xargs: clang-tidy: No such file or directory #84
Comments
|
Can you try installing clang-tidy using |
|
Alternatively, if you don't want to install clang-tidy you can do the following: |
|
[ 4%] Built target gtest |
|
after install clang-tidy, execute .sh file i got more errors. |
|
To me it seems as if your compiler is not correctly set |
|
compiler? what's the correct set? can you give me some tips? |
|
i run the .sh file in docker container. |
|
Is gcc and g++ installed? |
|
yes. here is the version info. (base) root@156f417d667f:/# g++ --version |
|
My bad not gcc, is the CUDA toolkit installed? |
|
i use this docker env to train my deep learning models, so I skipped the cuda env checking steps. in the container, i can use nvidia-smi, but not nvcc -V. in my sys path, there is the folder /usr/local/cuda---->/usr/local/cuda10.2. did this satisfied the conditions? |
That seems to be the issue, the CUDA compiler is not there. |
|
thank you, i will try that. |
|
i add the env variables in the .bashrc file, and now nvcc -V can print the normal info.Then i re-execute the .sh file, got the same question. |
|
even i install the 11.2 cuda version, it's still the same question. |
|
In our CI workflow, you see the exact setup steps that are required to get a bare ubuntu 18.04 docker to compile this code. Do you see a difference to your setup? |
|
the last two steps, i didn't execute the related commands. i don't know if this is the reason. i will try that tomorrow on my computer. thanks for the advice. |
|
Right, the last two steps are only static code analysis, nothing you need to run. |
|
can you provide a dockerfile that i can reproduce the env? or a docker image that i can docker pull from the docker hub? |
|
I have never used Docker, so I can't relly help much with that. |
|
I'm throwing a file together ad-hoc here from our setup instructions and CI, no guarantees for anything 😉 FROM ubuntu:18.04
ENV DEBIAN_FRONTEND noninteractive
RUN apt-get update && \
apt-get install -y --no-install-recommends \
libopencv-dev \
libglfw3-dev \
libglew-dev \
libglm-dev \
nvidia-cuda-toolkitThis should be enough to build the project. To execute our code, you'll need to also setup a CUDA runtime as described in the README inside this container. |
|
i'm stuck in this question for a few days. thank you for your kind advice. i will try that today. If I have any new information, I will reply here. |
|
Scanning dependencies of target dogm_spec still the same question…… |
|
I just tried to reproduce your problem on a pretty fresh Ubuntu 18.04 install, I can't. I did this: sudo apt install libopencv-dev
sudo apt install libglfw3-dev libglew-dev libglm-dev
sudo apt install nvidia-cuda-toolkit
export PATH=/usr/local/cuda/bin:$PATH
cd dynamic-occupancy-grid-map/dogm
mkdir build
cd build
cmake ..
make -jThis builds successfully. What commands are you trying? |
|
the difference i think is the installation of cuda. i use the installation commands on the nvidia website. I'm not sure what's the difference between the two installation methods. because i cannot connect to my remote computer, i will try your method tomorrow. another question. i use the .local_pipeline_ubuntu.sh according to the readme file to compile the project. is it the same function as "cd build; cmake ..; make -j"? |
|
|
|
use this command, i can build the project smoothly. but when i execute the build/demo/demo file, it occurs segmentation fault(core dumped). |
|
What is the error message just before the segfault? I guess there's a version mismatch between the CUDA versions with which you compile and with which you run.
Just execute it with |
|
it occurs like this: |
|
To determine your installed cuda and nvcc version, what is the output of these commands? nvcc --version
cat /usr/local/cuda/version.txt
ls -l /usr/local | grep cuda |
No description provided.
The text was updated successfully, but these errors were encountered: