* Re: PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook)
2003-09-01 2:45 PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook) Brown, Len
@ 2003-09-02 6:34 ` Stefan Winter
2003-09-02 18:25 ` Stefan Winter
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Stefan Winter @ 2003-09-02 6:34 UTC (permalink / raw)
To: Brown, Len, linux-kernel
Hello,
acpi=off does not make any difference. Yesterday I compiled a 2.5.25 kernel
and got some more info: during boot, the warning message
i8042.c: Couldnt get IRQ1 for KBD0
(or similar, the messages passed by quickly and, of course, scroll lock didn´t
work:-) appeared.
Plus, I have got a correction concerning the BIOS warning after reboot:
> > warns about "No interrupts from keyboard 0" and the BIOS
Actually, the warning is "No interrupts from TIMER 0." As a sidenote, the BIOS
in question seems somewhat broken. It reports overlapping MTRR regions and
sets the PCI cache line size incorrectly, according to the kernel.
Furthermore it reports bad refresh rates for my TFT display to the X server.
After all, I wouldn´t be surprised if this all relates to the f*cked up BIOS
(it is some "Insyde BIOS 1.07").
If it helps, I can go down the kernel versions to locate the version where it
broke.
Greetings,
Stefan Winter
> Stefan,
> Did earlier versions of 2.6 (or 2.5) work?
> Does booting 2.6.0-test4 with acpi=off make any difference?
>
> Curious that vanilla 2.4.x has an issue with kbd interrupts on this box,
> but SuSE 8.2 does not. Apparently SuSE 8.2 has a fix or workaround for
> this box that isn't in the baseline -- anybody know what it is?
>
> Thanks,
> -Len
>
> > [7.7.] Other information that might be relevant to the problem
> > (please look in /proc and include all information that you
> > think to be relevant):
> > sunshine:/proc # cat interrupts
> > CPU0
> > 0: 4574746 XT-PIC timer
> > 1: 14 XT-PIC i8042
> > 2: 0 XT-PIC cascade
> > 5: 2 XT-PIC ohci1394, VIA8233
> > 9: 5 XT-PIC acpi
> > 10: 4029 XT-PIC eth0
> > 11: 0 XT-PIC ehci_hcd
> > 12: 89 XT-PIC i8042
> > 14: 5523 XT-PIC ide0
> > 15: 45 XT-PIC ide1
> > NMI: 0
> > LOC: 4570927
> > ERR: 62338
> > MIS: 0
> >
> > [X.] Other notes, patches, fixes, workarounds:
> > the interrupt count on INT1,CPU0 (see /proc/interrupts above) does not
> > increase after key pressures. It looks like the interrupt
> > from the keyboard
> > isn´t caught. There is an issue on 2.4.x kernels that may
> > relate to that
> > problem: after running the kernel and restarting it with
> > "init 6", the BIOS
> > warns about "No interrupts from keyboard 0" and the BIOS
> > password entry
> > field doesn´t catch keypresses any more [this is only an
> > issue with the
> > vanilla kernel - SuSE 8.2´s patched kernel is fine]. An "init 0" and
> > re-poweron solves that issue in 2.4.x kernels. Maybe the 2.6. kernel
> > encounters that phenomenon earlier.
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook)
2003-09-01 2:45 PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook) Brown, Len
2003-09-02 6:34 ` Stefan Winter
@ 2003-09-02 18:25 ` Stefan Winter
2003-09-23 17:13 ` Stefan Winter
2003-09-29 7:00 ` [SOLVED] no keyboard and mouse on 2.6.0-test5 (notebook) Stefan Winter
3 siblings, 0 replies; 5+ messages in thread
From: Stefan Winter @ 2003-09-02 18:25 UTC (permalink / raw)
To: Brown, Len, linux-kernel
> Did earlier versions of 2.6 (or 2.5) work?
I have now localized the version that broke keyboard support for me. It worked
in 2.5.24 and broke in 2.5.25. After a quick look at the ChangeLog for 2.5.25
it seems that <vojtech@twilight.ucw.cz> has commited quite some stuff that
relates to keyboard.
