DPDK patches and discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: "Denis  Davidoglu" <denis.davidoglu@b-ulltech.com>
Cc: "Thomas Monjalon" <thomas@monjalon.net>,
	"dev@dpdk.org" <dev@dpdk.org>,
	"Volkan Atlı" <volkan.atli@b-ulltech.com>
Subject: Re: [PATCH] net: support Arista L2 headers
Date: Fri, 7 Feb 2025 09:54:06 -0800	[thread overview]
Message-ID: <20250207095406.2512f18d@hermes.local> (raw)
In-Reply-To: <GV1P190MB20415A820BF2090AE8151B2BA1F12@GV1P190MB2041.EURP190.PROD.OUTLOOK.COM>

On Fri, 7 Feb 2025 09:07:57 +0000
"Denis  Davidoglu" <denis.davidoglu@b-ulltech.com> wrote:

> From: Thomas Monjalon <thomas@monjalon.net>
> > I'm not familiar with these headers.
> > Are they standardized?
> > Is there a RFC?  
> 
> Arista Vendor Specific Protocol is not publicly disclosed yet. However, Wireshark already includes a dissector implementetation and its source code can serve for now as a reference:
> https://gitlab.com/wireshark/wireshark/-/blob/master/epan/dissectors/packet-avsp.c
> 
> > I'm not sure about adding vendor specific protocols.
> > What should be the policy here?  
> 
> Given the large number of registered EtherType™ assignments related to IEEE 802.3™ (ISO/IEC 8802-3) standard, supporting vendor-specific protocols should be considered. Though probably not in the way I organized it, mixing RFC-standardized and proprietary protocols in the same files. ./lib/net can have a new subdirectory for keeping them separate. 
> 
> Arista timestamp is particularly valuable for stock exchanges due to efficiency. The timestamp extension is already in use at German Eurex Exchange and it needs to undergo PoC trials at Borsa İstanbul, which utilizes Nasdaq infrastructure. Some parts of it rely on DPDK and this patch will be especially useful.


The rte_net_ptype() is intended to provide a software equivalent for the packet type
matching done in many smart NIC's. It is not meant to be a general purpose L2 packet
parser.

If a user needs to do this kind of packet processing, it belongs in the application
and all the data and metadata is exposed to do this.

      parent reply	other threads:[~2025-02-07 17:54 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-04 15:41 Denis  Davidoglu
2025-02-04 16:02 ` Thomas Monjalon
     [not found]   ` <GV1P190MB20415A820BF2090AE8151B2BA1F12@GV1P190MB2041.EURP190.PROD.OUTLOOK.COM>
2025-02-07 17:54     ` Stephen Hemminger [this message]

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=20250207095406.2512f18d@hermes.local \
    --to=stephen@networkplumber.org \
    --cc=denis.davidoglu@b-ulltech.com \
    --cc=dev@dpdk.org \
    --cc=thomas@monjalon.net \
    --cc=volkan.atli@b-ulltech.com \
    /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).