gst_buffer_get_meta: assertion `buffer != NULL' failed

[umlaeute]

gst_buffer_get_meta: assertion `buffer != NULL' failed

[umlaeute]

could you try with the latest "master" git repository? (post [b253016] commit)

[vvinothkumar]

I tried on the latest master (fd822cf). The error is still there.

On Tue, Mar 26, 2013 at 9:38 PM, umlaeute [email protected] wrote:

could you try with the latest "master" git repository? (post [b253016https://github.com/umlaeute/v4l2loopback/commit/b25301689b9]
commit)

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/39#issuecomment-15467846
.

[vvinothkumar]

Hi Umlaeute,

If you give some pointers, i can also work on it.

Thanks,
Vinoth Kumar

On Tue, Mar 26, 2013 at 9:55 PM, Vinothkumar Vivekananda
[email protected]:

I tried on the latest master (fd822cf). The error is still there.

On Tue, Mar 26, 2013 at 9:38 PM, umlaeute [email protected]:

could you try with the latest "master" git repository? (post [b253016https://github.com/umlaeute/v4l2loopback/commit/b25301689b9]
commit)

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/39#issuecomment-15467846
.

[vvinothkumar]

Hi Umlaeute,

I posted this bug in Gnome Bugzilla - GStreamer and got a reply.
"
v4l2: free all queued buffers
Don't just loop over the first num_queued buffers but loop over
all the buffers and check if they need to be freed. It is possible that
not all buffers are queued and then the entry in our array will be NULL.
Those buffers that are not queued were freed in stop().
"
Hope this might help you.

Thanks,
Vinoth Kumar

[vvinothkumar]

Gnome bugzilla link - https://bugzilla.gnome.org/show_bug.cgi?id=696651

Thanks,
Vinoth Kumar

[umlaeute]

do in understand correctly that the bug was really in gstreamer and has been fixed there?

[vvinothkumar]

Hi Umlaeute,

Gstreamer throws error when buffer used by v4l2 driver is not freed while
terminating the running pipeline. So, bug is with v4l2 driver.

Thanks,
Vinoth Kumar
On 12 Apr 2013 14:43, "umlaeute" [email protected] wrote:

do in understand correctly that the bug was really in gstreamer and has
been fixed there?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/39#issuecomment-16283213
.

[umlaeute]

i'm still unable to reproduce the problem. i'm aware that you cannot trigger the bug reliably every time, but at least you manage to trigger it "often". i have never seen/noticed the problem yet.

please give a detailed description what you do.
which parameters do you pass to the v4l2loopback module?
what's the content of the files in /sys/devices/virtual/video4linux/video1/ ...?

[vvinothkumar]

Hi Umlaeute,

*What i do, *
I am trying to stream the video from my system to my media server. The
reason i am in need of v4l2loopback driver is, i need two video devices.
One, is essentially used for streaming to media server and two, is used for
video preview at the user end. So i create a virtual video device and
writes the video device content into the virtual video device.

Parameters to v4l2loopback,
I am not passing any parameters to it. I just execute sudo modprobe -v v4l2loopback

Content of the files in /sys/devices/virtual/video4linux/video0/
'''
buffers - 4
dev - 81:0
format -
index - 0
max_openers - 10
name - Loopback video device 0
uevent - MAJOR=81 MINOR=0 DEVNAME=video0

power/
async - disabled
autosuspend_delay_ms -
control - auto
runtime_active_kids - 0
runtime_active_time - 0
runtime_enabled - disabled
runtime_status - unsupported
runtime_suspended_time - 0
runtime_usage - 0
'''

*Additional information, *
*System info, *
Release : Ubuntu 12.10 (quantal)
GNOME : 3.6.0 (Ubuntu 2012-10-04)
kernel : 3.5.0-30-generic (#51-Ubuntu SMP Tue May 14 18:49:52 UTC 2013)
cpu : 2 @1.83GHz
ram : 3029 MiB

