*BSD News Article 37492


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!yeshua.marcam.com!news.kei.com!sol.ctr.columbia.edu!argo!wpaul
From: wpaul@ctr.columbia.edu (Bill Paul)
Newsgroups: comp.os.386bsd.questions
Subject: Re: [FreeBSD/XFree86] Mouse woes
Date: 4 Nov 1994 19:49:42 GMT
Organization: Columbia University Center for Telecommunications Research
Lines: 121
Message-ID: <39e38n$m5k@sol.ctr.columbia.edu>
References: <397ag6$nr1@dcsun4.us.oracle.com> <398aps$pjl@sol.ctr.columbia.edu> <39cqq9$knv@dcsun4.us.oracle.com>
NNTP-Posting-Host: argo.ctr.columbia.edu
X-Newsreader: TIN [version 1.2 PL2]

Daring to challenge the will of the almighty Leviam00se, Bruce Caruthers
 (bcaruthe@us.oracle.com) had the courage to say:

: In article <398aps$pjl@sol.ctr.columbia.edu>,
: Bill Paul <wpaul@ctr.columbia.edu> wrote:
: >Daring to challenge the will of the almighty Leviam00se, Bruce Caruthers
: >(bcaruthe@us.oracle.com) had the courage to say:
: >

:     I have tried all of the above, and many other possible
: combinations, both from what the CMOS reports, what my DOS comm
: programs think, etc.  I have tried both using the COMx defines, as
: well as setting the values explicitly.
:     The asking about whether something needed to be attached was
: just a desperate hope.

Odd. Most systems have at least COM1 and COM2 installed by default.
Assuming there's nothing wrong with the hardware, they should be
detected.

:     The CMOS and comm programs think it is COM4 at 0x3e8, irq 5.  I
: tried that and many other combinations (as I mentioned above).

That's a little strange, since COM4 is usually set up at address 0x2e8.
Not to say it won't work that way: if you tell the kernel that this
is where the port's supposed to be, then that's where it'll look. One
must watch for conflicts though: if anything else tries to grap IRQ 5,
then things could get messy.

:     See above.  I have built many kernels (and not just for PC
: Unices).  I *am* considering replacing the multi-card.  (Although
: everything works fine for DOS/MSW/MSWinNT [except for 1280x1024 mode
: for the WinNT 3.5 S3 driver, where the mouse disappears, if that is
: an indicator of anything].)

Never having run WinNT, I couldn't say.

: >Bzzzt. I'm sorry, that's incorrect but thanks for playing. As a consolation
: >prize, I'll tell you what the real problem is.

:     Ah, the gentle kindnesses of the net overwhelm me.

We do try.

:     If it means anything, I can connect via cu to the port the mouse
: is one, and if I move the mouse, I get a hangup signal (actually, I
: am not sure I tried this at 1200 baud).  So, the mouse *is*
: connected at least somewhat (as I mentioned, it does sort of move at
: certain settings).

It's been my experience that if the mouse is indeed working correctly,
then you should be able to use cu (or tip, or kermit, or whatever) to
connect to the serial port (usually at 1200 baud), and moving the mouse
should produce a stream of characters. Pressing a button should generate
a character too. This is based on the two mice I've used: the original
one I had that broke, and the cheapo $10 one I got to replace it.
I wouldn't expect a SIGHUP.

: 	"Mouse Systems Mice come with either two or three buttons.
: 	 The two button mice are plug-compatible with the Microsoft
: 	 serial mouse.  Your application software should be
: 	 configured for a Microsoft serial mouse, and the buttons
: 	 will act identically to the Microsoft mouse."

:     I have the two button (beige) mouse.  I will give Microsoft,
: 1200 another try.

I didn't know they came in 2 button versions... on the other hand
the rodents I have are cheap clones, not real MouseSystems products.

: >2) You botched up your Xconfig file and you were too lazy to install
: >   the man pages and read the instructions to figure out where you went
: >   wrong. Check the mouse configuration section -- it should look like this:

:     Thanks for being so quick to assume I ask for help without
: trying anything first.

Unfortunately, this is usually the case. I can't begin to tell you how
many people post without first reading the instructions, or even looking
at previous posts. When you see the same question asked dozens of times
over, it gets to you.

:     I have read the Xconfig scripts (since they died during the
: autoconfig run) and man pages, and have tried the various options
: for Microsoft and MouseSystems (dtrclear, rtsclear, etc.) mice.
: I have the manual for my mouse, as well, which claims that it is
: 1200-8-N-1, as per standard mice, according to the Xconfig man
: pages.


: >#
: ># Mouse definition and related parameters
: >#
: >MouseSystems    "/dev/cua01"            <-- use your real mouse port here

:     /dev/tty0[01] (tried configuring for each address in various
: kernels -- no difference)

I prefer using the cua0? device nodes myself since it eliminates one
worry, namely that the clocal flag will be cleared properly. As a matter
of habit, I tend to reserve /dev/tty0? for use as dialin ports only
and use /dev/cua0? for everything else.

: >#  BaudRate     9600
:     That's what I have been playing with.

For most serial mice, 1200 is fine. You might actually get it to
work at 2400. Weirdness occurs at 9600 if your mouse isn't designed
for it.

: >#  Emulate3Buttons
:     Haven't gotten far enough to worry about this one.

If you'll be using a 2 button mouse, you'll need it. Otherwise you can
ignore it. As you may have seen in the man page, this causes the X server
to interpret both buttons pushed simultaneously as a press of the missing
middle button.

--
-Bill Paul
wpaul@ctr.columbia.edu