[Bug middle-end/58723] [4.9 Regression] ICE in lto_output_edge, at lto-cgraph.c:300 for OpenMP's simd reduction

[mpolacek at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58723

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mpolacek at gcc dot gnu.org

--- Comment #4 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Yeah, I hit exactly this as well, with DEF_INTERNAL_FN (UBSAN_NULL, ECF_LEAF |
ECF_NOTHROW).  The UBSAN_NULL is just like a macro, it expands to a GIMPLE_COND
later on.  It has to be leaf - otherwise it wouldn't be possible to have it in
the middle of a BB.  Temporarily, I disabled the tests with -flto.