http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53346
--- Comment #9 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2012-05-18 14:49:22 UTC --- > Unfortunately, the loop in original rnflow test still gets vectorized, with no > change in the runtime: Confirmed, at revision 187655 I still get -O2 21.8 -O2 -ftree-vectorize 27.9 -O3 25.2 -O3 -fno-tree-vectorize 21.4 Uneducated guess: is it possible that failed attempts to vectorize may mess up further optimizations?
