From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Ricardo Leitner Subject: Re: [PATCH stable 3.4 v2 1/2] ipv4: move route garbage collector to work queue Date: Tue, 19 Aug 2014 17:31:50 -0300 Message-ID: <53F3B436.8050407@redhat.com> References: <547688b69ee9a258cc04e4dad17a90ca6ab182d0.1408045325.git.mleitner@redhat.com> Reply-To: mleitner@redhat.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: hannes@redhat.com, netdev@vger.kernel.org, eric.dumazet@gmail.com To: davem@davemloft.net Return-path: Received: from mx1.redhat.com ([209.132.183.28]:23839 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750808AbaHSUb5 (ORCPT ); Tue, 19 Aug 2014 16:31:57 -0400 In-Reply-To: <547688b69ee9a258cc04e4dad17a90ca6ab182d0.1408045325.git.mleitner@redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: Em 14-08-2014 16:44, Marcelo Ricardo Leitner escreveu: > Currently the route garbage collector gets called by dst_alloc() if it > have more entries than the threshold. But it's an expensive call, that > don't really need to be done by then. > > Another issue with current way is that it allows running the garbage > collector with the same start parameters on multiple CPUs at once, which > is not optimal. A system may even soft lockup if the cache is big enough > as the garbage collectors will be fighting over the hash lock entries. > > This patch thus moves the garbage collector to run asynchronously on a > work queue, much similar to how rt_expire_check runs. > > There is one condition left that allows multiple executions, which is > handled by the next patch. > > Signed-off-by: Marcelo Ricardo Leitner > Acked-by: Hannes Frederic Sowa > --- > > Notes: > Hi, > > This set is needed for stables <= 3.4, as the IPv4 route cache was > removed after that. > > v1->v2: addressed indentation issue noticed by David Miller > > Thanks! > Hi Dave, Sorry to bother you but WDYT, is this patchset more appropriate? Thanks, Marcelo