linux-kernel - Re: [PATCH] media: imx283: Report correct V4L2_SEL_TGT

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20241031093859.GB2473@pendragon.ideasonboard.com>
Date: Thu, 31 Oct 2024 11:38:59 +0200
From: Laurent Pinchart <laurent.pinchart@...asonboard.com>
To: Stefan Klug <stefan.klug@...asonboard.com>
Cc: Umang Jain <umang.jain@...asonboard.com>,
	libcamera-devel@...ts.libcamera.org, linux-media@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Sakari Ailus <sakari.ailus@...ux.intel.com>,
	Kieran Bingham <kieran.bingham@...asonboard.com>,
	Mauro Carvalho Chehab <mchehab@...nel.org>
Subject: Re: [PATCH] media: imx283: Report correct V4L2_SEL_TGT_CROP

Hello,

On Thu, Oct 31, 2024 at 10:13:13AM +0100, Stefan Klug wrote:
> On Thu, Oct 31, 2024 at 11:07:00AM +0530, Umang Jain wrote:
> > On 30/10/24 10:04 pm, Stefan Klug wrote:
> > > The target crop rectangle is initialized with the crop of the default
> > > sensor mode. This is incorrect when a different sensor mode gets
> > > selected. Fix that by updating the crop rectangle when changing the
> > > sensor mode.
> > > 
> > > Signed-off-by: Stefan Klug <stefan.klug@...asonboard.com>
> > > ---
> > >   drivers/media/i2c/imx283.c | 4 ++++
> > >   1 file changed, 4 insertions(+)
> > > 
> > > diff --git a/drivers/media/i2c/imx283.c b/drivers/media/i2c/imx283.c
> > > index 3174d5ffd2d7..c8863c9e0ccf 100644
> > > --- a/drivers/media/i2c/imx283.c
> > > +++ b/drivers/media/i2c/imx283.c
> > > @@ -1123,6 +1123,7 @@ static int imx283_set_pad_format(struct v4l2_subdev *sd,
> > >   				 struct v4l2_subdev_state *sd_state,
> > >   				 struct v4l2_subdev_format *fmt)
> > >   {
> > > +	struct v4l2_rect *crop;
> > >   	struct v4l2_mbus_framefmt *format;
> > >   	const struct imx283_mode *mode;
> > >   	struct imx283 *imx283 = to_imx283(sd);
> > > @@ -1149,6 +1150,9 @@ static int imx283_set_pad_format(struct v4l2_subdev *sd,
> > >   	*format = fmt->format;
> > > +	crop = v4l2_subdev_state_get_crop(sd_state, IMAGE_PAD);
> > > +	*crop = mode->crop;
> > > +
> > 
> > One thing to note, is the crop for binning modes.
> > 
> > Do you need to report
> > 
> >     mode->crop.width / mode->hbin_ratio
> >     mode->crop.height / mode->vbin_ratio
> > 
> > for those modes?
> 
> Good point. I was naively assuming that it has the same semantics as we
> use for ScalerCrop in libcamera where it is explicitly stated that the
> coordinates are in sensor pixels without binning. That has the added
> advantage that we can deduce the binning factor from TGT_CROP and the
> actual output size. However I couldn't find a precise specification for
> that in the linux docs. 
> 
> Maybe Sakari or Laurent have a definiteve answer there?

This is not standardized in V4L2, and different drivers implement
different semantics. There's an ongoing effort to fix this, see
https://lore.kernel.org/r/20241011075535.588140-1-sakari.ailus@linux.intel.com.
Reviews are appreciated :-)

> > >   	return 0;
> > >   }

-- 
Regards,

Laurent Pinchart