From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Sakari Ailus <sakari.ailus@iki.fi>
Cc: linux-media@vger.kernel.org, hverkuil@xs4all.nl,
teturtia@gmail.com, dacohen@gmail.com, snjw23@gmail.com,
andriy.shevchenko@linux.intel.com, t.stanislaws@samsung.com,
tuukkat76@gmail.com, k.debski@gmail.com, riverful@gmail.com
Subject: Re: [PATCH v3 17/33] v4l: Implement v4l2_subdev_link_validate()
Date: Wed, 22 Feb 2012 11:14:49 +0100 [thread overview]
Message-ID: <2874788.a4NfdnO7z7@avalon> (raw)
In-Reply-To: <1329703032-31314-17-git-send-email-sakari.ailus@iki.fi>
Hi Sakari,
Thanks for the patch.
On Monday 20 February 2012 03:56:56 Sakari Ailus wrote:
> v4l2_subdev_link_validate() is the default op for validating a link. In V4L2
> subdev context, it is used to call a pad op which performs the proper link
> check without much extra work.
>
> Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
> ---
> Documentation/video4linux/v4l2-framework.txt | 12 +++++
> drivers/media/video/v4l2-subdev.c | 69
> ++++++++++++++++++++++++++ include/media/v4l2-subdev.h |
> 12 +++++
> 3 files changed, 93 insertions(+), 0 deletions(-)
>
> diff --git a/Documentation/video4linux/v4l2-framework.txt
> b/Documentation/video4linux/v4l2-framework.txt index f06c563..9d341bc
> 100644
> --- a/Documentation/video4linux/v4l2-framework.txt
> +++ b/Documentation/video4linux/v4l2-framework.txt
> @@ -312,6 +312,18 @@ If the subdev driver intends to process video and
> integrate with the media framework, it must implement format related
> functionality using
> v4l2_subdev_pad_ops instead of v4l2_subdev_video_ops.
>
> +In that case, the subdev driver may set the link_validate field to provide
> +its own link validation function. The link validation function is called
> for +every link in the pipeline where both of the ends of the links are
> V4L2 +sub-devices. The driver is still responsible for validating the
> correctness +of the format configuration between sub-devices and video
> nodes.
> +
> +If link_validate op is not set, the default function
> +v4l2_subdev_link_validate_default() is used instead. This function ensures
> +that width, height and the media bus pixel code are equal on both source
> and +sink of the link. Subdev drivers are also free to use this function to
> +perform the checks mentioned above in addition to their own checks. +
> A device (bridge) driver needs to register the v4l2_subdev with the
> v4l2_device:
>
> diff --git a/drivers/media/video/v4l2-subdev.c
> b/drivers/media/video/v4l2-subdev.c index ef27144..d2ec552 100644
> --- a/drivers/media/video/v4l2-subdev.c
> +++ b/drivers/media/video/v4l2-subdev.c
> @@ -379,6 +379,75 @@ const struct v4l2_file_operations v4l2_subdev_fops = {
> .poll = subdev_poll,
> };
>
> +#ifdef CONFIG_MEDIA_CONTROLLER
> +int v4l2_subdev_link_validate_default(struct v4l2_subdev *sd,
> + struct media_link *link,
> + struct v4l2_subdev_format *source_fmt,
> + struct v4l2_subdev_format *sink_fmt)
> +{
> + if (source_fmt->format.width != sink_fmt->format.width
> + || source_fmt->format.height != sink_fmt->format.height
> + || source_fmt->format.code != sink_fmt->format.code)
> + return -EINVAL;
> +
> + return 0;
> +}
> +EXPORT_SYMBOL_GPL(v4l2_subdev_link_validate_default);
> +
> +static struct v4l2_subdev_format
> +*v4l2_subdev_link_validate_get_format(struct media_pad *pad,
> + struct v4l2_subdev_format *fmt)
Why do you return a pointer to the passed struct v4l2_subdev_format instead of
just an error code ?
> +{
> + int rval;
I would have used ret instead of rval as that's the preferred variable name in
this file (OK, I'm cheating, there's a single instance of ret, and I wrote
that one ;-)).
