linux-kernel - Re: [PATCH 2/6] ipc: reorganize initialization of kern_ipc

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:   Thu, 5 Jul 2018 18:08:37 +0200
From:   Dmitry Vyukov <dvyukov@...gle.com>
To:     Manfred Spraul <manfred@...orfullife.com>
Cc:     Davidlohr Bueso <dave@...olabs.net>,
        Andrew Morton <akpm@...ux-foundation.org>,
        LKML <linux-kernel@...r.kernel.org>, 1vier1@....de,
        Kees Cook <keescook@...omium.org>,
        Michael Kerrisk <mtk.manpages@...il.com>
Subject: Re: [PATCH 2/6] ipc: reorganize initialization of kern_ipc_perm.seq

On Thu, Jul 5, 2018 at 5:12 PM, Manfred Spraul <manfred@...orfullife.com> wrote:
> Hi Dmitry,
>
> On 07/05/2018 10:36 AM, Dmitry Vyukov wrote:
>>
>> [...]
>> Hi Manfred,
>>
>> The series looks like a significant improvement to me. Thanks!
>>
>> I feel that this code can be further simplified (unless I am missing
>> something here). Please take a look at this version:
>>
>>
>> https://github.com/dvyukov/linux/commit/f77aeaf80f3c4ab524db92184d874b03063fea3a?diff=split
>>
>> This is on top of your patches. It basically does the same as your
>> code, but consolidates all id/seq assignment and dealing with next_id,
>> and deduplicates code re CONFIG_CHECKPOINT_RESTORE. Currently it's a
>> bit tricky to follow e.g. where exactly next_id is consumed and where
>> it needs to be left intact.
>> The only difference is that my code assigns new->id earlier. Not sure
>> if it can lead to anything bad. But if yes, then it seems that
>> currently uninitialized new->id is exposed. If necessary (?) we could
>> reset new->id in the same place where we set new->deleted.
>
> Everything looks correct for me, it is better than the current code.
> Except that you didn't sign off your last patch.

It was meant more like a review comment expressed in code. But I signed it now.

> As next step: Who can merge the patches towards linux-next?
> The only open point that I see are stress tests of the error codepaths.
>
> And:
> I don't think that the patches are relevant for linux-stable, correct?

I don't have a pressing need for stable (simply because KMSAN/KTSAN
are not backported to any stable releases, so we won't discover it
there).

View attachment "ipc.txt" of type "text/plain" (2541 bytes)