From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753211AbcJKQPe (ORCPT ); Tue, 11 Oct 2016 12:15:34 -0400 Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:57436 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752037AbcJKQPd (ORCPT ); Tue, 11 Oct 2016 12:15:33 -0400 Date: Tue, 11 Oct 2016 09:15:40 -0700 From: "Paul E. McKenney" To: riel@redhat.com Cc: linux-kernel@vger.kernel.org Subject: Re: Untested patch to recheck idle state for expedited grace periods Reply-To: paulmck@linux.vnet.ibm.com References: <20161011132849.GA21962@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20161011132849.GA21962@linux.vnet.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-TM-AS-GCONF: 00 X-Content-Scanned: Fidelis XPS MAILER x-cbid: 16101116-0008-0000-0000-000005C93F88 X-IBM-SpamModules-Scores: X-IBM-SpamModules-Versions: BY=3.00005893; HX=3.00000240; KW=3.00000007; PH=3.00000004; SC=3.00000186; SDB=6.00766967; UDB=6.00366912; IPR=6.00543109; BA=6.00004801; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009; ZB=6.00000000; ZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00012949; XFM=3.00000011; UTC=2016-10-11 16:15:31 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 16101116-0009-0000-0000-00003C0C345B Message-Id: <20161011161540.GA32060@linux.vnet.ibm.com> X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2016-10-11_09:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 spamscore=0 suspectscore=1 malwarescore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1609300000 definitions=main-1610110279 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Oct 11, 2016 at 06:28:49AM -0700, Paul E. McKenney wrote: > Hello, Rik, > > And it turns out that I did not in fact do the recheck at IPI time. > The (untested) patch below is an alleged fix. Thoughts? And it passes modest rcutorture testing, for whatever that might be worth. Thanx, Paul > ------------------------------------------------------------------------ > > commit e53e0b3e7b3c783962f9461bcb9aa8bc3e3a8688 > Author: Paul E. McKenney > Date: Tue Oct 11 06:09:59 2016 -0700 > > rcu: Make expedited grace periods recheck dyntick idle state > > Expedited grace periods check dyntick-idle state, and avoid sending > IPIs to idle CPUs, including those running guest OSes, and, on NOHZ_FULL > kernels, nohz_full CPUs. However, the kernel has been observed checking > a CPU while it was non-idle, but sending the IPI after it has gone > idle. This commit therefore rechecks idle state immediately before > sending the IPI, refraining from IPIing CPUs that have since gone idle. > > Reported-by: Rik van Riel > Signed-off-by: Paul E. McKenney > > diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h > index e99a5234d9ed..fe98dd24adf8 100644 > --- a/kernel/rcu/tree.h > +++ b/kernel/rcu/tree.h > @@ -404,6 +404,7 @@ struct rcu_data { > atomic_long_t exp_workdone1; /* # done by others #1. */ > atomic_long_t exp_workdone2; /* # done by others #2. */ > atomic_long_t exp_workdone3; /* # done by others #3. */ > + int exp_dynticks_snap; /* Double-check need for IPI. */ > > /* 7) Callback offloading. */ > #ifdef CONFIG_RCU_NOCB_CPU > diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h > index 24343eb87b58..d3053e99fdb6 100644 > --- a/kernel/rcu/tree_exp.h > +++ b/kernel/rcu/tree_exp.h > @@ -358,8 +358,10 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp, > struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); > struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); > > + rdp->exp_dynticks_snap = > + atomic_add_return(0, &rdtp->dynticks); > if (raw_smp_processor_id() == cpu || > - !(atomic_add_return(0, &rdtp->dynticks) & 0x1) || > + !(rdp->exp_dynticks_snap & 0x1) || > !(rnp->qsmaskinitnext & rdp->grpmask)) > mask_ofl_test |= rdp->grpmask; > } > @@ -377,9 +379,17 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp, > /* IPI the remaining CPUs for expedited quiescent state. */ > for_each_leaf_node_possible_cpu(rnp, cpu) { > unsigned long mask = leaf_node_cpu_bit(rnp, cpu); > + struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); > + struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); > + > if (!(mask_ofl_ipi & mask)) > continue; > retry_ipi: > + if (atomic_add_return(0, &rdtp->dynticks) != > + rdp->exp_dynticks_snap) { > + mask_ofl_test |= mask; > + continue; > + } > ret = smp_call_function_single(cpu, func, rsp, 0); > if (!ret) { > mask_ofl_ipi &= ~mask;