Driver failed probe after upgrading to Jetpack 6.2

Deqing_S · March 25, 2025, 7:52pm

Hi, we have a camera system that works OK on Jetpack 5.1.3. However, we’d like to upgrade the Jetpack to evaluate the performance boost, and the bug fix.

Our system consist of a Jetson Orin Nano, on a LI-ONX-CB-4CAM-FP carrier board, which has a FPGA to emulate an I2C mux (pca9548). The carrier board connects to 3 of DS90UB954, and each DS90UB954 connects to 2 cameras with coax cable, which contains AR0234 and DS90UB953.

Unfortunately, Leopard Imaging does not supply any help on Jetpack 6. So we are on our own.

I don’t think the MIPI signal matters before the camera start streaming. As for I2C signal flow, the I2C bus on 3180000 got multiplexed into 8 additional I2C buses. On 3 of the additional bus, there will be 3 devices: DS90UB954 de-serializers, and 2 sensors got I2C relayed by DS90UB954.

So far we’ve got the I2C mux working, and DS90UB954 driver working. But when we try to load AR0234 driver, the module does not probe.

We’ve tested the I2C signals. The Jetson never tried to send signal to AR0234 address, so the issue should be on software side. The device tree seems partially work, as we I run tree /sys/bus/i2c/devices/

/sys/bus/i2c/devices/
├── 1-0050 -> ../../../devices/platform/bus@0/3160000.i2c/i2c-1/1-0050
├── 1-0057 -> ../../../devices/platform/bus@0/3160000.i2c/i2c-1/1-0057
├── 2-0077 -> ../../../devices/platform/3180000.i2c/i2c-2/2-0077
├── 3-0021 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-3/3-0021
├── 3-0022 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-3/3-0022
├── 3-0030 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-3/3-0030
├── 4-0021 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-4/4-0021
├── 4-0022 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-4/4-0022
├── 4-0030 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-4/4-0030
├── 5-0021 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-5/5-0021
├── 5-0022 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-5/5-0022
├── 5-0030 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-5/5-0030
├── 6-0021 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-6/6-0021
├── 6-0022 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-6/6-0022
├── 6-0030 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-6/6-0030
├── i2c-0 -> ../../../devices/platform/bpmp/bpmp:i2c/i2c-0
├── i2c-1 -> ../../../devices/platform/bus@0/3160000.i2c/i2c-1
├── i2c-10 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-10
├── i2c-2 -> ../../../devices/platform/3180000.i2c/i2c-2
├── i2c-3 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-3
├── i2c-4 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-4
├── i2c-5 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-5
├── i2c-6 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-6
├── i2c-7 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-7
├── i2c-8 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-8
└── i2c-9 -> ../../../devices/platform/3180000.i2c/i2c-2/i2c-9

I can see the cameras appear at addresses such as 3-0021. In side that folder, when I do cat uevent

OF_NAME=custom_ar0234_a
OF_FULLNAME=/i2c@3180000/tca9548@77/i2c@0/custom_ar0234_a@21
OF_COMPATIBLE_0=onsemi,sh_ar0234
OF_COMPATIBLE_N=1
MODALIAS=of:Ncustom_ar0234_aT(null)Consemi,sh_ar0234

And those information matches the device tree. However, there is a waiting_for_supplier files, which content was 1. I think it should have something to do with the no-probing issue.

We’ve used the attached dtsi files with tegra234-p3768-0000+p3767-0003.dts to build the dtb.
dtsi.zip (4.4 KB)
tegra234-p3768-0000+p3767-0003.zip (25.8 KB)
And the decompiled dts
running_system_change_id.zip (26.2 KB)

We also tried to compare some device tree from Veye imaging for
Jetpack 6.2

github.com/veyeimaging/nvidia_jetson_veye_bsp

dtbs/Orin/JetPack_6.2_Linux_JETSON_ORIN_TARGETS/orin_nx_nano/tegra234-camera-rbpcv3-veye_mvcam.dtsi

fdf8d1d90

// SPDX-License-Identifier: GPL-2.0-only
// SPDX-FileCopyrightText: Copyright (c) 2023-2024, www.veye.cc. All rights reserved.

#include <dt-bindings/clock/tegra234-clock.h>
#include <dt-bindings/gpio/tegra234-gpio.h>


/*
    This dts is for MV and RAW series cameras :
    MV-MIPI-XXX
    RAW-MIPI-XXX 
*/
/ {
	fragment@0 {
		target-path = "/";
		__overlay__ {
		tegra-capture-vi  {

				num-channels = <2>;
				ports {

This file has been truncated. show original

and Jetpack 5.1.3

github.com/veyeimaging/nvidia_jetson_veye_bsp

dtbs/Orin/JetPack_5.1.3_Linux_JETSON_ORIN_TARGETS/dts%20dtb/common/t23x/p3768/kernel-dts/cvb/tegra234-camera-rbpcv3-veye_mvcam.dtsi

fdf8d1d90

/*
 * Copyright (c) 2020-2022, www.veye.cc.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/*
    This dts is for all VEYE series cameras :
    MV-MIPI-XXX

This file has been truncated. show original

with no success.

