Using Intel icpc compiler


I’m trying to specify the compiler for my nvcc.

$nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Sep__1_13:16:23_CDT_2017
Cuda compilation tools, release 9.0, V9.0.175
$icpc --version
icpc (ICC) 17.0.4 20170411
Copyright (C) 1985-2017 Intel Corporation.  All rights reserved.

To compile my file:

$ nvcc -o test -ccbin /usr/local/bin/icpc
nvcc fatal   : The version ('90000') of the host compiler ('clang') is not supported

How could I work around this?

Best Regards,

what is the result of

which icpc

are you on a mac? what OS is this?

$ which icpc

Yup! I’m on MacOSX version 10.12. OS Sierra.

I’d be happy if I could get help in configuring for g++ compiler as an alternate as well.

neither intel nor g++ are supported on MacOS for use with CUDA:

clang is the only option

My current clang is

$ cc --version
Apple LLVM version 9.0.0 (clang-900.0.37)
Target: x86_64-apple-darwin16.7.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin

I was trying to work around the incompatibility by either using intel or g++. However, the only option seems to be to downgrade to Apple LLVM version 8.1.0

Thanks for your help!