[PATCH v3 4/5] ARM: shmobile: BOCK-W: add VIN and ML86V7667 support

[Sergei Shtylyov]

Hello.

On 01-05-2013 13:09, Simon Horman wrote:

>> On 04/23/2013 09:32 PM, Sergei Shtylyov wrote:

>>> From: Vladimir Barinov <vladimir.barinov at cogentembedded.com>

>>> Add ML86V7667 platform devices on BOCK-W board, configure VIN0/1 pins, and
>>> register VIN0/1 devices with the ML86V7667 specific platform data.

>>> Signed-off-by: Vladimir Barinov <vladimir.barinov at cogentembedded.com>
>>> [Sergei: some macro/comment cleanup; updated the copyrights, removed duplicate
>>> #include, annotated all platform data as '__initdata'.]
>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov at cogentembedded.com>

>>> ---
>>> Changes since version 2:
>>> - removed duplicate #include <linux/pinctrl/machine.h>;
>>> - annotated all platform data as '__initdata' since they're kmemdup()'ed while
>>>    registering the platform devices anyway;
>>> - resolved rejects, refreshed the patch.

>>>   arch/arm/mach-shmobile/board-bockw.c |   39 +++++++++++++++++++++++++++++++++++
>>>   1 file changed, 39 insertions(+)

>>> Index: renesas/arch/arm/mach-shmobile/board-bockw.c
>>> ===================================================================
>>> --- renesas.orig/arch/arm/mach-shmobile/board-bockw.c
>>> +++ renesas/arch/arm/mach-shmobile/board-bockw.c

>> [...]

>>> @@ -80,6 +101,16 @@ static const struct pinctrl_map bockw_pi
>>>   				  "sdhi0_data4", "sdhi0"),
>>>   	PIN_MAP_MUX_GROUP_DEFAULT("sh_mobile_sdhi.0", "pfc-r8a7778",
>>>   				  "sdhi0_wp", "sdhi0"),
>>> +	/* VIN0 */
>>> +	PIN_MAP_MUX_GROUP_DEFAULT("rcar_vin.0", "pfc-r8a7778",
>>> +				  "vin0_clk", "vin0"),
>>> +	PIN_MAP_MUX_GROUP_DEFAULT("rcar_vin.0", "pfc-r8a7778",
>>> +				  "vin0_data8", "vin0"),
>>> +	/* VIN1 */
>>> +	PIN_MAP_MUX_GROUP_DEFAULT("rcar_vin.1", "pfc-r8a7778",
>>> +				  "vin1_clk", "vin1"),
>>> +	PIN_MAP_MUX_GROUP_DEFAULT("rcar_vin.1", "pfc-r8a7778",
>>> +				  "vin1_data8", "vin1"),
>>>   };

>>      It turned out that VIN1 signal VI1_CLK conflicts with Ether
>> signal ETH_REF_CLK.
>> So I'm a bit at a loss here: whether I should totally remove VIN1
>> device if I'm going
>> to add Ether device, or add a BOCK-W specific submenu to Kconfig, or
>> add driver
>> specific #ifdef's in the code to disable VIN1 if sh_eth driver is
>> enabled?...

> My preference, in order is:

> * Remove VIN1 (is it used?)

    At least video decoders are connected to both VIN0 and VIN1.

> * Kconfig-foo
> * #ifdef nastiness

    I took the 3rd approach but luckily it was possible without #ifdef, using 
*if* (!IS_ENABLED(CONFIG_SH_ETH)) statement. Working patch posted on Monday.

WBR, Sergei