We are out of clue to get the code work, any suggestions will be appreciated. Thanks

ShaneCCC · March 26, 2025, 5:23am

Dump the sensor is in /sys/firmware/devicetree/base/bus@0/i2c@3180000

ls -al /sys/firmware/devicetree/base/bus@0/i2c@3180000

Also, dump the device tree to confirm.

sudo dtc -I fs -O dts -o extracted_proc.dts /proc/device-tree

Deqing_S · March 26, 2025, 6:56pm

sh@rig35nano3:~/Documents/GitHub/SH002_CAM_DRIVER$ ls -al /sys/firmware/devicetree/base/bus@0/i2c@3180000
total 0
drwxr-xr-x   2 root root 0 Mar 26 14:54 .
drwxr-xr-x 103 root root 0 Mar 26 14:54 ..
-r--r--r--   1 root root 4 Mar 26 14:54 name
-r--r--r--   1 root root 4 Mar 26 14:54 phandle

Add the dumped tree is here
extracted_proc.zip (26.2 KB)

ShaneCCC · March 31, 2025, 7:04am

Didn’t see the tca9548@77 under the i2c@31800000, do you make sure the tca9548 probe successfully?

Deqing_S · March 31, 2025, 1:37pm

i2c@3180000 appears 2 times in the device tree.

but in the Jetpack5 version there is only one. let me see if it is a cause.

Deqing_S · March 31, 2025, 2:17pm

I removed the duplicated entry in device tree before make dtbs. The issue is still there.
The new device tree is.
extracted_proc.zip (26.1 KB)

ShaneCCC · April 1, 2025, 5:09am

Add message in the tca9548 probe driver to make sure it been loaded successfully.

Deqing_S · April 1, 2025, 5:05pm

How should I add message? It seems loaded automatically in dmesg

[    1.678919] i2c i2c-2: Added multiplexed i2c bus 3
[    1.679196] i2c i2c-2: Added multiplexed i2c bus 4
[    1.679487] i2c i2c-2: Added multiplexed i2c bus 5
[    1.679722] i2c i2c-2: Added multiplexed i2c bus 6
[    1.679802] i2c i2c-2: Added multiplexed i2c bus 7
[    1.679892] i2c i2c-2: Added multiplexed i2c bus 8
[    1.679974] i2c i2c-2: Added multiplexed i2c bus 9
[    1.680042] i2c i2c-2: Added multiplexed i2c bus 10
[    1.680047] pca954x 2-0077: registered 8 multiplexed busses for I2C switch pca9548

ShaneCCC · April 2, 2025, 3:00am

OK, that tell the pca954x driver is loaded.
You need to probe the i2c signal for the camera to confirm.

Deqing_S · April 6, 2025, 6:16am

Yes, the boot process will generate an access to address 77, and the i2c mux did generate ACK. This is confirmed on logic analyzer.
The i2c mux should be working properly both in hardware in linux. The DS90UB954 de-ser after the i2c mux also works well and the module can be loaded without problem. The only issue is the AR0234 driver refuse to probe.

ShaneCCC · April 7, 2025, 7:30am

OK, maybe review the device tree if any missing.

system · May 6, 2025, 7:03am

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

Topic		Replies	Views
Unable to load adv7180 module after updating device tree Jetson Orin NX camera , device-tree	20	130	August 13, 2024
Jetpack 6.0 - issues with soc csi2 camera Jetson Orin Nano camera	38	214	August 23, 2024
Enable camera sensor connected to i2c mux Jetson AGX Orin camera	26	801	July 30, 2024
[Jetpack-6.2] /dev/videox not created Jetson Orin Nano camera , jetson	26	70	March 25, 2025
Jetpack 5.1.1 and ar1335 cameras: e-Con e-CAM130A_CUXVR device tree issue Jetson AGX Xavier camera	11	992	September 27, 2023
Unable to detect multiple cameras on custom carrier board with Jetson orin NX Jetson Orin Nano camera , board-design , device-tree	17	176	September 20, 2024
Customer camera driver development Jetson Nano	33	2965	October 14, 2021
Custom camera device tree not found after flash \| JP6 Jetson Orin Nano camera , kernel , device-tree	13	211	October 22, 2024
Enabling pr_debug to analyze device-tree/driver binding Jetson AGX Orin device-tree	13	878	January 17, 2024
Sony EV9500M MIPI camera on JP6(36.3) Jetson Orin NX camera	12	106	January 14, 2025

Driver failed probe after upgrading to Jetpack 6.2

Related topics