CUDA distributions on Linux used to have a file named version.txt which read, e.g.:
CUDA Version 10.2.89
which is quite useful. However, as of CUDA 11.1, this file no longer exists.
How can I determine, on Linux and from the command line, and inspecting /path/to/cuda/toolkit, which exact version I'm looking at? Including the subversion?
(Answer due to @RobertCrovella's comment)
This will do the trick:
/path/to/cuda/toolkit/bin/nvcc --version \
| egrep -o "V[0-9]+.[0-9]+.[0-9]+" \
| cut -c2-
And of course, for the CUDA version currently chosen and configured to be used, just take the nvcc that's on the path:
nvcc --version | egrep -o "V[0-9]+.[0-9]+.[0-9]+" | cut -c2-
For example: You would get 11.2.67 for the download of CUDA 11.2 which was available this week on the NVIDIA website.
The full nvcc --version output would be:
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Nov_30_19:08:53_PST_2020
Cuda compilation tools, release 11.2, V11.2.67
Build cuda_11.2.r11.2/compiler.29373293_0