Unable to flash 3 Jetsen TX2 units with 4.2.2

I’m trying to manually flash 3 different TX2 dev kits, they’re all acting the same.

I have used various USB cables.

Host is a laptop running Ubuntu 18.04.02 Desktop.

$ uname -a
Linux minibeast 5.0.0-31-generic #33~18.04.1-Ubuntu SMP Tue Oct 1 10:20:39 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
$ lsusb
Bus 002 Device 003: ID 2109:0812 VIA Labs, Inc. VL812 Hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 006: ID 04f3:2379 Elan Microelectronics Corp. 
Bus 001 Device 005: ID 1532:0232 Razer USA, Ltd 
Bus 001 Device 024: ID 0955:7c18 NVidia Corp. 
Bus 001 Device 018: ID 2109:2812 VIA Labs, Inc. VL812 Hub
Bus 001 Device 0[code]03: ID 0cf3:e300 Atheros Communications, Inc. 
Bus 001 Device 002: ID 1bcf:2c9a Sunplus Innovation Technology Inc. 
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

When plugging in the device via USB and putting it into recovery mode I see the following in the syslog:

Oct 11 00:11:47 minibeast upowerd[11665]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1
Oct 11 00:11:49 minibeast kernel: [ 3616.020357] usb 1-7.1: new high-speed USB device number 28 using xhci_hcd
Oct 11 00:11:49 minibeast kernel: [ 3616.137365] usb 1-7.1: New USB device found, idVendor=0955, idProduct=7c18, bcdDevice= 0.00
Oct 11 00:11:49 minibeast kernel: [ 3616.137371] usb 1-7.1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
Oct 11 00:11:49 minibeast kernel: [ 3616.137374] usb 1-7.1: Product: APX
Oct 11 00:11:49 minibeast kernel: [ 3616.137377] usb 1-7.1: Manufacturer: NVIDIA Corp.
Oct 11 00:11:49 minibeast mtp-probe: checking bus 1, device 28: "/sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1"
Oct 11 00:11:49 minibeast mtp-probe: bus: 1, device: 28 was not an MTP device
Oct 11 00:11:50 minibeast upowerd[11665]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1
Oct 11 00:11:54 minibeast kernel: [ 3620.888935] usb 1-7.1: USB disconnect, device number 28
Oct 11 00:11:54 minibeast upowerd[11665]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1
Oct 11 00:11:55 minibeast kernel: [ 3622.480309] usb 1-7.1: new high-speed USB device number 29 using xhci_hcd
Oct 11 00:11:55 minibeast kernel: [ 3622.593605] usb 1-7.1: New USB device found, idVendor=0955, idProduct=7c18, bcdDevice= 0.00
Oct 11 00:11:55 minibeast kernel: [ 3622.593611] usb 1-7.1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
Oct 11 00:11:55 minibeast kernel: [ 3622.593614] usb 1-7.1: Product: APX
Oct 11 00:11:55 minibeast kernel: [ 3622.593617] usb 1-7.1: Manufacturer: NVIDIA Corp.
Oct 11 00:11:56 minibeast mtp-probe: checking bus 1, device 29: "/sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1"
Oct 11 00:11:56 minibeast mtp-probe: bus: 1, device: 29 was not an MTP device
Oct 11 00:11:56 minibeast upowerd[11665]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:14.0/usb1/1-7/1-7.1

when attempting to manually flash:

~/nvidia/nvidia_sdk/JetPack_4.2.2_Linux_GA_P3310/Linux_for_Tegra$ sudo ./flash.sh jetson-tx2 mmcblk0p1

###############################################################################
# L4T BSP Information:
# R32 , REVISION: 2.1
###############################################################################
Error: probing the target board failed.
       Make sure the target board is connected through 
       USB port and is in recovery mode.

Not sure what to do, but this seems related to Ubuntu 18.04 not properly detecting.

bump! Any ideas on how I might solve this issue?

hello phreak9i6,

we had verified multiple flash script files for flashing multiple modules, we tried to flash three device simultaneously for confirmation.

we had setup environment with multiple devices connect to host desktop directly.
could you please share your hardware environment setups, were you using usb-hub to connect several device with the host machine?
thanks

Jerry,

I’m sorry I haven’t represented this problem correctly. I am trying to flash 3 TX2 dev kits. One at a time. They all fail with the same issue separately.

hello phreak9i6,

those failure seems the device did not enter forced-recovery mode for flashing.
could you please set device enter forced-recovery mode and check the connection with lsusb,
for example, you should see NVidia Corp. in the list.

