Why does nvcc care what’s in the PATH if I am explicitly passing compiler path using -ccbin. This option should override whatever is in the PATH. This error is completely absurd…
I’ve found few machines where NVCC fails to compile CUDA examples with
nvcc fatal : Compiler ‘cl.exe’ in PATH different than the one specified with -ccbin
or simply errorlevel 1
Reason was in anaconda setting
HKEY_CURRENT_USER\Software\Microsoft\Command Processor\AutoRun
see
for explanation