[ tree-ssa] PRE problem

[law@redhat.com]

In message <1077310216.3169.210.camel@localhost.localdomain>, Diego Novillo wri
tes:
 >On Fri, 2004-02-20 at 15:31, law@redhat.com wrote:
 >
 >> It must be Friday...  Time to get away and recharge.
 >> 
 >So that you feel better, a checking-enabled bootstrap just ICE'd in
 >propagate_copy trying to build strstream.cc.   :)
I don't think may_propagate_copy needs/should be checking the decl_hard_reg
or occurs_in_abnormal_phi stuff for virtual operands.  It just doesn't make
any sense and gets in the way of doing useful stuff in the optimizers.

I think we'll just want to do the memory tag test for virtuals.

jeff