netdev - Re: tipc_udp_send_msg oops in 4.4 when setting link tolerance

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:   Wed, 15 Nov 2017 11:42:15 +0200
From:   Tommi Rantala <tommi.t.rantala@...ia.com>
To:     Jon Maloy <jon.maloy@...csson.com>,
        Ying Xue <ying.xue@...driver.com>,
        "David S. Miller" <davem@...emloft.net>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "tipc-discussion@...ts.sourceforge.net" 
        <tipc-discussion@...ts.sourceforge.net>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: tipc_udp_send_msg oops in 4.4 when setting link tolerance

On 14.11.2017 13:35, Jon Maloy wrote:
>> Found it, the missing patch is this one (9b3009604b8e does not help):
>>
>> commit d01332f1acacc0cb43a61f4244dd2b846d4cd585
>> Author: Richard Alpe <richard.alpe@...csson.com>
>> Date:   Mon Feb 1 08:19:56 2016 +0100
>>
>>       tipc: fix link attribute propagation bug
>>
>>
>> It does not apply as-is to 4.4, so backported it, see below.
>> Does it look good? I can send it forward to Greg for inclusion in 4.4.
> 
> Yes. I would be very grateful if you do.

OK, sent the patch to stable@...r.kernel.org

>> But with this patch included, I can easily reproduce the "BUG: Bad page state
>> in process git" issue also in 4.4 like this:
>>
>> $ tipc link set tolerance 100 link $LINKNAME $ cd /tmp && git clone
>> /path/to/linux-stable
>>
>> I can try to debug that a bit more to see if I can figure it out.
> 
> I would appreciate that. If this turns out to also be an already fixed bug (most likely) you can send that directly to GKH too.
> If you find a bug that requires a new patch send it to tipc-discussion for review first.
> 
> Finally, I want to say that I very much appreciate users like you, who try solving the problems themselves and contribute to the code.
> We are right now in an upgrade phase of TIPC, where we have been/are adding new features to TIPC (overlapping ring neighbor monitoring, group communication, new addressing model, performance improvements...) which inevitably entail some new bugs, but I expect the frequency of these changes/improvements to TIPC go down in a few months, and with that the number of fixes needed.

Thanks!

After more testing, I noticed that I could only reproduce the "BUG: Bad 
page state" in a single machine, so maybe something wrong with the 
machine, and not with tipc.

-Tommi