linux-ext4 - Re: [PATCH 1/1] ext4: fix crash on BUG_ON in ext4_alloc_group

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 for Android: free password hash cracker in your pocket

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

Message-ID: <0596a1ae-f47c-4b6f-8849-73e7cfe7ff39@sandeen.net>
Date: Thu, 26 Sep 2024 21:39:26 -0500
From: Eric Sandeen <sandeen@...deen.net>
To: Baokun Li <libaokun1@...wei.com>
Cc: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@...onical.com>,
 tytso@....edu, stable@...r.kernel.org,
 Andreas Dilger <adilger.kernel@...ger.ca>,
 Stéphane Graber <stgraber@...raber.org>,
 Christian Brauner <brauner@...nel.org>, linux-kernel@...r.kernel.org,
 linux-fsdevel@...r.kernel.org, linux-ext4@...r.kernel.org,
 Wesley Hershberger <wesley.hershberger@...onical.com>,
 Yang Erkun <yangerkun@...wei.com>, Jan Kara <jack@...e.cz>
Subject: Re: [PATCH 1/1] ext4: fix crash on BUG_ON in ext4_alloc_group_tables

On 9/26/24 8:51 PM, Baokun Li wrote:
> On 2024/9/27 0:29, Eric Sandeen wrote:
>> On 9/26/24 11:04 AM, Eric Sandeen wrote:
>>
>>  
>>> Can you explain what the 2 cases under
>>>
>>> /* Avoid allocating large 'groups' array if not needed */
>>>
>>> are doing? I *think* the first 'if' is trying not to over-allocate for the last
>>> batch of block groups that get added during a resize. What is the "else if" case
>>> doing?
>> (or maybe I had that backwards)
>>
>> Incidentally, the offending commit that this fixes (665d3e0af4d35ac) got turned
>> into CVE-2023-52622, so it's quite likely that distros have backported the broken
>> commit as part of the CVE game.
> The commit to fix CVE-2023-52622 is commit 5d1935ac02ca5a
> ("ext4: avoid online resizing failures due to oversized flex bg").

I'm sorry - you're right. 665d3e0af4d35ac was part of the original
series that included 5d1935ac02ca5a, but it was not the fix.

> This commit does not address the off by one issue above.

Agreed.

>>
>> So the followup fix looks a bit urgent to me.
>>
>> -Eric
> Okay, I'll send out the fix patch today.

thanks :)

-Eric

> 
> Regards,
> Baokun
> .
>