Camera driver failed

Hello everyone,

I’m currently working with Jetpack 5.0.2 preview on a Jetson AGX Orin. I’ve successfully created the device tree and the imager module IMX623. However, when I try to test it using the command v4l2-ctl --stream-mmap, I’m not receiving any data.
here are the dmesg and trace log :
dmesg.txt (3.7 KB)
tracelog.txt (16.5 KB)

If anyone could provide assistance or insights into this issue, I would greatly appreciate it. Thank you in advance for your help!

hello soufianehadi01,

could you please working with the latest release since we’ve several bug fixes to address camera stability issue.
besides, please share your camera device tree for reference, thanks

thank you for your reply
here is the device tree
tegra234-camera-imx623-a00.dtsi (23.2 KB)

this is incorrect.
please refer to Port Index section. there’re only 5 VI streams available.
please revise the mapping to connect CSI-G/H to VI stream-5.

	tegra-capture-vi {
			port@3 {
				status = "okay";
				reg = <3>;
				imx623_vi_in3: endpoint {
					status = "okay";
					port-index = <6>;
					bus-width = <4>;
					remote-endpoint = <&imx623_csi_out3>;

this meant the total active CSI lanes.
it should be 16 since you’ve four 4-lane config in your system.

	tegra-camera-platform {
		num_csi_lanes = <4>;

I changed that but I still have the same eror

 vi-output, imx6-2165    [005] ....   280.153246: tegra_channel_capture_setup: vnc_id 0 W 1936 H 1552 fmt c4
     kworker/2:2-140     [002] ....   280.216857: rtcpu_vinotify_event: tstamp:9450413854 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:302401656992 data:0x379d580010000000
     kworker/2:2-140     [002] ....   280.216859: rtcpu_vinotify_event: tstamp:9450413990 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:302401699648 data:0x0000000031000001
     kworker/2:2-140     [002] ....   280.216860: rtcpu_vinotify_event: tstamp:9450414141 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:302401717312 data:0x379d550010000000
     kworker/2:2-140     [002] ....   280.216860: rtcpu_vinotify_event: tstamp:9450414271 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:302401777696 data:0x0000000031000002
 vi-output, imx6-2165    [005] ....   282.969326: tegra_channel_capture_setup: vnc_id 0 W 1936 H 1552 fmt c4
     kworker/2:2-140     [002] ....   283.016839: rtcpu_vinotify_event: tstamp:9538061207 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:305217749536 data:0x379d580010000000
     kworker/2:2-140     [002] ....   283.016840: rtcpu_vinotify_event: tstamp:9538061343 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:305217792192 data:0x0000000031000001
     kworker/2:2-140     [002] ....   283.016841: rtcpu_vinotify_event: tstamp:9538061496 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:305217810976 data:0x379d550010000000
     kworker/2:2-140     [002] ....   283.016841: rtcpu_vinotify_event: tstamp:9538061627 cch:0 vi:0 tag:VIFALC_TDSTATE channel:0x23 frame:0 vi_tstamp:305217871328 data:0x0000000031000002

hello soufianehadi01,

it seems there’s no camera related signal on the CSI channel.
BTW, is it possible for moving to Jetpack-5.1.2 release verison?

Hello @JerryChang

Currently, we are using the setup below :

  • MAX9296A Deserializer (Through GMSL A)
  • MAX96717F Serializer
  • IMX623 Image sensor

Both of components are correctly detected though I2C, and the drivers are installed in each slave address, also the register table are correctly filled, and the registers are written successfully.

Do you think that the issue is comming from the image sensor by not sending the MIPI-CSI data ?
Or the transport part with serdes pair ?

Can you please give any suggestions for debugging ?

For the JetPack version, we’re currently using the 5.0.2, it works for another camera module we use, so i guess in our side the issue may come from the drivers or the device tree.


hello soufianehadi01,

it looks like image sensor is not sending the MIPI-CSI data.
may I also know the sensor output data-rate, is it larger than 1.5Gbps? if yes… that might be a known issue of deskew calibration timed out.

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