[Bug optimization/13875] [tree-ssa] missed jump thread optimization on the tree-level

["law at redhat dot com" <gcc-bugzilla at gcc dot gnu dot org>]

------- Additional Comments From law at redhat dot com  2004-03-09 23:37 -------
Subject: Re:  [tree-ssa] missed jump thread 
 optimization on the tree-level

In message <20040308235126.24783.qmail@sources.redhat.com>, "pinskia at gcc dot
 gnu dot org" writes:
 >
 >------- Additional Comments From pinskia at gcc dot gnu dot org  2004-03-08 2
 >3:51 -------
 >The problem here (I think) is that DOM cannot read inside a && yet, there is
 >already a testcase on the branch which is XFAIL'ed for the problem.
I'm not sure what you mean here, and optimizing this test really isn't a
job for the dominator optimizer.

The way I see to optimize this particular case would require us to determine
that there is an equivalence between a and b at the head of block #3 by 
knowing that a & b were equivalent on each incoming edge to block #3.  We
would then need to use that equivalence to thread the edge from 4->5 to
instead reach 6.

Note that block #3 is a merge point which really means this isn't a job
for the dominator optimizer.

A smart GVN pass _might_ be able to discover the equivalence.

Jeff



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13875