Hi Niklas, On Tue, Apr 14, 2020 at 8:39 PM Niklas <niklas.soderlund@xxxxxxxxxxxx> wrote: > > Hi Lad, > > I spent all day playing with different solutions to how to move forward > with this. My main problem is I have no setup where I can produce RAW > image formats to test. But reading the datasheet I see the problem you > are trying to solve. > Thank you for looking into this. > I think for now the best solution might be to in rvin_crop_scale_comp() > add a check for if the pixelformat is RAW and cut the value written to > VNIS_REG in half. The bpp for the format shall still be set to 1. > > > fmt = rvin_format_from_pixel(vin, vin->format.pixelformat); > stride = vin->format.bytesperline / fmt->bpp; > > if (vin->format.pixelformat == V4L2_PIX_FMT_SRGGB8) > stride /= 2; > > rvin_write(vin, stride, VNIS_REG); > > I would also add a nice big comment above the if () that explains why > the stride is cut in half for raw. > Agreed shall do that as above. Cheers, --Prabhakar > On 2020-04-07 11:56:23 +0200, Niklas wrote: > > Hi Lad, > > > > On 2020-04-06 18:20:33 +0100, Lad, Prabhakar wrote: > > > Did you manage to get the required information on this ? > > > > I'm still working on it, sorry for not completing it last week. I will > > let you know as soon as I can. > > > > -- > > Regards, > > Niklas Söderlund > > -- > Regards, > Niklas Söderlund
