OpenACC Compiling with AMD GPUs

Question

I am having trouble to compile it with AMD Radeon R7 M360. I have compiled it with multi-core as shown below, but can anybody please tell me how to compile it with GPU, I have tried -ta=amd64/radeon but none of it works. You can see code in this image. I compiled it with

pgc++ -V -Minfo=accel -acc -ta=multicore sum.cpp -o /home/fawad/Desktop/sum

main:

 7, Generating Multicore code
      9, #pragma acc loop gang
  7, Generating reduction(+:sum)
 10, Loop is parallelizable

I need someone to guide me how to compile with AMD. I am using the PGI Compiler.

Answer 1

For a variety of business and technical reasons, support for Radeon GPUs was removed from the PGI compilers on both Linux and Windows platforms starting with the PGI 17.1 release.

Answer 2

Use PGI 16, check your card with pgaccelinfo. My card is a hawaii archtecture so the target flag is: -ta=radeon:hawaii

ex: pgc++ -V -Minfo=accel -acc -ta=radeon:hawaii,nollvm