Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 25 Jul 2002 06:25:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 25 Jul 2002 06:25:57 -0400 Received: from pizda.ninka.net ([216.101.162.242]:40641 "EHLO pizda.ninka.net") by vger.kernel.org with ESMTP id ; Thu, 25 Jul 2002 06:25:57 -0400 Date: Thu, 25 Jul 2002 03:18:21 -0700 (PDT) Message-Id: <20020725.031821.123624987.davem@redhat.com> To: mingo@elte.hu Cc: torvalds@transmeta.com, linux-kernel@vger.kernel.org Subject: Re: Linux-2.5.28 From: "David S. Miller" In-Reply-To: References: <20020724.225921.108418454.davem@redhat.com> X-Mailer: Mew version 2.1 on Emacs 21.1 / Mule 5.0 (SAKAKI) 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 Content-Length: 817 Lines: 19 From: Ingo Molnar Date: Thu, 25 Jul 2002 11:28:15 +0200 (CEST) i think the networking code is a special case - nothing else relies on the interaction of timers and IRQ contexts in such a deep way. (which it does for performance reasons.) I'd say 99% of all cli()/sti() users are in the 'introduce a per-driver or per-subsystem lock' league Linus mentioned. I'm sure the serial drivers used to. Look at how they were using SERIAL_BH for example. RMK's stuff fixes that so wrt. the current state of affairs you're probably right. - 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/