[Bug tree-optimization/63836] [5 Regression] r217349 caused segfault building 178.galgel from cpu2000

hjl.tools at gmail dot com gcc-bugzilla@gcc.gnu.org
Wed Nov 12 18:25:00 GMT 2014


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63836

--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> ---
A simple testcase:

[hjl@gnu-mic-2 delta-fortran]$ cat foo.f90
      SUBROUTINE DTRSM ( NN, NB)
         DO 50 J = NN, 1, -NB
   50    CONTINUE
      end
[hjl@gnu-mic-2 delta-fortran]$
/export/project/git/gcc-regression/master/217349/usr/bin/gfortran -ffixed-form 
-ffixed-line-length-132 -DSPEC_CPU2000_LP64 -ffast-math -S  foo.f90 -O2
foo.f90:4:0:

       end
 ^
Error: non-trivial conversion at assignment
unsigned int
integer(kind=4)
_13 = _7;
foo.f90:4:0: internal compiler error: verify_gimple failed
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.



More information about the Gcc-bugs mailing list