DragonFlyBSD

On Sun, 28 Jan 2007, walt wrote:

I cleverly read the code and then added PSM_DEBUG to my kernel.

Now I see this during boot:

atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
 atkbd0: <AT Keyboard> irq 1 on atkbdc0
 kbd0 at atkbd0
 psm0: current command byte:0047
 psm0: failed to get data.
 psm0: <PS/2 Mouse> irq 12 on atkbdc0
 psm0: model Generic PS/2 mouse, device ID 0-00, 2 buttons
 psm0: config:00000000, flags:00000000, packet size:3
 psm0: syncmask:00, syncbits:00

And when moused starts up:
psm: status 00 02 64

And when attempting to re-start moused:
psm0: failed to enable the device (doopen).

Any ideas?

:On Sun, 28 Jan 2007, walt wrote:
:
:> Any ideas for debugging?
:
:I cleverly read the code and then added PSM_DEBUG to my kernel.
:
:Now I see this during boot:
:
: atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
: atkbd0: <AT Keyboard> irq 1 on atkbdc0
: kbd0 at atkbd0
: psm0: current command byte:0047
: psm0: failed to get data.
: psm0: <PS/2 Mouse> irq 12 on atkbdc0
: psm0: model Generic PS/2 mouse, device ID 0-00, 2 buttons
: psm0: config:00000000, flags:00000000, packet size:3
: psm0: syncmask:00, syncbits:00
:
:And when moused starts up:
:psm: status 00 02 64
:
:And when attempting to re-start moused:
:psm0: failed to enable the device (doopen).
:
:Any ideas?

I haven't had any luck reproducing this.  I can start moused, use
    vidcontrol -m on to test it, then I can kill moused and start it 
    again, and the mouse works again.

I then killed moused and started up X with a directly specified
    /dev/psm0, and that worked.  Then I terminated X and started it
    again, and that also worked.

-Matt
                    Matthew Dillon 
                    <dillon@backplane.com>

Sorry, I need to emend my original report: I said/implied the bug was
100% reproducible, but it's not.

I just reproduced it twice by doing the following:
#/etc/rc.d/moused restart

I do that eight or ten or more times, rapidly, before the bug is
triggered. The number of tries will vary depending (I think) on
the speed of repetition. Maybe doing it from a shellscript would
show it quicker.

And if it never reproduces on FreeBSD, porting some of updates from
FreeBSD may fix your problem. BTW what was the manufacturers and the
models of your mainboard and mouse?

Cheers.

Excellent clue, as always :o) After reproducing the same error on
FBSD I finally narrowed it down to my KVM -- when I plug the mouse
directly into the computer the bug disappears. Sorry for the noise.

Hi,

psm(4) was synced recently with FreeBSD. Among all the changes there were some
specific to KVM switches and in fact it solved my problems with my own KVM so I
encourage you to give it a try if you can.

Let me know.

Cheers,
Antonio Huete