From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753419AbbJ0Aoa (ORCPT ); Mon, 26 Oct 2015 20:44:30 -0400 Received: from e31.co.us.ibm.com ([32.97.110.149]:32963 "EHLO e31.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752013AbbJ0Ao2 (ORCPT ); Mon, 26 Oct 2015 20:44:28 -0400 X-IBM-Helo: d03dlp03.boulder.ibm.com X-IBM-MailFrom: paulmck@linux.vnet.ibm.com X-IBM-RcptTo: linux-kernel@vger.kernel.org;linux-rt-users@vger.kernel.org Date: Mon, 26 Oct 2015 17:44:22 -0700 From: "Paul E. McKenney" To: Josh Cartwright Cc: tglx@linutronix.de, bigeasy@linutronix.de, linux-rt-users@vger.kernel.org, linux-kernel@vger.kernel.org, Eric Dumazet , "David S. Miller" Subject: Re: [PATCH -rt] Revert "net: use synchronize_rcu_expedited()" Message-ID: <20151027004422.GH5105@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <1445886895-3692-1-git-send-email-joshc@ni.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1445886895-3692-1-git-send-email-joshc@ni.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15102700-8236-0000-0000-0000130D95EB Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Oct 26, 2015 at 02:14:55PM -0500, Josh Cartwright wrote: > This reverts commit be3fc413da9eb17cce0991f214ab019d16c88c41. > > While the use of synchronize_rcu_expedited() might make > synchronize_net() "faster", it does so at significant cost on RT > systems, as expediting a grace period forcibly preempts any > high-priority RT tasks (via the stop_machine() mechanism). > > Without be3fc413da9e reverted, we can observe a latency spike up to 30us > with cyclictest by rapidly unplugging/reestablishing an ethernet link. > > Cc: Eric Dumazet > Cc: Paul E. McKenney > Cc: David S. Miller > Signed-off-by: Josh Cartwright Hmmm... If I remember correctly, using expedited here resulted in impressive performance improvements in some important cases. Perhaps things have changed (I must defer to Eric), but if not, how about something like this instead? if (rtnl_is_locked() && !IS_ENABLED(CONFIG_PREMPT_RT_FULL)) synchronize_rcu_expedited(); else synchronize_rcu(); Alternatively, a boot-time option could be used: if (rtnl_is_locked() && !some_rt_boot_parameter) synchronize_rcu_expedited(); else synchronize_rcu(); I believe that the first alternative is better because it does the right thing without user intervention. The second would be preferred should distros want to offer full RT by default, but I am guessing thta most distros would be reluctant to do this for some time to come. Either way, these approaches have the advantage of giving RT users the latency they need, even in the face of networking configuration changes, while giving non-RT users the required performance of the networking configuration changes themselves. Thanx, Paul > --- > net/core/dev.c | 5 +---- > 1 file changed, 1 insertion(+), 4 deletions(-) > > diff --git a/net/core/dev.c b/net/core/dev.c > index f8c23de..869ef62 100644 > --- a/net/core/dev.c > +++ b/net/core/dev.c > @@ -6969,10 +6969,7 @@ EXPORT_SYMBOL(free_netdev); > void synchronize_net(void) > { > might_sleep(); > - if (rtnl_is_locked()) > - synchronize_rcu_expedited(); > - else > - synchronize_rcu(); > + synchronize_rcu(); > } > EXPORT_SYMBOL(synchronize_net); > > -- > 2.5.1 >