GCC and G++ for RS/6000 ?

[Ronald S. Woan]

In article <1990Aug4.225222.27970 at m2xenix.psg.com>, steven at m2xenix.psg.com (Steven Furber) writes:
> From what conversations I have had with the local IBM RS/6000 SE's, GNU
> products are not going to be ported or supported by IBM; EMACS was
> dropped because of the current lawsuit that is going on.

I think it would be better termed as a potential lawsuit by Unipress.

> Something that I pointed out to the SE who was at work last week was
> that the XLC compiler seems to choke on a lot of "standard" C code,
> making easy ports that work on other *NIX platforms a pain.  When he
> asked what I saw as a solution I suggestd GCC, since it would open up
> the doors to the rest of the GNU universe and the code written thereof.
> Since there is (from my understadning) no money that IBM can make off of
> a port, the SE doesn't see where IBM would get into it-- even if the
> lawsuit is settled.

I hate it when people say things like this without backing it up. Could you
post some examples of the "standard" (I take it this means K&R) C code 
that xlc chokes on? I have a feeling it is mainly due to a lack of
understanding of the header standards (did you look at the BSD porting
document in /usr/lpp/bos?). GCC would be nice, but then you would loose all
of the optimizations performed by xlc (very significant from recent
comp.arch postings) that cannot be handled by gcc in its current form. GCC
is great for 680?0 code but it leaves much to be desired for the current
generation of RISC processors, especially the IBM one.

					Ron

-- 
+-----All Views Expressed Are My Own And Are Not Necessarily Shared By------+
+------------------------------My Employer----------------------------------+
+ Ronald S. Woan                                     woan at peyote.cactus.org +
+ If nothing else works                              woan at soda.berkeley.edu +