XDP hardware hints discussion mail archive
 help / color / mirror / Atom feed
From: Jamal Hadi Salim <jhs@mojatatu.com>
To: "Toke Høiland-Jørgensen" <toke@redhat.com>,
	"John Fastabend" <john.fastabend@gmail.com>,
	"Jesper Dangaard Brouer" <jbrouer@redhat.com>,
	"Karlsson, Magnus" <magnus.karlsson@intel.com>,
	"Desouza, Ederson" <ederson.desouza@intel.com>
Cc: brouer@redhat.com,
	"xdp-hints@xdp-project.net" <xdp-hints@xdp-project.net>,
	Eelco Chaudron <echaudro@redhat.com>,
	Andrii Nakryiko <andrii@kernel.org>,
	"Fijalkowski, Maciej" <maciej.fijalkowski@intel.com>,
	"Burakov, Anatoly" <anatoly.burakov@intel.com>,
	tom Herbert <tom@sipanda.io>
Subject: [xdp-hints] Basic/Dumb question WAS(Re: Re: XDP-hints via local BTF info
Date: Mon, 22 Nov 2021 07:45:46 -0500	[thread overview]
Message-ID: <bff218f8-2a98-f235-67fe-f1797eba4704@mojatatu.com> (raw)
In-Reply-To: <871r3cdwng.fsf@toke.dk>

And it goes something like this:

Why does the metadata have to go in the DMA descriptors?
Our experience with the XDP metadata is: you start accessing
that there is a performance penalty (extra cache miss(es)).

Why is the metadata not encapped as part of the data? We
dont have MTU issues on receive since that is entirely a local matter;
meaning the hardware can expand the packet as much as it wants within
the boundaries of alloced DMA buffer space and XDP and any other
subsystem (TC for example) can take advantage of the metadata.
Then extracting metadata becomes a parser issue and you get rid of the
extra cache misses.
Another advantage is: If you make the metadata part the packet data
(and appropriately handle MTU issues) you can pass it around as well
to VMs, containers, other machines,etc

We have done something along these lines using TC IFE, see:
https://legacy.netdevconf.info/0.1/sessions/9.html

The BTF aspect for discoverability is still of value. I am wondering
if we can extend the IFE action to suck in some BTF info so we dont
have to create new kernel code...

cheers,
jamal

  reply	other threads:[~2021-11-22 12:45 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-17 17:22 [xdp-hints] " Jesper Dangaard Brouer
2021-11-17 20:07 ` [xdp-hints] " Karlsson, Magnus
2021-11-17 22:48   ` Toke Høiland-Jørgensen
2021-11-18  8:05     ` Karlsson, Magnus
2021-11-18 14:30       ` Jesper Dangaard Brouer
2021-11-18 14:57         ` Karlsson, Magnus
2021-11-18 15:18         ` John Fastabend
2021-11-19 14:53           ` Toke Høiland-Jørgensen
2021-11-22 12:45             ` Jamal Hadi Salim [this message]
2021-11-22 13:59               ` [xdp-hints] Re: Basic/Dumb question WAS(Re: " Toke Høiland-Jørgensen
2021-11-22 15:31                 ` Tom Herbert
2021-11-22 18:25                   ` Toke Høiland-Jørgensen
2021-11-22 12:57             ` [xdp-hints] " Alexander Lobakin
2021-11-24 11:54               ` Jesper Dangaard Brouer
2021-11-25 20:04                 ` Alexander Lobakin

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

  List information: https://lists.xdp-project.net/postorius/lists/xdp-hints.xdp-project.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bff218f8-2a98-f235-67fe-f1797eba4704@mojatatu.com \
    --to=jhs@mojatatu.com \
    --cc=anatoly.burakov@intel.com \
    --cc=andrii@kernel.org \
    --cc=brouer@redhat.com \
    --cc=echaudro@redhat.com \
    --cc=ederson.desouza@intel.com \
    --cc=jbrouer@redhat.com \
    --cc=john.fastabend@gmail.com \
    --cc=maciej.fijalkowski@intel.com \
    --cc=magnus.karlsson@intel.com \
    --cc=toke@redhat.com \
    --cc=tom@sipanda.io \
    --cc=xdp-hints@xdp-project.net \
    /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