All of lore.kernel.org
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Michael Grzeschik <m.grzeschik@pengutronix.de>
Cc: linux-usb@vger.kernel.org, balbi@kernel.org,
	paul.elder@ideasonboard.com, kernel@pengutronix.de,
	caleb.connolly@ideasonboard.com
Subject: Re: [PATCH 5/7] usb: gadget: uvc: only pump video data if necessary
Date: Tue, 5 Oct 2021 01:29:25 +0300	[thread overview]
Message-ID: <YVuARarduyZhwio8@pendragon.ideasonboard.com> (raw)
In-Reply-To: <20210930102717.15720-6-m.grzeschik@pengutronix.de>

Hi Michael,

Thank you for the patch.

On Thu, Sep 30, 2021 at 12:27:15PM +0200, Michael Grzeschik wrote:
> If the streaming endpoint is not enabled, the worker has nothing to do.
> In the case it still got enqueued, this patch ensueres that it will bail

s/it still got enqueued/buffers are still queued/
s/ensueres/ensures/

> out without handling any data.

Does this happen when uvc_function_set_alt() calls usb_ep_disable() ?
The current implementation will cause usb_ep_queue() (called from
uvcg_video_ep_queue(), from uvcg_video_pump()) to return an error in
that case, which will result in uvcg_queue_cancel() being called in
uvcg_video_pump(). With this patch, I believe this will still work fine
as userspace is notified of the stream off event and calls
VIDIOC_STREAMOFF, which in turn calls uvcg_video_enable(0) from
uvc_v4l2_streamoff() (or uvcg_video_enable(0) gets called from
uvc_v4l2_release() in the worst case if the application closes the
device). Could you confirm that your understanding matches this analysis
? If so,

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
> ---
>  drivers/usb/gadget/function/uvc_video.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/gadget/function/uvc_video.c b/drivers/usb/gadget/function/uvc_video.c
> index ccee35177411d..152647495fa61 100644
> --- a/drivers/usb/gadget/function/uvc_video.c
> +++ b/drivers/usb/gadget/function/uvc_video.c
> @@ -335,12 +335,12 @@ static void uvcg_video_pump(struct work_struct *work)
>  {
>  	struct uvc_video *video = container_of(work, struct uvc_video, pump);
>  	struct uvc_video_queue *queue = &video->queue;
> -	struct usb_request *req;
> +	struct usb_request *req = NULL;
>  	struct uvc_buffer *buf;
>  	unsigned long flags;
>  	int ret;
>  
> -	while (1) {
> +	while (video->ep->enabled) {
>  		/* Retrieve the first available USB request, protected by the
>  		 * request lock.
>  		 */
> @@ -390,6 +390,9 @@ static void uvcg_video_pump(struct work_struct *work)
>  		video->req_int_count++;
>  	}
>  
> +	if (!req)
> +		return;
> +
>  	spin_lock_irqsave(&video->req_lock, flags);
>  	list_add_tail(&req->list, &video->req_free);
>  	spin_unlock_irqrestore(&video->req_lock, flags);

-- 
Regards,

Laurent Pinchart

  parent reply	other threads:[~2021-10-04 22:29 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-30 10:27 [PATCH 0/7] usb: gadget: uvc: smaller fixes for stability Michael Grzeschik
2021-09-30 10:27 ` [PATCH 1/7] usb: gadget: uvc: consistently use define for headerlen Michael Grzeschik
2021-10-01  2:36   ` paul.elder
2021-10-04 22:08   ` Laurent Pinchart
2021-09-30 10:27 ` [PATCH 2/7] usb: gadget: uvc: rename function to be more consistent Michael Grzeschik
2021-10-01  2:41   ` paul.elder
2021-10-04 22:13   ` Laurent Pinchart
2021-09-30 10:27 ` [PATCH 3/7] usb: gadget: uvc: test if ep->desc is valid on ep_queue Michael Grzeschik
2021-10-01  2:47   ` paul.elder
2021-10-04 22:15   ` Laurent Pinchart
2021-09-30 10:27 ` [PATCH 4/7] usb: gadget: uvc: only schedule stream in streaming state Michael Grzeschik
2021-10-01  2:54   ` paul.elder
2021-10-04 22:21   ` Laurent Pinchart
2021-09-30 10:27 ` [PATCH 5/7] usb: gadget: uvc: only pump video data if necessary Michael Grzeschik
2021-10-01  3:03   ` paul.elder
2021-10-04 22:29   ` Laurent Pinchart [this message]
2021-10-17 21:11     ` Michael Grzeschik
2021-09-30 10:27 ` [PATCH 6/7] usb: gadget: uvc: ensure the vdev is unset Michael Grzeschik
2021-10-01  3:07   ` paul.elder
2021-10-04 22:31   ` Laurent Pinchart
2021-09-30 10:27 ` [PATCH 7/7] usb: gadget: udc: ensure the gadget is still available Michael Grzeschik
2021-10-04 22:41   ` Laurent Pinchart

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=YVuARarduyZhwio8@pendragon.ideasonboard.com \
    --to=laurent.pinchart@ideasonboard.com \
    --cc=balbi@kernel.org \
    --cc=caleb.connolly@ideasonboard.com \
    --cc=kernel@pengutronix.de \
    --cc=linux-usb@vger.kernel.org \
    --cc=m.grzeschik@pengutronix.de \
    --cc=paul.elder@ideasonboard.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.