From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262482AbTJBAOM (ORCPT ); Wed, 1 Oct 2003 20:14:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262672AbTJBAOM (ORCPT ); Wed, 1 Oct 2003 20:14:12 -0400 Received: from gateway-1237.mvista.com ([12.44.186.158]:31741 "EHLO av.mvista.com") by vger.kernel.org with ESMTP id S262482AbTJBAOL (ORCPT ); Wed, 1 Oct 2003 20:14:11 -0400 Message-ID: <3F7B6DCC.9090604@mvista.com> Date: Wed, 01 Oct 2003 17:14:04 -0700 From: George Anzinger Organization: MontaVista Software User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2) Gecko/20021202 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Yifan Zhu CC: linux-kernel@vger.kernel.org Subject: Re: TIMER_BH in 2.4 kernel References: <3F788DE4.8030505@unity.ncsu.edu> In-Reply-To: <3F788DE4.8030505@unity.ncsu.edu> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Yifan Zhu wrote: > Hi, > I'm looking at the 2.4 kernel. My understanding of the code is that the > TIMER_BH handler is executed when the hardware timer interrupt returns ( > after do_IRQ(), but before returning to the interrupted task ). Is that > right? > > Thanks for clarification! Unless there is a bh_lock in effect at the time, in which case it is executed when the lock is released. - -- George Anzinger george@mvista.com High-res-timers: http://sourceforge.net/projects/high-res-timers/ Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml