From: Michael Grzeschik <m.grzeschik@pengutronix.de>
To: laurent.pinchart@ideasonboard.com
Cc: linux-usb@vger.kernel.org, linux-media@vger.kernel.org,
dan.scally@ideasonboard.com, gregkh@linuxfoundation.org,
nicolas@ndufresne.ca, kernel@pengutronix.de
Subject: [PATCH 3/3] usb: gadget: uvc: rework pump worker to avoid while loop
Date: Mon, 11 Sep 2023 02:24:51 +0200 [thread overview]
Message-ID: <20230911002451.2860049-4-m.grzeschik@pengutronix.de> (raw)
In-Reply-To: <20230911002451.2860049-1-m.grzeschik@pengutronix.de>
The uvc_video_enable function is calling cancel_work_sync which will be
blocking as long as new requests will be queued with the while loop. To
ensure an earlier stop in the pumping loop in this particular case we
rework the worker to requeue itself on every requests. Since the worker
is already running prioritized, the scheduling overhad did not have real
impact on the performance.
Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
---
drivers/usb/gadget/function/uvc_video.c | 20 ++++++++++++++------
1 file changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/usb/gadget/function/uvc_video.c b/drivers/usb/gadget/function/uvc_video.c
index 52e3666b51f743..bfda91ca73ddc6 100644
--- a/drivers/usb/gadget/function/uvc_video.c
+++ b/drivers/usb/gadget/function/uvc_video.c
@@ -397,7 +397,7 @@ static void uvcg_video_pump(struct work_struct *work)
bool buf_done;
int ret;
- while (video->ep->enabled && uvc->state == UVC_STATE_STREAMING) {
+ if (video->ep->enabled && uvc->state == UVC_STATE_STREAMING) {
/*
* Retrieve the first available USB request, protected by the
* request lock.
@@ -409,6 +409,11 @@ static void uvcg_video_pump(struct work_struct *work)
}
req = list_first_entry(&video->req_free, struct usb_request,
list);
+ if (!req) {
+ spin_unlock_irqrestore(&video->req_lock, flags);
+ return;
+ }
+
list_del(&req->list);
spin_unlock_irqrestore(&video->req_lock, flags);
@@ -437,7 +442,7 @@ static void uvcg_video_pump(struct work_struct *work)
* further.
*/
spin_unlock_irqrestore(&queue->irqlock, flags);
- break;
+ goto out;
}
/*
@@ -470,20 +475,23 @@ static void uvcg_video_pump(struct work_struct *work)
/* Queue the USB request */
ret = uvcg_video_ep_queue(video, req);
spin_unlock_irqrestore(&queue->irqlock, flags);
-
if (ret < 0) {
uvcg_queue_cancel(queue, 0);
- break;
+ goto out;
}
/* Endpoint now owns the request */
req = NULL;
video->req_int_count++;
+ } else {
+ return;
}
- if (!req)
- return;
+ if (uvc->state == UVC_STATE_STREAMING)
+ queue_work(video->async_wq, &video->pump);
+ return;
+out:
spin_lock_irqsave(&video->req_lock, flags);
list_add_tail(&req->list, &video->req_free);
spin_unlock_irqrestore(&video->req_lock, flags);
--
2.39.2
prev parent reply other threads:[~2023-09-11 0:25 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-11 0:24 [PATCH 0/3] usb: gadget: uvc: restart fixes Michael Grzeschik
2023-09-11 0:24 ` [PATCH 1/3] usb: gadget: uvc: stop pump thread on video disable Michael Grzeschik
2023-09-11 4:35 ` kernel test robot
2023-09-11 8:05 ` kernel test robot
2023-09-11 0:24 ` [PATCH 2/3] usb: gadget: uvc: cleanup request when not in correct state Michael Grzeschik
2023-09-12 4:52 ` Avichal Rakesh
2023-09-15 23:32 ` Michael Grzeschik
2023-09-16 2:41 ` Avichal Rakesh
2023-09-16 23:23 ` Michael Grzeschik
2023-09-18 19:02 ` Avichal Rakesh
2023-09-18 21:43 ` Michael Grzeschik
2023-09-18 23:40 ` Avichal Rakesh
2023-09-19 8:08 ` Avichal Rakesh
2023-09-19 19:13 ` Michael Grzeschik
2023-09-19 19:55 ` Avichal Rakesh
2023-09-19 20:07 ` Michael Grzeschik
2023-09-19 20:22 ` Avichal Rakesh
2023-09-19 21:16 ` Michael Grzeschik
2023-09-20 20:15 ` Avichal Rakesh
2023-09-11 0:24 ` Michael Grzeschik [this message]
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=20230911002451.2860049-4-m.grzeschik@pengutronix.de \
--to=m.grzeschik@pengutronix.de \
--cc=dan.scally@ideasonboard.com \
--cc=gregkh@linuxfoundation.org \
--cc=kernel@pengutronix.de \
--cc=laurent.pinchart@ideasonboard.com \
--cc=linux-media@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=nicolas@ndufresne.ca \
/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