[fa.info-vax] AMBASSADOR Padding

info-vax (12/22/82)

>From dove@mit-dspg@mit-mc  Wed Dec 22 15:17:15 1982
Mail-From: ARPANET host MIT-MC rcvd at 22-Dec-82 1241-PST
To: info-vax@sri-csl
Remailed-Date: 22 Dec 1982 1336-PST
Remailed-From: the tty of Geoffrey S. Goodfellow  <Geoff5 at SRI-CSL>
Remailed-To: Info-VAX@SRI-CSL: ;

I just noticed some of this mail.  For your information, sending nulls
to an Ambassador is of no more value then sending any other character.
It CANNOT be padded as currently implemented.  That is because ALL
characters are saved in the input buffer on the assumption that they
may need to go to the printer.  Vt100's and the like discard nulls so
padding them is fine.  Thus sending padding chars to an AAA is just as
likely to overflow the input buffer as continuing to print.

Personally, I use mine on unix 2.8(1) at 19,200 but I have a special
version of termcap and have altered the tty driver so that delay chars
(those with the 0200 bit set that cause the driver to delay as many
tics as the remaining 7 bits of the char) can be used to HALT the tty
driver for the required time instead of having it send nulls.  THis
involves using the driver in LLITOUT mode and LLITIN mode instead of
RAW mode (where the 0200 chars would just pass through the driver
untouched) and is a good system because it doesn't keep your driver
busy sending nulls.  It does however cause more TIMEOUTS.  I sent the
necessary changes to Berkeley long ago.  I don't know if they are
available in current installations.

Unfortunately, I cannot help you with code as I am no longer involved
in system activities here.  However, someone at Berkeley has them.