/usr/local/cuda/bin was not created after reflashing with jetpack 6.2

I just reflashed my jetson orin agx developer kit with jetpack 6.2 with the sdk manager. After the installation, there was no bin directory inside /usr/local/cuda. Inside /usr/local/cuda there is only gds, include, lib64 and targets. How can I get the binary for nvcc?

Hi,
Here are some suggestions for the common issues:

1. Performance

Please run the below command before benchmarking deep learning use case:

$ sudo nvpmodel -m 0
$ sudo jetson_clocks

2. Installation

Installation guide of deep learning frameworks on Jetson:

3. Tutorial

Startup deep learning tutorial:

4. Report issue

If these suggestions don’t help and you want to report an issue to us, please attach the model, command/step, and the customized app (if any) with us to reproduce locally.

Thanks!

Here’s contents of ls /usr/local/cuda
bin, doc, EULA.txt, gds, lib64, nvvm, share, tools,
compute-sanitizer, DOCS, extras, include, nvml, README, targets, version.json.

What I noticed is that after flashing jetpack6.2 to orin agx dev kit is that sdkmanager then presented a dialogbox stating to logon to orin desktop and click through first boot. Then it will offer to install cuda and other SDKs.

So I think what you can do is attach the Orin(usbc) to host(usb) cable. Start your host and start sdkmanager. Then start your orin NOT in recovery mode, and after logging on to orin desktop, sdkmanager should offer to install the sdks including cuda, tensorrt, and whichever checkboxes you click.

Or, if you already did that with sdkmanager then you could
sudo apt install cuda-toolkit-12-6 cuda-nvcc-12-6

Hi,

If you want to get the dev package, please run the following command:

$ sudo apt install nvidia-jetpack-dev

Thanks.

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