From patchwork Thu Dec 22 16:18:52 2011 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jan Kara X-Patchwork-Id: 132865 Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 589F3B7142 for ; Fri, 23 Dec 2011 03:19:16 +1100 (EST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754404Ab1LVQTO (ORCPT ); Thu, 22 Dec 2011 11:19:14 -0500 Received: from cantor2.suse.de ([195.135.220.15]:60666 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752394Ab1LVQTO (ORCPT ); Thu, 22 Dec 2011 11:19:14 -0500 Received: from relay2.suse.de (nat.nue.novell.com [195.135.221.2]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mx2.suse.de (Postfix) with ESMTP id 5BFC78891E; Thu, 22 Dec 2011 17:19:13 +0100 (CET) Received: by quack.suse.cz (Postfix, from userid 1000) id 82972205D6; Thu, 22 Dec 2011 17:18:52 +0100 (CET) Date: Thu, 22 Dec 2011 17:18:52 +0100 From: Jan Kara To: Eric Sandeen Cc: Jan Kara , linux-ext4@vger.kernel.org Subject: Re: [PATCH] ext3: Don't warn from writepage when readonly inode is spotted after error Message-ID: <20111222161852.GB5639@quack.suse.cz> References: <1324569415-9824-1-git-send-email-jack@suse.cz> <4EF356B8.5010605@redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4EF356B8.5010605@redhat.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-ext4-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-ext4@vger.kernel.org On Thu 22-12-11 10:11:36, Eric Sandeen wrote: > On 12/22/11 9:56 AM, Jan Kara wrote: > > WARN_ON_ONCE(IS_RDONLY(inode)) tends to trip when filesystem hits error and is > > remounted read-only. This unnecessarily scares users (well, they should be > > scared because of filesystem error, but the stack trace distracts them from the > > right source of their fear ;-). We could as well just remove the WARN_ON but > > it's not hard to fix it to not trip on filesystem with errors and not use more > > cycles in the common case so that's what we do. > > I've seen that too and it is distracting. > > Looks good to me, aside from the rather long lines ;) Ah, right. My internal style checker had a coffee break I assume ;). Below is a version with long lines fixed. Honza From 0588350bf10c8dad70cc299971dfdeec10754b29 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 22 Dec 2011 16:49:05 +0100 Subject: [PATCH] ext3: Don't warn from writepage when readonly inode is spotted after error WARN_ON_ONCE(IS_RDONLY(inode)) tends to trip when filesystem hits error and is remounted read-only. This unnecessarily scares users (well, they should be scared because of filesystem error, but the stack trace distracts them from the right source of their fear ;-). We could as well just remove the WARN_ON but it's not hard to fix it to not trip on filesystem with errors and not use more cycles in the common case so that's what we do. CC: stable@kernel.org Signed-off-by: Jan Kara --- fs/ext3/inode.c | 24 +++++++++++++++++++++--- 1 files changed, 21 insertions(+), 3 deletions(-) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index a8d3217..fbf6599 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1623,7 +1623,13 @@ static int ext3_ordered_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); /* * We give up here if we're reentered, because it might be for a @@ -1698,7 +1704,13 @@ static int ext3_writeback_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); if (ext3_journal_current_handle()) goto out_fail; @@ -1741,7 +1753,13 @@ static int ext3_journalled_writepage(struct page *page, int err; J_ASSERT(PageLocked(page)); - WARN_ON_ONCE(IS_RDONLY(inode)); + /* + * We don't want to warn for emergency remount. The condition is + * ordered to avoid dereferencing inode->i_sb in non-error case to + * avoid slow-downs. + */ + WARN_ON_ONCE(IS_RDONLY(inode) && + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)); if (ext3_journal_current_handle()) goto no_write;