Hi Michael,
On Monday 29 April 2013 17:56:50 Michael Grzeschik wrote:
> On Mon, Feb 11, 2013 at 08:42:37PM +0100, Laurent Pinchart wrote:
> > On Monday 04 February 2013 11:58:27 Michael Grzeschik wrote:
> > > On Thu, Jan 17, 2013 at 04:23:51PM +0530, Bhupesh Sharma wrote:
> > > > This patch reworks the videobuffer management logic present in the UVC
> > > > webcam gadget and ports it to use the "more apt" videobuf2 framework
> > > > for video buffer management.
> > > >
> > > > To support routing video data captured from a real V4L2 video capture
> > > > device with a "zero copy" operation on videobuffers (as they pass from
> > > > the V4L2 domain to UVC domain via a user-space application), we need
> > > > to support USER_PTR IO method at the UVC gadget side.
> > > >
> > > > So the V4L2 capture device driver can still continue to use MMAP IO
> > > > method and now the user-space application can just pass a pointer to
> > > > the video buffers being dequeued from the V4L2 device side while
> > > > queueing them at the UVC gadget end. This ensures that we have a
> > > > "zero-copy" design as the videobuffers pass from the V4L2 capture
> > > > device to the UVC gadget.
> > > >
> > > > Note that there will still be a need to apply UVC specific payload
> > > > headers on top of each UVC payload data, which will still require a
> > > > copy operation to be performed in the 'encode' routines of the UVC
> > > > gadget.
> > > >
> > > > This patch also addresses one issue found out while porting the UVC
> > > >
> > > > gadget to videobuf2 framework:
> > > > - In case the usb requests queued by the gadget get completed
> > > > with a status of -ESHUTDOWN (disconnected from host),
> > > > the queue of videobuf2 should be cancelled to ensure that the
> > > > application space daemon is not left in a state waiting for
> > > > a vb2 to be successfully absorbed at the USB side.
> > > >
> > > > Signed-off-by: Bhupesh Sharma <bhupesh.sharma@xxxxxx>
> > > > ---
> > > >
> > > > drivers/usb/gadget/Kconfig | 1 +
> > > > drivers/usb/gadget/uvc_queue.c | 537 +++++++++++--------------------
> > > > drivers/usb/gadget/uvc_queue.h | 25 +--
> > > > drivers/usb/gadget/uvc_v4l2.c | 27 +--
> > > > drivers/usb/gadget/uvc_video.c | 18 +-
> > > > 5 files changed, 189 insertions(+), 419 deletions(-)
> >
> > [snip]
> >
> > > > diff --git a/drivers/usb/gadget/uvc_queue.c
> > > > b/drivers/usb/gadget/uvc_queue.c index 104ae9c..bd20fab 100644
> > > > --- a/drivers/usb/gadget/uvc_queue.c
> > > > +++ b/drivers/usb/gadget/uvc_queue.c
> >
> > [snip]
> >
> > > > -static int uvc_queue_waiton(struct uvc_buffer *buf, int nonblocking)
> > > > +static int uvc_queue_buffer(struct uvc_video_queue *queue,
> > > > + struct v4l2_buffer *buf)
> > > >
> > > > {
> > > >
> > > > - if (nonblocking) {
> > > > - return (buf->state != UVC_BUF_STATE_QUEUED &&
> > > > - buf->state != UVC_BUF_STATE_ACTIVE)
> > > > - ? 0 : -EAGAIN;
> > > > - }
> > > > + int ret;
> > > >
> > > > - return wait_event_interruptible(buf->wait,
> > > > - buf->state != UVC_BUF_STATE_QUEUED &&
> > > > - buf->state != UVC_BUF_STATE_ACTIVE);
> > > > + mutex_lock(&queue->mutex);
> > > > + ret = vb2_qbuf(&queue->queue, buf);
> > > > + mutex_unlock(&queue->mutex);
> > > > +
> > >
> > > How is the UVC_QUEUE_PAUSED handling supposed to be handled here?
> > >
> > > I see that this patch lost this hunk from uvc_queue_buffer():
> > > ret |= (queue->flags & uvc_queue_paused) != 0;
> > > queue->flags &= ~UVC_QUEUE_PAUSED;
> >
> > Good catch. This needs to be handled. The PAUSED flag needs to be set with
> > the irqlock held, this could be done in uvc_buffer_queue(). The return
> > value needs to be computed when setting the flag, so it would need to be
> > stored in the queue structure and read here. Adding a new UVC_QUEUE_PUMP
> > flag should do but will require taking the irqlock again to read and
> > clear that flag here. A new field is another possible solution.
>
> the handling of UVC_QUEUE_PAUSED is currently implemented like this:
>
> ret = (queue->flags & UVC_QUEUE_PAUSED) != 0
> queue->flags &= ~UVC_QUEUE_PAUSED;
>
> which is wrong as it overwrites the return value of vb2_qbuf
> which will lead to an stalled connection if the return value
> would be negative and got overwritten to "1".
Oops, indeed.
> What do you mean with the extra flag UVC_QUEUE_PUMP. I would like
> to implement a fixup patch, but need more information how to
> correctly handle the extra return value here.
Would the following patch fix the issue ?
>From f2349730e362f25e4a9661d1121a09ac02027889 Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
Date: Mon, 29 Apr 2013 22:18:01 +0200
Subject: [PATCH] usb: gadget/uvc: Fix error handling in uvc_queue_buffer()
The conversion to videobuf2 failed to check the return value of
vb2_qbuf(). Fix it.
Reported-by: Michael Grzeschik <mgr@xxxxxxxxxxxxxx>
Signed-off-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
---
drivers/usb/gadget/uvc_queue.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/usb/gadget/uvc_queue.c b/drivers/usb/gadget/uvc_queue.c
index e617047..0bb5d50 100644
--- a/drivers/usb/gadget/uvc_queue.c
+++ b/drivers/usb/gadget/uvc_queue.c
@@ -193,12 +193,16 @@ static int uvc_queue_buffer(struct uvc_video_queue *queue,
mutex_lock(&queue->mutex);
ret = vb2_qbuf(&queue->queue, buf);
+ if (ret < 0)
+ goto done;
+
spin_lock_irqsave(&queue->irqlock, flags);
ret = (queue->flags & UVC_QUEUE_PAUSED) != 0;
queue->flags &= ~UVC_QUEUE_PAUSED;
spin_unlock_irqrestore(&queue->irqlock, flags);
- mutex_unlock(&queue->mutex);
+done:
+ mutex_unlock(&queue->mutex);
return ret;
}
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
