[Bug tree-optimization/38785] [4.4/4.5/4.6/4.7 Regression] huge performance regression on EEMBC bitmnp01

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38785

--- Comment #28 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-01-16 13:14:22 UTC ---
Re-confirmed.  Regular PRE has got some kind of a cost-model (restricting
insertion to the case where we eliminate a redundancy on a fast path).
Partial-insertion has not been updated in a similar manner.