linux-kernel - Re: kmemleak splat on copy

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]

Date:   Fri, 17 Feb 2017 09:07:53 -0800
From:   Andy Lutomirski <luto@...capital.net>
To:     "Luis R. Rodriguez" <mcgrof@...nel.org>
Cc:     Michal Hocko <mhocko@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Ingo Molnar <mingo@...nel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Mateusz Guzik <mguzik@...hat.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: kmemleak splat on copy_process()

On Wed, Feb 8, 2017 at 5:37 PM, Luis R. Rodriguez <mcgrof@...nel.org> wrote:
> On Tue, Feb 07, 2017 at 09:03:43AM +0100, Michal Hocko wrote:
>> On Tue 07-02-17 02:37:02, Luis R. Rodriguez wrote:
>> > > From a quick check I do not see any leak there either.
>> >
>> > Then in that case what about:
>>
>> This just disables the kmemleak altogether which doesn't sound like a
>> good idea to me.
>
> Only for this case, but if that is also not desirable let us debug further.
> That or I think we could perhaps massage code to make it clearer to kmemleak
> things are good.
>

I'm not seeing the issue.  There should be a live pointer to stack at
all times, either in a local variable or in task->stack.  There's a
weird window in dup_task_struct in which we're stashing away
stack_vm_area, but stack itself should be okay, I think.  But maybe
there really is a race in which a kmemleak check right in the middle
of duplicating the task struct really can't see the stack pointer.