[U-Boot] [PATCH] compulab: eeprom: add default eeprom bus

[Nikita Kiryanov]

Gentle ping.

On 18/09/14 23:35, Albert ARIBAUD wrote:
> Hi Nikita,
>
> On Thu, 18 Sep 2014 16:21:07 +0300, Nikita Kiryanov
> <nikita at compulab.co.il> wrote:
>
>> Hi ALbert,
>>
>> On 18/09/14 15:26, Albert ARIBAUD wrote:
>>> Hi Igor,
>>>
>>> On Wed, 17 Sep 2014 17:18:49 +0300, Igor Grinberg
>>> <grinberg at compulab.co.il> wrote:
>>>
>>>> On 09/17/14 15:59, Nikita Kiryanov wrote:
>>>>> Add default eeprom bus setting.
>>>>> This addresses the trimslice compile error that was introduced
>>>>> with the addition of this setting.
>>>>>
>>>>> Cc: Albert ARIBAUD <albert.u.boot at aribaud.net>
>>>>> Cc: Igor Grinberg <grinberg at compulab.co.il>
>>>>> Signed-off-by: Nikita Kiryanov <nikita at compulab.co.il>
>>>>
>>>> Acked-by: Igor Grinberg <grinberg at compulab.co.il>
>>>>
>>>> Albert, this should be a temporary fix to not break the trimslice support.
>>>> We intend to rework this while switching Kconfig for all boards.
>>>>
>>>> Thanks!
>>>>
>>>>> ---
>>>>>    board/compulab/common/eeprom.c | 4 ++++
>>>>>    1 file changed, 4 insertions(+)
>>>>>
>>>>> diff --git a/board/compulab/common/eeprom.c b/board/compulab/common/eeprom.c
>>>>> index 85442cd..2df3ada 100644
>>>>> --- a/board/compulab/common/eeprom.c
>>>>> +++ b/board/compulab/common/eeprom.c
>>>>> @@ -15,6 +15,10 @@
>>>>>    # define CONFIG_SYS_I2C_EEPROM_ADDR_LEN	1
>>>>>    #endif
>>>>>
>>>>> +#ifndef CONFIG_SYS_I2C_EEPROM_BUS
>>>>> +#define CONFIG_SYS_I2C_EEPROM_BUS	0
>>>>> +#endif
>>>>> +
>>>>>    #define EEPROM_LAYOUT_VER_OFFSET	44
>>>>>    #define BOARD_SERIAL_OFFSET		20
>>>>>    #define BOARD_SERIAL_OFFSET_LEGACY	8
>>>>>
>>>
>>> a 'git grep CONFIG_SYS_I2C_EEPROM_BUS' shows that all values for
>>> CONFIG_SYS_I2C_EEPROM_BUS are set in include/configs/*.h.
>>>
>>> Therefore I would prefer that the temporary fix be in
>>> include/configs/trimslice.h header file too.
>>>
>>
>> Given that trimslice does not use the eeprom code, I feel that
>> it's cleaner to not populate its config file with irrelevant
>> defines. It's also consistent with the way CONFIG_SYS_I2C_EEPROM_ADDR
>> and CONFIG_SYS_I2C_EEPROM_ADDR_LEN are handled.
>
> Makes sense enough.
>
> Amicalement,
>

-- 
Regards,
Nikita Kiryanov