Recent gfortran regression on i686-pc-linux-gnu: Program aborted. Backtrace: #0 0xF7EC75C3 #1 0xF7EC903F #2 0xF7F80A67 #3 0x80488F4 in a3a1.2065.constprop.8 at class_optional_2.f90:0 FAIL: gfortran.dg/class_optional_2.f90 -O2 execution test Regressed between revision 195413 vs revision 195410 [1]. Added CCs of suspected comitters. [1] http://gcc.gnu.org/ml/gcc-regression/2013-01/msg00255.html
This is probably a duplicate of PR 55978 (which, however, was claimed to be due to r195125, which does not match the interval in comment 0).
As said in pr55978 comment #2, the bug is latent on some platforms (e.g., darwin) so it may not be triggered by the testsuite and may need to use valgrind to see it. As any latent bug it may be exposed in the test suite by unrelated changes (I still don't see it on darwin at r195432). Si I am marking the PR as a duplicate of pr55978. *** This bug has been marked as a duplicate of bug 55978 ***