linux-kernel - Re: [PATCH 5.10/5.15] io_uring: fix registered files leak

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] [day] [month] [year] [list]

Date: Thu, 14 Mar 2024 10:02:43 -0600
From: Jens Axboe <axboe@...nel.dk>
To: Fedor Pchelkin <pchelkin@...ras.ru>
Cc: Pavel Begunkov <asml.silence@...il.com>,
 Greg Kroah-Hartman <gregkh@...uxfoundation.org>, stable@...r.kernel.org,
 io-uring@...r.kernel.org, linux-kernel@...r.kernel.org,
 Alexey Khoroshilov <khoroshilov@...ras.ru>, lvc-project@...uxtesting.org,
 Nikita Zhandarovich <n.zhandarovich@...tech.ru>,
 Roman Belyaev <belyaevrd@...dex.ru>
Subject: Re: [PATCH 5.10/5.15] io_uring: fix registered files leak

On 3/14/24 9:55 AM, Fedor Pchelkin wrote:
> On 24/03/13 06:40PM, Jens Axboe wrote:
>> Hi,
>>
>> OK, here they are. Two patches attached for every stable kernel, that
>> gets rid of the remnants of the SCM related code:
>>
>> 5.4
>> 5.10 and 5.15 (same patches)
>> 6.1
>> 6.6
>> 6.7
>>
>> Would appreciate if Fedor and Pavel could give them a once over, but I
>> think they are all fine. It's just deleting the code...
> 
> Thank you, Jens!
> 
> FWIW, I think it's all good and it eliminates the reported problem
> obviously. Compiled and tested the repro with my kernel config.

Great, thanks for checking!

> Just a minor notice - stable rules declare two common ways for upstream
> patch mentioning in backports [1]. And the first one starts from
> lowercase. No big deal here definitely but maybe somebody has some
> handling of these two variants - by regexps or similar, I actually don't
> know. But I see in the git history that Greg also applies the variant
> you've used.

Honestly that doesn't matter, as long as the upstream commit is
referenced. I always do it this way, not my first stable rodeo.

-- 
Jens Axboe