Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 8/7/84; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!cbosgd!ucbvax!info-vax
From: info-vax@ucbvax.ARPA
Newsgroups: fa.info-vax
Subject: Re: Networking query
Message-ID: <2086@ucbvax.ARPA>
Date: Fri, 21-Sep-84 05:08:13 EDT
Article-I.D.: ucbvax.2086
Posted: Fri Sep 21 05:08:13 1984
Date-Received: Tue, 25-Sep-84 20:33:53 EDT
Sender: daemon@ucbvax.ARPA
Organization: University of California at Berkeley
Lines: 23

From: politik!paul@BERKELEY (Paul Vixie)

Richard:
	Thanks for the DECUS listing.  If it is indeed free, I might be able
to get my company to join (funny: I asked for 6MB of RAM for our site and
got it the same day, but decus is a "nonessential").
	READ?BLK!IO$M_TIMED with TIME=0 *does* do a "read whole typeahead
buffer".  And yes, I do make use of "how many characters are there in the
typeahead buffer" in another program.  The trouble with either of these
approaches is that they depend on polling, which usually means echo-delay
which drives me bananas.  Still, I suppose I could wait for a single-char
AST and then empty the whole buffer in one swipe...  I'll try that.
	My currentv approach has been to get the characters and put them
into a FIFO (two of them: one outgoing, one incoming) that is read in non-
AST mode following a WAKE call from the AST.  Works ok unless the system
is heavily loaded in which case the old "^G" <-> "\007" wars begin...  Possibly
with the two ideas taken together I can make things work.
	To the person who sent me TTLINK:  thank you, but it doesn't work
well on my heavily loaded development 750.  Drops characters.  Better than
Bell Wars, but still not quite on the spot.  Very nice code, though.

		Paul Vixie