[PATCH 0/6] Fix the format field order value for progressive subdevs

[PATCH 0/6] Fix the format field order value for progressive subdevs

[Laurent Pinchart]

Hello,

This patch set fixes five sensor drivers and one camera interface driver to
return a format field order value set to V4L2_FIELD_NONE instead of
V4L2_FIELD_ANY.

V4L2_FIELD_ANY is used by applications to notify the driver that they don't
care about the interlaced video field order. The value must never be returned
by drivers, they must instead select a default field order they support.

The six drivers fixed by this patch all forgot to initialize the field order,
resulting in V4L2_FIELD_ANY (=0) being returned. As all those drivers support
progressive video only, make them return V4L2_FIELD_NONE instead.

Cc: Andrzej Hajda <a.hajda@samsung.com>
Cc: Sakari Ailus <sakari.ailus@iki.fi>
Cc: Sangwook Lee <sangwook.lee@linaro.org>

Laurent Pinchart (6):
  v4l: noon010p30: Return V4L2_FIELD_NONE from pad-level set format
  v4l: s5k4ecgx: Return V4L2_FIELD_NONE from pad-level set format
  v4l: s5k5baf: Return V4L2_FIELD_NONE from pad-level set format
  v4l: s5k6a3: Return V4L2_FIELD_NONE from pad-level set format
  v4l: smiapp: Return V4L2_FIELD_NONE from pad-level get/set format
  v4l: s3c-camif: Return V4L2_FIELD_NONE from pad-level set format

 drivers/media/i2c/noon010pc30.c                  | 1 +
 drivers/media/i2c/s5k4ecgx.c                     | 1 +
 drivers/media/i2c/s5k5baf.c                      | 2 ++
 drivers/media/i2c/s5k6a3.c                       | 1 +
 drivers/media/i2c/smiapp/smiapp-core.c           | 3 +++
 drivers/media/platform/s3c-camif/camif-capture.c | 2 ++
 6 files changed, 10 insertions(+)

-- 
Regards,

Laurent Pinchart

[PATCH 1/6] v4l: noon010p30: Return V4L2_FIELD_NONE from pad-level set format

[Laurent Pinchart]

The sensor is progressive, always return the field order set to
V4L2_FIELD_NONE.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/i2c/noon010pc30.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/media/i2c/noon010pc30.c b/drivers/media/i2c/noon010pc30.c
index 271d0b7..7eae487 100644
--- a/drivers/media/i2c/noon010pc30.c
+++ b/drivers/media/i2c/noon010pc30.c
@@ -554,6 +554,7 @@ static int noon010_set_fmt(struct v4l2_subdev *sd, struct v4l2_subdev_fh *fh,
 	nf = noon010_try_fmt(sd, &fmt->format);
 	noon010_try_frame_size(&fmt->format, &size);
 	fmt->format.colorspace = V4L2_COLORSPACE_JPEG;
+	fmt->format.field = V4L2_FIELD_NONE;
 
 	if (fmt->which == V4L2_SUBDEV_FORMAT_TRY) {
 		if (fh) {
-- 
1.8.5.5

[PATCH 2/6] v4l: s5k4ecgx: Return V4L2_FIELD_NONE from pad-level set format

[Laurent Pinchart]

The sensor is progressive, always return the field order set to
V4L2_FIELD_NONE.

Cc: Sangwook Lee <sangwook.lee@linaro.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/i2c/s5k4ecgx.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/media/i2c/s5k4ecgx.c b/drivers/media/i2c/s5k4ecgx.c
index 2750de6..1fcc76f 100644
--- a/drivers/media/i2c/s5k4ecgx.c
+++ b/drivers/media/i2c/s5k4ecgx.c
@@ -594,6 +594,7 @@ static int s5k4ecgx_set_fmt(struct v4l2_subdev *sd, struct v4l2_subdev_fh *fh,
 	pf = s5k4ecgx_try_fmt(sd, &fmt->format);
 	s5k4ecgx_try_frame_size(&fmt->format, &fsize);
 	fmt->format.colorspace = V4L2_COLORSPACE_JPEG;
+	fmt->format.field = V4L2_FIELD_NONE;
 
 	if (fmt->which == V4L2_SUBDEV_FORMAT_TRY) {
 		if (fh) {
-- 
1.8.5.5

[PATCH 3/6] v4l: s5k5baf: Return V4L2_FIELD_NONE from pad-level set format

[Laurent Pinchart]

The sensor is progressive, always return the field order set to
V4L2_FIELD_NONE.

