From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261777AbTJHVbp (ORCPT ); Wed, 8 Oct 2003 17:31:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261788AbTJHVbp (ORCPT ); Wed, 8 Oct 2003 17:31:45 -0400 Received: from rth.ninka.net ([216.101.162.244]:6049 "EHLO rth.ninka.net") by vger.kernel.org with ESMTP id S261777AbTJHVbo (ORCPT ); Wed, 8 Oct 2003 17:31:44 -0400 Date: Wed, 8 Oct 2003 14:31:21 -0700 From: "David S. Miller" To: Marko Rauhamaa Cc: kuznet@ms2.inr.ac.ru, linux-kernel@vger.kernel.org, hadi@cyberus.ca, Robert.Olsson@data.slu.se Subject: Re: NAPI Race? Message-Id: <20031008143121.1efdd19d.davem@redhat.com> In-Reply-To: References: <200310081957.XAA01425@yakov.inr.ac.ru> X-Mailer: Sylpheed version 0.9.6 (GTK+ 1.2.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On 08 Oct 2003 14:17:31 -0700 Marko Rauhamaa wrote: > But do_softirq() explicitly refuses to run > the same softirq right away. Check current 2.6.x sources, it does loop a certain number of times even for the same softirq type. > Well, it almost does. You can blast a NAPI driver with a packet flood, > and the system is happy and responsive -- no interrupts are generated, > and packets are polled by ksoftirqd. However, you can find a packet rate > that will cause the CPU to spend virtually all of its time in NAPI. This situation can be created with non-NAPI drivers too. Alexey is trying to explain to you what the true cause of the problem is, and it's not NAPI, it's softirq starvation.