From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Ricardo Leitner Subject: Re: [PATCH stable v3.2 v3.4] ipv4: disable bh while doing route gc Date: Tue, 21 Oct 2014 17:51:56 -0200 Message-ID: <5446B95C.5040206@redhat.com> References: <6c3d6eca5d6a15c01393b010f2116bd169477c5a.1413215324.git.mleitner@redhat.com> <20141013.135127.1915115817707962111.davem@davemloft.net> <1413774581.31953.12.camel@decadent.org.uk> <5446AF2A.1030508@redhat.com> <1413919971.31953.39.camel@decadent.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Cc: David Miller , stable@vger.kernel.org, netdev@vger.kernel.org, hannes@stressinduktion.org To: Ben Hutchings Return-path: In-Reply-To: <1413919971.31953.39.camel@decadent.org.uk> Sender: stable-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 21-10-2014 17:32, Ben Hutchings wrote: > On Tue, 2014-10-21 at 17:08 -0200, Marcelo Ricardo Leitner wrote: >> On 20-10-2014 01:09, Ben Hutchings wrote: >>> On Mon, 2014-10-13 at 13:51 -0400, David Miller wrote: >>>> From: Marcelo Ricardo Leitner >>>> Date: Mon, 13 Oct 2014 14:03:30 -0300 >>>> >>>>> Further tests revealed that after moving the garbage collector to a work >>>>> queue and protecting it with a spinlock may leave the system prone to >>>>> soft lockups if bottom half gets very busy. >>>>> >>>>> It was reproced with a set of firewall rules that REJECTed packets. If >>>>> the NIC bottom half handler ends up running on the same CPU that is >>>>> running the garbage collector on a very large cache, the garbage >>>>> collector will not be able to do its job due to the amount of work >>>>> needed for handling the REJECTs and also won't reschedule. >>>>> >>>>> The fix is to disable bottom half during the garbage collecting, as it >>>>> already was in the first place (most calls to it came from softirqs). >>>>> >>>>> Signed-off-by: Marcelo Ricardo Leitner >>>>> Acked-by: Hannes Frederic Sowa >>>>> Acked-by: David S. Miller >>>>> Cc: stable@vger.kernel.org >>>> >>>> -stable folks, please integrate this directly, thanks! >>> >>> I've appplied this and the previous two patches mentioned ('ipv4: move >>> route garbage collector to work queue' and 'ipv4: avoid parallel route >>> cache gc executions'). But I didn't get the other two from you. The >>> last batch of networking fixes I received and applied was dated >>> 2014-08-07, and the next one I've seen is dated 2014-10-11 and has >>> nothing for 3.2 or 3.4. Did I miss one between these? >> >> Sorry to ask Ben but, where did you apply them? I'm not seeing the commits on >> linux-stable.git and couldn't find their summaries anywhere else. > > They're in a patch queue that I've just pushed to: > > git://git.kernel.org/pub/scm/linux/kernel/git/bwh/linux-3.2.y-queue.git Cool, thanks! Marcelo