https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58618
--- Comment #10 from Paul Thomas <pault at gcc dot gnu.org> --- (In reply to Dominique d'Humieres from comment #9) > With the patch in comment 8, the wrong-code is replaced with an ICE > > pr58618.f90:2:0: > > 2 | associate (x=>s(1)(1:1)) > | > internal compiler error: in fold_convert_loc, at fold-const.c:2425 > > (configured with --enable-checking=yes). That's weird - I am even more puzzled then :-( Thanks Paul
