lkml.org 
[lkml]   [2015]   [Mar]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH 3/3] media: omap3isp: ispvideo: use vb2_fop_mmap/poll
Date
Hi Prabhakar,

On Tuesday 24 February 2015 02:23:26 Laurent Pinchart wrote:
> On Monday 23 February 2015 20:19:33 Lad Prabhakar wrote:
> > From: "Lad, Prabhakar" <prabhakar.csengg@gmail.com>
> >
> > No need to reinvent the wheel. Just use the already existing
> > functions provided v4l-core.
> >
> > Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com>
> > ---
> >
> > drivers/media/platform/omap3isp/ispvideo.c | 30 ++++---------------------
> > 1 file changed, 4 insertions(+), 26 deletions(-)
> >
> > diff --git a/drivers/media/platform/omap3isp/ispvideo.c
> > b/drivers/media/platform/omap3isp/ispvideo.c index b648176..5dd5ffc 100644
> > --- a/drivers/media/platform/omap3isp/ispvideo.c
> > +++ b/drivers/media/platform/omap3isp/ispvideo.c
> > @@ -1277,37 +1277,13 @@ static int isp_video_release(struct file *file)
> > return ret;
> > }
> >
> > -static unsigned int isp_video_poll(struct file *file, poll_table *wait)
> > -{
> > - struct isp_video *video = video_drvdata(file);
> > - int ret;
> > -
> > - mutex_lock(&video->queue_lock);
> > - ret = vb2_poll(&video->queue, file, wait);
> > - mutex_unlock(&video->queue_lock);
> > -
> > - return ret;
> > -}
>
> This depends on patch 2/3, which can't be accepted as-is for now.
>
> > -static int isp_video_mmap(struct file *file, struct vm_area_struct *vma)
> > -{
> > - struct isp_video *video = video_drvdata(file);
> > - int ret;
> > -
> > - mutex_lock(&video->queue_lock);
> > - ret = vb2_mmap(&video->queue, vma);
> > - mutex_unlock(&video->queue_lock);
> > -
> > - return ret;
> > -}
>
> This should be good but has the side effect of removing locking in
> isp_video_mmap(). Now, I think that's the right thing to do, but it should
> be done in a separate patch first with a proper explanation. I can do so,
> or you can submit an additional patch.

After testing the change I realized it also depends on patch 2/3, as video-
>queue is only set at streamon time at the moment. I'm thus dropping the patch
from my queue.

> > static struct v4l2_file_operations isp_video_fops = {
> > .owner = THIS_MODULE,
> > .unlocked_ioctl = video_ioctl2,
> > .open = isp_video_open,
> > .release = isp_video_release,
> > - .poll = isp_video_poll,
> > - .mmap = isp_video_mmap,
> > + .poll = vb2_fop_poll,
> > + .mmap = vb2_fop_mmap,
> > };
> >
> > /* ----------------------------------------------------------------------
> > @@ -1389,6 +1365,8 @@ int omap3isp_video_register(struct isp_video
> > *video, struct v4l2_device *vdev)
> >
> > video->video.v4l2_dev = vdev;
> >
> > + /* queue isnt initalized */
> > + video->video.queue = &video->queue;
> > ret = video_register_device(&video->video, VFL_TYPE_GRABBER, -1);
> > if (ret < 0)
> > dev_err(video->isp->dev,

--
Regards,

Laurent Pinchart



\
 
 \ /
  Last update: 2015-03-08 21:41    [W:0.051 / U:7.336 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site