Re: [PATCH] media: Documentation: Fix frame interval calculation for raw camera sensors

[Jai Luthra <jai.luthra@ideasonboard.com>]

Quoting Sakari Ailus (2026-02-24 14:09:39)
> Hi Jai,
> 
> On Tue, Feb 24, 2026 at 10:55:33AM +0530, Jai Luthra wrote:
> > >         LLP = (analog_crop_width + HBLANK) / binning_timing_h
> > >         FLL = (analog_crop_height + VBLANK) / binning_timing_v
> > >         frame_interval = LLP * FLL / pixel_rate
> > > 
> > > I'm not 100% sure this is correct however, as the blankings should be
> > > expressed on a different clock domain that the pixel sampling rate,
> > > but I guess this is a reasonable approximation ?
> > 
> > What does these two extra controls really offer us?
> 
> The conclusion earlier on (well, maybe 10 years ago?) was that we should
> have had these to begin with, and not to use the blanking values to control
> the frame rate, largely because it requires a reference rectangle -- and
> the current drivers use a wrong one, mostly because the API doesn't offer
> the right one. The assumption back then was that everything would soon be
> SMIA (or now CCS) compliant so this wouldn't matter much. But here we
> are...
> 
> The VBLANK and HBLANK controls effectively also use the wrong reference
> (output size) in most drivers and that can't really be changed meaningfully
> without breaking things.
> 

Argh, makes sense.

> > 
> > All sensors we have seen thus far map their LLP/FLL (or equivalent HTOT/VTOT)
> > values with respect to the digital readout, and not the analogue pixel array.
> > 
> > If we add these two controls, we will have to support two different models for
> > frame interval calculation in the application layer too. Which I'm fine with if
> > it has a practical benefit, that is, it makes it easier to deal with some
> > particular sensor.
> 
> There are three models, effectively:
> 
> 1. CCS, where vblank and hblank are related to analogue crop rectangle, as in
>    hardware;
> 
> 2. Common raw sensor model, with LLP and FLL; and
> 
> 3. the rest of the existing drivers (vblank and hblank relative to format
>    on source pad).
> 

Thanks that's helpful summary. I will update v2 to reflect 1 and 3, while
details for common raw sensor model can be finalized later.

> In the case of frame length vs. blanking, I'd presume the difference in the
> user space will be relatively small.
> 

Indeed, switching to a different LLP/FFL control is easy, what's annoying
is having two different reference rectangles for the H/VBLANK controls.

> > 
> > And if exposing the LLP/FLL directly offer the same benefit, then that is
> > cleaner, as it leaves the HBLANK/VBLANK as-is in the new model.
> 
> -- 
> Kind regards,
> 
> Sakari Ailus

Thanks,
    Jai