https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81938
--- Comment #6 from Rimvydas (RJ) <rimvydas.jas at gmail dot com> --- (In reply to Dominique d'Humieres from comment #4) > Thanks for working on this issue. > > The patch in comment 2 fixes this PR along with the failures for > gfortran.dg/fmt_cache_1.f and gfortran.dg/fmt_cache_2.f reported in pr78672. > > Patches should be submitted to fort...@gcc.gnu.org and > gcc-patc...@gcc.gnu.org for review. Do you have write access to SVN? No, I do not have write access to gcc SVN. Just a part of technical staff at our institute working to ensure portability of our scientific models. If patch is OK(still think whole loop could be rewritten for better readability, but runtime libraries tend to be tricky) could we expect fix backported to GCC7 branch? Thanks in advance.