netdev - Re: [PATCH net] ipv6: fix race condition between ipv6_get_ifaddr and ipv6_del

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: <20240405134731.5db239c0@griffin>
Date: Fri, 5 Apr 2024 13:47:31 +0200
From: Jiri Benc <jbenc@...hat.com>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: netdev@...r.kernel.org, edumazet@...gle.com, Stephen Hemminger
 <stephen@...workplumber.org>, David Ahern <dsahern@...nel.org>
Subject: Re: [PATCH net] ipv6: fix race condition between ipv6_get_ifaddr
 and ipv6_del_addr

On Fri, 5 Apr 2024 13:31:52 +0200, Eric Dumazet wrote:
> refcount_t semantic should prevent this double transition to 0  ?

You're right. This was found in an old kernel where it manifested
by a crash. In not so ancient kernels there's a refcount_t WARN:

[  202.500654] refcount_t: addition on 0; use-after-free.
[  202.500665] WARNING: CPU: 0 PID: 1327 at lib/refcount.c:25 refcount_warn_saturate+0x74/0x110

I knew about that but I wrote the commit message on the old kernel and
when updating it I missed this spot. Sorry about that.

> Can you include a stack trace in the changelog ?

I don't think I have a stack trace from the latest net kernel; let me
get one.

> Otherwise patch looks good to me, thanks !
> 
> Reviewed-by: Eric Dumazet <edumazet@...gle.com>

Thanks for the review! v2 with the updated commit message coming on
Monday.

 Jiri