Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!genrad!panda!talcott!harvard!cmcl2!seismo!brl-tgr!gwyn
From: gwyn@brl-tgr.ARPA (Doug Gwyn <gwyn>)
Newsgroups: net.micro.att
Subject: Re: Pascal and C on a 3b2 with a 5620 DMD
Message-ID: <111@brl-tgr.ARPA>
Date: Tue, 26-Nov-85 23:55:18 EST
Article-I.D.: brl-tgr.111
Posted: Tue Nov 26 23:55:18 1985
Date-Received: Thu, 28-Nov-85 23:55:24 EST
References: <127@gt-cmmsr.UUCP>
Distribution: net
Organization: Ballistic Research Lab
Lines: 18

> 		pc -c pfile.p 			This does not work
> 		dmdcc cfile.c pfile.o

"dmdcc" is a cross-compiler, using header files for and
generating code for the unique DMD execution environment.
"pc" is a host compiler, generating code for use on the
UNIX host system, not in the DMD.  Except for the accident
that the DMD and the 3B2 have similar instruction sets,
there would be no chance of this combination working.

Now, if you can persuade "pc" to generate C source code
and then compile THAT with "dmdcc", so long as there were
no dependencies on host data structures or run-time support
you might be able to get the result to work.

My recommendation is to program the DMD in C, but only after
you get straight how its run-time environment differs from
the host's.  (There is quite a difference!)