[patch] LTO support for Mach-O (apple-darwin)

[Iain Sandoe]

On 14 May 2010, at 13:06, Steven Bosscher wrote:

> On Fri, May 14, 2010 at 7:48 AM, Iain Sandoe  
> <idsandoe@googlemail.com> wrote:
>> could someone OK this?
>> ...  it works for me on powerpc*-apple-darwin for several days...

Jack forwarded the patch to me ..  I've been testing it along with my  
other activities ..

> Works with --enable-lto?

yes - we need to hide it behind this until we iron out the big(ger)  
issues.

> I thought Jack reported lots of test suite
> failures still? Can you please confirm you have tested this patch,
> with --enable-lto, and don't have (too m)any test suite FAILs?

With this patch powerpc-apple-darwin9  fails in the same ways/places  
as i686-apple-darwin9.

m32 still has more fails than m64.

I am working on putting a small suite into ObjC and ObjC++ ...

... and whilst doing that trying to see what I can figure out about  
the m32 fails.

===

If you want to hold back the *-apple-darwin* then, fair enough, but I  
don't see that there's a huge difference between experimental on  
x86_64/powerpc64 and i686/powerpc :-)

I will do another powerpc64-apple-darwin9 build later...

(but more interested in trying to iron out the m32 fails).

Iain