> +
> + switch (media_entity_type(pad->entity)) {
> + case MEDIA_ENT_T_V4L2_SUBDEV:
> + fmt->which = V4L2_SUBDEV_FORMAT_ACTIVE;
> + fmt->pad = pad->index;
> + rval = v4l2_subdev_call(media_entity_to_v4l2_subdev(
> + pad->entity),
> + pad, get_fmt, NULL, fmt);
> + if (rval < 0)
> + return NULL;
> + return fmt;
> + default:
> + WARN(1, "Driver bug! Wrong media entity type %d, entity %s\n",
> + media_entity_type(pad->entity), pad->entity->name);
> + /* Fall through */
> + case MEDIA_ENT_T_DEVNODE_V4L:
> + return NULL;
> + }
> +}
> +
> +int v4l2_subdev_link_validate(struct media_link *link)
> +{
> + struct v4l2_subdev *sink;
> + struct v4l2_subdev_format _sink_fmt, _source_fmt;
> + struct v4l2_subdev_format *sink_fmt, *source_fmt;
> + int rval;
> +
> + source_fmt = v4l2_subdev_link_validate_get_format(
> + link->source, &_source_fmt);
> + if (!source_fmt)
> + return 0;
> +
> + sink_fmt = v4l2_subdev_link_validate_get_format(
> + link->sink, &_sink_fmt);
> + if (!sink_fmt)
> + return 0;
> +
> + sink = media_entity_to_v4l2_subdev(link->sink->entity);
> +
> + rval = v4l2_subdev_call(sink, pad, link_validate, link,
> + source_fmt, sink_fmt);
> + if (rval < 0 && rval != -ENOIOCTLCMD)
> + return rval;
> + return v4l2_subdev_link_validate_default(sink, link, source_fmt,
> + sink_fmt);
I don't think you should call v4l2_subdev_link_validate_default() if the
link_validate operation is implemented and returns that the link is valid. The
subdev might have weaker requirements than the default.
> +}
> +EXPORT_SYMBOL_GPL(v4l2_subdev_link_validate);
> +#endif /* CONFIG_MEDIA_CONTROLLER */
> +
> void v4l2_subdev_init(struct v4l2_subdev *sd, const struct v4l2_subdev_ops
> *ops) {
> INIT_LIST_HEAD(&sd->list);
> diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
> index d48dae5..f115608 100644
> --- a/include/media/v4l2-subdev.h
> +++ b/include/media/v4l2-subdev.h
> @@ -470,6 +470,11 @@ struct v4l2_subdev_pad_ops {
> struct v4l2_subdev_selection *sel);
> int (*set_selection)(struct v4l2_subdev *sd, struct v4l2_subdev_fh *fh,
> struct v4l2_subdev_selection *sel);
> +#ifdef CONFIG_MEDIA_CONTROLLER
> + int (*link_validate)(struct v4l2_subdev *sd, struct media_link *link,
> + struct v4l2_subdev_format *source_fmt,
> + struct v4l2_subdev_format *sink_fmt);
> +#endif /* CONFIG_MEDIA_CONTROLLER */
> };
>
> struct v4l2_subdev_ops {
> @@ -603,6 +608,13 @@ static inline void *v4l2_get_subdev_hostdata(const
> struct v4l2_subdev *sd) return sd->host_priv;
> }
>
> +#ifdef CONFIG_MEDIA_CONTROLLER
> +int v4l2_subdev_link_validate_default(struct v4l2_subdev *sd,
> + struct media_link *link,
> + struct v4l2_subdev_format *source_fmt,
> + struct v4l2_subdev_format *sink_fmt);
> +int v4l2_subdev_link_validate(struct media_link *link);
> +#endif /* CONFIG_MEDIA_CONTROLLER */
> void v4l2_subdev_init(struct v4l2_subdev *sd,
> const struct v4l2_subdev_ops *ops);
--
Regards,
Laurent Pinchart
next prev parent reply other threads:[~2012-02-22 10:14 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-20 1:56 [PATCH v3 0/33] V4L2 subdev and sensor control changes, SMIA++ driver and N9 camera board code Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 01/33] v4l: Introduce integer menu controls Sakari Ailus
2012-02-20 17:36 ` Sylwester Nawrocki
2012-02-20 1:56 ` [PATCH v3 02/33] v4l: Document " Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 03/33] vivi: Add an integer menu test control Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 04/33] v4l: VIDIOC_SUBDEV_S_SELECTION and VIDIOC_SUBDEV_G_SELECTION IOCTLs Sakari Ailus
2012-02-21 14:34 ` Laurent Pinchart
2012-02-23 5:49 ` Sakari Ailus
2012-02-21 16:15 ` Laurent Pinchart
2012-02-23 6:01 ` Sakari Ailus
2012-02-27 0:22 ` Laurent Pinchart
2012-02-27 0:57 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 05/33] v4l: vdev_to_v4l2_subdev() should have return type "struct v4l2_subdev *" Sakari Ailus
2012-02-21 14:37 ` Laurent Pinchart
2012-02-20 1:56 ` [PATCH v3 06/33] v4l: Check pad number in get try pointer functions Sakari Ailus
2012-02-21 14:42 ` Laurent Pinchart
2012-02-23 5:57 ` Sakari Ailus
2012-02-27 0:33 ` Laurent Pinchart
2012-02-27 12:27 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 07/33] v4l: Support s_crop and g_crop through s/g_selection Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 08/33] v4l: Add subdev selections documentation: svg and dia files Sakari Ailus
2012-02-21 15:00 ` Laurent Pinchart
2012-02-26 18:56 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 09/33] v4l: Add subdev selections documentation Sakari Ailus