Greetings,
Stefan Winter
>
> Curious that vanilla 2.4.x has an issue with kbd interrupts on this box,
> but SuSE 8.2 does not. Apparently SuSE 8.2 has a fix or workaround for
> this box that isn't in the baseline -- anybody know what it is?
>
> Thanks,
> -Len
>
> > [7.7.] Other information that might be relevant to the problem
> > (please look in /proc and include all information that you
> > think to be relevant):
> > sunshine:/proc # cat interrupts
> > CPU0
> > 0: 4574746 XT-PIC timer
> > 1: 14 XT-PIC i8042
> > 2: 0 XT-PIC cascade
> > 5: 2 XT-PIC ohci1394, VIA8233
> > 9: 5 XT-PIC acpi
> > 10: 4029 XT-PIC eth0
> > 11: 0 XT-PIC ehci_hcd
> > 12: 89 XT-PIC i8042
> > 14: 5523 XT-PIC ide0
> > 15: 45 XT-PIC ide1
> > NMI: 0
> > LOC: 4570927
> > ERR: 62338
> > MIS: 0
> >
> > [X.] Other notes, patches, fixes, workarounds:
> > the interrupt count on INT1,CPU0 (see /proc/interrupts above) does not
> > increase after key pressures. It looks like the interrupt
> > from the keyboard
> > isn´t caught. There is an issue on 2.4.x kernels that may
> > relate to that
> > problem: after running the kernel and restarting it with
> > "init 6", the BIOS
> > warns about "No interrupts from keyboard 0" and the BIOS
> > password entry
> > field doesn´t catch keypresses any more [this is only an
> > issue with the
> > vanilla kernel - SuSE 8.2´s patched kernel is fine]. An "init 0" and
> > re-poweron solves that issue in 2.4.x kernels. Maybe the 2.6. kernel
> > encounters that phenomenon earlier.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook)
2003-09-01 2:45 PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook) Brown, Len
2003-09-02 6:34 ` Stefan Winter
2003-09-02 18:25 ` Stefan Winter
@ 2003-09-23 17:13 ` Stefan Winter
2003-09-29 7:00 ` [SOLVED] no keyboard and mouse on 2.6.0-test5 (notebook) Stefan Winter
3 siblings, 0 replies; 5+ messages in thread
From: Stefan Winter @ 2003-09-23 17:13 UTC (permalink / raw)
To: Brown, Len, linux-kernel; +Cc: vojtech
Hi,
I investigated the below issue with keyboard and mouse not working under
2.6.0-testX further (still present in test5). For my box, it broke with the
patch from 2.5.32 to 2.5.33, specifically with the patch to file atkbd.c.
This change belongs to the Changelog entry
" More careful probe and init in atkbd.c and psmouse.c
to avoid triggering bugs in certain keyboards."
from vojtech at suse dot cz (cc´ed). I would fiddle with the patch to figure
out what exactly broke the keyboard support, but I don´t have a clue where to
start. Some advice in private mail would sure be of great help.
Am Montag, 1. September 2003 04:45 schrieb Brown, Len:
> Stefan,
> Did earlier versions of 2.6 (or 2.5) work?
> Does booting 2.6.0-test4 with acpi=off make any difference?
>
> Curious that vanilla 2.4.x has an issue with kbd interrupts on this box,
> but SuSE 8.2 does not. Apparently SuSE 8.2 has a fix or workaround for
> this box that isn't in the baseline -- anybody know what it is?
