Intel® C++ Compiler
Community support and assistance for creating C++ code that runs on platforms based on Intel® processors.

Can we offload openMp to the GPU ?

amit_l_
Beginner
544 Views

Hi,

I'm using Ubuntu 14.04.

1. Is there a way to use openMp and offload the parallel code into the GPU ? (intel HD graphics) ?

if yes:

2. which icc version do I need ? (can I do it with gcc ?)

3. which Intel processors are supported ?

 

Thanks

0 Kudos
3 Replies
TimP
Honored Contributor III
544 Views

You might start by reading some of the white papers already posted by Intel which your search engine ought to flag, e.g.

https://software.intel.com/en-us/articles/how-to-offload-computation-to-intelr-graphics-technology

https://software.intel.com/en-us/articles/pldi-tutorial-using-the-intelr-c-compiler-for-general-purpose-computation-offload-to-intelr

Perhaps the authors of the latter would answer as to whether there is released documentation, since they apparently presented on these subjects recently.

0 Kudos
amit_l_
Beginner
544 Views

Ok, It seems that there is a solution for CILK.

But what about openmp ?

0 Kudos
TimP
Honored Contributor III
544 Views

amit l. wrote:

Ok, It seems that there is a solution for CILK.

But what about openmp ?

That last URL I gave you states that the authors discussed OpenMP in their presentation.  I don't know the details.

0 Kudos
Reply