netdev - Re: [PATCH net-next] net: ip6_gre: Give ERSPAN a fill

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] [thread-next>] [day] [month] [year] [list]

Message-Id: <20190214.090847.1049635913680945391.davem@davemloft.net>
Date:   Thu, 14 Feb 2019 09:08:47 -0800 (PST)
From:   David Miller <davem@...emloft.net>
To:     petrm@...lanox.com
Cc:     netdev@...r.kernel.org, kuznet@....inr.ac.ru,
        yoshfuji@...ux-ipv6.org, lorenzo.bianconi@...hat.com
Subject: Re: [PATCH net-next] net: ip6_gre: Give ERSPAN a fill_info link op
 of its own

From: Petr Machata <petrm@...lanox.com>
Date: Wed, 13 Feb 2019 19:31:32 +0000

> In commit c706863bc890 ("net: ip6_gre: always reports o_key to
> userspace"), ip6gre and ip6gretap tunnels started reporting a TUNNEL_KEY
> output flag even if one was not configured at the device.
> 
> When an okey-less ip6gre or ip6gretap netdevice is created, it initially
> encapsulates the packets without okey. But any configuration change
> (even a non-change such as setting TOS to an already-configured value)
> then causes the okey flag from the reported configuration to be
> circulated back to actual configuration. From that point on, the device
> encapsulates packets with output key of 0.
> 
> The intention was to implement this behavior for ERSPAN devices, not for
> all ip6gre devices. The ERSPAN netdevice should really have its own
> fill_info callback. Add one.
> 
> Fixes: c706863bc890 ("net: ip6_gre: always reports o_key to userspace")
> CC: Lorenzo Bianconi <lorenzo.bianconi@...hat.com>
> Signed-off-by: Petr Machata <petrm@...lanox.com>

This commit you are fixing exists in the 'net' tree, therefore this is
a bug fix and should be targetted at 'net'.