From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265891AbTL3W1I (ORCPT ); Tue, 30 Dec 2003 17:27:08 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265872AbTL3WYQ (ORCPT ); Tue, 30 Dec 2003 17:24:16 -0500 Received: from imf20aec.mail.bellsouth.net ([205.152.59.68]:57550 "EHLO imf20aec.mail.bellsouth.net") by vger.kernel.org with ESMTP id S265880AbTL3WXi (ORCPT ); Tue, 30 Dec 2003 17:23:38 -0500 Subject: Re: no DRQ after issuing WRITE was Re: 2.4.23-uv3 patch set released From: Rob Love To: Linus Torvalds Cc: Marcelo Tosatti , Daniel Tram Lux , steve@drifthost.com, James Bourne , Linux Kernel , Gergely Tamas , Bartlomiej Zolnierkiewicz In-Reply-To: References: Content-Type: text/plain Message-Id: <1072823015.4350.40.camel@fur> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.5 (1.4.5-8) Date: Tue, 30 Dec 2003 17:23:35 -0500 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2003-12-30 at 16:57, Linus Torvalds wrote: > Is CONFIG_PREEMPT on in the cases, and is there really no locking > anywhere? Preempting in the middle of the data transfer phase sounds like > a fundamentally bad idea, and maybe the code needs a few preempt > disable/enable pairs somewhere? Is the kernel patched with kernel preemption? It is not in stock 2.4. Anyhow, if interrupts are disabled, preemption should be disabled (we check for that condition in both preempt_schedule() and return_from_intr). If interrupts are not disabled, then preempting would definitely be a bad thing. But I would think, for the same reasons you do not want to preempt, you would want interrupts disabled .. Rob Love