Segmentation fault using camera in Jetson TX2 (Jetpack 4.5.1)

Robotics & Edge Computing Jetson Systems Jetson TX2
, ,
1

Hello.

I would like to run a l4t-ml docker container on a Jetson TX2 (Jetpack 4.5.1) and save video files from a CSI camera.

Information about the development environment is below.

development environment 
$ cat /etc/os-release
NAME="Ubuntu"
VERSION="18.04.6 LTS (Bionic Beaver)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 18.04.6 LTS"
VERSION_ID="18.04"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=bionic
UBUNTU_CODENAME=bionic

$ cat /etc/nv_tegra_release
# R32 (release), REVISION: 5.1, GCID: 26202423, BOARD: t186ref, EABI: aarch64, DATE: Fri Feb 19 16:50:29 UTC 2021

$ v4l2-ctl -d /dev/video0 --list-formats-ext
ioctl: VIDIOC_ENUM_FMT
        Index       : 0
        Type        : Video Capture
        Pixel Format: 'BG10'
        Name        : 10-bit Bayer BGBG/GRGR
                Size: Discrete 2592x1944
                        Interval: Discrete 0.033s (30.000 fps)
                Size: Discrete 2592x1458
                        Interval: Discrete 0.033s (30.000 fps)
                Size: Discrete 1280x720
                        Interval: Discrete 0.008s (120.000 fps)

$ docker -v
Docker version 20.10.21, build 20.10.21-0ubuntu1~18.04.3

I wrote the following python code(captest_csi.py) and ran it on the docker container.

captest_csi.py 
# example
# 1280x960 5fps, output file's name 1280_960_5.mp4,  record 5sec
# python3 captest_csi.py 1280 960 5 video 1280_960_5.mp4

import cv2
import sys

if __name__ == '__main__':
    args = sys.argv
    width = args[1]
    height = args[2]
    fps = args[3]
    cap = cv2.VideoCapture(f'nvarguscamerasrc sensor-id=0\
            ! video/x-raw(memory:NVMM), width={width}, height={height}, format=(string)NV12, framerate=(fraction){fps}/1 \
            ! nvvidconv ! video/x-raw, width=(int){width}, height=(int){height}, format=(string)BGRx \
            ! videoconvert  \
            ! appsink', cv2.CAP_GSTREAMER)

    if cap.isOpened() is False:
        raise IOError("open failed")

    if args[4]=='video':
        fps = int(cap.get(cv2.CAP_PROP_FPS))
        w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
        h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
        print(str(fps) + "," + str(w) + "," + str(h))

        fourcc = cv2.VideoWriter_fourcc('m', 'p', '4', 'v')
        if len(args) == 6:
            filename = args[5]
        else:
            filename = "output.mp4"
        out = cv2.VideoWriter(filename, fourcc, fps, (w, h))

        fn = 0
        while fn<(fps*5):
            ret, frame = cap.read()
            if ret:
                out.write(frame)
            else:
                raise IOError("read failed")
            fn = fn+1

        cap.release()
        out.release()

    elif args[4]=='img':
        if len(args) == 6:
            filename = args[5]
        else:
            filename = "output.jpg"

        ret, frame = cap.read()
        if ret:
            cv2.imwrite(filename, frame)
        else:
            raise IOError("read failed")

        cap.release()

    else:
        print("argument invalid")
        exit()

$ docker run --runtime nvidia -it --rm  -v $PWD:/capture_test/ -v /tmp/argus_socket:/tmp/argus_socket --privileged --name captest nvcr.io/nvidia/l4t-ml:r32.5.0-py3
allow 10 sec for JupyterLab to start @ http://localhost:8888 (password nvidia)
JupterLab logging location:  /var/log/jupyter.log  (inside the container)
root@73ea4fe3f6f0:/# cd capture_test/
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 1280 960 5 video 1280_960_5.mp4
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 640 480 30 video 640_480_30.mp4
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 1280 960 30 video 1280_960_30.mp4

At this time, if the resolution and frame rate are too high, a segmentation fault will occur. For example, 1280x960 5fps and 640x480 30fps can output video normally, but 1280x960 30fps causes segmentation fault.

The following is the execution log.

log 
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 1280 960 5 video 1280_960_5.mp4
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 2592 x 1944 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 2592 x 1458 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 22000, max 358733000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 1 
   Output Stream W = 2592 H = 1458 
   seconds to Run    = 0 
   Frame Rate = 29.999999 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
5,1280,960
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 640 480 30 video 640_480_30.mp4
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 2592 x 1944 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 2592 x 1458 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 22000, max 358733000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 2 
   Output Stream W = 1280 H = 720 
   seconds to Run    = 0 
   Frame Rate = 120.000005 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
30,640,480
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 1280 960 30 video 1280_960_30.mp4
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 2592 x 1944 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 2592 x 1458 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 22000, max 358733000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 1 
   Output Stream W = 2592 H = 1458 
   seconds to Run    = 0 
   Frame Rate = 29.999999 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
30,1280,960
GST_ARGUS: Cleaning up
Segmentation fault (core dumped)

The 1280x960 30fps setting also outputs an image.

log 
root@73ea4fe3f6f0:/capture_test# python3 captest_csi.py 1280 960 30 img 1280_960_30.jpg
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 2592 x 1944 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 2592 x 1458 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 22000, max 358733000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 1 
   Output Stream W = 2592 H = 1458 
   seconds to Run    = 0 
   Frame Rate = 29.999999 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success

I checked with a print statement and found that the segmentation fault occurred in “cap.release()”.

How can I output a video file at 1280x960 30fps? Please advise.

3

Hi,

Have you tried the sample on a newer JetPack release? Ex. JetPack 4.6.3?
If not, could you give it a try?

Thanks.

4

Thanks for the reply.
I have not tried a newer jetpack yet. I already have another application running in this environment and don’t want to update too much.
Is there any way to fix the problem in this environment?

I will also consider making a backup of my current development environment and updating jetpack.
I will report back if I make any progress.

5

Hi,

Could you also check if your camera can support the configuration (ex. 1280x960 30fps)?
More, when you use the 1280x960 30fps setting, is cap.isOpened() success?

Thanks.

6

Hi
I think the camera supports this configuration because I did cv2.VideoCapture with 1280x960 30fps setting and output the image file with cv2.imwrite and got a normal image.
Also, cap.isOpened is True.

7

Hi,

Thanks.

We are going to reproduce this issue.
Will share more info with you later.

8

Hi,

Please upgrade your environment to JetPack 4.6.

We test this issue with JetPack 4.6.3 and l4t-ml:r32.7.1-py3.
The 1280x960 30fps setting can work correctly.

root@tegra-ubuntu:/capture_test# python3 captest_csi.py 1280 960 30 video 1280_960_30.mp4
nvbuf_utils: Could not get EGL display connection
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 2592 x 1944 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 2592 x 1458 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 34000, max 550385000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 16.000000; Exposure Range min 22000, max 358733000;

GST_ARGUS: Running with following settings:
   Camera index = 0
   Camera mode  = 1
   Output Stream W = 2592 H = 1458
   seconds to Run    = 0
   Frame Rate = 29.999999
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /opt/opencv/modules/videoio/src/cap_gstreamer.cpp (935) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
30,1280,960
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success

Thanks.

10

Thank you for your cooperation.
I will try to upgrade.

11

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.