$ lsusb
...
Bus 001 Device 103: ID 0955:7c18 NVidia Corp.

BTW,
there’s an alternative way to put your board into forced-recovery mode.

  1. disconnect the power to fully shutdown the device.
  2. connect the power, and also hold the recovery button.
  3. press the power button to power-on the device, keep few seconds then release recovery button.
  4. execute the script to flash the target.

Jerry,

I’m not sure I understand. In my copy paste I list the device shown by lsusb, it shows

0955:7c18 NVidia Corp.

Is this not what’s expected?

from what I can tell, the boards are definitely in forced-recovery mode. I see the device in lsusb.

I’ve attempted you alternative way to get into recovery mode but I have the same issue.

It seems that while it’s in USB recovery mode, Ubuntu 18.04 isn’t able to talk to the device properly. I’ve seen several other people post about a similar issue, but zero resolution.

All three of my TX2s are unusable, so any help is appreciated.

hello phreak9i6,

could you please check Topic 1064357,
please have a try to comment out get_fuse_level function calls in flash.sh for testing.
thanks

commenting out the line

get_fuse_level fuselevel hwchipid bootauth;

has no effect

after some poking around it looks like it fails the chk_soc_sanity() function.

I also found this discussion:

https://devtalk.nvidia.com/default/topic/1050471/-install-jet-pack-4-2-on-jetson-tx2/

and tried the same code change:

bootauth="";
if [ "${fuselevel}" = "" ]; then
        echo "hwchipid = ${hwchipid}";
        echo "get fuse level";
        get_fuse_level fuselevel hwchipid bootauth;
        echo "hwchipid = ${hwchipid}";

these are the results:

ken@minibeast:~/nvidia/nvidia_sdk/JetPack_4.2.2_Linux_GA_P3310/Linux_for_Tegra$ sudo ./flash.sh jetson-tx2 mmcblk0p1
###############################################################################
# L4T BSP Information:
# R32 , REVISION: 2.1
###############################################################################
hwchipid = 
get fuse level
hwchipid = 
Error: probing the target board failed.
       Make sure the target board is connected through 
       USB port and is in recovery mode.

So I’ve found an interesting timing issue.

If I quickly run the flash.sh after plugging in the usb cable while the jetson is in recovery mode it eventually starts.

It takes 2-3 attempts to work.

Then it hangs after

` tegrarcm_v2 --isapplet`

I followed the suggestion here:
https://devtalk.nvidia.com/default/topic/1031769/jetson-tx2/unable-install-any-jetpacks-jetson-tx2/post/5382140/#5382140

I’ve been flashing these devices with a Razer laptop, I’ve also found the flashing fails unless the system is powered.

hello phreak9i6,

several questions as below,

  1. were you able to flash JetPack-4.2.2.
  2. is your laptop running ubuntu-18.04 with a virtual machine.
  3. may I know had you fused those TX2 dev kits.

Jerry,

  1. I was yes, as per my previous posting.

  2. Laptop is running Ubuntu 18.04.2 natively.

  3. I’m not sure if they are fused. They were fresh out of the box, brand new.

hello phreak9i6,

may I know the hardware connection of your TX2 and Laptop, did you using USB-hub or OTG cables?
also, please add -r options into flash commands, it’ll skip building image but flash the board immediately.
for example,
$ sudo ./flash.sh -r jetson-tx2 mmcblk0p1

Hi phreak9i6,

Have you managed to get issue resolved?
Any result can be shared?

JerryChang,

As per my previous responses in post #14 and #11.

I was able to flash the firmware, it seems to be a timing issue. If I run the flash quickly (within a few seconds of plugging in the OTG cable) it works fine. Sometimes it takes several attempts to complete a flash process.

Laptop needs to be powered for some reason - 100% fail rate if the laptop is on battery.
Using an OTG cable on USB 3 ports. No Hub
Bare Metal OS install - No VM, No containers.