Skip site navigation (1)Skip section navigation (2)
Date:      20 Sep 1995 13:32:51 -0600
From:      "Owen Newnan" <onewnan@denitqm.ecte.uswc.uswest.com>
To:        "FreeBSD Bugs" <freebsd-bugs@freebsd.org>
Cc:        "John Bresler" <jmb@kryten.atinc.com_>, "Boyd Faulkner" <faulkner@mpd.tandem.com__>, "Brian Tao" <taob@gate.sinica.edu.tw>
Subject:   the mouse that snored
Message-ID:  <n1400518440.72842@denitqm.ecte.uswc.uswest.com>

next in thread | raw e-mail | index | archive | help
                       Subject:                               Time:8:07 AM
  OFFICE MEMO          the mouse that snored                  Date:9/20/95

I think I have a bug (else hardware problem), so I am forwarding this to the
bugs distribution rather than questions where I first raised it.

I am accused to have said:

> I rebuilt my kernel to pick up PS/2 mouse support, using the psm line out of
> LINT.  X came up with the mouse working and one of the first things I did in
> one of the windows was dump/restore my root fs to my backup root.  I made
the
> kernel again to pick up message queues and some other optimizations.  Kernel
> comes up fine, xdm comes up, but mouse is numb.  I back out the root
> filesystem using dd and the backup partition.  Same problem: X comes up but
> with nonfunctional mouse.  Obviously, something changed.  Perhaps X has
> populated some file in my root directory that isn't quite right.  Can anyone
> suggest what the problem might be, or diagnostic procedures I can use to
shake
> down the mouse?  BTW, if I try cat </dev/psm0, I always get "device
> temporarily unavailable."

Some additional information: The port is recognized at boot.  The XF86Config
shows /dev/psm0 as pointer device and PS/2 as protocol.  This driver is S3 and
the Mouse is a Logitech product.

After first experience the problem I restored root from backup (using dd). 
Then the mouse worked the first time, but not subsequent times.  Till I
restored the root again, in which case the mouse worked the first but not
subsequent times.

The problem occured with both xdm and startx.    Logs did not report any mouse
errors.   

I reconfigured startx with warpto specified in a function key.  This allowed
me to get off the root menu where I had been stuck (especially nasty with xdm)
and log out of X cleanly.  I started experimenting and found that sometimes
when I startx it works and sometimes not- even when I haven't tampered with
the config files in between.  I think I have a race condition.  Once
operational the mouse works fine every time throughout the X session.

This is circumstantial, but the frequency of the problem seems to vary
depending how I have .twmrc configured.  Lately its been working right pretty
consistently.  Anyhow I can circumvent by trying startx again.  I'm reporting
this to share the circumvention and information that might aid in debugging
down the road.

Two other users have reported a similar problem with PS/2 mouse (at least one
of them with S3) so this may well be related and I've copied them.  However,
in my case, once dead always dead, at least until the next time I startx...

Cheers!
Owen Newnan







Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?n1400518440.72842>