From: Jamal Hadi Salim <jhs-jkUAjuhPggJWk0Htik3J/w@public•gmane.org>
To: Thomas Graf <tgraf-G/eBtMaohhA@public•gmane.org>
Cc: ryazanov.s.a-Re5JQEeQqe8AvxtiuMwx3w@public•gmane.org,
jasowang-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org,
John Fastabend
<john.r.fastabend-ral2JQCrhuEAvxtiuMwx3w@public•gmane.org>,
Neil.Jerram-QnUH15yq9NYqDJ6do+/SaQ@public•gmane.org,
edumazet-hpIqsD4AKlfQT0dZR+AlfA@public•gmane.org,
andy-QlMahl40kYEqcZcGjlUOXw@public•gmane.org,
dev-yBygre7rU0TnMu66kgdUjQ@public•gmane.org,
nbd-p3rKhJxN3npAfugRpC6u6w@public•gmane.org,
f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public•gmane.org,
ronye-VPRAkNaXOzVWk0Htik3J/w@public•gmane.org,
jeffrey.t.kirsher-ral2JQCrhuEAvxtiuMwx3w@public•gmane.org,
ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public•gmane.org,
ben-/+tVBieCtBitmTQ+vhA3Yw@public•gmane.org,
buytenh-OLH4Qvv75CYX/NnBR394Jw@public•gmane.org,
alexander.h.duyck-ral2JQCrhuEAvxtiuMwx3w@public•gmane.org,
Jiri Pirko <jiri-rHqAuBHg3fBzbRFIqnYvSA@public•gmane.org>,
simon.horman-wFxRvT7yatFl57MIdRCFDg@public•gmane.org,
roopa-qUQiAmfTcIp+XZJcv9eMoEEOCMrvLtNR@public•gmane.org,
aviadr-VPRAkNaXOzVWk0Htik3J/w@public•gmane.org,
nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public•gmane.org,
vyasevic-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org,
nhorman-2XuSBdqkA4R54TAoqtyWWQ@public•gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public•gmane.org,
stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ@public•gmane.org,
dborkman-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org,
ebiederm-aS9lmoZGLiVWk0Htik3J/w@public•gmane.org,
davem-fT/PcQaiUtIeIZ0/mPfg9Q@public•gmane.org
Subject: Re: [patch net-next v2 8/9] switchdev: introduce Netlink API
Date: Sat, 20 Sep 2014 08:35:52 -0400 [thread overview]
Message-ID: <541D74A8.6080501@mojatatu.com> (raw)
In-Reply-To: <20140920115140.GA3777-FZi0V3Vbi30CUdFEqe4BF2D2FQJk+8+b@public.gmane.org>
On 09/20/14 07:51, Thomas Graf wrote:
> I fail to see the connection. You can use switch vendor SDK no matter
> how we define the kernel APIs. They already exist and have been
> designed in a way to be completely indepenedent from the kernel.
>
> Are you referring to vendor specific decisions in user space in
> general? I believe that the whole point of swdev is to provide *that*
> level of abstraction so decisions can be made in a vendor neutral way.
>
I am not against the swdev idea. I think we have disagreements
for the general classification/action interface how that should look
like - but that is resolvable with correct interfaces.
The vendor neutral way *already exists* via current netlink
abstractions that existing tools use. When we need to add new
interfaces then we should.
> I don't think anybody is saying that. P4 is likely a reality soon. But
> we definitely want hardware offload in a BPF world even if the hardware
> can't do BPF yet.
>
I dont think we have contradictions. We are speaking past each other.
You implied that in the future OVS s/w path might be based on BPF.
I implied BPF itself could be offloaded and stands on its own merit
and should work if we have the correct interface. As an example,
I dont care about P4 or OVS - but i have no problem if they use
the common interfaces provided by Linux. i.e
If i want to build a little cpu running the BPF instruction set
and use that as my offload then that interface should work and if
it doesnt i should provide extensions.
> Not sure I understand. OVS would be a user of eBPF just like tracing,
> xt_BPF, socket filter, ...
>
Ok, we are on the same page then.
> As I said, this can be argued about. It would require to push a lot of
> context into the kernel though. The FDB offload is relatively trivial
> in comparison to the complexity OVS user space can handle. I can't think
> of any reasons why to complicate the kernel further with OVS specific
> knowledge as long as we can guarantee the vendor abstraction.
