Currently I am using L4T 35.4.1. I can able to receive the video with this combination(Orin NX and Avermedia D133 carrier board). After This, I have loaded the device tree for the combination (Orin Nx and Orin Nano Super Developer Kit). But here i am not able to receive the video input(only green screen appears). I am getting the trace like VIFALC_TDSTATE(lane_polarity=6). When I change the lane_polarity=0 , this time I got this below debug trace
Is it possible for you to share which camera module are you using?
Where did you get the camera driver from ?
What other changes besides lane polarity are you applying to the device tree from one carrier board to the other?
Do you see any logs from the camera driver on dmesg ?
Seeing a green screen simply means that you are not getting buffers.
I would think that there might be differences in HW configuration from the devkit to the avermedia carrier. For example, if you had access to the schematics, I would suggest you check if CSI port routing looks the same for both carrier boards. I would also check DTB settings on the dtsi files from Avermedia and compare them to those of original Jetpack. There could be changes that point to differences in the carrier board.
For example, I would suggest you compare IMX219 device trees from both carriers. Those drivers shall be supported out of the box by both carriers, so you could have a good point of comparison between one carrier and the other.
Please keep us posted, and don’t hesitate to reach out if you need further support, we would love to help.
best regards,
Andrew
Embedded Software Engineer at ProventusNova
Interesting, and when you say you faced this same error, was it under the same circumstances ? In other words, were you porting the driver from one carrier to another ?
best regards,
Andrew
Embedded Software Engineer at ProventusNova
Actually earlier Avermedia board with ORIN Nx, I faced this same error.
Driver is same for both. Am I right?
Because I am using the same BSP, only device tree changes. Is my understanding is right?
Yes, I think you are right.
Driver should not change unless there are big Jetpack differences where the camera subsystem could require other driver logic or something.
And since the camera module is the same, I would not think there are differences on register table either.
So focusing on the device tree is the right approach for now I believe.
Do you have access to Avermedia’s device tree source code ?
Can you share IMX219 dtsi files for Avermedias and devkit ?
best regards,
Andrew
Embedded Software Engineer at ProventusNova
