http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47023
--- Comment #16 from janus at gcc dot gnu.org 2011-10-17 19:22:15 UTC --- (In reply to comment #7) > By the way, I think one should reject proc-pointers for both SIZEOF and > C_SIZEOF. For C_SIZEOF because they are not interoperable (only type(c_funptr) > is) and for SIZEOF as the pointee size and not the pointer size is returned, > which does not make sense for functions. (gfortran currently returns 1 byte.) For C_SIZEOF, proc-pointers are rejected since 4.6. Example code: use iso_c_binding procedure(), pointer :: pp print *, c_sizeof(pp) end Error: 'x' argument of 'c_sizeof' intrinsic at (1) must be an interoperable data entity About SIZEOF I don't worry right now. Since it is an extension, we can basically do whatever we like.
