Re: [PATCH] cmd: fdt: Add support for reading stringlist property values

[Marek Vasut <marex@denx.de>]

On 7/13/22 17:28, Simon Glass wrote:
> Hi Marek,
> 
> On Tue, 12 Jul 2022 at 05:47, Marek Vasut <marex@denx.de> wrote:
>>
>> On 7/12/22 12:58, Simon Glass wrote:
>>> Hi Marek,
>>>
>>> On Fri, 8 Jul 2022 at 15:50, Marek Vasut <marex@denx.de> wrote:
>>>>
>>>> The fdt command currently handles stringlists as strings in 'fdt get value'
>>>> subcommand. Since strings in FDT stringlists are separated by '\0', only
>>>> the first value gets inserted into the environment variable passed to the
>>>> 'fdt get value' command.
>>>>
>>>> Example, consider the following DT snippet:
>>>>
>>>> / { compatible = "foo", "bar" };
>>>>
>>>> The following command only reports the first string in stringlist:
>>>> => fdt get value var / compatible ; print var
>>>> foo
>>>>
>>>> It is not possible to assign list of null-terminated strings into U-Boot
>>>> environment variable. Add optional 'index' parameter to the subcommand
>>>> 'fdt get value <var> <path> <prop> [<index>]' which lets user specify which
>>>> string within the stringlist should be assigned into the 'var' variable.
>>>> The default value of 'index' is 0 in case it is not present. This way the
>>>> 'fdt' command API does not change and existing scripts are not broken.
>>>>
>>>> The following command now reports the Nth string in stringlist, counting
>>>> from zero:
>>>> => fdt get value var / compatible 1 ; print var
>>>> bar
>>>>
>>>> Signed-off-by: Marek Vasut <marex@denx.de>
>>>> Cc: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
>>>> Cc: Simon Glass <sjg@chromium.org>
>>>> Cc: Tom Rini <trini@konsulko.com>
>>>> ---
>>>>    cmd/fdt.c | 34 ++++++++++++++++++++++++++++------
>>>>    1 file changed, 28 insertions(+), 6 deletions(-)
>>>
>>> Can you please add docs and a test?
>>
>> I was expecting this kind of question ... there is no test for the FDT
>> command.
>>
>>> I am happy to do a starting point for a test if you like.
>>
>> You can implement the test for this while at it.
> 
> I sent a starting point for you, both docs and test.

Thank you.

Do I understand your remark correctly that you're blocking this 
particular patch from going in until there is a test ?