linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hans Verkuil <hverkuil@xs4all.nl>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: linux-media@vger.kernel.org, g.liakhovetski@gmx.de,
	prabhakar.csengg@gmail.com, Hans Verkuil <hans.verkuil@cisco.com>
Subject: Re: [RFC PATCH 6/8] v4l2-subdev: add v4l2_subdev_create_pad_configs
Date: Fri, 19 Dec 2014 14:28:19 +0100	[thread overview]
Message-ID: <549427F3.6090109@xs4all.nl> (raw)
In-Reply-To: <6152026.ecLNx03bSo@avalon>

Hi Laurent,

On 12/08/2014 12:50 AM, Laurent Pinchart wrote:
> Hi Hans,
> 
> Thank you for the patch.
> 
> On Thursday 04 December 2014 10:54:57 Hans Verkuil wrote:
>> From: Hans Verkuil <hans.verkuil@cisco.com>
>>
>> When a new subdevice is registered this new function is called to allocate
>> and initialize a pad_config array. This allows bridge drivers to use the
>> pad ops that require such an array as argument.
> 
> While I agree that a v4l2_subdev_create_pad_configs() function is needed, I 
> don't think the pad configs should be stored in the subdev structure. This 
> would defeat the whole point of having a try mechanism based on private 
> configs. Bridge drivers should instead create the configs on demand, use them, 
> and delete them afterwards.

I really don't like that. None of the current non-MC bridge drivers need this,
so why force lots of allocations and frees of pad configs if there is no need?

It all comes back to the fact that enum_mbus_code, enum_frame_size and
enum_frame_interval do not have a which field. If they had one, then bridge
drivers could call them with V4L2_SUBDEV_FORMAT_ACTIVE and a NULL pad_config
pointer. Unless of course they really need to use FORMAT_TRY, but in that
case they can set up the pad_config array themselves.

This would allow me to drop this patch and still keep the bridge drivers that
need this simple.

It's a bit of work to update the existing drivers to support the new which
field, but from what I've seen that doesn't look too difficult.

Can we discuss this on irc?

Regards,

	Hans

> 
>> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
>> ---
>>  drivers/media/v4l2-core/v4l2-device.c | 11 ++++++++++-
>>  drivers/media/v4l2-core/v4l2-subdev.c | 36 ++++++++++++++++++++++++++++++++
>>  include/media/v4l2-subdev.h           |  4 ++++
>>  3 files changed, 50 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/media/v4l2-core/v4l2-device.c
>> b/drivers/media/v4l2-core/v4l2-device.c index 015f92a..5c8d402 100644
>> --- a/drivers/media/v4l2-core/v4l2-device.c
>> +++ b/drivers/media/v4l2-core/v4l2-device.c
>> @@ -183,12 +183,16 @@ int v4l2_device_register_subdev(struct v4l2_device
>> *v4l2_dev, if (err)
>>  		goto error_unregister;
>>
>> +	err = v4l2_subdev_create_pad_configs(sd);
>> +	if (err)
>> +		goto error_unregister;
>> +
> 
> I don't thin
> 
>>  #if defined(CONFIG_MEDIA_CONTROLLER)
>>  	/* Register the entity. */
>>  	if (v4l2_dev->mdev) {
>>  		err = media_device_register_entity(v4l2_dev->mdev, entity);
>>  		if (err < 0)
>> -			goto error_unregister;
>> +			goto error_free_pad_configs;
>>  	}
>>  #endif
>>
>> @@ -198,6 +202,9 @@ int v4l2_device_register_subdev(struct v4l2_device
>> *v4l2_dev,
>>
>>  	return 0;
>>
>> +error_free_pad_configs:
>> +	kfree(sd->pad_configs);
>> +	sd->pad_configs = NULL;
>>  error_unregister:
>>  	if (sd->internal_ops && sd->internal_ops->unregistered)
>>  		sd->internal_ops->unregistered(sd);
>> @@ -282,6 +289,8 @@ void v4l2_device_unregister_subdev(struct v4l2_subdev
>> *sd)
>>
>>  	if (sd->internal_ops && sd->internal_ops->unregistered)
>>  		sd->internal_ops->unregistered(sd);
>> +	kfree(sd->pad_configs);
>> +	sd->pad_configs = NULL;
>>  	sd->v4l2_dev = NULL;
>>
>>  #if defined(CONFIG_MEDIA_CONTROLLER)
>> diff --git a/drivers/media/v4l2-core/v4l2-subdev.c
>> b/drivers/media/v4l2-core/v4l2-subdev.c index 3c8b198..a7b874e 100644
>> --- a/drivers/media/v4l2-core/v4l2-subdev.c
>> +++ b/drivers/media/v4l2-core/v4l2-subdev.c
>> @@ -560,6 +560,41 @@ int v4l2_subdev_link_validate(struct media_link *link)
>>  EXPORT_SYMBOL_GPL(v4l2_subdev_link_validate);
>>  #endif /* CONFIG_MEDIA_CONTROLLER */
>>
>> +int v4l2_subdev_create_pad_configs(struct v4l2_subdev *sd)
>> +{
>> +#if defined(CONFIG_VIDEO_V4L2_SUBDEV_API)
>> +	struct v4l2_subdev_format fmt = {
>> +		.which = V4L2_SUBDEV_FORMAT_ACTIVE,
>> +	};
>> +	unsigned pads = sd->entity.num_pads;
>> +	unsigned pad;
>> +	int err;
>> +
>> +	if (pads == 0)
>> +		return 0;
>> +	sd->pad_configs = kcalloc(pads, sizeof(*sd->pad_configs), GFP_KERNEL);
>> +	if (sd->pad_configs == NULL)
>> +		return -ENOMEM;
>> +	for (pad = 0; pad < pads; pad++) {
>> +		fmt.pad = pad;
>> +		err = v4l2_subdev_call(sd, pad, get_fmt, NULL, &fmt);
>> +
>> +		if (err && err != -ENOIOCTLCMD) {
>> +			kfree(sd->pad_configs);
>> +			sd->pad_configs = NULL;
>> +			return err;
>> +		}
>> +		sd->pad_configs[pad].try_fmt = fmt.format;
>> +		sd->pad_configs[pad].try_crop.width = fmt.format.width;
>> +		sd->pad_configs[pad].try_crop.height = fmt.format.height;
>> +		sd->pad_configs[pad].try_compose.width = fmt.format.width;
>> +		sd->pad_configs[pad].try_compose.height = fmt.format.height;
>> +	}
>> +#endif
>> +	return 0;
>> +}
>> +EXPORT_SYMBOL_GPL(v4l2_subdev_create_pad_configs);
>> +
>>  void v4l2_subdev_init(struct v4l2_subdev *sd, const struct v4l2_subdev_ops
>> *ops) {
>>  	INIT_LIST_HEAD(&sd->list);
>> @@ -571,6 +606,7 @@ void v4l2_subdev_init(struct v4l2_subdev *sd, const
>> struct v4l2_subdev_ops *ops) sd->grp_id = 0;
>>  	sd->dev_priv = NULL;
>>  	sd->host_priv = NULL;
>> +	sd->pad_configs = NULL;
>>  #if defined(CONFIG_MEDIA_CONTROLLER)
>>  	sd->entity.name = sd->name;
>>  	sd->entity.type = MEDIA_ENT_T_V4L2_SUBDEV;
>> diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
>> index 175dc4f..e9e8b15 100644
>> --- a/include/media/v4l2-subdev.h
>> +++ b/include/media/v4l2-subdev.h
>> @@ -611,6 +611,8 @@ struct v4l2_subdev {
>>  	char name[V4L2_SUBDEV_NAME_SIZE];
>>  	/* can be used to group similar subdevs, value is driver-specific */
>>  	u32 grp_id;
>> +	/* Used by bridge drivers to allocate a pad config array */
>> +	struct v4l2_subdev_pad_config *pad_configs;
>>  	/* pointer to private data */
>>  	void *dev_priv;
>>  	void *host_priv;
>> @@ -689,6 +691,8 @@ int v4l2_subdev_link_validate_default(struct v4l2_subdev
>> *sd, struct v4l2_subdev_format *sink_fmt);
>>  int v4l2_subdev_link_validate(struct media_link *link);
>>  #endif /* CONFIG_MEDIA_CONTROLLER */
>> +
>> +int v4l2_subdev_create_pad_configs(struct v4l2_subdev *sd);
>>  void v4l2_subdev_init(struct v4l2_subdev *sd,
>>  		      const struct v4l2_subdev_ops *ops);
> 

