linux-kernel - Re: [GIT] Networking

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:   Fri, 7 Aug 2020 11:29:14 -0700
From:   John Stultz <john.stultz@...aro.org>
To:     Christoph Hellwig <hch@....de>
Cc:     Eric Dumazet <eric.dumazet@...il.com>,
        David Miller <davem@...emloft.net>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        netdev <netdev@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Todd Kjos <tkjos@...gle.com>,
        Amit Pundir <amit.pundir@...aro.org>,
        Bjorn Andersson <bjorn.andersson@...aro.org>
Subject: Re: [GIT] Networking

On Fri, Aug 7, 2020 at 12:19 AM Christoph Hellwig <hch@....de> wrote:
>
> On Thu, Aug 06, 2020 at 11:23:34PM -0700, John Stultz wrote:
> > So I've finally rebase-bisected it down to:
> >   a31edb2059ed ("net: improve the user pointer check in init_user_sockptr")
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a31edb2059ed4e498f9aa8230c734b59d0ad797a
> >
> > And reverting that from linus/HEAD (at least from this morning) seems
> > to avoid it.
> >
> > Seems like it is just adding extra checks on the data passed, so maybe
> > existing trouble from a different driver is the issue here, but it's
> > not really clear from the crash what might be wrong.
> >
> > Suggestions would be greatly appreciated!
>
> I think the sockpt optimization is just a little to clever for its
> own sake, as also chown by the other issue pointed out by Eric.
>
> Can you try this revert that just goes back to the "boring" normal
> version for everyone?

Yes! This seems to avoid the crash and networking looks ok.

Tested-by: John Stultz <john.stultz@...aro.org>

thanks
-john