All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 0/1] hw/arm/aspeed: Add Fuji machine type
@ 2021-09-01 22:59 pdel
  2021-09-01 22:59 ` [PATCH 1/1] " pdel
  2021-09-02  8:12 ` [PATCH 0/1] " Cédric Le Goater
  0 siblings, 2 replies; 8+ messages in thread
From: pdel @ 2021-09-01 22:59 UTC (permalink / raw)
  Cc: clg, joel, qemu-devel, qemu-arm, patrick, f4bug, peter.maydell,
	andrew, Peter Delevoryas

From: Peter Delevoryas <pdel@fb.com>

Adding a new Aspeed AST2600 machine type, uses MAC3 for ethernet1 and
UART1 for the serial console, which is different than the existing
ast2600-evb. Otherwise though, my usage so far hasn't required a
different set of hardware strap registers or anything, so I just
inherited the rest of the configuration from the ast2600-evb. If
preferred, I can eliminate the inheritance and just declare Fuji
completely independently.

I included more info in the commit message, but just for convenience,
here's the DTS link and a link to an image I'm providing for the
acceptance test.

https://github.com/facebook/openbmc-uboot/blob/openbmc/helium/v2019.04/arch/arm/dts/aspeed-bmc-facebook-fuji.dts
https://github.com/peterdelevoryas/openbmc/releases/download/fuji-v0.1-alpha/fuji.mtd

I'm not sure exactly what the requirements are on the image url
provided, or the requirements of the image itself.

The existing OpenBMC acceptance tests mostly use images from the Linux
Foundation repository's releases, e.g.

https://github.com/openbmc/openbmc/releases/download/2.9.0/obmc-phosphor-image-romulus.static.mtd

Although, I do see one image from a regular user's repo:

https://github.com/hskinnemoen/openbmc/releases/download/20200711-gsj-qemu-0/obmc-phosphor-image-gsj.static.mtd.gz

So maybe it's not that unreasonable? I also might be able to organize a
release link on the official Facebook OpenBMC Github.

As far as the actual requirements of the acceptance test, I didn't use
the existing "do_test_arm_aspeed" because Fuji was sufficiently
different (different U-Boot version, kernel address, CPU, SPI control
register, doesn't use systemd), but I could try unifying them if there's
interest or suggestions on how to do so.

By the way, Fuji takes a long time to boot on my Macbook:

(19/38) tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_ast2600_fuji_openbmc: PASS (85.13 s)

The next longest boot time is <20 seconds:

17/38) tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_ast2500_romulus_openbmc_v2_9_0: PASS (17.46 s)

Not sure if that's a problem for the CIT infra time or not.

Thanks,
Peter

Peter Delevoryas (1):
  hw/arm/aspeed: Add Fuji machine type

 hw/arm/aspeed.c                        | 14 ++++++++++++++
 tests/acceptance/boot_linux_console.py | 25 +++++++++++++++++++++++++
 2 files changed, 39 insertions(+)

-- 
2.30.2



^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2021-09-02 16:47 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-09-01 22:59 [PATCH 0/1] hw/arm/aspeed: Add Fuji machine type pdel
2021-09-01 22:59 ` [PATCH 1/1] " pdel
2021-09-02  8:12 ` [PATCH 0/1] " Cédric Le Goater
2021-09-02 15:12   ` Peter Delevoryas
2021-09-02 15:46     ` Patrick Williams
2021-09-02 16:19       ` Cédric Le Goater
2021-09-02 16:23         ` Patrick Williams
2021-09-02 16:27           ` Cédric Le Goater

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.