[Bug tree-optimization/19038] [4.0 Regression] out-of ssa causing loops to have more than one BB

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2004-12-24 12:55 -------
Possible work-arounds/hacks:
- Don't copy-prop into a PHI argument when the root variable of the
  PHI result is the same as the root variable of the copy that is
  being replaced.  Seems silly that this would be necessary.
- Try to insert copies before conditional jumps if the jump condition
  does not depend on anything in the copy.
- Figure out why we don't coalesce temp1 and temp2.  Can we coalesce
  them?  If so, why are we failing to do it?



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19038