[libcamera-devel] [PATCH] libcamera: pipeline: simple: Add support for ST's DCMIPP

[Hugues FRUCHET]

Hi Laurent,

DCMIPP kernel driver has been submitted but no feedback received yet:
https://www.spinics.net/lists/linux-media/msg218378.html
Would be nice to get at least a first level of review before pushing a 
new version.

For GC2145, we are in the middle of the bridge with two versions of 
driver that can be found on the net: a pinephone version  with a 
parallel interface support and our driver which is a CSI-2 version (we 
have a small amount of support from GalaxyCore, not amazing but better 
than nothing...)

Best regards,

On 5/2/23 14:33, Laurent Pinchart wrote:
> On Tue, May 02, 2023 at 01:29:31PM +0100, Kieran Bingham via libcamera-devel wrote:
>> Quoting Daniel Scally via libcamera-devel (2023-05-02 12:13:57)
>>> The STM32 contains a camera pipeline known as the DCMIPP (Digital
>>> Camera-Memory Interface Pixel Processor) which receives data from a
>>> parallel interface and dumps the post-processed data to memory. The
>>> pipeline is capable of some processing in the form of downscaling
>>> captured data through cropping or skipping the sensor's output.
>>>
>>> The simple pipeline handler is quite capable of handling the DCMIPP
>>> given its operation is handled entirely through configuring the pads
>>> of a media graph, so add support for the driver to the pipeline's
>>> supportedDevices array.
>>>
>>> Signed-off-by: Daniel Scally <dan.scally at ideasonboard.com>
>>> ---
>>> Support for the dcmipp has been tested on an stm32mp135f-dk with a gc2145
>>> camera sensor. Note that the dcmipp driver **is not** in mainline at the moment,
>>> and nor is the gc2145's driver. @Hugues, @Alain, is there a plan to move those
>>> drivers upstream?
>>
>> As I understand it, the drivers are open, and as long as there is an
>> expectation that they are going upstream - that's 'enough' to get this
>> merged in libcamera I believe.
> 
> We don't have to wait for the dcmipp driver to land upstream before
> merging this, but I'd like to understand the upstreaming plan first.
> 
>> It's also a really easily 'line' to maintain ;-)
>>
>>> A few changes to ST's kernel were needed to fix some format propagation problems
>>> and some missing controls for the GC2145 for this to work. Those can be found in
>>> my tree linked below [1], along with some fixes to the CSI-2 to Parallel bridge
>>> driver which _is_ upstream and which I posted to linux-media [2].
>>>
>>> [1] https://github.com/djrscally/linux/tree/v5.15-stm32mp
>>> [2] https://lore.kernel.org/linux-media/20230502103547.150918-1-dan.scally@ideasonboard.com/T/#t
>>>
>>>   src/libcamera/pipeline/simple/simple.cpp | 1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/src/libcamera/pipeline/simple/simple.cpp b/src/libcamera/pipeline/simple/simple.cpp
>>> index e1f8b989..01516bc3 100644
>>> --- a/src/libcamera/pipeline/simple/simple.cpp
>>> +++ b/src/libcamera/pipeline/simple/simple.cpp
>>> @@ -194,6 +194,7 @@ static const SimplePipelineInfo supportedDevices[] = {
>>>          { "mxc-isi", {} },
>>>          { "qcom-camss", {} },
>>>          { "sun6i-csi", {} },
>>> +       { "dcmipp", {} },
>>
>> Alphabetical sort order ;-)
>>
>> With that...
>>
>> Reviewed-by: Kieran Bingham <kieran.bingham at ideasonboard.com>
>>
>>>   };
>>>   
>>>   } /* namespace */
>