From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933683Ab3E1Vcj (ORCPT ); Tue, 28 May 2013 17:32:39 -0400 Received: from mx1.redhat.com ([209.132.183.28]:21270 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757394Ab3E1Vcf (ORCPT ); Tue, 28 May 2013 17:32:35 -0400 Message-ID: <51A52257.4060905@redhat.com> Date: Tue, 28 May 2013 17:32:07 -0400 From: Rik van Riel User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130514 Thunderbird/17.0.6 MIME-Version: 1.0 To: Eric Dumazet CC: Rafael Aquini , Ben Greear , Francois Romieu , atomlin@redhat.com, netdev@vger.kernel.org, davem@davemloft.net, edumazet@google.com, pshelar@nicira.com, mst@redhat.com, alexander.h.duyck@intel.com, sergei.shtylyov@cogentembedded.com, linux-kernel@vger.kernel.org Subject: Re: [Patch v2] skbuff: Hide GFP_ATOMIC page allocation failures for dropped packets References: <1369601101-23057-1-git-send-email-atomlin@redhat.com> <20130527224149.GA4384@electric-eye.fr.zoreil.com> <51A4D4AD.2010507@candelatech.com> <20130528161518.GC11614@optiplex.redhat.com> <1369758577.3301.543.camel@edumazet-glaptop> In-Reply-To: <1369758577.3301.543.camel@edumazet-glaptop> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 05/28/2013 12:29 PM, Eric Dumazet wrote: > On Tue, 2013-05-28 at 13:15 -0300, Rafael Aquini wrote: > >> The real problem seems to be that more and more the network stack (drivers, perhaps) >> is relying on chunks of contiguous page-blocks without a fallback mechanism to >> order-0 page allocations. When memory gets fragmented, these alloc failures >> start to pop up more often and they scare ordinary sysadmins out of their paints. >> > > Where do you see that ? > > I see exactly the opposite trend. > > We have less and less buggy drivers, and we want to catch last > offenders. These backtraces would still get printed out for drivers that DO do the right thing and fall back to smaller allocations. The initial failed large allocation would cause a backtrace.