http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58359
--- Comment #12 from Anatoly Sinyavin <a.sinyavin at samsung dot com> ---
Does it mean that my solution is not accepted?
If it's so I am going to think about two approaches
- vectorizer should ignore that path (Richard Biener 2013-09-09 08:22:53
UTC)
- replacing the GIMPLE_COND, leading to a block that contains only
a call to __builtin_unreachable, with an ASSERT_EXPR, or range information
on this SSA_NAME (now that we store it), or anything without control flow (Marc
Glisse 2013-10-24 12:03:18 UTC)
