Hello!
Fernando Paredes <[email protected]> wrote:
FP> Applied these patches. Nothing while tail'ing /var/log/messages. Nothing
FP> in the root console that I can see either.
FP> Patched the source to 2.6.6. Still get the same lockups, totally random.
FP> Any more ideas?
Not sure if I have exact problem like you do, but at least I have something
similar. Once in a while keyboard suddenly stopps working, touchpad still work
though (I have Toshiba Satellite Pro (centrino based) laptop here).
I figured out that if I leave the keyboard for some time (up to 2 minutes),
it starts to work again, at least this was the case with XFree 4.4,
during those no-keyboard times, mouse cursor was moving with small jumps
(when keyboard works it moves smoothly).
I upgraded to FC2 (and hence to xorg X server) today, and lockup happened
once already, the "wait for some time" strategy did not work, so I
remembered initially I thought this was something bad pressed on keyboard
(btw, usually lockups happens when I press several keys at the same time by
accident (but not always when I do this)), so I was pressing various
modifier keys and Fn-Fx stuff. (Fn is one key, Fx is set of functional keys).
So today after some waiting I jut pressed alt+ctrl+shifts in some
conbinations and it unfroze almost immediately to my surprise.
Also I can add that sometimes when I press more than one alphanumeric key,
I get this sort of message from kernel:
"atkbd.c: Keyboard on isa0060/serio0 reports too many keys pressed."
Also sometimes (not always) after the freeze/unfreeze I get various messages
from atkbd.c, here are some examples:
atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0).
atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly.
(numbers may differ from time to time)
atkbd.c: Unknown key pressed (translated set 0, code 0x1d on isa0060/serio0).
atkbd.c: Use 'setkeycodes 1d <keycode>' to make it known.
(numbers may differ from time to time voth in keycode and in translated set)
atkbd.c: Unknown key pressed (translated set 0, code 0x22 on isa0060/serio0).
(again numbers may vary)
atkbd.c: Failed to enable keyboard on isa0060/serio0
(this is some ancient message from October 25th, 2004, seen it only once,
just prior to reeboot, according to logs, do not remember details).
And also I often see multiple
"input: AT Translated Set 2 keyboard on isa0060/serio0" messages.
And new ones are appearing after the freeze/unfreeze. (not every time, though).
May be that will help someone.
Bye,
Oleg
On 2:15:12 pm 06/10/04 Oleg Drokin <[email protected]> wrote:
> Hello!
>
> Fernando Paredes <[email protected]> wrote:
>
> FP> Applied these patches. Nothing while tail'ing /var/log/messages.
> Nothing FP> in the root console that I can see either.
> FP> Patched the source to 2.6.6. Still get the same lockups, totally
> random. FP> Any more ideas?
>
> Not sure if I have exact problem like you do, but at least I have
> something similar. Once in a while keyboard suddenly stopps working,
> touchpad still work though (I have Toshiba Satellite Pro (centrino
> based) laptop here). I figured out that if I leave the keyboard for
> some time (up to 2 minutes), it starts to work again, at least this
> was the case with XFree 4.4, during those no-keyboard times, mouse
> cursor was moving with small jumps (when keyboard works it moves
> smoothly). I upgraded to FC2 (and hence to xorg X server) today, and
> lockup happened once already, the "wait for some time" strategy did
> not work, so I remembered initially I thought this was something bad
> pressed on keyboard
I have had similar issues with a toshiba laptop keyboard with 2.6+ kernels
for awhile. I have found that repeating the last key combination pressed
will "unlock" it. No logs or dmesg entries are produced when the lockup
occurs.
Its a Toshiba Satellite 1410-S173, currently running 2.6.7-rc2-mm2
\__ Jason Munro
\__ [email protected]
\__ http://hastymail.sourceforge.net/
To the benefit of developers, here are the tracebacks for today. I
haven't experienced any lockups so far, BUT i have experienced two weird
experiences:
1) The space key seemed to get stuck and it unstock by pressing <backspace>
2) The 'n' key stuck and unstuck itself, displaying about 5 'n' on my
screen.
For both events I have an identified backtrace from 'dmesg' (thanks, to
the traceback code in atkbd.c). There is another event there in the logs
where I didn't notice in the keyboard.
As someone in this thread expressed earlier, it seems that the driver is
requesting a keyboard reconnect. This shouldn't be hapenning, right?
Here are the traces with their desrcibing title:
----> Space key stuck until I pressed "BACKSPACE" <-----
serio: RESCAN || RECONNECT requested: 1!
Call Trace:
[<c02b068c>] serio_queue_event+0x95/0x97
[<c02b06a5>] serio_rescan+0x17/0x1b
[<c02af4d7>] atkbd_interrupt+0x465/0x550
[<c0113333>] scheduler_tick+0x1f/0x530
[<c02b073f>] serio_interrupt+0x7b/0x7d
[<c02b0f80>] i8042_interrupt+0xd0/0x14c
[<c01058b9>] handle_IRQ_event+0x3a/0x64
[<c0105c41>] do_IRQ+0xdc/0x1c0
=======================
[<c0104194>] common_interrupt+0x18/0x20
[<c022c2b0>] acpi_processor_idle+0x13c/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c022c174>] acpi_processor_idle+0x0/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c01020a9>] cpu_idle+0x2e/0x37
[<c04066b0>] start_kernel+0x15d/0x17a
[<c0406402>] unknown_bootoption+0x0/0x129
atkbd.c: Unknown key released (translated set 2, code 0x39 on
isa0060/serio0).
atkbd.c: Use 'setkeycodes 39 <keycode>' to make it known.
input: AT Translated Set 2 keyboard on isa0060/serio0
----> Some random event, nothing notable happened to the keyboard: <-----
serio: RESCAN || RECONNECT requested: 1!
