Re: [PATCH v2 1/8] media: v4l: ctrls: add a control for flash/strobe duration

[Sakari Ailus]

Hi Richard, Dave,

On Tue, Mar 25, 2025 at 09:20:29AM +0100, Richard Leitner wrote:
> Hi Dave, Hi Sakari
> 
> On Wed, Mar 19, 2025 at 10:06:28AM +0000, Sakari Ailus wrote:
> > Hi Dave,
> > 
> > On Tue, Mar 18, 2025 at 04:39:05PM +0000, Dave Stevenson wrote:
> > > Hi Sakari
> > > 
> > > On Tue, 18 Mar 2025 at 15:11, Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> wrote:
> > > >
> > > > Hi Richard,
> > > >
> > > > On Tue, Mar 18, 2025 at 03:46:18PM +0100, Richard Leitner wrote:
> 
> ...
> 
> > > > > The ov9282 driver uses the sensor in global shutter mode.
> > > > >
> > > > > I totally agree with your statement. This pattern is only useful for
> > > > > global shutter operation.
> > > >
> > > > I think (nearly?) all supported sensors use a rolling shutter.
> > > 
> > > You've got at least two other global shutter sensors supported in
> > > mainline - Omnivision ov7251 and Sony imx296.
> > > Patches have been posted for OnSemi ar0144 (Laurent) and ar0234
> > > (Dongcheng), which are also both global shutter sensors.
> > > 
> > > So yes they are in the minority, but not that uncommon.
> > 
> > Thanks for the list. I briefly discussed this with Laurent and I agree with
> > him this information would probably be best kept with userspace (libcamera)
> > without trying to enumerate it from the kernel (albeit CCS might be an
> > exception, but that's an entirely different discussion then). Only changing
> > the global/rolling configuration would require a control.
> 
> Thanks for the feeback and clarification!
> 
> So am I understanding this correctly that the flash/strobe duration
> approach in this series is basically fine?

I'd think so, yes.

> 
> If so I will send a v3 later today.

-- 
Regards,

Sakari Ailus