* Re: [ANN] netdev call - Jul 29th
2025-07-28 15:53 [ANN] netdev call - Jul 29th Jakub Kicinski
@ 2025-07-28 20:45 ` Daniel Golle
2025-07-28 20:48 ` Sean Anderson
2025-07-29 18:23 ` [ANN] netdev call - Jul 29th (meeting notes + TSC) Jakub Kicinski
1 sibling, 1 reply; 4+ messages in thread
From: Daniel Golle @ 2025-07-28 20:45 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Sean Anderson, Simon Horman, Andrew Lunn, David S . Miller,
Eric Dumazet, Paolo Abeni, Maxime Chevallier, Russell King,
Vineeth Karumanchi, Heiner Kallweit, Kory Maincent,
Christian Marangi, Lei Wei, Michal Simek, Radhey Shyam Pandey,
Robert Hancock, John Crispin, Felix Fietkau, Robert Marko,
Frank Wunderlich, linux-kernel, netdev-driver-reviewers
Hi!
On Mon, Jul 28, 2025 at 08:53:56AM -0700, Jakub Kicinski wrote:
> Hi!
>
> The bi-weekly call is scheduled for tomorrow at 8:30 am (PT) /
> 5:30 pm (~EU), at https://bbb.lwn.net/rooms/ldm-chf-zxx-we7/join
>
> We have one agenda topic - Arkadiusz Kubalewski says:
>
> Wanted to meet and discuss the need for clock id on microchip driver
> provided by Ivan, I hope we could have a meeting to brainstorm some
> ideas and decide next steps on how to deal with such cases.
>
> I can also provide a quick update on the netdev foundation activities.
Christian Marangi and me would like to join you this week in order to
discuss how to proceed with the groundworks for standalone PCS drivers.
See also:
https://lore.kernel.org/all/20250709135216.GA721198@horms.kernel.org/
I know this is kinda short notice, but it would be great to add this to
the agenda and decide in which direction we should go.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [ANN] netdev call - Jul 29th (meeting notes + TSC)
2025-07-28 15:53 [ANN] netdev call - Jul 29th Jakub Kicinski
2025-07-28 20:45 ` Daniel Golle
@ 2025-07-29 18:23 ` Jakub Kicinski
1 sibling, 0 replies; 4+ messages in thread
From: Jakub Kicinski @ 2025-07-29 18:23 UTC (permalink / raw)
To: netdev, netdev-driver-reviewers
On Mon, 28 Jul 2025 08:53:56 -0700 Jakub Kicinski wrote:
> The bi-weekly call is scheduled for tomorrow at 8:30 am (PT) /
> 5:30 pm (~EU), at https://bbb.lwn.net/rooms/ldm-chf-zxx-we7/join
>
> We have one agenda topic - Arkadiusz Kubalewski says:
>
> Wanted to meet and discuss the need for clock id on microchip driver
> provided by Ivan, I hope we could have a meeting to brainstorm some
> ideas and decide next steps on how to deal with such cases.
>
> I can also provide a quick update on the netdev foundation activities.
Agenda items for next meeting (Aug 12th):
- standalone phylink PCS drivers: as part of phylink? or as standalone
subsystem? or not at all?
Meeting notes
DPLL
- Arkadiusz: The standalone DPLL driver from Ivan doesn't have a
unique clock ID, which is supposed to be the handle for the whole DPLL
devices. The clock ID works for time appliances. The standalone DPLL
device doesn't have a network card, so no MAC address, or ID.
- Ivan: the device can be connected to multiple controllers, and drive
multiple clocks. There’s no ID on the device, no serial even. We use
a random number for now, devlink can be used to change it.
- Andrew: it’s an i2c device? (yes, i2c or SPI) Can we use the address?
- Ivan: yes, but the i2c address is static, so if there are multiple
buses or a mux the address will be the same for all devices in the
system.
- Andrew: we can include the full bus address (with the bus).
- Jakub: can we define the relationship to the netdev port in the DT?
- Ivan: we could extend the DT note which contains the pins to include the MAC phandle.
- Andrew: you can use the DSA definition as the sample to steal ideas from.
- Ivan: the DPLL can drive multiple NICs, which device to pick for the clock ID.
- Andrew: a diagram will be very helpful; Jakub: prefer using an explicit annotation.
Next step: send an RFC to the DT list..
We also had a netdev foundation TSC meeting today, here are the notes:
CI migration / HW lab:
- Simon: we sent the RFQ out, colo is 10-15k, managed solution is
double that at least, last minute we got an offer to host the lab
for free.
- Andrew: having a commitment for the free offer for at least a year
would be good
- Johannes: the company is considering joining the Governing Board
too, so they should stick around
- Eric: what is the location?
- Simon: Chicago and Amsterdam, none of them have our people local
The HW lab project ($70k regardless of colo selection) is approved for
funding.
Small budget items (<$1k/yr total - Google Drive, Plausible subscription):
- Pay for a Google Drive for TSC docs and Plausible for web analytics
Approved
New members:
- The company trying to join made us realize we don't have a process defined.
- There may be contentious applications in future, do we want some control?
- We have a voting system for electing TSC members, but not to Governing Board.
Revisit in next meeting.
Rate / rank other projects on GH:
- We need a shepherd per project first - assign tickets in github
- TSC members to "like" projects they want to see done
- Johannes to add more labels/milestones to mark what's already in progress
Revisit in next meeting.
How do we engage with the Governing Board:
- Simon: the GB has a chair elected. Whenever we have a proposal we
contact the chair and ask for them to convene.
- Willem: does the board has a regular meeting?
- Simon: the cadence is not decided, the default is every 3mo. We can
ask to expedite, or even to meet monthly.
- Willem: if we’re trying to get them to convene we should ask about
the membership vote right now. Avoid delays.
- Ask LF if membership approval is allowed. Ask for simple majority vote.
- Simon: LF will be involved in the email exchange with GB, we can go
direct to GB. Probably want a ⅔ majority, simple majority is too
weak, veto is too strong.
- Jakub: what about renewals. How about also optional ⅔ vote to block renewals,
if no vote requested renewals auto-approved.
Ask the board for: HW lab, small items, and the membership vote amendment.
^ permalink raw reply [flat|nested] 4+ messages in thread