netdev - Re: [bpf-next v3 8/9] bpf: Provide helper to do forwarding lookups in kernel FIB table

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: <867e4a5a-8d53-b142-10f8-485c6dec2859@gmail.com>
Date:   Thu, 10 May 2018 23:30:00 -0700
From:   David Ahern <dsahern@...il.com>
To:     Mathieu Xhonneux <m.xhonneux@...il.com>
Cc:     netdev@...r.kernel.org, borkmann@...earbox.net, ast@...nel.org,
        David Miller <davem@...emloft.net>, shm@...ulusnetworks.com,
        roopa@...ulusnetworks.com, brouer@...hat.com, toke@...e.dk,
        john.fastabend@...il.com
Subject: Re: [bpf-next v3 8/9] bpf: Provide helper to do forwarding lookups in
 kernel FIB table

On 5/10/18 12:27 PM, Mathieu Xhonneux wrote:
> I'm quite interested in this helper to implement OAM features (through
> other hooks, e.g. the BPF LWT hook). Do you have an idea about how it
> behaves with ECMP routes (with IPv4 and/or IPv6) ? In IPv6, I'm
> guessing that the returned gateway address is always a link-local
> address ?

ECMP works with BPF and forwarding just like it does with the full stack
-- the traffic should be distributed based on the L3 or L4 hash
algorithm and the number of links.

As for the IPv6 gateway question, it returns the gateway of the route in
the FIB -- ie., link local or global address. No assumptions are
necessary or made for the BPF helper.