From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751075AbXBUEtr (ORCPT ); Tue, 20 Feb 2007 23:49:47 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751116AbXBUEtr (ORCPT ); Tue, 20 Feb 2007 23:49:47 -0500 Received: from ns2.suse.de ([195.135.220.15]:41418 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751075AbXBUEtq (ORCPT ); Tue, 20 Feb 2007 23:49:46 -0500 From: Nick Piggin To: Linux Memory Management , Andrew Morton Cc: Linux Kernel , Nick Piggin , Benjamin Herrenschmidt Message-Id: <20070221023656.6306.246.sendpatchset@linux.site> Subject: [patch 0/6] fault vs truncate/invalidate race fix Date: Wed, 21 Feb 2007 05:49:38 +0100 (CET) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org The following set of patches are based on current git. These fix the fault vs invalidate and fault vs truncate_range race for filemap_nopage mappings, plus those and fault vs truncate race for nonlinear mappings. These patches fix silent data corruption that we've had several people hitting in SUSE kernels. Our kernels have similar patches to lock the page over page fault, and no problem. I've also got rid of the horrible populate API, and integrated nonlinear pages properly with the page fault path. Downside is that this adds one more vector through which the buffered write deadlock can occur. However this is just a very tiny one (pte being unmapped for reclaim), compared to all the other ways that deadlock can occur (unmap, reclaim, truncate, invalidate). I doubt it will be noticable. At any rate, it is better than data corruption. I hope these can get merged (at least into -mm) soon. Thanks, Nick -- SuSE Labs