Mainline regressions on powerpc-apple-darwin

[Andrew Pinski <pinskia at physics dot uc dot edu>]

Here are all of the mainline regressions on powerpc-apple-darwin, one of them
is a ABI problem.

I think I filed a bug for at least a couple of these.  Please who ever is responable
for breaking these, please fix them as I would like the regression tester to stop
barking at me for not fixing the regression.

native g++.sum g++.dg/gcov/gcov-1.C
	* gcov problem

native g++.sum g++.old-deja/g++.abi/cxa_vec.C
native g++.sum g++.old-deja/g++.brendan/new3.C
	* no overloading of operator new
	* most likely just need to add xfail powerpc-apple-darwin[78]*

native g++.sum g++.old-deja/g++.other/decl3.C
	* unknown

native g++.sum g++.old-deja/g++.other/warn7.C
	* unused parameter

native gcc.sum gcc.c-torture/compile/labels-1.c
native gcc.sum gcc.c-torture/compile/labels-2.c
native gcc.sum gcc.c-torture/compile/20021108-1.c
	* labels after returns (PR 15014)

native gcc.sum gcc.c-torture/compile/920501-4.c
	* ICE due to a patch (PR 15921)

native gcc.sum gcc.c-torture/execute/931004-12.c
	* ABI causing bug (PR 15923 and PR 14262)

native gcc.sum gcc.dg/altivec-varargs-1.c
	* gimplifier and __builtin_va_start and TREE_ADDRESSABLE (PR 15885)

native gcc.sum gcc.c-torture/execute/va-arg-22.c
	* unknown (most likely the same problem as altivec-varargs-1.c but I have not looked into it yet)

native gcc.sum gcc.dg/c99-complit-2.c
	* ICE in gimplifer after an error (happens after Kenner's big patch on June 21)

native gcc.sum gcc.dg/c90-const-expr-3.c
native gcc.sum gcc.dg/c99-const-expr-3.c
	* unknown, most likely a testcase bug

native gcc.sum gcc.misc-tests/gcov-4.c
	* gcov problem

native objc.sum objc.dg/const-str-3.m
native objc.sum objc.dg/const-str-4.m
	* two Objective-C strings are not the same even though they should be.

native objc.sum objc.dg/const-str-5.m
	* Objective-C string not considered constant


Thanks,
Andrew Pinski