From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754572AbYJCPWs (ORCPT ); Fri, 3 Oct 2008 11:22:48 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754771AbYJCPWi (ORCPT ); Fri, 3 Oct 2008 11:22:38 -0400 Received: from mga05.intel.com ([192.55.52.89]:56061 "EHLO fmsmga101.fm.intel.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1754025AbYJCPWi (ORCPT ); Fri, 3 Oct 2008 11:22:38 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.33,357,1220252400"; d="scan'208";a="623036574" Message-ID: <48E638B5.9040204@linux.intel.com> Date: Fri, 03 Oct 2008 08:22:29 -0700 From: Arjan van de Ven User-Agent: Thunderbird 2.0.0.17 (Windows/20080914) MIME-Version: 1.0 To: Jan Kasprzak CC: linux-kernel@vger.kernel.org Subject: Re: IRQ balancing on a router References: <20081003132117.GM16624@fi.muni.cz> <20081003063857.76b7b61a@linux.intel.com> <20081003145737.GQ16624@fi.muni.cz> In-Reply-To: <20081003145737.GQ16624@fi.muni.cz> 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 Jan Kasprzak wrote: > Arjan van de Ven wrote: > : Jan Kasprzak wrote: > : > The result is > : > that the CPU which receives IRQs for the uplink interface > : > is 100 % busy (softirq mostly), while the other one is 90% idle. > : > : one of the hard cases for irqbalance is that irqbalance doesn't have a > : way to find out the actual cpu time spend in the handlers. For > : networking it makes an estimate just based on the number of packets > : (which is better than nothing)... but that breaks down if you have an > : non-symmetry in CPU costs per packet like you have. > : > : The good news is that irqthreads at least have the potential to solve > : this "lack of information"; if not, we could consider doing a form of > : microaccounting for irq handlers.... > > I am not sure whether this would help. In my case, the most of the > in-kernel CPU time is not spend in the irq handler per se, but in softirq > (i.e. checking the packet against iptables rules). there is some consideration of making softirqs that are raised run as part of the irq thread. or at least thoughts in that direction.