The device cann't start

Hi ,we used R32 (release), REVISION: 5.2, GCID: 27767740, BOARD: t210ref.
We have a serious problem: the nvidia module cann’t start, no reaction at all when power on. The module had run sometimes normally, but it suddenly won’t turn on.
We connet the uart, but without any information.
We had found five moduls on the production line.
We used emmc mode, and if we re-burn the firmware, it will work ok, but we don’t know if it’s going to suddenly turn off
This is an urgent matter, we need your help!

Hi yong.lu

Are you using the devkit or custom board?
You mean some modules from production can not boot up, but some modules can?

If re-flash could work fine w/o error, I would suggest re-flash them all again.

We would need console log for further check.
Try following instruction for serial console:
Jetson Nano Style - Serial Console

It is ourself board, and many modules can bootup, and the modules in question had run aging video test, when do other tests , they need reboot, we found them cann’t boot up again.
We connected the uart port,but without any information.
They didn’t have problems to begin with, so we are afraid of serious hidden problems.

When you doing this test, is there the serial UART log output?

So, boot up issue always occurs after reboot?

When you doing this test, is there the serial UART log output?
Normally, there are lot of serial UART logs output, includle boot up logs.
Now, the serial UART is without any information, so we don’t know what to do next?

So, boot up issue always occurs after reboot?
Yes , it occurs after reboot!

Could you provide the UART log before reboot to check if there’s any error?

we will suggest our retest colleague connect the serial UART for testing the case!
Brfore found the problem, we didn’t connect the UART, and when it occurd, connected the UART, there are nothing output!

Hi, we want to just flash the cbbot.bin for the boot up failed board, and it flash failed.
We got the flash log:
1s1191_update_cboot_failed.log (4.1 KB).
It had these errors:
[1216_10:57:30][0010.105] Loading NvTbootBootloaderDTB
[1216_10:57:30][0010.121] Verifying NvTbootBootloaderDTB in SecurePKC mode
[1216_10:57:30][0010.141] NvTbootBootloaderDTB is corrupted!
[1216_10:57:30][0010.145] Error in NvTbootLoadBinary: 0x14 !
[1216_10:57:30][0010.149] Bootloader DTB Load Address: 0x83000000
[1216_10:57:30][0010.154] Bootloader DTB invalid, fallback to kernel DTB.
What does this mean?

It looks something wrong with your cboot image.
How do you flash cboot.bin only?

Could you flash all on failed board successufully?
Or just flash cpu-bootloader failed?

we used the cmd to flash cboot.bin on normal board, it flash succseefully. The cmd is:
sudo ./flash.sh -k EBT -u ./Product03/rsa_priv.pem jetson-nano-devkit-emmc-smpsq mmcblk0p1.

We had tested to flash all image files on the failed board , it can flash successufully, and boot up ok!

Where is the rsa_priv.pem come from?

Could you try the following command instead?
sudo ./flash.sh -r -k cpu-bootloader <board> mmcblk0p1

Our board had burned efuse with PKC key. we are Nano product, we had checked the cmd on another case(How to flash cboot only). And we had tested the cmd on normal board, I think it is ok.

There are some users encounter the similar issue, but they resolve with different host PC.
Unable to flash: Error: probing the target board failed. and boot: Error in NvTbootLoadBinary: 0x14 - #17 by linuxdev

Are you using the VM with Ubuntu or standalone x86-64 Ubuntu?
and what’s your Ubuntu version?

We didn’t used VM ,we know VM’s error.
We had flashed successfully on normal board.
I think maybe the emmc had bad block, because the board could bootup after reflash all Image bin.

We want to check the emmc bad block by flash cboot.bin only.
It flash cboot.bin failed, and had the error logs.

Do you still have these errors when you flash cboot.bin only?

yes!

Hi yong.lu,

Could you help using R32.7.3 released recently to check if this issue still exists?
Jetson Linux R32.7.3 | NVIDIA Developer

Do you doubt where?

Hi @yong.lu

Sorry to jump in. What is the exact issue now?

It looks like you can flash the board and boot up ok on 12/16. Is this still an issue?