"nvrtc: error: failed to load builtins for compute_72", but compute_70 works

Accelerated Computing CUDA CUDA Programming and Performance
1

I have a system running Red Hat Enterprise Linux 7.7. The system is equipped with an RTX 2080 Ti GPU and has CUDA 10.1 installed. I am trying to run a program that uses NVRTC for run-time compilation. The program queries the compute capability of the device (7.2) and uses that to construct an -arch=compute_72 option for the call to nvrtcCompileProgram(). Unfortunately, this fails with an error message:

nvrtc: error: failed to load builtins for compute_72.

If I modify the program to use =arch=compute_70, then things work, but I’m wary of whether this can have some performance implications.

What could be going wrong?

2

You can use the gencode argument to use the Compute 7.0 architecture and to generate SASS assembly code specifically for the 7.2 version of it.

-gencode arch=compute_70,code=sm_72

I haven’t really played with the NVRTC interface yet, so I do not quite know what is possible with it and what isn’t.

Christian

3

RTX2080Ti is compute capability 7.5 not 7.2

Based on what you have described here, you shouldn’t be using compute capability 7.2 at all.

4

You’re right, the program was wrong to pick compute_72.

(Now, for mysterious reasons, using compute_75 makes it run much slower than when using compute_60 - but that’s a different concern.)