On Related Note; Any news for OpenMP with LLVM?

As of April 2013, the stuff that was done was:

   Implemented and committed
      -fopenmp option
      #pragma omp threadprivate
      Parsing and semantic analysis, AST representation

From slide 20 of this presentation: http://llvm.org/devmtg/2013-04/bokhanko-bataev-slides.pdf

AMD's OpenMP implementation required support from both Clang and the LLVM IR, while Intel’s latest solution does it all in Clang. Intel’s Clang only design appears to be on its way to becoming the standard. Performance wise, LLVM needs all the help it can get to catch up to gcc and Intel’s compilers. Hopefully this is the right technical decision and not to appeasing some obstructive member of the LLVM Developer Group committee. Any compiler guys follow this topic?

With so many projects using OpenMP, this is great news.

The threading affinity stuff (2.5.2 in the spec) looks pretty awesome.