From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933053AbWKNAWa (ORCPT ); Mon, 13 Nov 2006 19:22:30 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933240AbWKNAW3 (ORCPT ); Mon, 13 Nov 2006 19:22:29 -0500 Received: from ns1.suse.de ([195.135.220.2]:30105 "EHLO mx1.suse.de") by vger.kernel.org with ESMTP id S933053AbWKNAW3 (ORCPT ); Mon, 13 Nov 2006 19:22:29 -0500 From: NeilBrown To: Andrew Morton Date: Tue, 14 Nov 2006 11:22:23 +1100 X-face: [Gw_3E*Gng}4rRrKRYotwlE?.2|**#s9D Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org The patches which enable reading from raid5 without going through the stripe cache have a bug which causes corruption when reading from a raid6. It might be appropriate to put out a hotfix for rc5-mm1 which reverts md-enable-bypassing-cache-for-reads.patch The follow 4 patches fix this bug and a few other bugs I found while re-reviewing and re-testing this code. There are actually about 9 separate bugs here, but I grouped some of them to avoid having lots of tiny patches. NeilBrown [PATCH 001 of 4] md: Fix innocuous bug in raid6 stripe_to_pdidx [PATCH 002 of 4] md: Fix newly introduced read-corruption with raid6. [PATCH 003 of 4] md: Misc fixes for aligned-read handling. [PATCH 004 of 4] md: Fix a couple more bugs in raid5/6 aligned reads