*BSD News Article 19027


Return to BSD News archive

Newsgroups: comp.os.386bsd.bugs
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!dog.ee.lbl.gov!hellgate.utah.edu!fcom.cc.utah.edu!cs.weber.edu!terry
From: terry@cs.weber.edu (A Wizard of Earth C)
Subject: Re: Problems with patchkit 0.2.4
Message-ID: <1993Jul29.205550.25212@fcom.cc.utah.edu>
Sender: news@fcom.cc.utah.edu
Organization: Weber State University, Ogden, UT
References: <BLYMN.93Jul25173659@siren.awadi.com.au> <1993Jul25.220104.13519@fcom.cc.utah.edu> <CAsruq.9LD@cyb.cojones.com>
Date: Thu, 29 Jul 93 20:55:50 GMT
Lines: 59

In article <CAsruq.9LD@cyb.cojones.com> loodvrij%cyb@fredbox.cts.com writes:
>>In any case, it is not safe to use CTS/RTS flow control until the driver
>>has been fixed, and it hasn't been fixed yet.  When* it has been fixed,
>>*then* your advice will be applicable.
>
>OK.  Is this a general comment on UNIX drivers, or specifically {Net,386}bsd
>ones?  If the former, then how does the rest of the world do it?  As far as
>serious dial-up networking goes it seems the world uses Telebit as often as
>not.  Are there any plans to fix the driver?  Would the fix be trivial?

In order:
1)	The latter (the *BSD driver specifically).
2)	Very little of the rest of the world gets it right.  SCO and Sun
	are exceptions, and even then, the Sun requires a page of patch
	instructions from Telebit (by SCO I mean the SCO supplied driver,
	not multiport board drivers).
3)	I believe there are people working on the driver right now; my
	comments are about the reliability of the current driver and
	should not be taken to reflect on their efforts.
4)	The fixes are non-trivial (or it wouldn't be an issue); I'd
	estimate about a week of work (that's 40 hours, not elapsed
	work+non-work time) if I were to fix it.  I can't distribute
	code right now, so that's not an option.  I think it'd be
	about the same time (or less) if someone like CGD or BDE or
	someone else who knew the drivers did it.

>>Until then, don't set the connection to the modem to 19200 baud unless you
>>*know* the modem will *always* be able to push 1920 characters a second
>>down the line.
>
>What baud rate would you recommend setting it to for the time being?

What baud rate is the modem vendor willing to guarantee you without flow
control?  Generally in the 2400 to 4800 range if there is sufficient memory
in the modem; up to 14.4 for a MCM (multicarrier modulation) modem with
enough memory (like Telebit) if you use a non-silence-compressing phone
carrier (like AT&T or TelAmerica). Of course most serial ports aren't set
up with the controls to set the UART clock divider for 14.4.  The ioctl()'s
generaly go 2400/4800/9600/19200/38400.

I'd say 9600 for a Telebit, and that or less for other vendors, depending
on the amount of RAM, line turnaround, compression, and modulation techniques
being used.

If you live in a US West Digital Service area (like me), I'd say forget the
modem and go ISDN if they'll still endpoint you on something like WestNET;
fractional FrameRelay is also an option (and is standard between US West
and other vendors, unlike their ISDN).  That'll get you 64K or 19.2K,
respectively.

Of course if you use the modem for something other than SL/IP, you'll
still want a modem.


					Terry Lambert
					terry@icarus.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.