Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!mit-eddie!husc6!harvard!spdcc!dyer
From: dyer@spdcc.UUCP
Newsgroups: comp.windows.x
Subject: Re: X on Xenix/386
Message-ID: <526@spdcc.COM>
Date: Tue, 17-Feb-87 09:35:33 EST
Article-I.D.: spdcc.526
Posted: Tue Feb 17 09:35:33 1987
Date-Received: Wed, 18-Feb-87 04:08:29 EST
References: <8702152056.AA17802@tcgould.tn.cornell.edu> <267@gssc.UUCP>
Reply-To: dyer@spdcc.COM (Steve Dyer)
Organization: S.P. Dyer Computer Consulting, Cambridge MA
Lines: 20

>NOOOOOOOOOO!!!!!!!  DO NOT COUNT ON X ONLY RUNNING ON 4.X MACHINES!!!!!
>
>We have ported X to System V Xenix/386, and intend on doing more of the
>similar.  If you want your application to run in all environments, follows
>standard "portable code" procedures and use generic calls, or use ifdefs, etc.

I'd like to know what IPC model(s) you chose for implementing X on System V
Xenix/386.  Do you rely on external network cards like the Excelan TCP/IP
board for all the IPC?  I had understood that using their on-board 'sockets'
for IPC within a single machine was unusually inefficient (warning, this
was gleaned from an Excelan techie at Uniforum, and not from direct experience.)
Do you optimize intramachine operations using System V facilities like
shared memory?

Finally, my Intel 386 card is coming soon; once I get my hands on SCO's
386 Xenix, how can I then obtain X?
-- 
Steve Dyer
dyer@harvard.HARVARD.EDU
dyer@spdcc.COM aka {linus,wanginst,bbnccv,harvard,ima,ihnp4}!spdcc!dyer