709110657.GL2476@hirez.programming.kicks-ass.net> <20180709123457.GM3593@linux.vnet.ibm.com> <20180709125516.GO2476@hirez.programming.kicks-ass.net> <20180709130227.GQ2512@hirez.programming.kicks-ass.net> <20180709142932.GO3593@linux.vnet.ibm.com> <20180709152632.GX2476@hirez.programming.kicks-ass.net> <20180709163432.GV3593@linux.vnet.ibm.com> <1531162254.26547.3.camel@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1531162254.26547.3.camel@infradead.org> User-Agent: Mutt/1.5.21 (2010-09-15) X-TM-AS-GCONF: 00 x-cbid: 18070920-0052-0000-0000-0000030B6535 X-IBM-SpamModules-Scores: X-IBM-SpamModules-Versions: BY=3.00009340; HX=3.00000241; KW=3.00000007; PH=3.00000004; SC=3.00000266; SDB=6.01058976; UDB=6.00543453; IPR=6.00836896; MB=3.00022076; MTD=3.00000008; XFM=3.00000015; UTC=2018-07-09 20:32:25 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 18070920-0053-0000-0000-00005D4D90D7 Message-Id: <20180709203441.GE3593@linux.vnet.ibm.com> X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2018-07-09_07:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 malwarescore=0 suspectscore=0 phishscore=0 bulkscore=0 spamscore=0 clxscore=1015 lowpriorityscore=0 mlxscore=0 impostorscore=0 mlxlogscore=999 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1806210000 definitions=main-1807090232 Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jul 09, 2018 at 07:50:54PM +0100, David Woodhouse wrote: > > > On Mon, 2018-07-09 at 09:34 -0700, Paul E. McKenney wrote: > > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c > > index 51919985f6cf..33b0a1ec0536 100644 > > --- a/kernel/rcu/tree.c > > +++ b/kernel/rcu/tree.c > > @@ -2496,6 +2496,10 @@ void rcu_check_callbacks(int user) > >  { > >   trace_rcu_utilization(TPS("Start scheduler-tick")); > >   raw_cpu_inc(rcu_data.ticks_this_gp); > > + if (smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs)) && > > +    !is_idle_task(current)) > > + set_tsk_need_resched(current); > > OK, so this will make KVM (and various other code) see that > need_resched() is true, and they'll call cond_resched() or something > else that might not actually schedule another task, but will at least > end up in rcu_all_qs()... > > > + __this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); > > ... which bails out immediately and does nothing, because that's set to > false? > > Am I missing something? If this is the idle task, RCU will detect that as a quiescent state via its dyntick-idle mechanism. In which case, there is no point in leaving .rcu_urgent_qs being true. If this is not the idle task, the scheduler will invoke rcu_note_context_switch(), which will in turn invoke rcu_sched_qs(), rcu_preempt_qs(), or rcu_qs(), depending on kernel version and configuration. This will happen independently of .rcu_urgent_qs, so it is OK to set .rcu_urgent_qs to false. And doing so reduces the overhead of the next cond_resched(). I may end up using rcu_is_cpu_rrupt_from_idle() instead of is_idle_task() at some point, but the former eases backporting. And the only difference is if someone has a long loop within an _rcuidle tracepoint used in the idle loop, and where that loop check need_resched(). Which currently seems to be the empty set. And I should treat interruption of a usermode task the same as that of an idle task. In the PREEMPT case, ->rcu_read_lock_nesting better be zero (lockdep would have complained), so the quiescent state will be reported. In the !PREEMPT case, user=1 directly causes reporting of a quiescent state. So here are the possible code paths when .rcu_urgent_qs is set to true: 1. A context switch will record the quiescent state and clear .rcu_urgent_qs. (The failure to do the clearing in current -rcu for PREEMPT builds is a performance bug that I need to fix.) 2. A cond_resched() will cause rcu_all_qs() to be invoked, which will record a quiescent state and clear .rcu_urgent_qs. 3. With the patch below, a scheduling-clock interrupt of a non-idle non-userspace task will force a reschedule, which will result in #1 above happening. However, I should avoid setting .rcu_urgent_qs to false when it is already false, shouldn't I? So how about the following instead? I am doing some light testing and will let you know how that goes. Thanx, Paul ------------------------------------------------------------------------ diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 51919985f6cf..c3b688c7127a 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2496,6 +2496,15 @@ void rcu_check_callbacks(int user) { trace_rcu_utilization(TPS("Start scheduler-tick")); raw_cpu_inc(rcu_data.ticks_this_gp); + /* The load-acquire pairs with the store-release setting to true. */ + if (smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) { + /* Idle already is a quiescent state. */ + if (!is_idle_task(current) && !user) { + set_tsk_need_resched(current); + set_preempt_need_resched(); + } + __this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); + } rcu_flavor_check_callbacks(user); if (rcu_pending()) invoke_rcu_core();