Hi,
Does anyone have experience running CUDA on VM with Linux OS (using Oracle Virtual box) ?
Best regards,
–Elan.
It probably won’t work for actually running CUDA codes, unless that particular solution has a GPU passthrough capability.
Hi txbob,
Thanks for the info. If that is the case I have to write CUDA code on Windows OS since I don’t have a linux machine. Will it be possible to use Nsight Eclipse on IDEs other than Visual Studio such as Eclipse? If yes, is there any link that I can follow to setup the initial settings for Eclipse?
Best regards,
–Elan.
No, the only supported environment for CUDA development on windows is VS. VS2013 community edition can be downloaded at no cost.
I’ve installed the VS2015 Community version. However, I don’t see Cuda Linker in the project properties. Is there any links that I can refer to, to correctly configure the project so that CUDA code can be compiled and run on VS2015 Community version?
Which CUDA version are you trying to use?
I suggest you carefully follow the windows installation guide for whichever CUDA version you are trying to use
I have installed CUDA 7.5 and Nsight prior to installing VS2015 community. It seems that VS2015 doesn’t know that CUDA and nsight is already available.
VS 2015 is not officially supported by CUDA 7.5
Furthermore, VS needs to be installed before CUDA.
Please follow the directions in the CUDA windows installation guide.
Which version of CUDA is supported by VS2015 community version?
I installed CUDA ver8.0 (after installing VS2015 community), however when I build the CUDA samples I’m getting the following errors:
Severity Code Description Project File Line Suppression State
Error C1083 Cannot open include file: ‘corecrt.h’: No such file or directory clock C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\include\crtdefs.h 10
CUDA 8.0 is the only CUDA version that officially supports VS 2015.