Cc: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/i2c/s5k5baf.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/media/i2c/s5k5baf.c b/drivers/media/i2c/s5k5baf.c
index 2d768ef..564f05f 100644
--- a/drivers/media/i2c/s5k5baf.c
+++ b/drivers/media/i2c/s5k5baf.c
@@ -1313,6 +1313,8 @@ static int s5k5baf_set_fmt(struct v4l2_subdev *sd, struct v4l2_subdev_fh *fh,
 	const struct s5k5baf_pixfmt *pixfmt;
 	int ret = 0;
 
+	mf->field = V4L2_FIELD_NONE;
+
 	if (fmt->which == V4L2_SUBDEV_FORMAT_TRY) {
 		*v4l2_subdev_get_try_format(fh, fmt->pad) = *mf;
 		return 0;
-- 
1.8.5.5

[PATCH 4/6] v4l: s5k6a3: Return V4L2_FIELD_NONE from pad-level set format

[Laurent Pinchart]

The sensor is progressive, always return the field order set to
V4L2_FIELD_NONE.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/i2c/s5k6a3.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/media/i2c/s5k6a3.c b/drivers/media/i2c/s5k6a3.c
index 7bc2271..c11a408 100644
--- a/drivers/media/i2c/s5k6a3.c
+++ b/drivers/media/i2c/s5k6a3.c
@@ -115,6 +115,7 @@ static void s5k6a3_try_format(struct v4l2_mbus_framefmt *mf)
 
 	fmt = find_sensor_format(mf);
 	mf->code = fmt->code;
+	mf->field = V4L2_FIELD_NONE;
 	v4l_bound_align_image(&mf->width, S5K6A3_SENSOR_MIN_WIDTH,
 			      S5K6A3_SENSOR_MAX_WIDTH, 0,
 			      &mf->height, S5K6A3_SENSOR_MIN_HEIGHT,
-- 
1.8.5.5

[PATCH 5/6] v4l: smiapp: Return V4L2_FIELD_NONE from pad-level get/set format

[Laurent Pinchart]

The SMIA++ sensors are progressive, always return the field order set to
V4L2_FIELD_NONE.

Cc: Sakari Ailus <sakari.ailus@iki.fi>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/i2c/smiapp/smiapp-core.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/media/i2c/smiapp/smiapp-core.c b/drivers/media/i2c/smiapp/smiapp-core.c
index 06fb032..c741546 100644
--- a/drivers/media/i2c/smiapp/smiapp-core.c
+++ b/drivers/media/i2c/smiapp/smiapp-core.c
@@ -1554,6 +1554,7 @@ static int __smiapp_get_format(struct v4l2_subdev *subdev,
 		fmt->format.code = __smiapp_get_mbus_code(subdev, fmt->pad);
 		fmt->format.width = r->width;
 		fmt->format.height = r->height;
+		fmt->format.field = V4L2_FIELD_NONE;
 	}
 
 	return 0;
@@ -1687,6 +1688,7 @@ static int smiapp_set_format(struct v4l2_subdev *subdev,
 	fmt->format.code = __smiapp_get_mbus_code(subdev, fmt->pad);
 	fmt->format.width &= ~1;
 	fmt->format.height &= ~1;
+	fmt->format.field = V4L2_FIELD_NONE;
 
 	fmt->format.width =
 		clamp(fmt->format.width,
@@ -2674,6 +2676,7 @@ static int smiapp_open(struct v4l2_subdev *sd, struct v4l2_subdev_fh *fh)
 		try_fmt->width = sensor->limits[SMIAPP_LIMIT_X_ADDR_MAX] + 1;
 		try_fmt->height = sensor->limits[SMIAPP_LIMIT_Y_ADDR_MAX] + 1;
 		try_fmt->code = mbus_code;
+		try_fmt->field = V4L2_FIELD_NONE;
 
 		try_crop->top = 0;
 		try_crop->left = 0;
-- 
1.8.5.5

[PATCH 6/6] v4l: s3c-camif: Return V4L2_FIELD_NONE from pad-level set format

[Laurent Pinchart]

