https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118471
--- Comment #4 from anlauf at gcc dot gnu.org --- (In reply to Tobias Burnus from comment #2) > Seems as if a special case for Fortran pointers is needed, where the a->span > has to be replaced by the byte size of the base type. > > (Needs some special casing for polymorphic types, and deferred-length > character/derived type vars - and be it by just not replacing it by the byte > size but keeping a->span.) If we are talking performance, handling intrinsic types only will likely be good enough. I'd expect significant overhead for derived types. For rank > 1, going further, e.g. loop collapse, might be attractive, too.
