From: Jesper Dangaard Brouer via iovisor-dev <iovisor-dev-9jONkmmOlFHEE9lA1F8Ukti2O/JbrIOy@public.gmane.org>
To: "iovisor-dev-9jONkmmOlFHEE9lA1F8Ukti2O/JbrIOy@public.gmane.org"
<iovisor-dev-9jONkmmOlFHEE9lA1F8Ukti2O/JbrIOy@public.gmane.org>
Cc: Jakub Kicinski
<jakub.kicinski-wFxRvT7yatFl57MIdRCFDg@public.gmane.org>,
"netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
John Fastabend
<john.r.fastabend-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
Edward Cree <ecree-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>,
Simon Horman
<simon.horman-wFxRvT7yatFl57MIdRCFDg@public.gmane.org>,
Rana Shahout <ranas-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
Ari Saha <as754m-60p5jsuXm+c@public.gmane.org>,
Tariq Toukan <tariqt-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Subject: XDP seeking input from NIC hardware vendors
Date: Thu, 7 Jul 2016 12:42:45 +0200 [thread overview]
Message-ID: <20160707124245.6d95635a@redhat.com> (raw)
Would it make sense from a hardware point of view, to split the XDP
eBPF program into two stages.
Stage-1: Filter (restricted eBPF / no-helper calls)
Stage-2: Program
Then the HW can choose to offload stage-1 "filter", and keep the
likely more advanced stage-2 on the kernel side. Do HW vendors see a
benefit of this approach?
The generic problem I'm trying to solve is parsing. E.g. that the
first step in every XDP program will be to parse the packet-data,
in-order to determine if this is a packet the XDP program should
process.
Actions from stage-1 "filter" program:
- DROP (like XDP_DROP, early drop)
- PASS (like XDP_PASS, normal netstack)
- MATCH (call stage-2, likely carry-over opaque return code)
The MATCH action should likely carry-over an opaque return code, that
makes sense for the stage-2 program. E.g. proto id and/or data offset.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
Author of http://www.iptv-analyzer.org
LinkedIn: http://www.linkedin.com/in/brouer
next reply other threads:[~2016-07-07 10:42 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-07 10:42 Jesper Dangaard Brouer via iovisor-dev [this message]
2016-07-07 15:18 ` XDP seeking input from NIC hardware vendors Fastabend, John R
[not found] ` <D6BB30FE66EA894C9F13C9E3CDDF00F564E5FB81-5FK+k9557ZBqS6EAlXoojrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2016-07-07 16:12 ` Jakub Kicinski via iovisor-dev
2016-07-07 17:53 ` Tom Herbert via iovisor-dev
[not found] ` <CALx6S36BADKByJAYQLMXBx1NEDaqn6fdqsCk-OdgNo5vgHrO1Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-07-07 21:33 ` John Fastabend via iovisor-dev
2016-07-08 2:22 ` Alexei Starovoitov via iovisor-dev
[not found] ` <20160708022210.GA12244-+o4/htvd0TDFYCXBM6kdu7fOX0fSgVTm@public.gmane.org>
2016-07-08 4:05 ` John Fastabend via iovisor-dev
[not found] ` <577F2689.4010602-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2016-07-08 4:28 ` Alexei Starovoitov via iovisor-dev
2016-07-08 13:44 ` Jakub Kicinski via iovisor-dev
2016-07-08 15:19 ` Jesper Dangaard Brouer via iovisor-dev
[not found] ` <20160708171943.0e1ce8d7-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-07-08 16:07 ` Jakub Kicinski via iovisor-dev
2016-07-08 16:45 ` John Fastabend via iovisor-dev
[not found] ` <577FD8A5.8020700-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2016-07-08 17:51 ` Jakub Kicinski via iovisor-dev
2016-07-09 11:27 ` Jesper Dangaard Brouer via iovisor-dev
2016-07-12 2:24 ` Alexei Starovoitov
[not found] ` <20160712022423.GA47757-+o4/htvd0TDFYCXBM6kdu7fOX0fSgVTm@public.gmane.org>
2016-07-12 19:13 ` John Fastabend via iovisor-dev
[not found] ` <5785413D.4050901-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2016-07-12 19:49 ` Jakub Kicinski via iovisor-dev
2016-07-12 20:32 ` Jesper Dangaard Brouer via iovisor-dev
[not found] ` <20160712223231.202cd122-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-07-26 13:31 ` Thomas Monjalon via iovisor-dev
2016-07-26 16:08 ` [iovisor-dev] " Tom Herbert
[not found] ` <CALx6S35XjCsG5EmiYBpbGk9NckQbe4VbNSGLqV7h+d16PgNGKg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-07-26 17:53 ` John Fastabend via iovisor-dev
[not found] ` <5797A381.90406-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2016-07-26 18:42 ` Jesper Dangaard Brouer via iovisor-dev
2016-07-26 18:58 ` Tom Herbert via iovisor-dev
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160707124245.6d95635a@redhat.com \
--to=iovisor-dev-9jonkmmolfhee9la1f8ukti2o/jbrioy@public.gmane.org \
--cc=as754m-60p5jsuXm+c@public.gmane.org \
--cc=brouer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=ecree-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org \
--cc=jakub.kicinski-wFxRvT7yatFl57MIdRCFDg@public.gmane.org \
--cc=john.r.fastabend-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
--cc=ranas-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
--cc=simon.horman-wFxRvT7yatFl57MIdRCFDg@public.gmane.org \
--cc=tariqt-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).