Thank you for the report on this. It does appear this should be implemented in the GCC 4.7 header/libs used when targeting the Intel Xeon Phi™ and thus a non-zero value should be returned. I reported this to compiler Development (see internal tracking id below). I am not sure whether this is a compiler/run-time issue or if it falls on the MPSS side but I will keep you posted on what I learn.
(Internal tracking id: DPD200257008)
