[U-Boot] [PATCHv5 4/4] Add support the Avionic Design Meerkat COM and Kein Baseboard

[Stephen Warren]

On 09/06/2016 12:14 PM, Julian Scheel wrote:
> On 06.09.16 19:15, Stephen Warren wrote:
>> On 09/05/2016 07:29 AM, Julian Scheel wrote:
>>> Add support for platforms based on the Meerkat COM module. Includes
>>> support
>>> for the minimal reference platform called Kein Baseboard, which in
>>> fact is
>>> sufficient to run most existing Meerkat carriers.

>>> diff --git a/board/avionic-design/common/pinmux-config-meerkat.h
>>> b/board/avionic-design/common/pinmux-config-meerkat.h
>>
>>> +/*
>>> + * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
>>> + * Copyright (c) 2015, Avionic Design GmbH
>>> + *
>>> + * SPDX-License-Identifier: GPL-2.0+
>>> + */
>>
>> Can you please also add support for this board to tegra-pinmux-scripts,
>> so that anyone can generate this file? That will also allow you to
>> re-generate the file using the latest version of tegra-pinmux-scripts
>> which will add (a) the missing "this file is auto-generated" notice, (b)
>> the GPIO initialization table, (c) the MIPI padctl initialization table.
>
> I'll discuss this internally. We have never used those scripts at all,
> but did handwrite this code, based from Jetson code.

Hmm. Do you use the Excel pinmux spreadsheet that NVIDIA syseng 
supplies? That performs a variety of error-checking, so I'd certainly 
advocate using that. If you do, then supporting tegra-pinmux-script is a 
simple matter of saving the spreadsheet as CSV and running the import 
script in tegra-pinmux-scripts. If not, I worry that the spreadsheet's 
error-checking rules haven't been validated.

>>> +CONFIG_USB_STORAGE=y
>>
>> For Jetson TK1, this is defined in include/configs/jetson-tk1.h. I'd
>> expect the two board configs to work the same way.
>
> Actually we have moved this to defconfig, as we decided that all options
> that are Kconfig options by now shall be selectable via Kconfig by the
> user.

Who's "we" here; has this conversion happened in mainline? If so, it 
looks like the author of the conversion commit forgot to convert all the 
existing configuration headers and defconfigs, so someone should go back 
and fix up that commit. If the conversion hasn't happened in mainline, I 
imagine it would be welcome so feel free to send it, but if this series 
gets submitted before the conversion is submitted, it can't assume the 
conversion has happened.