-
What is your question? environment:
But I got compile error when compiling
I also checked path could anyone help me fix the compile issue? thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 6 comments
-
Hi @hsgui, can you provide details of how you are compiling? |
Beta Was this translation helpful? Give feedback.
-
/usr/local/gcc12/bin/g++ cuvs_hello.cc -o cuvs_hello -I /usr/local/cuda/include/ -I /data/workspace/conda/envs/cuvs/include -L /data/workspace/conda/envs/cuvs/lib/ -L /usr/local/cuda/lib64/ -DSPDLOG_FMT_EXTERNAL=1 besides, I'm building cuvs from C++ source file from this wiki:
|
Beta Was this translation helpful? Give feedback.
-
I think i need upgrade the cuda toolkit maybe? the but my [root@xxx cuda-demo]# nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Feb__7_19:32:13_PST_2023
Cuda compilation tools, release 12.1, V12.1.66
Build cuda_12.1.r12.1/compiler.32415258_0 |
Beta Was this translation helpful? Give feedback.
-
Using a more recent CUDA toolkit is one option, but not the option RAPIDS relies on. We use cmake to fetch CCCL from GitHub. Building manually you will need to provide a -I line to point to CCCL include directory. Using cmake will save you a lot of trouble. You could base it off of the cmakelists.txt from the RMM tests or benchmarks. |
Beta Was this translation helpful? Give feedback.
-
thanks very much @harrism I'll play with cmake |
Beta Was this translation helpful? Give feedback.
-
I'm going to convert this to a discussion. You can "Mark as answer" a comment that helps if / when you are satisfied. Thanks! |
Beta Was this translation helpful? Give feedback.
I'm going to convert this to a discussion. You can "Mark as answer" a comment that helps if / when you are satisfied. Thanks!