Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 SMI; site sun.uucp
Path: utzoo!linus!decvax!decwrl!sun!guy
From: guy@sun.uucp (Guy Harris)
Newsgroups: net.bugs.uucp
Subject: Re: One-second break is too long
Message-ID: <2795@sun.uucp>
Date: Fri, 13-Sep-85 20:36:50 EDT
Article-I.D.: sun.2795
Posted: Fri Sep 13 20:36:50 1985
Date-Received: Sun, 15-Sep-85 05:03:48 EDT
References: <24@pyramid.UUCP>
Organization: Sun Microsystems, Inc.
Lines: 17

The original V7 code for transmitting breaks did so by setting the speed to
some strange value and writing a pile of NULs.  This was because the V7 TTY
driver didn't give you access to the break-generation hardware that was in
the PDP-11s (and lots of other machines') terminal muxes.

The S3, S5R1, and S5R2 code just does a TCSBRK system call which turns on
the mux's break bit for 1/4 sec and then turns it off again.  S5R2V2 may do
it differently (I don't have the source handy).

4.1BSD and 4.2BSD have no call to send a timed break; they just have the
ability to turn the break bit on and off.  I suspect the code for the 4.2BSD
UUCP may be a mix of the V7 code and code that used TIOC[SG]BRK.  The way
it's written there may be a historical dreg.

OK, Peter, what does honey danber do?

	Guy Harris