Path: utzoo!utgpu!water!watmath!clyde!bellcore!decvax!yale!husc6!mit-eddie!rich
From: rich@eddie.MIT.EDU (Richard Caloggero)
Newsgroups: comp.sys.apollo
Subject: Re: SR10 and me
Message-ID: <8186@eddie.MIT.EDU>
Date: 17 Feb 88 21:02:26 GMT
References: <8802090007.AA09431@ELI.CS.YALE.EDU> <1670@ssc-vax.UUCP> <3a569abe.c366@apollo.uucp>
Reply-To: rich@eddie.MIT.EDU (Richard Caloggero)
Organization: MIT, EE/CS Computer Facilities, Cambridge, MA
Lines: 47

In article <3a569abe.c366@apollo.uucp> mishkin@apollo.UUCP (Nathaniel Mishkin) writes:
>Now I know that there may not be many "Aegis" diehards out there in this
>audience, but for those that are, well, don't worry.  DOMAIN/OS has some
>changes that may irk you, but we broke nothing that didn't have to be
>broken.  We have compatibility support so that virtually all "old" programs
>will run, un-changed, un-recompiled, un-relinked.  "Aegis" has not gone
>away.  From my perspective, it's really sort of funny to think that way.
>To most people in R&D, "Aegis" simply means the OS kernel.

     Well, ok, so AEGIS hasn't gone away, but what of its distinctive
     features (acls, object-oriented file system, protected subsystems,
nice linker and object library handling...); have these things
changed?  I would venture to guess that most of these things would not
*have* to change in order to implement the various system interfaces
on top of them.  What of it!  Can anyone from Apollo post a list of the
changes made to the kernel at SR10?

      Now, a question/plea!!!  I've sent mail to various people at
Apollo about this before, but I'm not sure if I've ever posted to this
group on this topic, so bare with me if this is an old flame!  I'm
blind, and thus can't use the DM.  I thus log in via an SIO line
(siomonit/siologin).  Csh likes this arranhment just fine, but various
other programs (emt is the bigest culprate) act strangely (differently)
when run from an sio line.  Also, the tty driver itself is not fully
implemented ("stty -tabs ctlecho prterase" is effectively a no-op).
Actually, emt works fine if it is run from the node on which you are
attached , implying that the node must have at least 2 sio ports on it.
Since we have no such node, I must crp to  another, and run emt there.
Now, everything is fine till you go into remote mode.  At that point,
emt checks its input, and if it is a crp mailbox, it sends down a
special record which is intended to be interpreted by the frob who is
passing characters to crp (in most cases the dm).  This record says
"put the primary input device in raw mode with no echo" so the remote
host can do its own input processing.  The dm pays attention to such
"spm escape sequences" and handles them properly.  However, the sio
driver does not handle these control messages.  Thus, my question is:
does anyone have a frob which will take care of this problem (how about
a hacked up tty driver).  Is their an *easy* (nothing is ever easy) way
of doing this (maybe using pty's -- having  siologin run some frob that
just forks of a csh, and handles these spm escape sequences  ...).

		Thanx for putting up with my ramblings!
-- 
						-- Rich (rich@eddie.mit.edu).
	The circle is open, but unbroken.
	Merry meet, merry part,
	and merry meet again.