netdev - Re: [PATCH net] net, ip_tunnel: fix interface lookup with no key

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 for Android: free password hash cracker in your pocket

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

Message-ID: <67629ca1-ad66-735a-ecfc-28531b079c40@6wind.com>
Date:   Thu, 26 Mar 2020 19:01:20 +0100
From:   Nicolas Dichtel <nicolas.dichtel@...nd.com>
To:     William Dauchy <w.dauchy@...teo.com>, netdev@...r.kernel.org
Cc:     pshelar@...ira.com
Subject: Re: [PATCH net] net, ip_tunnel: fix interface lookup with no key

Le 25/03/2020 à 16:03, William Dauchy a écrit :
> when creating a new ipip interface with no local/remote configuration,
> the lookup is done with TUNNEL_NO_KEY flag, making it impossible to
> match the new interface (only possible match being fallback or metada
> case interface); e.g: `ip link add tunl1 type ipip dev eth0`
> 
> If we consider `key` being zero in ipip case, we might consider ok to
> go through this last loop, and make it possible to match such interface.
> In fact this is what is done when we create a gre interface without key
> and local/remote.
> 
> context being on my side, I'm creating an extra ipip interface attached
> to the physical one, and moving it to a dedicated namespace.
> 
> Fixes: c54419321455 ("GRE: Refactor GRE tunneling code.")
> Signed-off-by: William Dauchy <w.dauchy@...teo.com>
> ---
>  net/ipv4/ip_tunnel.c | 3 ---
>  1 file changed, 3 deletions(-)
> 
> diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
> index 74e1d964a615..f6578bcadbed 100644
> --- a/net/ipv4/ip_tunnel.c
> +++ b/net/ipv4/ip_tunnel.c
> @@ -142,9 +142,6 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn,
>  			cand = t;
>  	}
>  
> -	if (flags & TUNNEL_NO_KEY)
> -		goto skip_key_lookup;
> -
Hmm, removing this test may break some existing scenario. This flag is part of
the UAPI (for gre). Suppose that a tool configures a gre tunnel, leaves the key
uninitialized and set this flag. After this patch, the lookup may return
something else.


Regards,
Nicolas