Hi! I run
test CUDA to ensure that everything works as expected. But some errors occurred as below. Are these errors related to the test file or the configuration of the system? How can I resolve them? Thanks!
I have also tried the same thing, like configuring Nvidia driver and CUDA toolkit again on another virtual machine, but still got the same errors.
And here is the CUDA version info.
That’s a bug in the CUDA.jl test suite; it looks like
llvm.nvvm.wmma.m8n32k16.load.a.col.stride.u8 is only supported on Turing (sm_75+). Please file an issue, including this information and a full backtrace in textual form (i.e. not a screenshot). Also make sure to note which version of CUDA.jl you are using (versioninfo() currently does not report that).
Ok, I will do that. Thanks!