netdev - Re: [PATCH v2 bpf-next 06/17] net: Add IFLA_XDP_EGRESS for XDP programs in the egress path

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: <20200424115616.77a5689c@carbon>
Date:   Fri, 24 Apr 2020 11:56:16 +0200
From:   Jesper Dangaard Brouer <brouer@...hat.com>
To:     David Ahern <dsahern@...nel.org>
Cc:     netdev@...r.kernel.org, davem@...emloft.net, kuba@...nel.org,
        prashantbhole.linux@...il.com, jasowang@...hat.com,
        toke@...hat.com, toshiaki.makita1@...il.com, daniel@...earbox.net,
        john.fastabend@...il.com, ast@...nel.org, kafai@...com,
        songliubraving@...com, yhs@...com, andriin@...com,
        dsahern@...il.com, David Ahern <dahern@...italocean.com>,
        brouer@...hat.com
Subject: Re: [PATCH v2 bpf-next 06/17] net: Add IFLA_XDP_EGRESS for XDP
 programs in the egress path

On Thu, 23 Apr 2020 20:11:37 -0600
David Ahern <dsahern@...nel.org> wrote:

> Running programs in the egress path, on skbs or xdp_frames, does not
> require driver specific resources like Rx path. Accordingly, the
> programs can be run in core code, so add xdp_egress_prog to net_device
> to hold a reference to an attached program.

I disagree.  The TX path does need driver specific resources, most
importantly information about the TX-queue that was used.

That said, I think this patch is the right design, to place this more
centrally in the net-core code, as driver changes are harder to
maintain and generally painful (I speak from experience ;-)).

After this patchset goes in, we need continue this work, and find a
solution for the XDP-redirect overflow problem, and TX-queue selection.
We might still have some driver work ahead, as I think we can change the
ndo_xdp_xmit() API, to give us the information I'm looking for.

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer