[comp.unix.i386] Two blind mice

bill@ssbn.WLK.COM (Bill Kennedy) (09/04/89)

I wailed a week or so ago about being unable to make a Logitech
mouse work with VP/ix and 386/ix 1.0.6.  Vernon Schryver (among
others) was most helpful and inspired me to experiment some more,
including moving the bus mouse IRQ and juggling COM ports.  I have
both serial and bus mice and I'm unable to get either of them to
function properly.  One at a time:

I assigned ttyi04 (Computone AT-8 card) to COM1 and COM1MOUSE and
was not able to get MOUSE.COM to find the mouse although COMCHECK
(and my line monitor) said there was rustling in the COM1 bushes.
I changed from MOUSE.COM in AUTOEXEC.BAT to DEVICE=MOUSE.SYS in
CONFIG.SYS.  Behold!  The rustling was the mouse.  The problem is
that it doesn't move the cursor correctly in MTRACK40, the button
lights turn on and off as the mouse moves and the cursor careens
around the screen, only generally in the direction the mouse is
moving.  So I'm able to get them to coexist but not function with
the smart card and MOUSE.SYS.

When I made a kernel with the bus mouse assigned to IRQ4 (I have
completely disabled COM1) it worked OK until I went into VP/ix
which signed on with an ERROR 6 (no such device) on /dev/mouse.
The node is there, 666 permissions.  MOUSE.COM complained (with
justification) that the mouse wasn't responding and both COMCHECK
and MTRACK40 confirmed this.  After I changed to MOUSE.SYS in the
CONFIG.SYS the mouse driver installs without a complaint, but the
kernel (/dev/mouse) isn't passing anything through to it.  Oh yeah,
I disabled the asy driver for COM1 as well as disabling the hardware.

So I have two blind mice.  I prowled around in the /usr/vpix/etc/conf
directory and found a bus mouse driver source, the UNIX part and IEM
part.  The readme.doc file is very clear, these examples are not
meant to be compiled.  They're right, two files, ptypes.h and v86error.h
are nowhere to be found.  I did find gdi.h but no where near where it
would be found to compile bmouse.c.  I also see where bmouse.c has the
IRQ and SIGNAL number both hard coded to 5, perhaps mouse.o that talks
to /dev/mouse also has IRQ5 hard coded.  Vernon's bus mouse works on
IRQ3, he and I each have a Wangtek tape that really wants to live on 5.
I have also tried IRQ2(9), but the system won't boot at all (no boot
prompt even) with the bus mouse set there.

Sorry to be so lengthy, but I wanted to save net.helpers from suggesting
something that I have already tried and I wanted to explain what I tried
so that someone could tell me what I might have screwed up.  Oh yes, both
mice work just fine in DOS, either COM port, or the bus mouse.  Thanks,
-- 
Bill Kennedy  usenet      {attctc,att,cs.utexas.edu,sun!daver}!ssbn!bill
              internet    bill@ssbn.WLK.COM   or attmail!ssbn!bill

campbell@redsox.bsw.com (Larry Campbell) (09/05/89)

In article <1220@ssbn.WLK.COM> bill@ssbn.WLK.COM (Bill Kennedy) writes:
-I wailed a week or so ago about being unable to make a Logitech
-mouse work with VP/ix and 386/ix 1.0.6.

This is known not to work, which you might have discovered by phoning
Interactive (as I did).  They claim to have fixed it in 2.0.2.  I haven't
had time to install 2.0.2 yet, so I don't know if it's really fixed.

I do suspect, however, that you're wasting your time with 1.0.6.
-- 
Larry Campbell                          The Boston Software Works, Inc.
campbell@bsw.com                        120 Fulton Street
wjh12!redsox!campbell                   Boston, MA 02146

izen@amelia.nas.nasa.gov (Steven H. Izen) (09/05/89)

In article <1395@redsox.bsw.com> (Larry Campbell) writes:

>In article <1220@ssbn.WLK.COM> bill@ssbn.WLK.COM (Bill Kennedy) writes:
>-I wailed a week or so ago about being unable to make a Logitech
>-mouse work with VP/ix and 386/ix 1.0.6.

>This is known not to work, which you might have discovered by phoning
>Interactive (as I did).  They claim to have fixed it in 2.0.2.  I haven't
>had time to install 2.0.2 yet, so I don't know if it's really fixed.

If you buy ISC's X-windows, you get a driver /dev/logi for the Logitech bus
mouse.  I'm using it with vpix on 2.0.1 and have had no problems.  Just
replace /dev/mouse references in vpix.cnf with /dev/logi.  Even the /dev/mouse
ioctls
seem to work!

I was told by ISC that they only provide the driver when you buy X-windows.
But that was a while ago.
-- 
Steve Izen: {sun,decvax,uunet}!cwjcc!skybridge!izen386!steve
or steve%izen386.uucp@skybridge.scl.cwru.edu
or izen@cwru.cwru.edu		"My second bike is a car."

johnl@esegue.segue.boston.ma.us (John R. Levine) (09/05/89)

In article <1395@redsox.bsw.com> (Larry Campbell) writes:

>In article <1220@ssbn.WLK.COM> bill@ssbn.WLK.COM (Bill Kennedy) writes:
>-I wailed a week or so ago about being unable to make a Logitech
>-mouse work with VP/ix and 386/ix 1.0.6.

>This is known not to work, which you might have discovered by phoning
>Interactive (as I did).  They claim to have fixed it in 2.0.2.  ...

I run a Logitech serial mouse under 2.0.2 and it works just fine.  I plugged
it into the first serial port, uncommented the COM1MOUSE line in the VP/ix
configuration file, put the mouse driver into the VP/ix CONFIG.sys (you did
that didn't you?) and it works like a charm with MS Word and other such mouse-
loving programs.  Both Logitech's and MS's mouse drivers seem to work.

The only fly in this ointment is that although the serial driver is fine for
VP/ix, it has serious bugs that make it impossible to use it for bidirectional
uucp.  There is a PD replacement from Jim Murray that solves the uucp
but doesn't yet have the hooks that VP/ix needs.  Interactive is supposed to
have a new version of their async driver this week, I hope that fixes it.
-- 
John R. Levine, Segue Software, POB 349, Cambridge MA 02238, +1 617 492 3869
{ima|lotus}!esegue!johnl, johnl@ima.isc.com, Levine@YALE.something
Massachusetts has 64 licensed drivers who are over 100 years old.  -The Globe

jon@wilbur.uucp (Jonathan Broome) (09/06/89)

In article <3065@amelia.nas.nasa.gov> Steven H. Izen writes:
[ quotes other's articles about problems with Logitech mice, then adds ]
: If you buy ISC's X-windows, you get a driver /dev/logi for the Logitech bus
: mouse.  I'm using it with vpix on 2.0.1 and have had no problems.  Just
: replace /dev/mouse references in vpix.cnf with /dev/logi.  Even the /dev/mouse
: ioctls seem to work!
: 
: I was told by ISC that they only provide the driver when you buy X-windows.
: But that was a while ago.

ISC's 386/ix version 2.0.2 comes with full support for the Logitech
old-style bus mouse (model "P7").  [It appears that Logitech "fixed"
some hardware "flaws" in the new "Series 9" mouse, introducing several
incompatibilities in the process.]  The driver works with VP/ix as a
Logitech mouse and everything.  ISC's X product also includes the driver
for use with pre-2.0.2 systems (though VP/ix didn't know of Logitech in
earlier versions, and just emulates the Microsoft hardware.)  The logi
kernel driver is fully compatible with the Microsoft driver, so it should
come as no surprise that the ioctls work.

 ---Jonathan Broome
 jon@wilbur.uucp

Disclaimer: I speak for the organization listed in the "Organization:" line!