2012-02-21 16:41 ` Laurent Pinchart
2012-02-26 21:42 ` Sakari Ailus
2012-02-28 11:42 ` Laurent Pinchart
2012-03-02 12:24 ` Sakari Ailus
2012-03-02 17:54 ` Laurent Pinchart
2012-03-02 18:01 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 10/33] v4l: Mark VIDIOC_SUBDEV_G_CROP and VIDIOC_SUBDEV_S_CROP obsolete Sakari Ailus
2012-02-21 16:42 ` Laurent Pinchart
2012-02-20 1:56 ` [PATCH v3 11/33] v4l: Image source control class Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 12/33] v4l: Image processing " Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 13/33] v4l: Document raw bayer 4CC codes Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 14/33] v4l: Add DPCM compressed formats Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 15/33] media: Add link_validate() op to check links to the sink pad Sakari Ailus
2012-02-22 10:05 ` Laurent Pinchart
2012-02-23 15:04 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 16/33] v4l: Improve sub-device documentation for pad ops Sakari Ailus
2012-02-22 10:06 ` Laurent Pinchart
2012-02-20 1:56 ` [PATCH v3 17/33] v4l: Implement v4l2_subdev_link_validate() Sakari Ailus
2012-02-22 10:14 ` Laurent Pinchart [this message]
2012-02-23 16:07 ` Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 18/33] v4l: Allow changing control handler lock Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 19/33] omap3isp: Support additional in-memory compressed bayer formats Sakari Ailus
2012-02-20 1:56 ` [PATCH v3 20/33] omap3isp: Move definitions required by board code under include/media Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 21/33] omap3: add definition for CONTROL_CAMERA_PHY_CTRL Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 22/33] omap3isp: Assume media_entity_pipeline_start may fail Sakari Ailus
2012-02-22 10:48 ` Laurent Pinchart
2012-02-26 1:08 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 23/33] omap3isp: Add lane configuration to platform data Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 24/33] omap3isp: Add information on external subdev to struct isp_pipeline Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 25/33] omap3isp: Introduce omap3isp_get_external_info() Sakari Ailus
2012-02-22 10:55 ` Laurent Pinchart
2012-02-26 1:09 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 26/33] omap3isp: Default link validation for ccp2, csi2, preview and resizer Sakari Ailus
2012-02-22 11:01 ` Laurent Pinchart
2012-02-25 1:34 ` Sakari Ailus
2012-02-26 23:14 ` Laurent Pinchart
2012-02-26 23:40 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 27/33] omap3isp: Implement proper CCDC link validation, check pixel rate Sakari Ailus
2012-02-22 11:11 ` Laurent Pinchart
2012-02-25 1:42 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 28/33] omap3isp: Move setting constaints above media_entity_pipeline_start Sakari Ailus
2012-02-22 11:12 ` Laurent Pinchart
2012-02-25 1:46 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 29/33] omap3isp: Configure CSI-2 phy based on platform data Sakari Ailus
2012-02-22 11:21 ` Laurent Pinchart
2012-02-25 1:49 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 30/33] omap3isp: Add resizer data rate configuration to resizer_set_stream Sakari Ailus
2012-02-22 11:24 ` Laurent Pinchart
2012-02-26 1:10 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 31/33] omap3isp: Remove isp_validate_pipeline and other old stuff Sakari Ailus
2012-02-22 11:26 ` Laurent Pinchart
2012-02-25 1:52 ` Sakari Ailus
2012-02-20 1:57 ` [PATCH v3 32/33] smiapp: Add driver Sakari Ailus
2012-02-27 15:38 ` Laurent Pinchart
2012-02-29 5:41 ` Sakari Ailus
2012-02-29 9:35 ` Laurent Pinchart
2012-02-29 10:00 ` Sylwester Nawrocki
2012-03-01 14:01 ` Sakari Ailus
2012-03-01 14:56 ` Laurent Pinchart
2012-02-20 1:57 ` [PATCH v3 33/33] rm680: Add camera init Sakari Ailus
2012-02-27 1:06 ` Laurent Pinchart
2012-02-28 19:05 ` Sakari Ailus
2012-02-20 2:03 ` [PATCH v3 0/33] V4L2 subdev and sensor control changes, SMIA++ driver and N9 camera board code Sakari Ailus
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=2874788.a4NfdnO7z7@avalon \
--to=laurent.pinchart@ideasonboard.com \
--cc=andriy.shevchenko@linux.intel.com \
--cc=dacohen@gmail.com \
--cc=hverkuil@xs4all.nl \
--cc=k.debski@gmail.com \
--cc=linux-media@vger.kernel.org \
--cc=riverful@gmail.com \
--cc=sakari.ailus@iki.fi \
--cc=snjw23@gmail.com \
--cc=t.stanislaws@samsung.com \
--cc=teturtia@gmail.com \
--cc=tuukkat76@gmail.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox