🚀 go-pugleaf

>In article <1991Jun22.145936.14935@monu6.cc.monash.edu.au> anthony@monu6.cc.monash.edu.au (Anthony Baxter) writes:
>> Has anyone had any luck using g++ on a Pyramid? I've managed to get
>> it compiled, but it produces assembler with labels like __$_FooBar,
>> and the Pyramid as doesnt allow $ as a valid character in a label.
>> Has anyone had any luck with this, or could anyone tell me where
>> I might begin looking in the code to fix this?
>> (btw, its a Pyramid MIS-2/01, running OSx 5.1a-910305)

and

In article <1991Jun23.015823.25711@comp.vuw.ac.nz> jonathan@beta.isor.vuw.ac.nz (Jonathan Stone) answers:
> The short answer is, unfortunately, that it may not be worth your
> while trying to get g++ working on a Pyramid.
>
>Enjoy,
>--Jonathan

I #defined NO_DOLLAR_IN_LABEL somewhere in the g++-Sources, I used the
GNU-ld that is distributed with the GNU-binutils, plus a few patches
(may those patches are in the distribution by now, I didnt check
recently), and I applied the VUW-patches from ftp.isor.vuw.ac.nz (by
Hand, to g++-1.39, btw). With all this, only 4 of the libg++ - tests
will coredump, opposed to ``all but 2'' with unmodified g++.
And then, we got a handful of sparc-stations, and nobody is interested
in working on the pyramid anymore :-} (actually a Nixdorf Targon/35 M50,
alias pyramid 98x, OSx4.4).

Not so much to enjoy in this case :-}

S.P.