public inbox for linux-media@vger.kernel.org
 help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
To: Hans Verkuil <hverkuil@xs4all.nl>
Cc: linux-media@vger.kernel.org, Hans Verkuil <hans.verkuil@cisco.com>
Subject: Re: [PATCHv13 04/28] media-request: add media_request_get_by_fd
Date: Tue, 8 May 2018 07:38:13 -0300	[thread overview]
Message-ID: <20180508073813.2b121eba@vento.lan> (raw)
In-Reply-To: <bad8f0a4-4e78-865a-9083-6dafb03722a1@xs4all.nl>

Em Tue, 8 May 2018 09:34:11 +0200
Hans Verkuil <hverkuil@xs4all.nl> escreveu:

> On 05/07/2018 07:01 PM, Mauro Carvalho Chehab wrote:
> > Em Thu,  3 May 2018 16:52:54 +0200
> > Hans Verkuil <hverkuil@xs4all.nl> escreveu:
> >   
> >> From: Hans Verkuil <hans.verkuil@cisco.com>
> >>
> >> Add media_request_get_by_fd() to find a request based on the file
> >> descriptor.
> >>
> >> The caller has to call media_request_put() for the returned
> >> request since this function increments the refcount.
> >>
> >> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
> >> ---
> >>  drivers/media/media-request.c | 32 ++++++++++++++++++++++++++++++++
> >>  include/media/media-request.h | 24 ++++++++++++++++++++++++
> >>  2 files changed, 56 insertions(+)
> >>
> >> diff --git a/drivers/media/media-request.c b/drivers/media/media-request.c
> >> index c216c4ab628b..edc1c3af1959 100644
> >> --- a/drivers/media/media-request.c
> >> +++ b/drivers/media/media-request.c
> >> @@ -218,6 +218,38 @@ static const struct file_operations request_fops = {
> >>  	.release = media_request_close,
> >>  };
> >>  
> >> +struct media_request *
> >> +media_request_get_by_fd(struct media_device *mdev, int request_fd)
> >> +{
> >> +	struct file *filp;
> >> +	struct media_request *req;
> >> +
> >> +	if (!mdev || !mdev->ops ||
> >> +	    !mdev->ops->req_validate || !mdev->ops->req_queue)
> >> +		return ERR_PTR(-EPERM);
> >> +
> >> +	filp = fget(request_fd);
> >> +	if (!filp)
> >> +		return ERR_PTR(-ENOENT);
> >> +
> >> +	if (filp->f_op != &request_fops)
> >> +		goto err_fput;
> >> +	req = filp->private_data;
> >> +	if (req->mdev != mdev)
> >> +		goto err_fput;
> >> +
> >> +	media_request_get(req);  
> > 
> > Hmm... this function changes the req struct (by calling kref_get) without
> > holding neither the mutex or the spin lock...  
> 
> kref_get is atomic, so why would you need to add additional locking?
> 
> Neither can the request be 'put' by another process before media_request_get()
> is called due to the fget() above: while the fd has a refcount > 0 the
> request refcount is also > 0. I'll add some comments to clarify this.

I see. Yeah, please add a comment here.

Thanks,
Mauro





> 
> >   
> >> +	fput(filp);
> >> +
> >> +	return req;
> >> +
> >> +err_fput:
> >> +	fput(filp);
> >> +
> >> +	return ERR_PTR(-ENOENT);
> >> +}
> >> +EXPORT_SYMBOL_GPL(media_request_get_by_fd);
> >> +
> >>  int media_request_alloc(struct media_device *mdev,
> >>  			struct media_request_alloc *alloc)
> >>  {
> >> diff --git a/include/media/media-request.h b/include/media/media-request.h
> >> index e39122dfd717..997e096d7128 100644
> >> --- a/include/media/media-request.h
> >> +++ b/include/media/media-request.h
> >> @@ -86,6 +86,24 @@ static inline void media_request_get(struct media_request *req)
> >>   */
> >>  void media_request_put(struct media_request *req);
> >>  
> >> +/**
> >> + * media_request_get_by_fd - Get a media request by fd
> >> + *
> >> + * @mdev: Media device this request belongs to
> >> + * @request_fd: The file descriptor of the request
> >> + *
> >> + * Get the request represented by @request_fd that is owned
> >> + * by the media device.
> >> + *
> >> + * Return a -EPERM error pointer if requests are not supported
> >> + * by this driver. Return -ENOENT if the request was not found.
> >> + * Return the pointer to the request if found: the caller will
> >> + * have to call @media_request_put when it finished using the
> >> + * request.  
> > 
> > ... so, it should be said here how this should be serialized, in order
> > to avoid it to be destroyed by a task while some other task might be
> > trying to instantiate it.  
> 
> This doesn't need any serialization. If the request_fd is found, then
> it will return the request with the request refcount increased.
> 
> I also do not understand what you mean with "some other task might be
> trying to instantiate it".
> 
> I think there is some misunderstanding here.
> 
> Regards,
> 
> 	Hans
> 
> 
> >   
> >> + */
> >> +struct media_request *
> >> +media_request_get_by_fd(struct media_device *mdev, int request_fd);
> >> +
> >>  /**
> >>   * media_request_alloc - Allocate the media request
> >>   *
> >> @@ -107,6 +125,12 @@ static inline void media_request_put(struct media_request *req)
> >>  {
> >>  }
> >>  
> >> +static inline struct media_request *
> >> +media_request_get_by_fd(struct media_device *mdev, int request_fd)
> >> +{
> >> +	return ERR_PTR(-EPERM);
> >> +}
> >> +
> >>  #endif
> >>  
> >>  /**  
> > 
> > 
> > 
> > Thanks,
> > Mauro
> >   
> 



