https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123782
--- Comment #10 from Ivan Pribec <ivan.pribec at gmail dot com> --- > Does someone want to ask the Intel people for their opinion? I asked and received the reply "the compile time error makes sense to me that "any part other than final" refers to "a" in "a%x", and "a" it's a 1-d array of size 5, so its rank is 1, and it's hard for the compiler to justify if "a%x" is contiguous at compile time when assigning it to an explicitly declared CONTIGUOUS pointer." and the further justification that gfortran also flagged this as an error.
