linux-kernel - Re: [RFC 2/3] futex: exit: Print a warning when futex

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: <90124b02-76b5-d775-ac7d-a2e9d7677fa1@redhat.com>
Date:   Fri, 22 Apr 2022 10:23:50 -0400
From:   Nico Pache <npache@...hat.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Ingo Molnar <mingo@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Darren Hart <dvhart@...radead.org>,
        Davidlohr Bueso <dave@...olabs.net>,
        André Almeida <andrealmeid@...labora.com>,
        Arjan van de Ven <arjan@...radead.org>,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        Matthew Wilcox <willy@...radead.org>
Subject: Re: [RFC 2/3] futex: exit: Print a warning when futex_cleanup fails



On 4/21/22 16:53, Thomas Gleixner wrote:
> On Thu, Apr 21 2022 at 15:05, Nico Pache wrote:
> 
>> The futex_cleanup routines currently fails silently.
>>
>> Allow the futex_cleanup routines to fail more verbosely so we can
>> leave a message behind for easier debugging/error detecting.
> 
> This is a free ticket to spam dmesg for any unpriviledged user.
As Matthew pointed out, we'd want to rate limit this. I was also thinking we can
limit this to debug kernels so unpriviledged users can not cause dmesg spamming
in production configs.
> 
>> Fixes: 0771dfefc9e5 ("[PATCH] lightweight robust futexes: core")
> 
> There is nothing to fix, really.
Fair enough, its been part of the design since inceptions.
> 
> Robust futexes are best effort as I explained you before and spamming
> dmesg won't help anything.
It would have helped find the OOM/Robust Futex issue more quickly. It may also
help detect breakages in the pthread code (or any other users of robust futex)
if someone is doing something incorrectly.

Just a thought though. If you think its pointless to proceed down this path I
wont push forward.

Cheers,
-- Nico