The bridge driver doesn't support interlaced formats, always return the
field order set to V4L2_FIELD_NONE.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 drivers/media/platform/s3c-camif/camif-capture.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/media/platform/s3c-camif/camif-capture.c b/drivers/media/platform/s3c-camif/camif-capture.c
index deba425..cc89133 100644
--- a/drivers/media/platform/s3c-camif/camif-capture.c
+++ b/drivers/media/platform/s3c-camif/camif-capture.c
@@ -1271,6 +1271,7 @@ static int s3c_camif_subdev_get_fmt(struct v4l2_subdev *sd,
 	}
 
 	mutex_unlock(&camif->lock);
+	mf->field = V4L2_FIELD_NONE;
 	mf->colorspace = V4L2_COLORSPACE_JPEG;
 	return 0;
 }
@@ -1319,6 +1320,7 @@ static int s3c_camif_subdev_set_fmt(struct v4l2_subdev *sd,
 	v4l2_dbg(1, debug, sd, "pad%d: code: 0x%x, %ux%u\n",
 		 fmt->pad, mf->code, mf->width, mf->height);
 
+	mf->field = V4L2_FIELD_NONE;
 	mf->colorspace = V4L2_COLORSPACE_JPEG;
 	mutex_lock(&camif->lock);
 
-- 
1.8.5.5

Re: [PATCH 0/6] Fix the format field order value for progressive subdevs

[Sylwester Nawrocki]

Hi Laurent,

On 05/26/2014 09:05 PM, Laurent Pinchart wrote:
> Hello,
>
> This patch set fixes five sensor drivers and one camera interface driver to
> return a format field order value set to V4L2_FIELD_NONE instead of
> V4L2_FIELD_ANY.
>
> V4L2_FIELD_ANY is used by applications to notify the driver that they don't
> care about the interlaced video field order. The value must never be returned
> by drivers, they must instead select a default field order they support.
>
> The six drivers fixed by this patch all forgot to initialize the field order,
> resulting in V4L2_FIELD_ANY (=0) being returned. As all those drivers support
> progressive video only, make them return V4L2_FIELD_NONE instead.
>
> Cc: Andrzej Hajda <a.hajda@samsung.com>
> Cc: Sakari Ailus <sakari.ailus@iki.fi>
> Cc: Sangwook Lee <sangwook.lee@linaro.org>
>
> Laurent Pinchart (6):
>    v4l: noon010p30: Return V4L2_FIELD_NONE from pad-level set format
>    v4l: s5k4ecgx: Return V4L2_FIELD_NONE from pad-level set format
>    v4l: s5k5baf: Return V4L2_FIELD_NONE from pad-level set format
>    v4l: s5k6a3: Return V4L2_FIELD_NONE from pad-level set format
>    v4l: smiapp: Return V4L2_FIELD_NONE from pad-level get/set format
>    v4l: s3c-camif: Return V4L2_FIELD_NONE from pad-level set format

For patches 1...4, 6:

Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>

>   drivers/media/i2c/noon010pc30.c                  | 1 +
>   drivers/media/i2c/s5k4ecgx.c                     | 1 +
>   drivers/media/i2c/s5k5baf.c                      | 2 ++
>   drivers/media/i2c/s5k6a3.c                       | 1 +
>   drivers/media/i2c/smiapp/smiapp-core.c           | 3 +++
>   drivers/media/platform/s3c-camif/camif-capture.c | 2 ++
>   6 files changed, 10 insertions(+)

--
Regards,
Sylwester

Re: [PATCH 5/6] v4l: smiapp: Return V4L2_FIELD_NONE from pad-level get/set format

[Sakari Ailus]

Hi Laurent,

Laurent Pinchart wrote:
> The SMIA++ sensors are progressive, always return the field order set to
> V4L2_FIELD_NONE.
>
> Cc: Sakari Ailus <sakari.ailus@iki.fi>
> Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

Acked-by: Sakari Ailus <sakari.ailus@linux.intel.com>

-- 
Regards,

Sakari Ailus
sakari.ailus@iki.fi