Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!crdgw1!uunet!hayes!tnixon
From: tnixon@hayes.uucp
Newsgroups: comp.dcom.modems
Subject: Re: Hardware Flow Control - How does it work?
Message-ID: <3964.2842629d@hayes.uucp>
Date: 28 May 91 14:01:01 GMT
References: <1991May25.061834.4893@colnet.uucp>
Organization: Hayes Microcomputer Products, Norcross, GA
Lines: 40

In article <1991May25.061834.4893@colnet.uucp>, res@colnet.uucp (Rob
Stampfli) writes: 

> There is an involved discussion going on right now in comp.sys.3b1 about
> why some people are having problems making hardware flow control work,
> specifically with the Unix-PC.  Generally, there seems to be two
> implementations of hardware flow control, as shown in the following excerpt
> from "Managing UUCP and Usenet" by O'Reilly & Associates (brackets mine).
> 
> My question is, which of the two implementations are modems currently
> designed to work with?  Which does Unix provide? 

I can't tell you what Unix provides, but I can tell you what today's 
error control modems do.  It is as "Managing UUCP and Usenet" 
describes for "full duplex".  If the modem is ready to accept data, 
it asserts CTS; if it is unable to receive data, it turns off CTS.  
When the DTE is ready to accept data, it turns on RTS; when it is 
unable to accept data, it turns off RTS.  RTS and CTS are totally 
independent in this scheme -- each controls flow coming from the 
opposite direction, and neither depends on the other (unlike the 
half-duplex use of the signals).

I should note, for completeness (since I am actively involved in 
standards committees) that when "RTS" is used this way, it really 
_isn't_ "RTS", but "RTR".  That is, not "Request to Send", but 
"Ready To Receive".  The use of this signal for flow control 
purposes is described by CCITT V.24 as circuit number 133.  We've 
defined it as appearing on the SAME PIN AS RTS, since (a) you never 
use both the RTS and RTR function at the same time, and (b) most 
DTEs are able to drive that pin as an output, so it is convenient to 
use it instead of some other pin.  CCITT V.42 specifically calls out 
the use of RTR/CTS flow control, _not_ RTS/CTS.  Note that the use 
of CTS for flow control _is_ permitted under the standardized 
definitions of the circuit.

-- 
Toby Nixon, Principal Engineer    | Voice   +1-404-840-9200  Telex 151243420
Hayes Microcomputer Products Inc. | Fax     +1-404-447-0178  CIS   70271,404
P.O. Box 105203                   | UUCP uunet!hayes!tnixon  AT&T    !tnixon
Atlanta, Georgia  30348  USA      | Internet       hayes!tnixon@uunet.uu.net