All of lore.kernel.org
 help / color / mirror / Atom feed
* Aspeed platforms: MAC address provisioning
@ 2022-05-06  6:04 Patrick Rudolph
  2022-05-10  1:28 ` Joel Stanley
  0 siblings, 1 reply; 2+ messages in thread
From: Patrick Rudolph @ 2022-05-06  6:04 UTC (permalink / raw)
  To: openbmc

Hi,
I'm looking for the generic "OpenBMC way" of provisioning the BMC MAC
address on platforms with an integrated NIC. In the past there has
been the u-boot "u-boot-initial-env" target to create an environment,
but it's no longer used on u-boot-common-aspeed-sdk_2019.04.

What's the default/common/preferred way of provisioning an unique MAC
address in production? Should it be stored in an I2C EEPROM or should
the u-boot env be created from the running system?

Regards,
Patrick Rudolph
B.Sc. Electrical Engineering
System Firmware Developer

9elements GmbH, Kortumstraße 19-21, 44787 Bochum, Germany
Email: patrick.rudolph@9elements.com
Phone:  +49 234 68 94 188

Sitz der Gesellschaft: Bochum
Handelsregister: Amtsgericht Bochum, HRB 17519
Geschäftsführung: Sebastian Deutsch, Eray Basar

Datenschutzhinweise nach Art. 13 DSGVO

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Aspeed platforms: MAC address provisioning
  2022-05-06  6:04 Aspeed platforms: MAC address provisioning Patrick Rudolph
@ 2022-05-10  1:28 ` Joel Stanley
  0 siblings, 0 replies; 2+ messages in thread
From: Joel Stanley @ 2022-05-10  1:28 UTC (permalink / raw)
  To: Patrick Rudolph; +Cc: OpenBMC Maillist

On Fri, 6 May 2022 at 06:04, Patrick Rudolph
<patrick.rudolph@9elements.com> wrote:
>
> Hi,
> I'm looking for the generic "OpenBMC way" of provisioning the BMC MAC
> address on platforms with an integrated NIC. In the past there has
> been the u-boot "u-boot-initial-env" target to create an environment,
> but it's no longer used on u-boot-common-aspeed-sdk_2019.04.

I wasn't aware of this. One option you have is to add this
functionality to the new SDK.

>
> What's the default/common/preferred way of provisioning an unique MAC
> address in production? Should it be stored in an I2C EEPROM or should
> the u-boot env be created from the running system?

Take a look at first-boot-set-mac.

AFAIK there's code that populates the Inventory with the MAC addresses
from the VPD.

There's another service, first-boot-set-mac, that reads the inventory
and configures the network service with the mac addresses. This
persists the mac address in the network configuration (eg
/etc/systemd/network/00-bmc-eth0.network) and optionally calls
fw_setenv to configure the u-boot environment:

 https://github.com/openbmc/phosphor-networkd/blob/master/src/ethernet_interface.cpp#L995

 https://github.com/openbmc/phosphor-networkd/blob/master/src/ethernet_interface.cpp#L1206

A design goal was to not have to read the I2C EEPROM and parse the VPD
in u-boot, hence doing it on first boot.

Cheers,

Joel

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2022-05-10  1:29 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-05-06  6:04 Aspeed platforms: MAC address provisioning Patrick Rudolph
2022-05-10  1:28 ` Joel Stanley

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.