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 2023-10-04
Date: Mon, 9 Oct 2023 16:33:31 -0700	[thread overview]
Message-ID: <20231009163331.79e523f6@hermes.local> (raw)


Members Attending
-----------------
Aaron Conole
Bruce Richardson
Hemant Agrawal
Honappa Nagarahalli
Jerin Jacob
Kevin Traynor
Konstantin Ananyev
Maxime Coquelin
Stephen Hemminger
Thomas Monjalon

NOTE: The technical board meetings are on every second Wednesday at 3 pm UTC.
Meetings are public, and DPDK community members are welcome to attend.
This meeting was special, in Dublin on September 11th before the summit.
Agenda and minutes can be found at http://core.dpdk.org/techboard/minutes

Next meeting will be on Wednesday 2023-Oct-18 @ 3pm UTC, and will
be chaired by Thomas.

Agenda Items
============

Bugzilla
--------
Ajit has said he will continue maintaining Bugzilla.
What additional maintenance is necessary?
The bug backlog has grown and it is not being used as much as it could be.
Would be good to have (automated?) status report to mailing list.
Need to make sure that all bugs are assigned to correct owners.

Techboard Minutes
-----------------
Last entry on DPDK techboard minutes is from May 17.
Need to fill in with later minutes.
There will be gap in summer because a couple meetings did not have quorum.

Power Management
----------------
There was a discussion of power management in DPDK.
Thomas raised the issue that we don't describe power management in the
web site and marketing which leads to assumptions that DPDK does not
do any power management.

Kevin said that recent OpenVswitch changes dramatically improved power
consumption. He is writing a blog describing details.

Many network equipment vendors use (or base their design) off of testpmd
application which has no power management integration.
There is a proposed patch which is not integrated to do some sleeps in testpmd.
   http://inbox.dpdk.org/dev/20230316151438.186241-2-aharivel@redhat.com/
   https://patchwork.dpdk.org/project/dpdk/list/?series=27422&state=%2A&archive=both
Still needs followup. Perhaps a more general and complete solution.

Power management in VM's is even more difficult. Guest really does not
have a good way to tell host desired power profile. Right now, the method
is to use interrupts which has long latency.
There has be some ongoing work on QEMU side:
   https://patchew.org/QEMU/20230616140830.316655-1-aharivel@redhat.com/

Power management also depends on work load and application.
The high impact changes are in the application; and there is not
a simple API for doing PM in application. There is a library for
power management in DPDK but it is not complete enough.

Upcoming Joint Meeting
----------------------
There is an upcoming joint meeting of Tech and Governing board
on 17th Oct @ 10am EDT, 7am PDT, 1400 UTC.


                 reply	other threads:[~2023-10-09 23:33 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=20231009163331.79e523f6@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).