lukasplevac/cuda-samples
1
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2025-04-10 18:22:11 +01:00
Code Issues Packages Projects Releases Wiki Activity

Add automotive Linux build instructions

Browse Source
This commit is contained in:
Rob Armstrong 2025-01-27 10:16:20 -08:00
parent 9824a63101
commit be5012ef69
1 changed files with 18 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
README.md
View File

@ -104,6 +104,24 @@ make -j$(nproc)
```
Transfer the built binaries to the Tegra device and execute them there.
### Building for Automotive Linux Platforms
These platforms require additional information to be passed to CMake on the command line to ensure proper resolution of all necessary include and library files.
Instead of being in the default location, `/usr/local/cuda/include` or `/usr/local/cuda/lib64`, you must point to architecture-specific paths:
`/usr/local/cuda/<ARCH>/targets/aarch64-linux/lib`
and
`/usr/local/cuda-12.8/<ARCH>/include`
An example build might look like this:
```
$ mkdir build
$ cd build
$ cmake -DCMAKE_CUDA_COMPILER=/usr/local/cuda/bin/nvcc -DCMAKE_LIBRARY_PATH=/usr/local/cuda/orin/lib64/ -DCMAKE_INCLUDE_PATH=/usr/local/cuda/orin/include -DBUILD_TEGRA=True ..
```
### QNX
Note that in the current branch sample cross-compilation for QNX is not fully validated. This placeholder will be updated in the