Re: [PATCH v2 0/7] Add support of OV9655 camera

[H. Nikolaus Schaller]

Hi,

> Am 18.07.2017 um 21:52 schrieb Sakari Ailus <sakari.ailus@xxxxxx>:
> 
> On Tue, Jul 18, 2017 at 12:53:12PM +0000, Hugues FRUCHET wrote:
>> 
>> 
>> On 07/18/2017 02:17 PM, H. Nikolaus Schaller wrote:
>>> Hi,
>>> 
>>>> Am 18.07.2017 um 13:59 schrieb Hans Verkuil <hverkuil@xxxxxxxxx>:
>>>> 
>>>> On 12/07/17 22:01, Sylwester Nawrocki wrote:
>>>>> Hi Hugues,
>>>>> 
>>>>> On 07/03/2017 11:16 AM, Hugues Fruchet wrote:
>>>>>> This patchset enables OV9655 camera support.
>>>>>> 
>>>>>> OV9655 support has been tested using STM32F4DIS-CAM extension board
>>>>>> plugged on connector P1 of STM32F746G-DISCO board.
>>>>>> Due to lack of OV9650/52 hardware support, the modified related code
>>>>>> could not have been checked for non-regression.
>>>>>> 
>>>>>> First patches upgrade current support of OV9650/52 to prepare then
>>>>>> introduction of OV9655 variant patch.
>>>>>> Because of OV9655 register set slightly different from OV9650/9652,
>>>>>> not all of the driver features are supported (controls). Supported
>>>>>> resolutions are limited to VGA, QVGA, QQVGA.
>>>>>> Supported format is limited to RGB565.
>>>>>> Controls are limited to color bar test pattern for test purpose.
>>>>> 
>>>>> I appreciate your efforts towards making a common driver but IMO it would be
>>>>> better to create a separate driver for the OV9655 sensor.  The original driver
>>>>> is 1576 lines of code, your patch set adds half of that (816).  There are
>>>>> significant differences in the feature set of both sensors, there are
>>>>> differences in the register layout.  I would go for a separate driver, we
>>>>> would then have code easier to follow and wouldn't need to worry about possible
>>>>> regressions.  I'm afraid I have lost the camera module and won't be able
>>>>> to test the patch set against regressions.
>>>>> 
>>>>> IMHO from maintenance POV it's better to make a separate driver. In the end
>>>>> of the day we wouldn't be adding much more code than it is being done now.
>>>> 
>>>> I agree. We do not have great experiences in the past with trying to support
>>>> multiple variants in a single driver (unless the diffs are truly small).
>>> 
>>> Well,
>>> IMHO the diffs in ov965x are smaller (but untestable because nobody seems
>>> to have an ov9650/52 board) than within the bq27xxx chips, but I can dig out
>>> an old pdata based separate ov9655 driver and extend that to become DT compatible.
>>> 
>>> I had abandoned that separate approach in favour of extending the ov965x driver.
>>> 
>>> Have to discuss with Hugues how to proceed.
>>> 
>>> BR and thanks,
>>> Nikolaus
>>> 
>> 
>> As Sylwester and Hans, I'm also in flavour of a separate driver, the 
>> fact that register set seems similar but in fact is not and that we 
>> cannot test for non-regression of 9650/52 are killer for me to continue 
>> on a single driver.
>> We can now restart from a new fresh state of the art sensor driver 
>> getting rid of legacy (pdata, old gpio, etc...).
> 
> Agreed. I bet the result will look cleaner indeed although this wasn't one
> of the complex drivers.

I finally managed to find the bug why mplayer did select-timeout on the GTA04.
Was a bug in pinmux setup of the GTA04 for the omap3isp.

And I have resurrected our years old 3.12 camera driver, which was based on the
MT9P031 code. It was already separate from ov9650/52.

I have extended it to support DT by including some parts of Hugues' work.

It still needs some cleanup and discussion but will be a simple patch (one
for ov9655.c + Kconfig + Makefile) and one for bindings (I hope it includes
all your comments).

I will post v1 in the next days.

BR,
Nikolaus