Re: [PATCH 7/7] meta: Add barebox bootloader support

[Jan Kiszka <jan.kiszka@siemens.com>]

On 10.02.22 09:45, Anton Mikanovich wrote:
> 10.02.2022 11:05, Jan Kiszka wrote:
>> On 10.02.22 08:58, Anton Mikanovich wrote:
>>>>> diff --git a/meta/recipes-bsp/barebox/barebox_2021.04.0-phy5.bb
>>>>> b/meta/recipes-bsp/barebox/barebox_2021.04.0-phy5.bb
>>>>> new file mode 100644
>>>>> index 0000000..ed8842c
>>>>> --- /dev/null
>>>>> +++ b/meta/recipes-bsp/barebox/barebox_2021.04.0-phy5.bb
>>>>> @@ -0,0 +1,13 @@
>>>>> +# This software is a part of ISAR.
>>>>> +# Copyright (C) 2022 ilbers GmbH
>>>>> +
>>>>> +inherit dpkg
>>>>> +
>>>>> +require barebox.inc
>>>>> +
>>>>> +SRC_URI +=
>>>>> "https://git.phytec.de/barebox/snapshot/barebox-2021.04.0-phy5.tar.bz2
>>>>> \
>>>> Do we really want to or have to promote downstream BSP anti-patterns
>>>> here? Upstream barebox does not work?
>>>>
>>>> I'm mentioning this as I worked hard to avoid that bad practice for the
>>>> TF-A, OP-TEE, and U-Boot examples we provide.
>>> Upstream barebox will probably need some patching.
>>> But it should be not so hard to adapt, so I can try.
>> If upstream barebox is incomplete then why use it? Upstream U-Boot works
>> on that board (that's why we used it so far).
>>
>> Jan
>>
> Upstream barebox actually boots ok, but when using env generated by
> phytec yocto it tries to call some extended commands. So there can be
> two ways: to patch barebox or to change environment.
> 

OK, so what was missing in the integration so far? What is that
yocto-generated env?

Jan

-- 
Siemens AG, Technology
Competence Center Embedded Linux