>
I disagree. OVS maybe complex in that sense (I am sorry i am making
an assumption based on what you are saying) but i dont think there is
any other kernel subsystem that has this challenge.
Note: i am pointing to fdb only because it carries the concept of "put
this in hardware and/or software". I agree the fdb maybe reasonably
simpler.
cheers,
jamal
next prev parent reply other threads:[~2014-09-20 12:35 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-19 13:49 [patch net-next v2 0/9] introduce rocker switch driver with hardware accelerated datapath api Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 1/9] net: rename netdev_phys_port_id to more generic name Jiri Pirko
[not found] ` <1411134590-4586-2-git-send-email-jiri-rHqAuBHg3fBzbRFIqnYvSA@public.gmane.org>
2014-09-19 13:54 ` Jeff Kirsher
2014-09-19 13:49 ` [patch net-next v2 3/9] rtnl: expose physical switch id for particular device Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 4/9] net-sysfs: " Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 5/9] net: introduce dummy switch Jiri Pirko
[not found] ` <1411134590-4586-6-git-send-email-jiri-rHqAuBHg3fBzbRFIqnYvSA@public.gmane.org>
2014-09-20 5:21 ` Florian Fainelli
2014-09-20 7:37 ` Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 6/9] switchdev: add basic support for flow matching and actions Jiri Pirko
2014-09-20 5:32 ` Florian Fainelli
2014-09-20 7:28 ` Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 7/9] switchdev: add swdev features Jiri Pirko
2014-09-19 13:49 ` [patch net-next v2 8/9] switchdev: introduce Netlink API Jiri Pirko
2014-09-19 15:25 ` Jamal Hadi Salim
2014-09-19 15:49 ` Jiri Pirko
2014-09-19 17:57 ` Jamal Hadi Salim
2014-09-19 22:12 ` John Fastabend
2014-09-19 22:18 ` Jamal Hadi Salim
2014-09-20 5:39 ` Florian Fainelli
2014-09-20 8:25 ` Jiri Pirko
2014-09-20 8:17 ` Jiri Pirko
2014-09-20 10:19 ` Jamal Hadi Salim
2014-09-20 11:01 ` Thomas Graf
2014-09-20 11:32 ` Jamal Hadi Salim
2014-09-20 11:51 ` Thomas Graf
[not found] ` <20140920115140.GA3777-FZi0V3Vbi30CUdFEqe4BF2D2FQJk+8+b@public.gmane.org>
2014-09-20 12:35 ` Jamal Hadi Salim [this message]
2014-09-22 7:53 ` Jiri Pirko
[not found] ` <20140922075337.GA1828-6KJVSR23iU488b5SBfVpbw@public.gmane.org>
2014-09-22 11:48 ` Jamal Hadi Salim
2014-09-20 5:36 ` Florian Fainelli
2014-09-20 8:14 ` Jiri Pirko
2014-09-20 10:53 ` Thomas Graf
2014-09-20 22:50 ` Alexei Starovoitov
2014-09-22 8:13 ` Thomas Graf
2014-09-22 15:10 ` Tom Herbert
2014-09-22 22:17 ` Thomas Graf
[not found] ` <20140922221727.GA4708-FZi0V3Vbi30CUdFEqe4BF2D2FQJk+8+b@public.gmane.org>
2014-09-22 22:40 ` Tom Herbert
2014-09-22 22:53 ` Thomas Graf
2014-09-22 23:07 ` Tom Herbert
2014-09-23 1:36 ` John Fastabend
2014-09-23 7:19 ` Thomas Graf
2014-09-23 11:09 ` Jamal Hadi Salim
[not found] ` <CA+mtBx9ZVQ5r5Hzy9-uEnk+iu+HKkOP4+VANC06Xf8VvTxktwQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-09-23 9:18 ` Thomas Graf
2014-09-23 1:54 ` Alexei Starovoitov
2014-09-23 2:16 ` Tom Herbert
2014-09-23 4:11 ` Andy Gospodarek
2014-09-23 10:11 ` Thomas Graf
2014-09-23 15:32 ` Or Gerlitz
2014-09-24 13:32 ` Thomas Graf
2014-09-26 20:03 ` Or Gerlitz
2014-09-26 21:02 ` Thomas Graf
2014-09-23 9:52 ` Thomas Graf
2014-09-20 3:41 ` Roopa Prabhu
2014-09-20 8:09 ` Jiri Pirko
2014-09-20 12:39 ` Roopa Prabhu
2014-09-20 8:10 ` Scott Feldman
2014-09-20 10:31 ` Jamal Hadi Salim
[not found] ` <DDC24110-C3F5-470F-B9BE-1D1792415D1E-qUQiAmfTcIp+XZJcv9eMoEEOCMrvLtNR@public.gmane.org>
2014-09-20 12:51 ` Roopa Prabhu
2014-09-20 17:21 ` Scott Feldman
2014-09-20 17:38 ` Jiri Pirko
2014-09-21 1:30 ` Roopa Prabhu
2014-09-19 13:49 ` [patch net-next v2 9/9] rocker: introduce rocker switch driver Jiri Pirko
[not found] ` <1411134590-4586-1-git-send-email-jiri-rHqAuBHg3fBzbRFIqnYvSA@public.gmane.org>
2014-09-19 13:49 ` [patch net-next v2 2/9] net: introduce generic switch devices support Jiri Pirko
2014-09-19 14:15 ` [patch net-next v2 0/9] introduce rocker switch driver with hardware accelerated datapath api David Laight
[not found] ` <063D6719AE5E284EB5DD2968C1650D6D17495CC6-VkEWCZq2GCInGFn1LkZF6NBPR1lH4CV8@public.gmane.org>
2014-09-19 14:20 ` Jiri Pirko
2014-09-20 5:37 ` Florian Fainelli
-- strict thread matches above, loose matches on Subject: below --
2014-09-23 3:43 [patch net-next v2 8/9] switchdev: introduce Netlink API Alexei Starovoitov
2014-09-23 20:57 ` Tom Herbert
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=541D74A8.6080501@mojatatu.com \
--to=jhs-jkuajuhpggjwk0htik3j/w@public$(echo .)gmane.org \
--cc=Neil.Jerram-QnUH15yq9NYqDJ6do+/SaQ@public$(echo .)gmane.org \
--cc=alexander.h.duyck-ral2JQCrhuEAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=andy-QlMahl40kYEqcZcGjlUOXw@public$(echo .)gmane.org \
--cc=aviadr-VPRAkNaXOzVWk0Htik3J/w@public$(echo .)gmane.org \
--cc=ben-/+tVBieCtBitmTQ+vhA3Yw@public$(echo .)gmane.org \
--cc=buytenh-OLH4Qvv75CYX/NnBR394Jw@public$(echo .)gmane.org \
--cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public$(echo .)gmane.org \
--cc=dborkman-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=dev-yBygre7rU0TnMu66kgdUjQ@public$(echo .)gmane.org \
--cc=ebiederm-aS9lmoZGLiVWk0Htik3J/w@public$(echo .)gmane.org \
--cc=edumazet-hpIqsD4AKlfQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public$(echo .)gmane.org \
--cc=jasowang-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=jeffrey.t.kirsher-ral2JQCrhuEAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=jiri-rHqAuBHg3fBzbRFIqnYvSA@public$(echo .)gmane.org \
--cc=john.r.fastabend-ral2JQCrhuEAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=nbd-p3rKhJxN3npAfugRpC6u6w@public$(echo .)gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=nhorman-2XuSBdqkA4R54TAoqtyWWQ@public$(echo .)gmane.org \
--cc=nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public$(echo .)gmane.org \
--cc=ronye-VPRAkNaXOzVWk0Htik3J/w@public$(echo .)gmane.org \
--cc=roopa-qUQiAmfTcIp+XZJcv9eMoEEOCMrvLtNR@public$(echo .)gmane.org \
--cc=ryazanov.s.a-Re5JQEeQqe8AvxtiuMwx3w@public$(echo .)gmane.org \
--cc=simon.horman-wFxRvT7yatFl57MIdRCFDg@public$(echo .)gmane.org \
--cc=stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ@public$(echo .)gmane.org \
--cc=tgraf-G/eBtMaohhA@public$(echo .)gmane.org \
--cc=vyasevic-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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