assertify darwin, and some missed rs6000 bits

[Nathan Sidwell]

Mike Stump wrote:
> On Sunday, May 1, 2005, at 11:32  AM, Nathan Sidwell wrote:

>>     * config/darwin-crt2.c (darwin_unwind_dyld_remove_image_hook): Use
>>     gcc_assert or gcc_unreachable as appropriate.
> 
> 
> mrs $ grep abort libgcc2.c
> /* Don't use `fancy_abort' here even if config.h says to use it.  */
> #ifdef abort
> #undef abort
>     abort ();
>     abort ();
>     abort ();
>     abort ();
> 
> ?  I'd rather think about this some more or have you do libgcc2 first.  
> Offhand, I don't think I don't think I'd want this bit.  crt2 is a 
> special and I don't think I'd want any dependency from it to anything 
> other than those already present.

the aborts in libgcc2 really should be abort (one of my goals is to try
and remove the #define abort fancy_abort, to make this kind of thing
more obvious).  It might be that the #undef it contains is a relic from
when it included config.h rather than tconfig.h.  darwin-crt2.c already
includes the correct header to get the runtime version of gcc_assert --
which just calls abort.  I shall revisit darwin-crt2.c later though.

> Ok for the rest of the patch.
Thanks

nathan

-- 
Nathan Sidwell    ::   http://www.codesourcery.com   ::     CodeSourcery LLC
nathan@codesourcery.com    ::     http://www.planetfall.pwp.blueyonder.co.uk