From mboxrd@z Thu Jan 1 00:00:00 1970 From: Neil Horman Subject: Re: Multicast packet loss Date: Sat, 31 Jan 2009 11:03:33 -0500 Message-ID: <20090131160333.GC23100@localhost.localdomain> References: <49833DBC.7040607@athenacr.com> <20090130200330.GA12659@hmsreliant.think-freely.org> <49837F56.2020502@athenacr.com> <49838213.90700@cosmosbay.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Kenny Chang , netdev@vger.kernel.org To: Eric Dumazet Return-path: Received: from charlotte.tuxdriver.com ([70.61.120.58]:41432 "EHLO smtp.tuxdriver.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752187AbZAaQDh (ORCPT ); Sat, 31 Jan 2009 11:03:37 -0500 Content-Disposition: inline In-Reply-To: <49838213.90700@cosmosbay.com> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, Jan 30, 2009 at 11:41:23PM +0100, Eric Dumazet wrote: > Kenny Chang a =E9crit : > > Ah, sorry, here's the test program attached. > >=20 > > We've tried 2.6.28.1, but no, we haven't tried the 2.6.28.2 or the > > 2.6.29.-rcX. > >=20 > > Right now, we are trying to step through the kernel versions until = we > > see where the performance drops significantly. We'll try 2.6.29-rc= soon > > and post the result. >=20 > 2.6.29-rc contains UDP receive improvements (lockless) >=20 > Problem is multicast handling was not yet updated, but could be :) >=20 >=20 > I was asking you "cat /proc/interrupts" because I believe you might > have a problem NIC interrupts being handled by one CPU only (when hav= ing problems) >=20 That would be expected (if irqbalance is running), and desireable, sinc= e spreading high volume interrupts like NICS accross multiple cores (or m= ore specifically multiple L2 caches), is going increase your cache line mis= s rate significantly and decrease rx throughput. Although you do have a point here, if the system isn't running irqbalan= ce, and the NICS irq affinity is spread accross multiple L2 caches, that would = be a point of improvement performance-wise. =20 Kenny, if you could provide the /proc/interrupts info along with /proc/= cpuinfo and your stats that I asked about earlier, that would be a big help. Regards Neil