Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 28 Nov 2001 20:40:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 28 Nov 2001 20:40:07 -0500 Received: from lightning.swansea.linux.org.uk ([194.168.151.1]:7172 "EHLO the-village.bc.nu") by vger.kernel.org with ESMTP id ; Wed, 28 Nov 2001 20:39:51 -0500 Subject: Re: [PATCH] remove BKL from drivers' release functions To: haveblue@us.ibm.com (David C. Hansen) Date: Thu, 29 Nov 2001 01:47:47 +0000 (GMT) Cc: rmk@arm.linux.org.uk (Russell King), linux-kernel@vger.kernel.org In-Reply-To: <3C059087.9070900@us.ibm.com> from "David C. Hansen" at Nov 28, 2001 05:33:59 PM X-Mailer: ELM [version 2.5 PL6] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: From: Alan Cox Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > Does everyone agree that we need to get the BKL out of common areas like > this? For starters, what about adding a pair of spinlocks for block Ideally the BKL itself should die. > devices and character devices to take the place of the BKL in > serializing opens? Or, should we make it the driver's responsibility > completely? It needs to be the drivers job, to be documented as such and to be implemented properly in some drivers. In paticular there are some extremely interesting closedown races in existing drivers where it goes CPU1 CPU2 release [do slow thing] open ioctl setting stuff up slow thing done trash the chip setup turn the chip off return ouch bang splat Oops!!! Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/