Path: utzoo!utgpu!watmath!att!rutgers!ucsd!ucbvax!NORTHSTAR.DARTMOUTH.EDU!tobeye
From: tobeye@NORTHSTAR.DARTMOUTH.EDU (Anthony Edwards)
Newsgroups: comp.soft-sys.andrew
Subject: Re: problems with hc -g option on RT aos
Message-ID: <EZMkUoy1eVVa00q=EL@northstar.dartmouth.edu>
Date: 16 Nov 89 19:19:16 GMT
References: <8ZKnuc600Uk4E47Kor@andrew.cmu.edu>
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 17

> Excerpts from mail: 10-Nov-89 Re: problems with hc -g opt.. David
Anderson@andrew.cm (283+0)

> I do not believe we've tried compiling atk with hc 2.1n. Because of a
> variety of problems in early versions of hc 2.1 we've continued to use
> hc 1.4 here at the ITC. I've heard from a group who seems happier with
hc 2.1s, though I don't know if they've compiled atk with it.

--david

Here at NORTHSTAR, we use hc1.4 all the time, but I find that certain
files of the ATK Beta tape will not compile.  So I switch over to
hc2.1('s' maybe) for those files.  (The files that choke up look like
they were created with a code generator).  I don't use hc2.1s to do the
whole compile because a) I don't trust it and b) it can't compile
certain files that hc1.4 can...  The joy of hc on RT's...
   - Anthony Edwards