From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964835AbWAFWeL (ORCPT ); Fri, 6 Jan 2006 17:34:11 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932640AbWAFWeL (ORCPT ); Fri, 6 Jan 2006 17:34:11 -0500 Received: from cantor.suse.de ([195.135.220.2]:24556 "EHLO mx1.suse.de") by vger.kernel.org with ESMTP id S932275AbWAFWeC (ORCPT ); Fri, 6 Jan 2006 17:34:02 -0500 From: Andi Kleen To: paulmck@us.ibm.com Subject: Re: [PATCH, RFC] RCU : OOM avoidance and lower latency Date: Fri, 6 Jan 2006 21:57:41 +0100 User-Agent: KMail/1.8.2 Cc: Eric Dumazet , Alan Cox , Linus Torvalds , linux-kernel@vger.kernel.org, "David S. Miller" , Dipankar Sarma , Manfred Spraul , netdev@vger.kernel.org References: <20060105235845.967478000@sorel.sous-sol.org> <43BEA693.5010509@cosmosbay.com> <20060106202626.GA5677@us.ibm.com> In-Reply-To: <20060106202626.GA5677@us.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200601062157.42470.ak@suse.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Friday 06 January 2006 21:26, Paul E. McKenney wrote: > If not, it may be worthwhile to limit the number of times that > rt_run_flush() runs per RCU grace period. Problem is that without rt_run_flush new routes and route attribute changes don't get used by the stack. If RCU takes long and routes keep changing this might be a big issue. As a admin I would be certainly annoyed if the network stack ignored my new route for some unbounded time. Perhaps a better way would be to just exclude dst entries in RCU state from the normal accounting and assume that if the system really runs short of memory because of this the results would trigger quiescent states more quickly, freeing the memory again. -Andi