Tegrarcm_v2 reset board after each download

Robotics & Edge Computing Jetson Systems Jetson Orin Nano
, ,
1

Hello,
For nano evb, with l4t 3644,
I’m following https://forums.developer.nvidia.com/t/failing-with-custom-keys-for-encrypting-nvme-on-jetson-orin-nano/270934/13 for some issue about secure boot .
one of the steps there is " l4t_initrd_flash…", there are tow probleam:
1)during the process, the board (nano) got reset and exit from recovery mode, what lead to failures. i can handle this by reset manuall the board (unplug power) while the “force recovery” shorted to ground.
2) at the end of l4t_init_flash, it call tegrarcm_v2, for download images. but there also, after the first image it download, the board got reset, exit from recovery, and the operation failed ,according to the logs, it tried it manually :
"sudo ./tegrarcm_v2 --new_session --chip 0x23 0 --uid --download bct_br br_bct_BR.bct --download mb1 mb1_t234_prod_aligned_sigheader.bin.encrypt --download psc_bl1 psc_bl1_t234_prod_aligned_sigheader.bin.encrypt --download bct_mb1 mb1_bct_MB1_sigheader.bct.encrypt
"
it got reset after bct_br download, and cant continue to the next images.

How can I solve it?

Thanks

3

Hi Moti_Shani,

Are you using the devkit or custom board for Orin Nano?
What’s the Jetpack version in use?

Do you mean Jetpack 6.2.1.(L4T r36.4.4)?

Do you want to enable the disk encryption for rootfs or secureboot for bootloader?

Please share the exact command you run and the full flash log for further check.

4

Hey

I’m using the official devkit for Orin Nano

Building the l4t r36.4.4 (Jetson_Linux_R36.4.4_aarch64).

For now only secure boot for bootlaoder

Attached the full logs from host:
nvidia_nano_l4t_initrd_flash_fauil_log.txt (4.7 KB)
And the serial log from the nano (it start printing right after the tegrarcm_v2 executed)
nvidia_nano_tegrarcm_v2_SERIAL_LOG.txt (34.5 KB)

As you can see from the serial log, my main issue is booting hanged due to some UEFI key invalidation (this is what i try to solve by Failing with custom keys for encrypting NVME on Jetson Orin Nano - #13 by stefankmfr0) but i’m stock at the last part - the flashing

Thanks!

5

There should be 3 steps to flash the device and you just shared the result of the last step.

I would also like to check the log for step 1 and step 2 to generate image for both internal QSPI and external NVMe.

Do you mean the secure boot(i.e. PKC, SBK) or UEFI Secureboot(i.e. UEFI key conf)?

[   0.0406 ] ERROR: might be timeout in USB write.
Error: Return value 3
Command tegrarcm_v2 --instance 3-1 --new_session --chip 0x23 0 --uid --download bct_br br_bct_BR.bct --download mb1 mb1_t234_prod_aligned_sigheader.bin.encrypt --download psc_bl1 psc_bl1_t234_prod_aligned_sigheader.bin.encrypt --download bct_mb1 mb1_bct_MB1_sigheader.bct.encrypt
Cleaning up...

From the log you shared, it seems you have USB connection issue.
Please use another USB cable or the port from host to check if it could help.

6

Hey

  1. I dont have nvme on the nano, only sdcard

  2. secureboot PKC and UEFI (by the error from the serial log, it have issue with the signed image)

3)I pretty sure it is not cable issue, when the nano in recovery mode its recognized fine, the problem is that during the tegrarcm_v2 itself, the board perform reboot and get out from the recovery, it can be seen in the log file of the serial output. if I reset the board (unpluged and pluged) with RECOVERY_PIN shorted to GRND, it goes back to recovery and recognized well

I attached the full log for the whole process
NVIDIA_FULL_LOGS_FOR_FLASH_SECURED_FAILE_AT_TEGRARCM.txt (550.3 KB)
(as described at the topic above), The only workaround i foubd to prevent l4t_initrd_flash failure due to “rcm not found” is to reset the nano each time it performed a cold reboot and went out from recovery mode. this workaround can’t work for tegrarcm_v2 command, which reboot the board during the proceesss, so it can load only BCR and fail for the next image

Seems that my problem, Is to find a way to make the nano not to reboot after every time the tools talk to him

Thanks!

7

Could you try to flash w/o enabling Secureboot and UEFI secureboot clarify if the issue relates to them?

Not really, sometimes it can be recognized in force recovery state correctly before flash and would still hit USB timeout issue due to the unstable connection/signal.
Are you using standalone Ubuntu as host PC to flash the board?
Or you are using something like VM/docker/WSL2?

Several reboots during flash are expected so that I suggested for the stable USB connection.

9

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