Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!mips!moosehead.mips.com!trevc
From: trevc@moosehead.mips.com (Trevor Cotton)
Newsgroups: comp.sys.mips
Subject: Re: Problems with Specialix 3.11 (Risc/OS 4.51 - M/120)
Message-ID: <710@spim.mips.COM>
Date: 5 Mar 91 01:40:34 GMT
References: <8467@dmshq.mn.org>
Sender: news@mips.COM
Reply-To: trevc@mips.com
Organization: Mips Computer Systems Inc., Sunnyvale, CA
Lines: 67
Nntp-Posting-Host: moosehead.mips.com

In article <8467@dmshq.mn.org>, pnessutt@dmshq.mn.org (Bob Monio) writes:
> One of our customers has installed the latest update of the Specialix
> driver on his M/120 (Version 3.11 running under Risc/OS 4.51).  In doing
> so we found that it cured the terminal hangs and in conjunction with wcat
> (a patch given to us to help fix problems with lpsched) we were able to
> clear up the truncated output to his printers.
> 
> But we found some behavior that has created a few problems.  It seems
> that the default behavior of the ports is now -ixon, -ixoff, and
> -ixany.  Since his printers use xon/xoff flow control only these
> needed to be turned on in order to allow flow control to work
> correctly.  Since some of his printed output is spooled and the rest
> is not, we figured we would put a shell script in place to keep the
> port open with ixon, ixany, and ixoff on.  Once running, the ports tty
> characteristics change correctly.  Any output sent to the port via
> cat, etc., works fine.  But once a spooled job gets sent to the
> printer and printed, the characteristics of the port are reset to the
> original default settings.  This occurs even with the shell script
> still running.  Since this behavior is inconsistent with behavior
> we've had in past with other machines, etc., this would seem to be a
> bug.
> 
> For the time being we've managed to get the software vendor of
> the package he is using to spool to that particular printer, which
> has bypassed the problem.  But there are other customers that use
> M/120s who will be installing this driver soon that are not going to
> be as lucky.  Is there any information available on when a patch
> and/or release update will be out with this problem fixed?
> 
> Thanks for your help.
> 
>  -Bob
> 
> -- 
>  Robert A. Monio                       "Sometimes I wish that people would 
>  International Quality Institute, Inc.  take this war more seriously.  Peace
>  pnessutt@dmshq.mn.org                  is going to be a hell of a letdown."
>  ..uunet!rosevax!sialis!dmshq!pnessutt             -- George S. Patton, Jr.

This is a 'known problem' with Specialix 3.11 for RISC/os.
Rather than hold up shipments of 3.11 ( which cured other, more serious 
problems with modems etc ) it was decided to fix this specific problem in a 
subsequent release, and work began on that release as soon as the problem
was discovered.

Version 3.12 of the Specialix driver for RISC/os is currently in QA in 
Specialix UK. This version sets up the port defaults to be the same as for
'MIPS supplied' serial ports.

Users of the Unix print spoolers should configure them to set all the
port characteristics that they rely on using /etc/printcap for lpd or
the model 
file for lp. 
( this is a good habit to get into anyway )

For other uses of the serial ports, a workaround is a shell script or small
program that opens the port, sets the line parameters as required, and then
goes to sleep, keeping the line open.
( again, applications should not really rely on 'defaults', but I know that
many do )

--trevc--

  Trevor Cotton, MIPS Computer Systems Inc.
  MS 6-05, 930 DeGuigne Drive, Sunnyvale, CA 94086
  Tel: +1 408 524 7286  Fax: +1 408 524 7521
  Email: {wyse,ames,decwrl,pyramid}!mips!trevc    trevc@mips.com