[Bug fortran/83042] [8.0 regression] ICE on valid code

2017-11-18 Thread juergen.reuter at desy dot de
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83042 --- Comment #4 from Jürgen Reuter --- Created attachment 42647 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42647&action=edit v3 of reproducer, still smaller

[Bug fortran/83042] [8.0 regression] ICE on valid code

2017-11-18 Thread juergen.reuter at desy dot de
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83042 --- Comment #3 from Jürgen Reuter --- Created attachment 42646 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42646&action=edit v2 of reproducer, much smaller

[Bug fortran/83042] [8.0 regression] ICE on valid code

2017-11-18 Thread juergen.reuter at desy dot de
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83042 --- Comment #2 from Jürgen Reuter --- Shorter version with only 3 files with 189 lines in total. For the ICE to appear it is important that the modules reside in separate files. (?)

[Bug fortran/83042] [8.0 regression] ICE on valid code

2017-11-18 Thread juergen.reuter at desy dot de
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83042 --- Comment #1 from Jürgen Reuter --- Have to extend the range to r253161 as possible earliest svn version.