[fa.info-vax] Broadcast in passall

info-vax (12/05/82)

>From GZ@MIT-MC  Sat Dec  4 15:54:20 1982
Mail-From: ARPANET host MIT-MC rcvd at 4-Dec-82 0814-PST
To: info-vax@SRI-CSL
Remailed-Date:  4 Dec 1982 1047-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

Does anybody have a patch for VMS 3.0 that allows receiving broadcast
messages in passall mode?

info-vax (12/22/82)

>From TK.FONER@MIT-OZ  Wed Dec 22 08:21:28 1982
Mail-From: ARPANET host MIT-ML rcvd at 21-Dec-82 2324-PST
To: Gail Zacharias <GZ@MIT-MC>
Cc: Info-VAX@SRI-CSL, Tk.Foner@MIT-OZ
Reply-To: Foner at MIT-MC
In-Reply-To: The message of 4 Dec 1982  11:15-EST from Gail Zacharias <GZ at MIT-MC>
Remailed-Date: 22 Dec 1982 0609-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

Date: Saturday, 4 December 1982  11:15-EST
    From: Gail Zacharias <GZ at MIT-MC>
    To:   info-vax at SRI-CSL
    Re:   broadcast in passall
    Remailed-date: 4 Dec 1982 1047-PST
    Remailed-from: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
    Remailed-to: Info-VAX@SRI-CSL: ;

    Does anybody have a patch for VMS 3.0 that allows receiving broadcast
    messages in passall mode?

I have a such a patch, guaranteed to work on VAX/VMS Release 2.5 and
earlier releases.  It is not tested on VAX/VMS Release 3.0 and later
releases, and probably needs to be changed.  The documentation
describing the patch is very clear, so anyone with 3.0 or 3.1 fiche
can dive in and find the new location to patch very quickly, assuming
that TTDRIVER.EXE has not changed grossly.

The patch is to TTDRIVER itself, and does not require any new
programming once the patch is made.

I cannot get the patch this instant, due to problems with dead hosts
and possibly funny Chaosnet connections here at MIT.  However, I gave
this patch to several Info-VAX readers many months ago.  If anyone
still has the patch online, please speak up.

Have fun, folks.

						<LNF>

info-vax (12/22/82)

>From GZ@MIT-MC  Wed Dec 22 15:05:46 1982
Mail-From: ARPANET host MIT-MC rcvd at 22-Dec-82 0816-PST
To: Tk.Foner@MIT-OZ
Cc: Info-VAX@SRI-CSL
In-Reply-To: The message of 22 Dec 1982  02:26-EST from Leonard N. Foner <TK.FONER at MIT-OZ>
Remailed-Date: 22 Dec 1982 1021-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

I had been using your patch for 2.5.  TTDRIVER has apparently been rewritten
for 3.0 and I am asking about equivalent patches for that.

I have your old patch in MC:GZ;PASSALL PATCH if anybody wants to FTP a copy.

info-vax (12/24/82)

>From coar.umass@UDel-Relay  Thu Dec 23 22:17:17 1982
Mail-From: ARPANET host SANDIA rcvd at 23-Dec-82 1703-PST
Mail-From: ARPANET site UDEL-RELAY rcvd at 23-Dec-82 1759-MST
To: info-vax.umass@UDel-Relay
Cc: coar.umass@UDel-Relay
Via:  UMASS-ECE; 23 Dec 82 19:50-EST
Remailed-Date: 23 Dec 1982 2045-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

Concerning patching the terminal driver (VMS) in order to receive
broadcast messages in passall mode: I myself am violently against
such a patch, at least for V3.x. Version 3 of VMS includes a method
of handling these messages (see the I/O User's Guide section on
terminal/mailbox interaction), and I think it should be used, rather
than re-invented. I can see the need for such a thing on older versions
of VMS, but patching DEC-supplied drivers is ALWAYS risky, particularly
when the driver is a fairly new release: your patch may interfere with
subsequent DEC-released patches.

If anyone needs details on this broadcast-trapping (or would like
a copy of the routine I wrote to enable Emacs on VMS to catch them)
I am always willing to spew forth what I know.

People waiting for symbiont specs: please be patient; I am putting
them together, but this is a very busy time of year. I haven't
fergotten youse, no fear!

	ken
	coar.umass@udel-relay

info-vax (12/25/82)

>From STEW@MIT-MC  Fri Dec 24 22:07:48 1982
Mail-From: ARPANET host SANDIA rcvd at 24-Dec-82 1705-PST
Mail-From: ARPANET site MIT-MC rcvd at 24-Dec-82 1801-MST
To: coar.umass@UDEL-RELAY
Cc: INFO-VAX@MIT-MC
In-Reply-To: The message of 23 Dec 82 15:55-EDT (Thu) from the Golux <coar.umass at UDel-Relay>
Remailed-Date: 24 Dec 1982 2103-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

I would like your emacs patches for broadcast trapping.

There is still a problem with this, which is that the same mailbox is
used for broadcast messages and for other ttdriver-related things.
Furthermore, there may be only one mailbox associated with the
terminal.  So if you've got something trapping broadcast messages,
then you can't run something else which needs the associated mailbox
feature for other purposes.  LIB$SPAWN is one such beast.  What is
needed is a linked list of associated mailboxes.  When the terminal
driver wants to send a msg, it sends it to the most recently
established mailbox.  If this mailbox does not want to handle this
message, then it must be able to get a channel on the next mbx in the
list so it can "resignal" the message.  I plan on submitting this in
the form of an SIR.  Anyone have any comments on this?

Anyway, I have a program which I spawn/nowait which traps broadcasts
and writes them on the bottom line of the terminal.  Currently, I
spawn it before running emacs (via command procedure) and kill it when
emacs exits.  Due to the above problem, you can't spawn anything else
while it is running.  If anyone wants this program, I'd be glad to put
it on MC.

Stew