From mboxrd@z Thu Jan  1 00:00:00 1970
From: David Ahern <dsahern@gmail.com>
Subject: Re: [Bug 199643] New: UBSAN: Undefined behaviour in
 ./include/net/route.h:240:2
Date: Thu, 7 Jun 2018 17:53:35 -0700
Message-ID: <e6d18d55-2e35-b888-040f-b31117e38daf@gmail.com>
References: <20180508.001251.2189714817072679386.davem@davemloft.net>
 <b0e36641-b84f-0bde-4c2d-125a412b5bb2@gmail.com>
 <20180607170623.6e5d56b5@cakuba.netronome.com>
 <20180607.201142.389815374334636334.davem@davemloft.net>
 <e05416a5-b4fe-6e9f-cefc-86848fb146cf@gmail.com>
 <20180607174923.13a11d08@cakuba.netronome.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Cc: David Miller <davem@davemloft.net>, stephen@networkplumber.org,
        netdev@vger.kernel.org
To: Jakub Kicinski <jakub.kicinski@netronome.com>,
        Eric Dumazet <eric.dumazet@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pf0-f180.google.com ([209.85.192.180]:38193 "EHLO
        mail-pf0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752273AbeFHAxj (ORCPT
        <rfc822;netdev@vger.kernel.org>); Thu, 7 Jun 2018 20:53:39 -0400
Received: by mail-pf0-f180.google.com with SMTP id b74-v6so5745318pfl.5
        for <netdev@vger.kernel.org>; Thu, 07 Jun 2018 17:53:39 -0700 (PDT)
In-Reply-To: <20180607174923.13a11d08@cakuba.netronome.com>
Content-Language: en-US
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On 6/7/18 5:49 PM, Jakub Kicinski wrote:
> On Thu, 7 Jun 2018 17:28:59 -0700, Eric Dumazet wrote:
>> On 06/07/2018 05:11 PM, David Miller wrote:
>>> From: Jakub Kicinski <jakub.kicinski@netronome.com>
>>> Date: Thu, 7 Jun 2018 17:06:23 -0700
>>>   
>>>> [  293.213661]  ip_send_unicast_reply+0x1b67/0x1d0e  
>>>
>>> This calls ip_setup_cork() which can NULL out the 'rt' route
>>> pointer.  Hmmm... :-/
>>>   
>>
>>
>> UBSAN seems unhappy  with dst being NULL in :
>>
>> dst_release(&rt->dst);
>>
>> But the code obviously is ready for dst being NULL, it is even documented :)
> 
> Oh, so the code depends on dst being the first member?  Would it make
> sense to just cast the pointer instead?
> 

I've been going the other way with 'rt to dst' and 'dst to rt'
transformations.

Perhaps UBSAN should be updated to understand that NULL + 0 is ok.