On 5/3/2014 1:43 AM, Mark Farnell wrote:
Suppose if I write a C++ application that contains hints to offload
part of the code to the Intel Xeon phi (MIC), then what compile do I
pass to allow generation of code that offload the code to MIC? Also,
if I want to have a native MIC binary which the whole application runs
inside the MIC card, what compile flag should I use instead?
Finally, does gcc-4.9 support the same set of MIC-related keywords as
the Intel Cluster Studio compiler, including the offloading keyword,
and keywords that mark variables as shared between the CPU and MIC?
All communications I've seen indicate that gcc development toward
supporting MIC is based on OpenMP 4.0, and currently the target clauses
support only host (when -fopenmp is set).
As far as I've seen, gcc support for MIC native operation remains
minimal, without auto-vectorization or openmp.
--
Tim Prince
