In the IBM single-source OpenMP compiler (IBM XL C/C++ Alpha Edition
for Multicore Acceleration for Linux v0.9 of the Single-Source) there is
a limitation documented in the README:
"...The OpenMP parallel regions will be processor-agnostic, and as such
processor-specific intrinsics will not be available for use inside them."
Is this limitation removed in version 10.1 of the XL C/C++ for Multicore
Acceleration for Linux?
I hope to call routines in a GCC-compiled library from within OpenMP