linux-kernel - Re: Possible race in dev_coredumpm()-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] [day] [month] [year] [list]

Date:   Thu, 14 Apr 2022 13:31:49 +0200
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Mukesh Ojha <quic_mojha@...cinc.com>
Cc:     gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org,
        sboyd@...nel.org, johannes@...solutions.net, rafael@...nel.org
Subject: Re: Possible race in dev_coredumpm()-del_timer() path

On Thu, Apr 14 2022 at 16:50, Mukesh Ojha wrote:
> On Thu, Apr 14, 2022 at 12:38:13PM +0200, Thomas Gleixner wrote:
>> So, yes this needs serialization of some sort.
>
> Thanks for understanding the problem.
> Can the patch mentioned at below link helps with the first problem ?
>
> https://lore.kernel.org/lkml/57a04278-0a60-cc7d-7ce8-a75c2befd568@quicinc.com/

Something like that.

>> Same problem vs. disabled_store().
>
> you mean, while userspace is reading the data and suddenly disable_store() done from
> sysfs.

No, that's not a problem because the reader holds a reference, but it's
the same problem vs. initialization:

  device_add()
                        disable_store()
                          devcd_free()
                           mod_delayed_work()
                           ....
Thanks,

        tglx