From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH 00/16] Remove the ipv4 routing cache Date: Thu, 26 Jul 2012 19:30:58 +0200 Message-ID: <1343323858.2626.11783.camel@edumazet-glaptop> References: <1343290414.2626.11181.camel@edumazet-glaptop> <20120726.011853.973075299731735416.davem@davemloft.net> <1343291278.2626.11188.camel@edumazet-glaptop> <20120726.014736.1066206957795563053.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: David Miller , netdev@vger.kernel.org To: Alexander Duyck Return-path: Received: from mail-bk0-f46.google.com ([209.85.214.46]:37393 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751670Ab2GZRbE (ORCPT ); Thu, 26 Jul 2012 13:31:04 -0400 Received: by bkwj10 with SMTP id j10so1424165bkw.19 for ; Thu, 26 Jul 2012 10:31:03 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 2012-07-26 at 10:18 -0700, Alexander Duyck wrote: > I tested this patch and it looks like it runs, but still has the same > performance issue. I did some digging into the annotation for > ip_route_intput_noref and it seems like the issue is that I am hitting > the dst_hold call in __mkroute_input. David suggested a percpu cache. nh_rth_input would be allocated by alloc_percpu(struct dst *) I can work on this.