https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102003
--- Comment #9 from anlauf at gcc dot gnu.org --- (In reply to Paul Thomas from comment #8) > If the offending line is replaced with > > integer :: lm > lm = p% c% len > > then ifx is happy. We're not here to please ifx, are we? ;-) In the past I reported a couple of inquiry issues (%kind, %len) at the Intel forum, but cannot find this one right now. Looks like I forgot; there might have been too many... My last strange finding in this regard was: https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-issue-with-kind-inquiry-of-complex-variable-when-passing-to/m-p/1645651
