netdev - Re: [PATCH net 1/2] ip: fix dflt addr selection for connected nexthop

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: <806e5672-f890-fa06-040b-2d59ed80d9e3@6wind.com>
Date:   Tue, 12 Jul 2022 10:45:55 +0200
From:   Nicolas Dichtel <nicolas.dichtel@...nd.com>
To:     Paolo Abeni <pabeni@...hat.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Eric Dumazet <edumazet@...gle.com>,
        David Ahern <dsahern@...nel.org>
Cc:     netdev@...r.kernel.org, stable@...r.kernel.org,
        Edwin Brossette <edwin.brossette@...nd.com>
Subject: Re: [PATCH net 1/2] ip: fix dflt addr selection for connected nexthop


Le 12/07/2022 à 10:11, Paolo Abeni a écrit :
[snip]
>>
>> ip -n foo link add name veth0 type veth peer name veth1 netns bar
>> ip -n foo link set veth0 address 00:09:c0:26:05:82
>> ip -n foo link set veth0 arp off
> 
> It looks like the 'arp off'/fixed mac address is not relevant for the
> test case, could you please drop it, so that the example and the self-
> test are more clean?Good point, I will remove these lines.

[snip]
>> Fixes: 597cfe4fc339 ("nexthop: Add support for IPv4 nexthops")
> 
> Why that commit? It looks like fib_check_nh() used SCOPE_HOST for nongw
> next hop since well before ?!?
Yes, but with "standard" route, ie when the nexthop objects are not used, the
scope used, at the end, is the one specified in the netlink message.
With nexthop object, the user cannot specify the scope and we end up in this
problem.
Thus, the problem exists only with nexthop object only.

I tried to explain this difference in the commit log, but maybe it's not clear
enough.

> 
> Otherwise LGTM.


Thank you,
Nicolas