Make error with Xavier

[ChienLiu]

have someone successfully built with Nvidia Xavier or other Jetpack platforms? I met a make error, Does someone have any suggestion?

this error is like that:

[ 46%] Linking CXX shared library ../../lib/libnvonnxparser.so /usr/bin/ld: cannot find -lcudart /usr/bin/ld: cannot find -lcublas collect2: error: ld returned 1 exit status onnx-tensorrt/CMakeFiles/nvonnxparser.dir/build.make:235: recipe for target '../lib/libnvonnxparser.so' failed make[2]: *** [../lib/libnvonnxparser.so] Error 1 CMakeFiles/Makefile2:92: recipe for target 'onnx-tensorrt/CMakeFiles/nvonnxparser.dir/all' failed make[1]: *** [onnx-tensorrt/CMakeFiles/nvonnxparser.dir/all] Error 2 Makefile:129: recipe for target 'all' failed make: *** [all] Error 2

I know it is a error about cuda path, but how can I modify the path cmake using?

thanks.

[meteorshowers]

you can communicate with me. I am doing the same thing. Wechat 16600286930 @ChienLiu

[facerless]

sudo ln -s /usr/local/cuda/lib64/libcudart.so /usr/lib/libcudart.so It works for me！You can try

[ElonKou]

Maybe you can try in this issue #39

[zhaoxuyan]

sudo ln -s /usr/local/cuda/lib64/libcudart.so /usr/lib/libcudart.so It works for me！You can try

It works for me too~ thanks a lot !

[yanlongbinluck]

Hi，what is the environment in Xavier？ Jetpack4.2 is cuda10，ubuntu18.04... by default. is it ok? Do we need to configure the following environment on Xavier? pytorch 1.0-1.1 ubuntu 1604 TensorRT 5.0 onnx-tensorrt v5.0 cuda 9.0 Thanks.

[eagle717]

@yanlongbinluck cuda10, u18.04 with JP42 is already ok for main branch

[yanlongbinluck]

@yanlongbinluck cuda10, u18.04 with JP42 is already ok for main branch

Yeah. Thank you.