From: Manish Katiyar Subject: Re: [PATCH] ext4: Fix missing iput for root inode in case of all failed mount paths. Date: Mon, 7 Feb 2011 15:31:41 -0800 Message-ID: References: <032DF412-6B37-4EB7-A687-8216CB1E2BAB@dilger.ca> <20110207232343.GA26289@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Cc: Andreas Dilger , ext4 To: "Ted Ts'o" Return-path: Received: from mail-qy0-f174.google.com ([209.85.216.174]:60493 "EHLO mail-qy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755204Ab1BGXcW (ORCPT ); Mon, 7 Feb 2011 18:32:22 -0500 Received: by qyj19 with SMTP id 19so1934466qyj.19 for ; Mon, 07 Feb 2011 15:32:03 -0800 (PST) In-Reply-To: <20110207232343.GA26289@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Mon, Feb 7, 2011 at 3:23 PM, Ted Ts'o wrote: > On Sat, Jan 29, 2011 at 09:40:34PM -0800, Manish Katiyar wrote: >> >> Any feedback on this ? > > The patch looks good, but it doesn't fix bug #26752. > > The root cause of that bug is the fact that ext4_sync_fs() and > ext4_put_super() assumes that sbi has been initialized. Hmm.. Actually I was looking at generic_super_shutdown() which I thought would call ext4_put_super() only if sb->root is set, and since I was setting sb->root explicitly to NULL in my patch in case of failures. -- Thanks - Manish