From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ipmail03.adl6.internode.on.net ([150.101.137.143]:29801 "EHLO ipmail03.adl6.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750805AbeEKCKD (ORCPT ); Thu, 10 May 2018 22:10:03 -0400 Received: from discord.disaster.area ([192.168.1.111]) by dastard with esmtp (Exim 4.80) (envelope-from ) id 1fGxVB-0006xh-IN for linux-xfs@vger.kernel.org; Fri, 11 May 2018 12:09:45 +1000 Received: from dave by discord.disaster.area with local (Exim 4.91) (envelope-from ) id 1fGxVB-0002gI-FH for linux-xfs@vger.kernel.org; Fri, 11 May 2018 12:09:45 +1000 From: Dave Chinner Subject: [PATCH 0/2] xfs: handle mount failures more cleanly Date: Fri, 11 May 2018 12:09:41 +1000 Message-Id: <20180511020943.10132-1-david@fromorbit.com> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: linux-xfs@vger.kernel.org Hi folks, This is the XFS side of the "fs: don't scan the inode cache before SB_BORN is set" changes to handle mount failures more cleanly. This addresses teh use-after-free problems that can occur when xfs_fs_fill_super() fails, frees the xfs_mount structure but then leaves the VFS superblock pointing to it. Th efirst patch is the mount delay sysfs option I've been using to test this. I'll post the fstest that uses it in a minute. Cheers, Dave.