Unable to detect CUDA version during model compilation for LLVM

Troubleshooting
#1

Hello, I’m compiling model using just LLVM support for arm device. And I’m getting following information:

[12:47:29] /tvm/src/target/target_kind.cc:164: Warning: Unable to detect CUDA version, default to "-mcpu=sm_20" instead
[12:47:29] /tvm/src/target/target_kind.cc:190: Warning: Unable to detect ROCm compute arch, default to "-mcpu=gfx900" instead
[12:47:29] /tvm/src/target/target_kind.cc:204: Warning: Unable to detect ROCm version, assuming >= 3.5
[12:47:29] /tvm/src/target/target_kind.cc:164: Warning: Unable to detect CUDA version, default to "-mcpu=sm_20" instead
[12:47:29] /tvm/src/target/target_kind.cc:190: Warning: Unable to detect ROCm compute arch, default to "-mcpu=gfx900" instead
[12:47:29] /tvm/src/target/target_kind.cc:204: Warning: Unable to detect ROCm version, assuming >= 3.5

Should I worried about these warning messages?

1 Like