compdoc wrote:
>> The problem is that I have some other hardware that does not work all
>> that well after suspend/hibernate and so I can have either that hardware
>> (M-Audio 1010) working and the mouse at risk of failing or the 1010 not
>> working.
>>
>
> So what you're saying is, while the M-Audio 1010 is working the ps/2 stuff
> fails, and when the M-Audio 1010 has stopped working due to
> suspend/hibernate the PS/2 stuff works fine. Coincidence?
>
> I find it odd that anything would interfere with a ps/2 keyboard, since it's
> on such a low and well known IRQ. Keyboard is 1 and mouse is 12.
>
> APIC is enabled in the bios?
>
>
>
It is odd as PS2 interfaces and 8042's are as old as the hills.
Following your reply I monitored /proc/interrupts (via a remote ssh
session) and I can see the count of keyboard interrupts rising on CPU2
and the mouse ones on CPU3. When the failure occurs the counts stop
rising. If I suspend/resume the machine the count then rises again as
the PS2 devices are working once more.
So it seems like the 8042 stops interrupting the processor, or the
processor/apic is ignoring them, or something along those lines. The
evidence is that keyboard & mouse cease functioning and the interrupt
count stops rising.
Noticeable things about this are:-
* The fault only occurs with X running
* Both keyboard and mouse are affected
* USB mouse still works
* Fault vanishes after suspend/resume
* cpu0 has a few interrupts logged and then the count rises on cpu2
(keyboard) and cpu3 (mouse)
As far as I can see there are no APIC settings in the BIOS on this
motherboard
I did notice that one of my two Delta 1010's share an interrupt with the
video card so I will need to move that card to another slot
xorg.conf includes
Section "InputDevice"
# generated from data in "/etc/sysconfig/keyboard"
Identifier "Keyboard0"
Driver "keyboard"
Option "XkbLayout" "gb"
Option "XkbModel" "pc105"
EndSection
Section "InputDevice"
# generated from default
Identifier "Mouse0"
Driver "mouse"
Option "Protocol" "auto"
Option "Device" "/dev/input/mice"
Option "Emulate3Buttons" "no"
Option "ZAxisMapping" "4 5"
EndSection
Cat'ing /dev/input/mice produces a stream of characters. I was waiting
for the system to fail while doing that before sending this message but
the mouse/keyboard are still working. Its Heisenberg's Uncertainty
Principle at work.....
Maybe I'll need to brew up some code to log the traffic to the 8042
Anyone else seen this?
Thanks
Ken
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
--
users mailing list
users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
