Problem in flash os to tx1 board


I got a new Jetson TX1 board, and then I followed the guide to setup Jetpack environment on my vmware.

vmware host: ubuntu 14.04

I download latest and then run it, download and install the relate package step by step.

after several steps, it shows that host configuration is complete and it will start to flash OS to TX1 board. but I met the problem that vmware can’t find the USB controller, so that i had to stop the flash OS step.

Now, I have fixed the USB issue, I would like to continute the flash OS step…

can you share me the steps? thanks!

JetPack has its own instructions, but if you want a minimal flash using command line:

  1. Unpack driver package
  2. cd Linux_for_Tegra
  3. sudo -s
  4. cd rootfs
  5. tar xjf /where/ever/it/is/Tegra_Linux_Sample-Root-Filesystem_R24.2.1_aarch64.tbz2
  6. cd ..
  7. ./
  8. # Make sure Jetson is connected and in recovery mode.
  9. -S 14580MiB jetson-tx1 mmcblk0p1
  10. # Be patient, it's slow.
  11. exit

Yes, I agree with you that JetPack has its own instructions, so I am thinking that whether I can continute from my breakpoint ( Host installation is complete and start to set Jetson to recovery mode ), otherthan the minimal steps.

so what is your recommend option for my issue?

  1. re-run and install again.
  2. follow your command line steps


Command line is good for testing because it is the minimal set of software to involve. I suspect you’re running into issues of vmware setup (e.g., buffer sizes, USB modes), and the quickest way to figure that out would indeed be on command line (if it fails, you only need repeat the “sudo ./ -S 14580MiB jetson-tx1 mmcblk0p1” step after restarting recovery mode). In the end you’ll want JetPack to work, but since is a subset of what JetPack does, I’d start there anyway. Once command line works, move on to JetPack.

yes, with the command line steps, I can flash the board, and then I also re-run JetPack and it also can flash the board now. but the JetPack hung in below log after flash board done:

==========log ============

[ 770.7062 ]
[ 770.7128 ] tegradevflash --write BCT P2180_A00_LP4_DSC_204Mhz.bct
[ 770.7144 ] Cboot version 00.01.0000
[ 770.7785 ] Writing partition BCT with P2180_A00_LP4_DSC_204Mhz.bct
[ 770.7938 ] […] 100%
[ 770.9561 ]
[ 770.9562 ] Flashing completed

[ 770.9562 ] Coldbooting the device
[ 770.9576 ] tegradevflash --reboot coldboot
[ 770.9591 ] Cboot version 00.01.0000
[ 771.0618 ]
*** The target t210ref has been flashed successfully. ***
Reset the board to boot from internal eMMC.

Finished Flashing OS
Determining the IP address of target…


May I know what happen at this point? HOST PC is trying to find the tx1 board?

and then I check “/home/ubuntu/” folder, I can’t find the sample code here.

Please ignor my above question :), I have found the root cause for that issue, it due to the Vmware network setting, because Vmware is in private network, TX1 board can’t ping Vmware success.

BUT, I met a new issue :(

=============New Issue============

After complete the network, I run “”, then the JetPack on host can go on for the configuration, but during the configuration, maybe because my network not so stable, it shows logs below, can you show me what I should do? thanks!

---------Log start---------
ssh: connect to host port 22: Network is unreachable
lost connection
Copying cuda-repo-l4t-8-0-local_8.0.34-1_arm64.deb file to target…
ssh: connect to host port 22: Network is unreachable
lost connection

ssh: connect to host port 22: Network is unreachable

Error: CUDA cannot be installed on device. This may be caused by other apt-get command running on device when installing CUDA. Please use apt-get command in a terminal to make sure following packages are installed correctly on device before continuing:
cuda-toolkit-8-0 libgomp1 libfreeimage-dev libopenmpi-dev openmpi-bin
After these packages are installed on device, press Enter key to continue

----------Log End-------------