Re: [PATCH v2 00/23] media: ov5640: Rework the clock tree programming for MIPI

[Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>]

On 10/02/2022 19:11, Jacopo Mondi wrote:
> Hi Tomi
> 
> On Thu, Feb 10, 2022 at 03:00:21PM +0200, Tomi Valkeinen wrote:
>> On 10/02/2022 13:04, Jacopo Mondi wrote:
>>> v1:
>>> https://patchwork.linuxtv.org/project/linux-media/list/?series=7249
>>
>> You could rather point to lore.kernel.org, so that the intro letter and the
>> discussions are also visible.
> 
> Sure, here you go!
> https://lore.kernel.org/linux-media/20220131143245.128089-1-jacopo@jmondi.org/
> 
>>
>>> A branch for testing based on the most recent media-master is available at
>>> https://git.sr.ht/~jmondi_/linux #jmondi/media-master/ov5640-v2
>>>
>>> If anyone with a DVP setup could verify I have not broken their use case
>>> I would very much appreciate that :)
>>>
>>> v1 -> v2:
>>> - rework the modes definition to process the full pixel array
>>> - rework get_selection to report the correct BOUND and DEFAULT targets
>>> - implement init_cfg
>>> - minor style changes as suggested by Laurent
>>> - test with 1 data lane
>>
>> Very nice! I tested this on TI's DRA76 EVM (CSI-2). UYVY and RGB565, with
>> the following resolutions: 160 120, 176 144, 320 240, 640 480, 720 480, 720
>> 576, 1024 768, 1280 720, 1920 1080.
> 
> Great! A 2 data lanes setup I assume ? Have you been able to test the
> framerate as well ?

Yes, 2 data lanes. I can see that with 640x480 and 1024x768 I get ~30fps.

Why does ov5640_enum_frame_interval() return an error if ov5640 is a csi 
sensor?

  Tomi