https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113621
--- Comment #1 from anlauf at gcc dot gnu.org --- (In reply to seurer from comment #0) > This appears to be a problem just on big endian. This is only for -m32, right? > Program received signal SIGSEGV: Segmentation fault - invalid memory > reference. > > Backtrace for this error: > #0 0xf7f803f3 in ??? > #1 0x10000c68 in __m_char_MOD_three_val > at > /home/seurer/gcc/git/gcc-test2/gcc/testsuite/gfortran.dg/optional_absent_10. > f90:97 > #2 0x10001bcb in __m_char_MOD_test_char > at > /home/seurer/gcc/git/gcc-test2/gcc/testsuite/gfortran.dg/optional_absent_10. > f90:19 > #3 0x1000269b in p > at > /home/seurer/gcc/git/gcc-test2/gcc/testsuite/gfortran.dg/optional_absent_10. > f90:218 > #4 0x100026f3 in main > at > /home/seurer/gcc/git/gcc-test2/gcc/testsuite/gfortran.dg/optional_absent_10. > f90:214 > Segmentation fault We likely mishandle the passing-by-value for character(len=4). There previously was an issue with character(len=1), see pr110419, fixed by Mikael. This also occurred only on Power big endian for -m32.
