DPDK patches and discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: dev@dpdk.org
Subject: Minutes of Technical Board Meeting, 2021-12-01
Date: Thu, 2 Dec 2021 11:25:06 -0800	[thread overview]
Message-ID: <20211202112506.68acaa1a@hermes.local> (raw)


Members Attending
=================
- Aaron
- Bruce
- David
- Ferruh
- Hemant
- Honnappa
- Jerin
- Konstantin
- Maxime
- Olivier
- Stephen (Chair)
- Thomas

NOTE: The technical board meetings every second Wednesday at
https://meet.jit.si/DPDK at 3 pm UTC.
Meetings are public, and DPDK community members are welcome to attend.

NOTE: Next meeting will be on Wednesday 2021-12-01 @3pm UTC, and will
be chaired by Thomas.

1. Governing board
==================
Jerin's slides for governing board were reviewed with no new feedback.

2. Review the default maximum number of cores
=============================================
Continued discussion about the default number of cores from last meeting.
The proposed patch only addressed AMD EPYC platform but the issue is more
general since today's limit of 128 cores will be on many high end
two socket x86 system. A suggestion is to increase this to default
maximum 256 cores for all x86 platforms; to avoid any out of the
box user experience issues.

Users can remap lcores today from the command line to run with up
to 128 cores used (out of a much larger set).  But this requires users
explicitly configuring the command line. Suggestion was made to
have an new command line flag to automatically create the mapping;
based on the lcore command line.

Something like the following
  dpdk-XXX --cpu-offset 128 --lcores 0-2
would be same as:
  dpdk-XXX -lcores 0@128,1,@129,2@130

Olivier will investigate

3. Review of outdated examples.
===============================
Discussion of future status of examples that are not being maintained:
  - ioat was removed in 21.11.
  - ntb should be moved to different directory to indicate that
    it is vendor specific.
  - vmdq examples might be candidate for deprecation.
    Bruce to check with Intel
  - performance-thread should be deprecated and removed.
  - flow filtering and flow classify examples overlap.

4. LTS release maintenance
==========================
The current LTS release and maintenance schedule on web site needs update.
The plan is for LTS releases to come immediately after the main DPDK
release. Now that main DPDK releases are three times a year, the LTS
release schedule needs to be updated as well.

The other issues is extending the LTS releases to 3 years instead
of just 2. General consensus was that this is good idea but will
require additional resources from LTS maintainers and vendor validation
teams. Will wait for confirmation before committing to longer cycle.

For the 21.11 LTS, Kevin was approved by Techboard as LTS maintainer.

5. KNI deprecation
==================
Deprecation was previously discussed by TB, discused in more detail.
Ferruh proposed time line is:
 - 22.11 move to kmod repository
   KNI library and example won't be built by default
 - 23.11 final removal
Approved, Ferruh to follow up with more patches.

                 reply	other threads:[~2021-12-02 19:25 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20211202112506.68acaa1a@hermes.local \
    --to=stephen@networkplumber.org \
    --cc=dev@dpdk.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).