------- Comment #7 from pault at gcc dot gnu dot org 2007-05-11 13:13 ------- (In reply to comment #6)
> So I suspect that the problem lies in the interfacing around > gfc_conv_function_call but I just don't see it yet. Indeed this is the case: In Tobias' #5, changing the declaration for 'words' to - character (len=5) :: words(0:1) = (/"two ","three"/) produces a code that works correctly. Looking at the code, the hidden calculation of len=sum(len_trim(words)), made by the interface for the function call has an offset of 0, rather than -1. Now to find out why! Paul -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31867
