Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264738AbTE1QWa (ORCPT ); Wed, 28 May 2003 12:22:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264793AbTE1QWa (ORCPT ); Wed, 28 May 2003 12:22:30 -0400 Received: from smtpzilla2.xs4all.nl ([194.109.127.138]:25609 "EHLO smtpzilla2.xs4all.nl") by vger.kernel.org with ESMTP id S264738AbTE1QW3 (ORCPT ); Wed, 28 May 2003 12:22:29 -0400 Date: Wed, 28 May 2003 18:35:15 +0200 (CEST) From: Roman Zippel X-X-Sender: roman@serv To: "David S. Miller" cc: mika.penttila@kolumbus.fi, , Andrew Morton , , , Subject: Re: [patch] cache flush bug in mm/filemap.c (all kernels >= 2.5.30(at least)) In-Reply-To: <20030527.142233.71088632.davem@redhat.com> Message-ID: References: <20030526.153415.41663121.davem@redhat.com> <20030527.142233.71088632.davem@redhat.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 989 Lines: 25 Hi, On Tue, 27 May 2003, David S. Miller wrote: > The point I don't like about update_mmu_cache() is that it's called > _after_ set_pte(). Practically it's maybe not a problem right now, but > the cache synchronization should happen before set_pte(). > > update_mmu_cache() is specifically supposed to always occur before > anyone could try to use the mapping created. > > If this is ever violated, it will be fixed because it is a BUG(). set_pte() establishes the mapping, this means another cpu can get the pte and start reading the data e.g. into the instruction cache, but if that cpu still has dirty data in the data cache (e.g. a write() or a disk read), the following update_mmu_cache() might be too late. bye, Roman - 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/