*While executing the pipeline, *

GST_DEBUG=3 gst-launch-1.0 videotestsrc ! v4l2sink device=/dev/video0

Setting pipeline to PAUSED ...
libv4l2: error getting pixformat: Invalid argument
Pipeline is PREROLLING ...
0:00:00.045666755  9285  0x88886c0 WARN                 default
gstutils.c:3622:gst_pad_create_stream_id_printf_valist:<videotestsrc0:src>
Creating random stream-id, consider implementing a deterministic way of
creating a stream-id
0:00:00.046099993  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YUYV@48x32: 0
0:00:00.046229483  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YUYV@48x32: 0
0:00:00.046289967  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at UYVY@48x32: 0
0:00:00.046341232  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YU12@48x32: 0
0:00:00.046396618  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YV12@48x32: 0
0:00:00.046452073  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YVU9@48x32: 0
0:00:00.046502640  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YUV9@48x32: 0
0:00:00.046553486  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at RGB4@48x32: 0
0:00:00.046605379  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at BGR4@48x32: 0
0:00:00.046655876  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at RGB3@48x32: 0
0:00:00.046705884  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at BGR3@48x32: 0
0:00:00.046755682  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at GREY@48x32: 0
0:00:00.046807366  9285  0x88886c0 WARN                    v4l2
gstv4l2object.c:1790:gst_v4l2_object_probe_caps_for_format_and_size:<v4l2sink0>
Unknown frame interval type at YVYU@48x32: 0
Pipeline is PREROLLED ...
Setting pipeline to PLAYING ...
New clock: GstSystemClock
^CCaught interrupt -- handling interrupt.
Interrupt: Stopping pipeline ...
Execution ended after 5507443258 ns.
Setting pipeline to PAUSED ...
Setting pipeline to READY ...

(gst-launch-1.0:9285): GStreamer-CRITICAL **: gst_buffer_get_meta:
assertion `buffer != NULL' failed
Caught SIGSEGV accessing address 0xc

[umlaeute]

thanks for the quick reply.
could you also post the content of the files in /sys/devices/virtual/video4linux/video1/ while the pipeline is running?

[vvinothkumar]

Content of /sys/devices/virtual/video4linux/, i have video0 alone.
But i have video0 and video1 in /sys/devices/virtual/video4linux/video0/subsystem/.

##
vinothkumar@vinothkumar-ThinkPad-T60:/sys/devices/virtual/video4linux$ ls -R
.:
video0

./video0:
buffers  dev  format  index  max_openers  name  power  subsystem  uevent

./video0/power:
async                 control              runtime_active_time
 runtime_status          runtime_usage
autosuspend_delay_ms  runtime_active_kids  runtime_enabled
 runtime_suspended_time
##

Do you want the content of symlink
/sys/devices/virtual/video4linux/video0/subsystem/video1 ?

Thanks,
Vinoth

[umlaeute]

sorry to be unclear, what i really want is the information in /sys about the device your are using as loopback device. (since your original post was about /dev/video1 i continued to use that, but it seems that you switched to /dev/video0 in the meantime). i'm also not really interested in the information in any subdirectories (power/ and subsystem/).
so, if you use /dev/video0 as your sink, i'd be interested in the output of:

grep . /sys/devices/virtual/video4linux/video0/*

[vvinothkumar]

Really sorry, I should have mentioned my sink device. Here you go,

vinothkumar@vinothkumar-ThinkPad-T60:~$ grep . /sys/devices/virtual/video4linux/video0/*
/sys/devices/virtual/video4linux/video0/buffers:2
/sys/devices/virtual/video4linux/video0/dev:81:0
/sys/devices/virtual/video4linux/video0/format:YUYV:320x240@30
/sys/devices/virtual/video4linux/video0/index:0
/sys/devices/virtual/video4linux/video0/max_openers:10
/sys/devices/virtual/video4linux/video0/name:Loopback video device 0
grep: /sys/devices/virtual/video4linux/video0/power: Is a directory
grep: /sys/devices/virtual/video4linux/video0/subsystem: Is a directory
/sys/devices/virtual/video4linux/video0/uevent:MAJOR=81
/sys/devices/virtual/video4linux/video0/uevent:MINOR=0
/sys/devices/virtual/video4linux/video0/uevent:DEVNAME=video0

[umlaeute]

excuse my ignorance and lack of response, but looking at the gstreamer commit logs i'm very much under the impression, that the bug was indeed with gstreamer, and has been fixed for the 1.0.7 release.
this can also be confirmed in the ChangeLog of gstreamer-plugins-good and the 1.0.7 milestone

[vvinothkumar]

Hi Umlaeute,

Thanks. Hope this solves my issue. Let me try with 1.0.7 release and update
you ASAP.

Thanks,
Vinoth

On Sat, Jul 20, 2013 at 12:51 AM, umlaeute [email protected] wrote:

excuse my ignorance and lack of response, but looking at the gstreamer
commit logshttp://cgit.freedesktop.org/gstreamer/gst-plugins-good/commit/?id=14300e58059ff5a6ca6cf2b2942804f4f64b8ad0i'm very much under the impression, that the bug was indeed with gstreamer,
and has been fixed for the 1.0.7 release.
this can also be confirmed in the ChangeLog of gstreamer-plugins-good and
the 1.0.7 milestonehttps://bugzilla.gnome.org/buglist.cgi?product=GStreamer&target_milestone=1.0.7

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/39#issuecomment-21271431
.