>
> Thanks,
> -Len
>
> > [7.7.] Other information that might be relevant to the problem
> > (please look in /proc and include all information that you
> > think to be relevant):
> > sunshine:/proc # cat interrupts
> > CPU0
> > 0: 4574746 XT-PIC timer
> > 1: 14 XT-PIC i8042
> > 2: 0 XT-PIC cascade
> > 5: 2 XT-PIC ohci1394, VIA8233
> > 9: 5 XT-PIC acpi
> > 10: 4029 XT-PIC eth0
> > 11: 0 XT-PIC ehci_hcd
> > 12: 89 XT-PIC i8042
> > 14: 5523 XT-PIC ide0
> > 15: 45 XT-PIC ide1
> > NMI: 0
> > LOC: 4570927
> > ERR: 62338
> > MIS: 0
> >
> > [X.] Other notes, patches, fixes, workarounds:
> > the interrupt count on INT1,CPU0 (see /proc/interrupts above) does not
> > increase after key pressures. It looks like the interrupt
> > from the keyboard
> > isn´t caught. There is an issue on 2.4.x kernels that may
> > relate to that
> > problem: after running the kernel and restarting it with
> > "init 6", the BIOS
> > warns about "No interrupts from timer 0" and the BIOS
> > password entry
> > field doesn´t catch keypresses any more [this is only an
> > issue with the
> > vanilla kernel - SuSE 8.2´s patched kernel is fine]. An "init 0" and
> > re-poweron solves that issue in 2.4.x kernels. Maybe the 2.6. kernel
> > encounters that phenomenon earlier.
^ permalink raw reply [flat|nested] 5+ messages in thread
* [SOLVED] no keyboard and mouse on 2.6.0-test5 (notebook)
2003-09-01 2:45 PROBLEM: no keyboard and mouse on 2.6.0-test4 (notebook) Brown, Len
` (2 preceding siblings ...)
2003-09-23 17:13 ` Stefan Winter
@ 2003-09-29 7:00 ` Stefan Winter
3 siblings, 0 replies; 5+ messages in thread
From: Stefan Winter @ 2003-09-29 7:00 UTC (permalink / raw)
To: Brown, Len, linux-kernel, vojtech
Hi,
2.6.0-test6 fixes the below issue. Thanks guys!
Stefan Winter
Am Montag 01 September 2003 04:45 schrieb Brown, Len:
> Stefan,
> Did earlier versions of 2.6 (or 2.5) work?
> Does booting 2.6.0-test4 with acpi=off make any difference?
>
> Curious that vanilla 2.4.x has an issue with kbd interrupts on this box,
> but SuSE 8.2 does not. Apparently SuSE 8.2 has a fix or workaround for
> this box that isn't in the baseline -- anybody know what it is?
>
> Thanks,
> -Len
>
> > [7.7.] Other information that might be relevant to the problem
> > (please look in /proc and include all information that you
> > think to be relevant):
> > sunshine:/proc # cat interrupts
> > CPU0
> > 0: 4574746 XT-PIC timer
> > 1: 14 XT-PIC i8042
> > 2: 0 XT-PIC cascade
> > 5: 2 XT-PIC ohci1394, VIA8233
> > 9: 5 XT-PIC acpi
> > 10: 4029 XT-PIC eth0
> > 11: 0 XT-PIC ehci_hcd
> > 12: 89 XT-PIC i8042
> > 14: 5523 XT-PIC ide0
> > 15: 45 XT-PIC ide1
> > NMI: 0
> > LOC: 4570927
> > ERR: 62338
> > MIS: 0
> >
> > [X.] Other notes, patches, fixes, workarounds:
> > the interrupt count on INT1,CPU0 (see /proc/interrupts above) does not
> > increase after key pressures. It looks like the interrupt
> > from the keyboard
> > isn´t caught. There is an issue on 2.4.x kernels that may
> > relate to that
> > problem: after running the kernel and restarting it with
> > "init 6", the BIOS
> > warns about "No interrupts from keyboard 0" and the BIOS
> > password entry
> > field doesn´t catch keypresses any more [this is only an
> > issue with the
> > vanilla kernel - SuSE 8.2´s patched kernel is fine]. An "init 0" and
> > re-poweron solves that issue in 2.4.x kernels. Maybe the 2.6. kernel
> > encounters that phenomenon earlier.
^ permalink raw reply [flat|nested] 5+ messages in thread