regressions with the 20060304 snapshot

[Paul Thomas]

Dominique and Steve,

This one came up last time I posted Polyhedron results for gfortran.

Searching on -ftree-loop-linear on Bugzilla yields:

19910 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19910> 	nor 	P2 	
	dberlin@gcc.gnu.org 	REOP 	
	[4.2 regression] ICE with -ftree-loop-linear
24309 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24309> 	nor 	P2 	
	
	NEW 	
	[4.1/4.2 Regression] ICE with -O3 -ftree-loop-linear
25211 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25211> 	nor 	P2 	
	
	NEW 	
	[4.1/4.2 Regression] verify_ssa ICE with -Os -ftree-loop-...
25449 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25449> 	nor 	P3 
i686-pc-linux-gnu 	
	UNCO 	
	[4.0] endless loop in nbench neural net with -ftree-loop-...
25937 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25937> 	nor 	P3 
powerpc64-linux 	
	NEW 	
	ICE with -O2 -ftree-loop-linear
26435 <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26435> 	nor 	P3 
powerpc64-linux 	
	NEW 	
	[4.1/4.2 regression] ICE with -O1 -ftree-loop-linear and ...

6 bugs found.

I understand that the gfortran problem is one and the same.

Paul