[Bug fortran/59654] Broken function table with complex OO use case

[Thomas.L.Clune at nasa dot gov]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59654

--- Comment #5 from tlcclt <Thomas.L.Clune at nasa dot gov> ---
Ignore that output from Valgind.  I was logged out between building and
running, and failed to reload the appropriate environment.   With the correct
environment, I get a clean run from valgrind:

valgrind ./a.out
==9886== Memcheck, a memory error detector
==9886== Copyright (C) 2002-2009, and GNU GPL'd, by Julian Seward et al.
==9886== Using Valgrind-3.5.0 and LibVEX; rerun with -h for copyright info
==9886== Command: ./a.out
==9886==
--9886-- WARNING: Serious error when reading debug info
--9886-- When reading debug info from /gpfsm/dhome/tclune/a.out:
--9886-- Ignoring non-Dwarf2/3 block in .debug_info
--9886-- WARNING: Serious error when reading debug info
--9886-- When reading debug info from
/usr/local/other/SLES11.1/gcc/4.9/lib64/libgfortran.so.3.0.0:
--9886-- Ignoring non-Dwarf2/3 block in .debug_info

…

--9886-- Ignoring non-Dwarf2/3 block in .debug_info
--9886-- WARNING: Serious error when reading debug info
--9886-- When reading debug info from
/usr/local/other/SLES11.1/gcc/4.9/lib64/libquadmath.so.0.0.0:
--9886-- Ignoring non-Dwarf2/3 block in .debug_info
 Error - incorrect number of tests were run.
==9886==
==9886== HEAP SUMMARY:
==9886==     in use at exit: 1 bytes in 1 blocks
==9886==   total heap usage: 22 allocs, 21 frees, 11,835 bytes allocated
==9886==
==9886== LEAK SUMMARY:
==9886==    definitely lost: 0 bytes in 0 blocks
==9886==    indirectly lost: 0 bytes in 0 blocks
==9886==      possibly lost: 0 bytes in 0 blocks
==9886==    still reachable: 1 bytes in 1 blocks
==9886==         suppressed: 0 bytes in 0 blocks
==9886== Rerun with --leak-check=full to see details of leaked memory
==9886==
==9886== For counts of detected and suppressed errors, rerun with: -v
==9886== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 2 from 2)