netdev - Re: [PATCH] ipv6: remove min MTU check for tunnel destinations

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]

Message-ID: <0738c72c20d60fd2ffa68573951d0ee3@codeaurora.org>
Date:   Thu, 10 May 2018 19:17:42 +0530
From:   Ashwanth Goli <ashwanth@...eaurora.org>
To:     David Miller <davem@...emloft.net>
Cc:     netdev@...r.kernel.org, pabeni@...hat.com, dsahern@...il.com,
        netdev-owner@...r.kernel.org
Subject: Re: [PATCH] ipv6: remove min MTU check for tunnel destinations

On 2018-05-01 21:52, David Miller wrote:
> From: Ashwanth Goli <ashwanth@...eaurora.org>
> Date: Mon, 30 Apr 2018 00:36:47 +0530
> 
>> With 749439bfac "fix udpv6 sendmsg crash caused by too small MTU"
>> tunnel dst's that report a MTU less than IPV6_MIN_MTU are broken
>> even for packets that are smaller than IPV6_MIN_MTU.
>> 
>> According to rfc2473#section-7.1
>> 
>>     if the original IPv6 packet is equal or  smaller  than  the
>>     IPv6 minimum link MTU, the tunnel entry-point node
>>     encapsulates the original packet, and subsequently
>>     fragments the resulting IPv6 tunnel packet into IPv6
>>     fragments that do not exceed the Path MTU to the tunnel
>>     exit-point.
>> 
>> This patch drops the MTU check for tunnel destinations.
>> 
>> Signed-off-by: Ashwanth Goli <ashwanth@...eaurora.org>
> 
> RFC 2473 is generally about ipv6 tunnels....
> 
>> -	if (mtu < IPV6_MIN_MTU)
>> +	if (!(rt->dst.flags & DST_XFRM_TUNNEL) && mtu < IPV6_MIN_MTU)
>>  		return -EINVAL;
> 
> But the check you are adding is specifically checking only IPSEC
> tunnels.
> 
> If what you say is true in your commit message, this test must
> more generally trigger for all ipv6 tunnel types, not just IPSEC
> ones.
> 
> If IPSEC tunnels are being targetting in this patch intentionally,
> that needs to be explained in the commit message.

My intention is to fix the issue for IPSEC tunnels. Will resend by 
changing the commit text.