Thanks,
Mauro

  reply	other threads:[~2018-05-08 10:38 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-03 14:52 [PATCHv13 00/28] Request API Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 01/28] v4l2-device.h: always expose mdev Hans Verkuil
2018-05-04 10:51   ` Sakari Ailus
2018-05-07 15:46     ` Mauro Carvalho Chehab
2018-05-08  8:34       ` Hans Verkuil
2018-05-16  3:40   ` Laurent Pinchart
2018-05-03 14:52 ` [PATCHv13 02/28] uapi/linux/media.h: add request API Hans Verkuil
2018-05-04 10:51   ` Sakari Ailus
2018-05-18 14:49   ` Laurent Pinchart
2018-05-03 14:52 ` [PATCHv13 03/28] media-request: implement media requests Hans Verkuil
2018-05-04 12:27   ` Sakari Ailus
2018-05-08 10:21     ` Mauro Carvalho Chehab
2018-05-08 10:52       ` Sakari Ailus
2018-05-08 12:41         ` Mauro Carvalho Chehab
2018-05-08 13:21           ` Sakari Ailus
2018-05-24 11:19       ` Hans Verkuil
2018-05-24  9:26     ` Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 04/28] media-request: add media_request_get_by_fd Hans Verkuil
2018-05-04 12:26   ` Sakari Ailus
2018-05-07 17:01   ` Mauro Carvalho Chehab
2018-05-08  7:34     ` Hans Verkuil
2018-05-08 10:38       ` Mauro Carvalho Chehab [this message]
2018-05-03 14:52 ` [PATCHv13 05/28] media-request: add media_request_object_find Hans Verkuil
2018-05-04 12:43   ` Sakari Ailus
2018-05-07 17:05     ` Mauro Carvalho Chehab
2018-05-24  9:36       ` Hans Verkuil
2018-05-08 10:54     ` Sakari Ailus
2018-05-24  9:28     ` Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 06/28] v4l2-dev: lock req_queue_mutex Hans Verkuil
2018-05-07 17:20   ` Mauro Carvalho Chehab
2018-05-08  7:45     ` Hans Verkuil
2018-05-08 10:45       ` Mauro Carvalho Chehab
2018-05-24  9:51         ` Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 07/28] videodev2.h: add request_fd field to v4l2_ext_controls Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 08/28] v4l2-ctrls: v4l2_ctrl_add_handler: add from_other_dev Hans Verkuil
2018-05-03 14:52 ` [PATCHv13 09/28] v4l2-ctrls: prepare internal structs for request API Hans Verkuil
2018-05-07 17:35   ` Mauro Carvalho Chehab
2018-05-08  7:49     ` Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 10/28] v4l2-ctrls: alloc memory for p_req Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 11/28] v4l2-ctrls: use ref in helper instead of ctrl Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 12/28] v4l2-ctrls: add core request support Hans Verkuil
2018-05-07 18:06   ` Mauro Carvalho Chehab
2018-05-08  8:07     ` Hans Verkuil
2018-05-08 10:49       ` Mauro Carvalho Chehab
2018-05-24 10:27         ` Hans Verkuil
2018-05-16 10:19   ` Sakari Ailus
2018-05-16 10:46     ` Sakari Ailus
2018-05-16 10:55     ` Hans Verkuil
2018-05-16 11:18   ` Sakari Ailus
2018-05-03 14:53 ` [PATCHv13 13/28] v4l2-ctrls: support g/s_ext_ctrls for requests Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 14/28] videodev2.h: Add request_fd field to v4l2_buffer Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 15/28] vb2: store userspace data in vb2_v4l2_buffer Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 16/28] videobuf2-core: embed media_request_object Hans Verkuil
2018-05-08  9:54   ` Mauro Carvalho Chehab
2018-05-03 14:53 ` [PATCHv13 17/28] videobuf2-core: integrate with media requests Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 18/28] videobuf2-v4l2: " Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 19/28] videobuf2-core: add request helper functions Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 20/28] videobuf2-v4l2: add vb2_request_queue/validate helpers Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 21/28] v4l2-mem2mem: add vb2_m2m_request_queue Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 22/28] Documentation: v4l: document request API Hans Verkuil
2018-05-18 14:46   ` Laurent Pinchart
2018-05-24  4:32     ` Tomasz Figa
2018-05-24  7:55       ` Hans Verkuil
2018-05-24 14:46     ` Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 23/28] media: vim2m: add media device Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 24/28] vim2m: use workqueue Hans Verkuil
2018-05-04 11:34   ` Sakari Ailus
2018-05-03 14:53 ` [PATCHv13 25/28] vim2m: support requests Hans Verkuil
2018-05-17 20:41   ` Sakari Ailus
2018-05-03 14:53 ` [PATCHv13 26/28] vivid: add mc Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 27/28] vivid: add request support Hans Verkuil
2018-05-03 14:53 ` [PATCHv13 28/28] RFC: media-requests: add debugfs node Hans Verkuil
2018-05-08 10:26 ` [PATCHv13 00/28] Request API Mauro Carvalho Chehab

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=20180508073813.2b121eba@vento.lan \
    --to=mchehab+samsung@kernel.org \
    --cc=hans.verkuil@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox