netdev - Re: VRF: All router multicast entry(FF02:2) not added to VRF Dev but added on VLAN Dev

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>] [day] [month] [year] [list]

Message-ID: <ea4a8fbe-70c9-ead6-62b0-0be90959ccd8@gmail.com>
Date:   Sat, 21 Mar 2020 15:16:34 -0600
From:   David Ahern <dsahern@...il.com>
To:     Sukumar Gopalakrishnan <sukumarg1973@...il.com>,
        netdev@...r.kernel.org
Subject: Re: VRF: All router multicast entry(FF02:2) not added to VRF Dev but
 added on VLAN Dev

On 3/20/20 2:33 AM, Sukumar Gopalakrishnan wrote:
> Hi David,
> 
> 
> Kernel Version : 4.14.170
> 
> 
> Initiating ping to All Router address(FF02::2) with DIP as FF02:2. (
> decoded packet attached).
> 
>  
> 
> when the packet reaches
> 
> vrf_ip6_rcv
> 
>   ipv6_ndisc_frame() returns FALSE since the ICMPV6 TYPE is 128 ( echo
> request)

But need_strict is true since it is a mcast address, so the device
change should not happen. As I recall this is one of those fundamental
use cases (ping to ff02::*) that has worked since day 1 for VRF.

> 
>   Changing ingress VLAN dev to VRF DEV on skb->dev.
> 
>      
>