linux-ext4 - [PATCH] xfstests: add regression test for extent corruption on ext4

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87pqhrs8il.fsf@dmbot.sw.ru>
Date:	Fri, 21 Oct 2011 00:38:58 +0400
From:	Dmitry Monakhov <dmonakhov@...nvz.org>
To:	linux-fsdevel@...r.kernel.org, xfs@....sgi.com,
	linux-ext4@...r.kernel.org, jack@...e.cz, hch@...radead.org,
	aelder@....com, tytso@....edu
Subject: [PATCH] xfstests: add regression test for extent corruption on ext4

On Thu, 20 Oct 2011 13:40:15 +0400, Dmitry Monakhov <dmonakhov@...nvz.org> wrote:
> On Wed, 19 Oct 2011 14:29:41 +0400, Dmitry Monakhov <dmonakhov@...nvz.org> wrote:
> > xfstests: fsstress should kill children tasks before exit
> > xfstests: add different logging option to fsstress
> > xfstests: add fallocate support to fsstress
> > xfstests: fsstress add FS_IOC_{SET,GET}FLAGS operations
> > xfstests: Dump inode info when possible
> > xfstests: add fiemap operation
> > 
> > # Following testcase will fail on recent kernel in case of ext4
> > # Some fixes are already available here:
> > # http://www.spinics.net/lists/linux-ext4/msg27857.html
> > xfstests: add new stress test
> FYI: i've run 264'th test on recent tytso/ext4.git(dcf2d804ed6ffe)
> and have got new type of fs-corruption, i haven't seen this before,
> at this moment I dont have a fix for that:
> Pass 1: Checking inodes, blocks, and sizes
> Inode 176 has an invalid extent
>       (logical block 236, invalid physical block 0, len 0)
> Clear? no
> 
> Inode 22293 has an invalid extent
>       (logical block 531, invalid physical block 0, len 0)
> Clear? no
> 
> Pass 2: Checking directory structure
> Pass 3: Checking directory connectivity
> Pass 4: Checking reference counts
> Pass 5: Checking group summary information
> ...
Ok, this one appears to be easy. eh_entries updated after
ext4_ext_dirty() was called inside ext4_ext_rm_leaf().
Testcase attached, will send a fix for the issue in a minute.

View attachment "0001-xfstests-add-regression-test-for-extent-corruption-o.patch" of type "text/plain" (3771 bytes)