This is a note to let you know that I've just added the patch titled
media: em28xx-v4l: give back all active video buffers to the vb2 core properly on streaming stop
to the 3.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
media-em28xx-v4l-give-back-all-active-video-buffers-to-the-vb2-core-properly-on-streaming-stop.patch
and it can be found in the queue-3.10 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 627530c32a43283474e9dd3e954519410ffa033a Mon Sep 17 00:00:00 2001
From: Frank Schaefer <fschaefer.oss@xxxxxxxxxxxxxx>
Date: Sat, 9 Aug 2014 06:37:20 -0300
Subject: media: em28xx-v4l: give back all active video buffers to the vb2 core properly on streaming stop
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
From: Frank Schaefer <fschaefer.oss@xxxxxxxxxxxxxx>
commit 627530c32a43283474e9dd3e954519410ffa033a upstream.
When a new video frame is started, the driver takes the next video buffer from
the list of active buffers and moves it to dev->usb_ctl.vid_buf / dev->usb_ctl.vbi_buf
for further processing.
On streaming stop we currently only give back the pending buffers from the list
but not the ones which are currently processed.
This causes the following warning from the vb2 core since kernel 3.15:
...
------------[ cut here ]------------
WARNING: CPU: 1 PID: 2284 at drivers/media/v4l2-core/videobuf2-core.c:2115 __vb2_queue_cancel+0xed/0x150 [videobuf2_core]()
[...]
Call Trace:
[<c0769c46>] dump_stack+0x48/0x69
[<c0245b69>] warn_slowpath_common+0x79/0x90
[<f925e4ad>] ? __vb2_queue_cancel+0xed/0x150 [videobuf2_core]
[<f925e4ad>] ? __vb2_queue_cancel+0xed/0x150 [videobuf2_core]
[<c0245bfd>] warn_slowpath_null+0x1d/0x20
[<f925e4ad>] __vb2_queue_cancel+0xed/0x150 [videobuf2_core]
[<f925fa35>] vb2_internal_streamoff+0x35/0x90 [videobuf2_core]
[<f925fac5>] vb2_streamoff+0x35/0x60 [videobuf2_core]
[<f925fb27>] vb2_ioctl_streamoff+0x37/0x40 [videobuf2_core]
[<f8e45895>] v4l_streamoff+0x15/0x20 [videodev]
[<f8e4925d>] __video_do_ioctl+0x23d/0x2d0 [videodev]
[<f8e49020>] ? video_ioctl2+0x20/0x20 [videodev]
[<f8e48c63>] video_usercopy+0x203/0x5a0 [videodev]
[<f8e49020>] ? video_ioctl2+0x20/0x20 [videodev]
[<c039d0e7>] ? fsnotify+0x1e7/0x2b0
[<f8e49012>] video_ioctl2+0x12/0x20 [videodev]
[<f8e49020>] ? video_ioctl2+0x20/0x20 [videodev]
[<f8e4461e>] v4l2_ioctl+0xee/0x130 [videodev]
[<f8e44530>] ? v4l2_open+0xf0/0xf0 [videodev]
[<c0378de2>] do_vfs_ioctl+0x2e2/0x4d0
[<c0368eec>] ? vfs_write+0x13c/0x1c0
[<c0369a8f>] ? vfs_writev+0x2f/0x50
[<c0379028>] SyS_ioctl+0x58/0x80
[<c076fff3>] sysenter_do_call+0x12/0x12
---[ end trace 5545f934409f13f4 ]---
...
Many thanks to Hans Verkuil, whose recently added check in the vb2 core unveiled
this long standing issue and who has investigated it further.
Signed-off-by: Frank Schäfer <fschaefer.oss@xxxxxxxxxxxxxx>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@xxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/media/usb/em28xx/em28xx-video.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
--- a/drivers/media/usb/em28xx/em28xx-video.c
+++ b/drivers/media/usb/em28xx/em28xx-video.c
@@ -696,13 +696,16 @@ static int em28xx_stop_streaming(struct
}
spin_lock_irqsave(&dev->slock, flags);
+ if (dev->usb_ctl.vid_buf != NULL) {
+ vb2_buffer_done(&dev->usb_ctl.vid_buf->vb, VB2_BUF_STATE_ERROR);
+ dev->usb_ctl.vid_buf = NULL;
+ }
while (!list_empty(&vidq->active)) {
struct em28xx_buffer *buf;
buf = list_entry(vidq->active.next, struct em28xx_buffer, list);
list_del(&buf->list);
vb2_buffer_done(&buf->vb, VB2_BUF_STATE_ERROR);
}
- dev->usb_ctl.vid_buf = NULL;
spin_unlock_irqrestore(&dev->slock, flags);
return 0;
@@ -724,13 +727,16 @@ int em28xx_stop_vbi_streaming(struct vb2
}
spin_lock_irqsave(&dev->slock, flags);
+ if (dev->usb_ctl.vbi_buf != NULL) {
+ vb2_buffer_done(&dev->usb_ctl.vbi_buf->vb, VB2_BUF_STATE_ERROR);
+ dev->usb_ctl.vbi_buf = NULL;
+ }
while (!list_empty(&vbiq->active)) {
struct em28xx_buffer *buf;
buf = list_entry(vbiq->active.next, struct em28xx_buffer, list);
list_del(&buf->list);
vb2_buffer_done(&buf->vb, VB2_BUF_STATE_ERROR);
}
- dev->usb_ctl.vbi_buf = NULL;
spin_unlock_irqrestore(&dev->slock, flags);
return 0;
Patches currently in stable-queue which might be from fschaefer.oss@xxxxxxxxxxxxxx are
queue-3.10/media-em28xx-v4l-give-back-all-active-video-buffers-to-the-vb2-core-properly-on-streaming-stop.patch
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html