state: eMMC boot0-partitions backend for Linux

state: eMMC boot0-partitions backend for Linux

[Christoph Fritz]

Hi,

 I would like to use emmc-hardware-partition boot1 as a state backend,
this has the charm of keeping the big hw-partition clean while using SLC
for e.g. serial numbers.

Within barebox using an emmc-hardware-partition boot1 as a state backend
just works!

But trying to hand over this backend to Linux fails right before
starting the kernel:

  WARNING: Failed to fixup node in of_state_fixup+0x1/0x288: No such device

I guess dt-utils also has no support for using /dev/mmcblk1boot0 as a
backend (by default Linux doesn't allow to write to boot0/1) ?

What would be the best way to add this feature?

Any ideas?

 -- Christoph


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

Re: state: eMMC boot0-partitions backend for Linux

[Ahmad Fatoum]

Hello Christoph,

On 11/3/19 4:45 PM, Christoph Fritz wrote:
> Hi,
> 
>  I would like to use emmc-hardware-partition boot1 as a state backend,
> this has the charm of keeping the big hw-partition clean while using SLC
> for e.g. serial numbers.
> Within barebox using an emmc-hardware-partition boot1 as a state backend
> just works!
> 
> But trying to hand over this backend to Linux fails right before
> starting the kernel:
> 
>   WARNING: Failed to fixup node in of_state_fixup+0x1/0x288: No such device

I ran into this before, but with the barebox environment.
I 'solved' it by moving the environment out of the boot partition..
See d5285342d87 ("mci: skip of_partitions_register_fixup for boot partitions").

> I guess dt-utils also has no support for using /dev/mmcblk1boot0 as a
> backend (by default Linux doesn't allow to write to boot0/1) ?

Yes.

> What would be the best way to add this feature?
> 
> Any ideas?

How about:
- formalize the eMMC boot partitions binding with
  a compatible = "barebox,emmc-bootpart" and change barebox to fix it up
  into the kernel's device tree.
- have barebox_state handle this properly. You can check out how RAUC
  (LGPL-2.1) does it.

Cheers
Ahmad

> 
>  -- Christoph
> 
> 
> _______________________________________________
> barebox mailing list
> barebox@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/barebox
> 


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | https://www.pengutronix.de/ |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox