Re: [Qemu-devel] [PATCH] qcow2: Fix refcount block allocation during qcow2_allocate_cluster_at()

[Kevin Wolf]

Am 23.04.2012 01:35, schrieb Marcelo Tosatti:
> On Sun, Apr 22, 2012 at 08:18:49PM -0300, Marcelo Tosatti wrote:
>> On Fri, Apr 20, 2012 at 03:56:01PM +0200, Kevin Wolf wrote:
>>> Refcount block allocation and refcount table growth rely on
>>> s->free_cluster_index pointing to somewhere after the current
>>> allocation. Change qcow2_allocate_cluster_at() to fulfill this
>>> assumption.
>>>
>>> Without this change it could happen that a newly allocated refcount
>>> block and the allocated data block point to the same area in the image
>>> file, causing data corruption in the long run.
>>>
>>> This fixes a bug that became first visible after commit 250196f1.
>>>
>>> Signed-off-by: Kevin Wolf <address@hidden>
>>
>> Kevin,
>>
>> This patch fixes explicit filesystem errors (qemu-img check also OK), 
>> but autotest is still failing, see attached screenshot. It is not
>> reproducible without
>>
>> f081987ad20a8c8dc391deded55161ea8d38be5f
> 
> Sorry, i meant
> 
> commit 250196f19c6e7df12965d74a5073e10aba06c802
> Author: Kevin Wolf <address@hidden>
> Date:   Fri Mar 2 14:10:54 2012 +0100
> 
>     qcow2: Reduce number of I/O requests

The screenshot doesn't really give a lot of information, but let's
assume that _something_ must have been corrupted... Can you try finding
the corrupted file (e.g. using rpm -V) and see in which way it differs
from the real one?

Kevin