  reply	other threads:[~2014-12-19 13:28 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-04  9:54 [RFC PATCH 0/8] Removing duplicate video/pad ops Hans Verkuil
2014-12-04  9:54 ` [RFC PATCH 1/8] v4l2 subdevs: replace get/set_crop by get/set_selection Hans Verkuil
2014-12-12 13:24   ` Sakari Ailus
2014-12-16 14:49   ` Prabhakar Lad
2014-12-04  9:54 ` [RFC PATCH 2/8] v4l2-subdev: drop get/set_crop pad ops Hans Verkuil
2014-12-12 13:27   ` Sakari Ailus
2014-12-16 14:52   ` Prabhakar Lad
2014-12-04  9:54 ` [RFC PATCH 3/8] v4l2-subdev: drop unused op enum_mbus_fmt Hans Verkuil
2014-12-07 22:57   ` Laurent Pinchart
2014-12-12 13:28   ` Sakari Ailus
2014-12-16 14:54   ` Prabhakar Lad
2014-12-18 22:08   ` Guennadi Liakhovetski
2014-12-19 11:37     ` Hans Verkuil
2014-12-04  9:54 ` [RFC PATCH 4/8] v4l2-subdev: replace v4l2_subdev_fh by v4l2_subdev_pad_config Hans Verkuil
2014-12-07 23:37   ` Laurent Pinchart
2014-12-04  9:54 ` [RFC PATCH 5/8] media/i2c/Kconfig: drop superfluous MEDIA_CONTROLLER Hans Verkuil
2014-12-07 23:38   ` Laurent Pinchart
2014-12-19 11:44     ` Hans Verkuil
2014-12-19 12:18       ` Laurent Pinchart
2014-12-19 12:21         ` Hans Verkuil
2014-12-20 18:56           ` Laurent Pinchart
2014-12-04  9:54 ` [RFC PATCH 6/8] v4l2-subdev: add v4l2_subdev_create_pad_configs Hans Verkuil
2014-12-07 23:50   ` Laurent Pinchart
2014-12-19 13:28     ` Hans Verkuil [this message]
2014-12-04  9:54 ` [RFC PATCH 7/8] v4l2-subdev: remove enum_framesizes/intervals Hans Verkuil
2014-12-04  9:54 ` [RFC PATCH 8/8] v4l2-subdev: remove g/s_crop and cropcap from video ops Hans Verkuil
2014-12-12 13:22   ` Sakari Ailus
2014-12-17  6:45 ` [RFC PATCH 0/8] Removing duplicate video/pad ops Prabhakar Lad

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=549427F3.6090109@xs4all.nl \
    --to=hverkuil@xs4all.nl \
    --cc=g.liakhovetski@gmx.de \
    --cc=hans.verkuil@cisco.com \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-media@vger.kernel.org \
    --cc=prabhakar.csengg@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;
as well as URLs for NNTP newsgroup(s).