teensmaio.blogg.se

Identifying which GPU card is installed and what version

NVRM version: NVIDIA UNIX x86_64 Kernel Module 367.48 Sat Sep 3 18:21: You can also use the kernel to run a CUDA version check: ~ $ cat /proc/driver/nvidia/version } Identifying which CUDA driver version is installed and active in the kernel Int driver_version = 0, runtime_version = 0 You can find a full example of using cudaDriverGetVersion() here: #include The API call gets the CUDA version from the active driver, currently loaded in Linux or Windows. When you’re writing your own code, figuring out how to check the CUDA version, including capabilities is often accomplished with the cudaDriverGetVersion () API call. Figure out which one is the relevant one for you, and modify the environment variables to match, or get rid of the older versions. Note that if the nvcc version doesn’t match the driver version, you may have multiple nvccs in your PATH. This means that we have CUDA version 8.0.61 installed. You can check nvcc -version to get the CUDA compiler version, which matches the toolkit version: ~ $ nvcc -versionĬopyright (c) 2005-2016 NVIDIA CorporationĬuda compilation tools, release 8.0, V8.0.61 If you have installed the CUDA toolkit but which nvcc returns no results, you might need to add the directory to your path. If that appears, your NVCC is installed in the standard directory. You should see something like /usr/bin/nvcc. Run which nvcc to find if nvcc is installed properly. Check if CUDA is installed and it’s location with NVCC There are several ways and steps you could check which CUDA version is installed on your Linux box.