Don't know the answer to the question about int vs. size_t for min|max_ind reducers. Note that Intel cilk(tm) plus uses size_t in several contexts where they must be cast to (int) in order to attain satisfactory performance, but the reducers in question don't give good performance, so may be using size_t internally.
(reposting) a comment was added to the bugzilla hinting that gcc cilkplus is likely not to get future support, following earlier comments that they didn't know whether Intel remained interested.
