All of lore.kernel.org
 help / color / mirror / Atom feed
From: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
To: Hans Verkuil <hansverk@cisco.com>
Cc: Hans Verkuil <hverkuil@xs4all.nl>,
	linux-media@vger.kernel.org,
	Hans Verkuil <hans.verkuil@cisco.com>
Subject: Re: [PATCH 9/9] mt9t112: initialize left and top
Date: Mon, 4 May 2015 09:54:54 +0200 (CEST)	[thread overview]
Message-ID: <Pine.LNX.4.64.1505040949450.9253@axis700.grange> (raw)
In-Reply-To: <55472333.7000204@cisco.com>

On Mon, 4 May 2015, Hans Verkuil wrote:

> On 05/04/2015 09:40 AM, Guennadi Liakhovetski wrote:
> > On Mon, 4 May 2015, Hans Verkuil wrote:
> > 
> >> On 05/03/2015 11:02 PM, Guennadi Liakhovetski wrote:
> >>> Hi Hans,
> >>>
> >>> On Sun, 3 May 2015, Hans Verkuil wrote:
> >>>
> >>>> From: Hans Verkuil <hans.verkuil@cisco.com>
> >>>>
> >>>> The left and top variables were uninitialized, leading to unexpected
> >>>> results.
> >>>>
> >>>> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
> >>>> ---
> >>>>  drivers/media/i2c/soc_camera/mt9t112.c | 3 ++-
> >>>>  1 file changed, 2 insertions(+), 1 deletion(-)
> >>>>
> >>>> diff --git a/drivers/media/i2c/soc_camera/mt9t112.c b/drivers/media/i2c/soc_camera/mt9t112.c
> >>>> index de10a76..02190d6 100644
> >>>> --- a/drivers/media/i2c/soc_camera/mt9t112.c
> >>>> +++ b/drivers/media/i2c/soc_camera/mt9t112.c
> >>>> @@ -952,7 +952,8 @@ static int mt9t112_set_fmt(struct v4l2_subdev *sd,
> >>>>  	struct v4l2_mbus_framefmt *mf = &format->format;
> >>>>  	struct i2c_client *client = v4l2_get_subdevdata(sd);
> >>>>  	struct mt9t112_priv *priv = to_mt9t112(client);
> >>>> -	unsigned int top, left;
> >>>> +	unsigned int top = priv->frame.top;
> >>>> +	unsigned int left = priv->frame.left;
> >>>
> >>> I don't think this is needed. We don't care about left and top in 
> >>> mt9t112_set_fmt().
> >>
> >> On further analysis you are correct, it will work with random left/top
> >> values. But I think it is 1) very unexpected and 2) bad form to leave it
> >> with random values.
> >>
> >> I prefer to keep this patch, unless you disagree.
> > 
> > Sorry, but I do. Assigning those specific values to left and top makes the 
> > code even more confusing, it makes it look like that makes any sense, 
> > whereas it doesn't. If anything we can add a comment there. Or we can pass 
> > NULL and make sure to catch it somewhere down the line.
> > 
> 
> What about this:
> 
> 	unsigned int top = 0; /* don't care */
> 	unsigned int left = 0; /* don't care */

This would be my third preference. My first preference is just a comment. 
My second preference is adding

	if (!start)
		return;

in the middle of soc_camera_limit_side() and using NULL in 
mt9t112_set_fmt(). I really dislike meaningless operations, also when they 
fix compiler warnings, but here even the compiler is happy:)

Thanks
Guennadi

> >>>>  	int i;
> >>>>  
> >>>>  	if (format->pad)
> >>>> -- 
> >>>> 2.1.4
> >>>>
> >>
> 

      reply	other threads:[~2015-05-04  7:55 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-03  9:54 [PATCH 0/9] soc-camera sensor improvements Hans Verkuil
2015-05-03  9:54 ` [PATCH 1/9] imx074: don't call imx074_find_datafmt() twice Hans Verkuil
2015-05-03 17:54   ` Guennadi Liakhovetski
2015-05-04  7:22     ` Hans Verkuil
2015-05-03  9:54 ` [PATCH 2/9] mt9m001: avoid calling mt9m001_find_datafmt() twice Hans Verkuil
2015-05-03 17:56   ` Guennadi Liakhovetski
2015-05-03  9:54 ` [PATCH 3/9] mt9v022: avoid calling mt9v022_find_datafmt() twice Hans Verkuil
2015-05-03 17:57   ` Guennadi Liakhovetski
2015-05-03  9:54 ` [PATCH 4/9] ov2640: avoid calling ov2640_select_win() twice Hans Verkuil
2015-05-03 18:19   ` Guennadi Liakhovetski
2015-05-04  7:25     ` Hans Verkuil
2015-05-03  9:54 ` [PATCH 5/9] ov5642: avoid calling ov5642_find_datafmt() twice Hans Verkuil
2015-05-03 20:24   ` Guennadi Liakhovetski
2015-05-04  7:26     ` Hans Verkuil
2015-05-03  9:54 ` [PATCH 6/9] ov772x: avoid calling ov772x_select_params() twice Hans Verkuil
2015-05-03 20:30   ` Guennadi Liakhovetski
2015-05-03  9:54 ` [PATCH 7/9] ov9640: avoid calling ov9640_res_roundup() twice Hans Verkuil
2015-05-03 20:34   ` Guennadi Liakhovetski
2015-05-03  9:54 ` [PATCH 8/9] ov9740: avoid calling ov9740_res_roundup() twice Hans Verkuil
2015-05-03 20:47   ` Guennadi Liakhovetski
2015-05-04 10:20     ` Hans Verkuil
2015-05-03  9:54 ` [PATCH 9/9] mt9t112: initialize left and top Hans Verkuil
2015-05-03 21:02   ` Guennadi Liakhovetski
2015-05-03 21:09     ` Guennadi Liakhovetski
2015-05-04  7:31     ` Hans Verkuil
2015-05-04  7:40       ` Guennadi Liakhovetski
2015-05-04  7:43         ` Hans Verkuil
2015-05-04  7:54           ` Guennadi Liakhovetski [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Pine.LNX.4.64.1505040949450.9253@axis700.grange \
    --to=g.liakhovetski@gmx.de \
    --cc=hans.verkuil@cisco.com \
    --cc=hansverk@cisco.com \
    --cc=hverkuil@xs4all.nl \
    --cc=linux-media@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.