------- Comment #1 from burnus at gcc dot gnu dot org 2010-01-14 11:30 ------- Hmm, without test case, it is quite difficult to debug. Though, maybe Jerry gets an idea from the backtrace.
Can't you write out "fmtstr" & create a program with "fmtstr" set such, set anzarg2 and call the same WRITE statement? Even if it does not work, knowing the "fmtstr" might help -- I could imaging the it has something to do with the format-cache patch - in that case also 4.4 would be affected. I think, at the end we really need a small testcase - either by reducing the big program or by dreaming up a new test case based on the clues one gets from backtracing & looking at the source. Reducing for run-time problems is unfortunately not as easy as for internal compiler errors (for which http://gcc.gnu.org/wiki/A_guide_to_testcase_reduction helps a lot, though it also works for run-time problems.) -- burnus at gcc dot gnu dot org changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jvdelisle at gcc dot gnu dot | |org Keywords| |wrong-code Known to fail| |4.5.0 Known to work| |4.3.5 Summary|SIGSEGV at |[4.5 Regression] SIGSEGV at |libgfortran/io/format.c:111 |libgfortran/io/format.c:111 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42742
