AMD/ROCm "AOMP" Compiler Enables OpenMP 5.0 By Default, Preps For More Upstreaming
Written by Michael Larabel in Radeon on 30 June 2020 at 07:30 PM EDT. 1 Comment
RADEON --
AOMP 11.6-2 is out this evening as the final Radeon Open Compute update for H1'2020. This is AMD's LLVM Clang downstream focused on providing OpenMP offloading to Radeon GPUs.

AOMP is ROCm's OpenMP focused compiler based on Clang that launched with ROCm 3.0 at the tail end of last year. Since then AOMP releases have come fairly frequently and AMD has been working to upstream the code where possible/relevant into upstream LLVM/Clang.

With today's AOMP 11.6-2 release, the latest LLVM 11 sources as of earlier this month are used. On top of rebasing against the latest upstream, the big change with this release is OpenMP 5.0 support is now enabled by default. OpenMP 4.5 support can still be targeted using the -fopenmp-version=45 switch.

This release also has work in preparing to upstream the ROCm compiler into mainline LLVM, cleanups to the code to also help in upstreaming, and various other fixes and improvements. OpenMP 5.0 support is great but it's equally exciting that they are preparing to upstream more significant portions of their Radeon GPU compiler code into upstream LLVM.

AOMP 11.6-2 downloads in source form as well as RPMs for both RHEL/CentOS and SUSE Linux Enterprise Server and Ubuntu/Debian packages via GitHub.
Related News
About The Author
Author picture

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter or contacted via MichaelLarabel.com.

Popular News This Week