Call Trace:
[<c02b068c>] serio_queue_event+0x95/0x97
[<c02b06a5>] serio_rescan+0x17/0x1b
[<c02af4d7>] atkbd_interrupt+0x465/0x550
[<c0113333>] scheduler_tick+0x1f/0x530
[<c02b073f>] serio_interrupt+0x7b/0x7d
[<c02b0f80>] i8042_interrupt+0xd0/0x14c
[<c01058b9>] handle_IRQ_event+0x3a/0x64
[<c0105c41>] do_IRQ+0xdc/0x1c0
=======================
[<c0104194>] common_interrupt+0x18/0x20
[<c022c2b0>] acpi_processor_idle+0x13c/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c022c174>] acpi_processor_idle+0x0/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c01020a9>] cpu_idle+0x2e/0x37
[<c04066b0>] start_kernel+0x15d/0x17a
[<c0406402>] unknown_bootoption+0x0/0x129
----> 'n' key stuck for 2 seconds (4 'n's appeared..actually there were
two events here, not sure if they're related) <-----
Call Trace:
[<c02b068c>] serio_queue_event+0x95/0x97
[<c02b06a5>] serio_rescan+0x17/0x1b
[<c02af4d7>] atkbd_interrupt+0x465/0x550
[<c0113333>] scheduler_tick+0x1f/0x530
[<c02b073f>] serio_interrupt+0x7b/0x7d
[<c02b0f80>] i8042_interrupt+0xd0/0x14c
[<c01058b9>] handle_IRQ_event+0x3a/0x64
[<c0105c41>] do_IRQ+0xdc/0x1c0
=======================
[<c0104194>] common_interrupt+0x18/0x20
[<c022c2b0>] acpi_processor_idle+0x13c/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c010201e>] default_idle+0x0/0x27
[<c01020a9>] cpu_idle+0x2e/0x37
[<c04066b0>] start_kernel+0x15d/0x17a
[<c0406402>] unknown_bootoption+0x0/0x129
serio: RESCAN || RECONNECT requested: 1!
Call Trace:
[<c02b068c>] serio_queue_event+0x95/0x97
[<c02b06a5>] serio_rescan+0x17/0x1b
[<c02b070a>] serio_interrupt+0x46/0x7d
[<c02b0f80>] i8042_interrupt+0xd0/0x14c
[<c02b1112>] i8042_timer_func+0x0/0x23
[<c02b1131>] i8042_timer_func+0x1f/0x23
[<c011ea46>] run_timer_softirq+0xcb/0x1b0
[<c012615d>] rcu_process_callbacks+0x108/0x114
[<c011ab75>] __do_softirq+0x81/0x83
[<c0106590>] do_softirq+0x43/0x52
=======================
[<c0105cd2>] do_IRQ+0x16d/0x1c0
[<c0104194>] common_interrupt+0x18/0x20
[<c022c2b0>] acpi_processor_idle+0x13c/0x1cb
[<c010201e>] default_idle+0x0/0x27
[<c010201e>] default_idle+0x0/0x27
[<c01020a9>] cpu_idle+0x2e/0x37
[<c04066b0>] start_kernel+0x15d/0x17a
[<c0406402>] unknown_bootoption+0x0/0x129
Jason Munro wrote:
>On 2:15:12 pm 06/10/04 Oleg Drokin <[email protected]> wrote:
>
>
>>Hello!
>>
>>Fernando Paredes <[email protected]> wrote:
>>
>>FP> Applied these patches. Nothing while tail'ing /var/log/messages.
>>Nothing FP> in the root console that I can see either.
>>FP> Patched the source to 2.6.6. Still get the same lockups, totally
>>random. FP> Any more ideas?
>>
>>Not sure if I have exact problem like you do, but at least I have
>>something similar. Once in a while keyboard suddenly stopps working,
>>touchpad still work though (I have Toshiba Satellite Pro (centrino
>>based) laptop here). I figured out that if I leave the keyboard for
>>some time (up to 2 minutes), it starts to work again, at least this
>>was the case with XFree 4.4, during those no-keyboard times, mouse
>>cursor was moving with small jumps (when keyboard works it moves
>>smoothly). I upgraded to FC2 (and hence to xorg X server) today, and
>>lockup happened once already, the "wait for some time" strategy did
>>not work, so I remembered initially I thought this was something bad
>>pressed on keyboard
>>
>>
>
>I have had similar issues with a toshiba laptop keyboard with 2.6+ kernels
>for awhile. I have found that repeating the last key combination pressed
>will "unlock" it. No logs or dmesg entries are produced when the lockup
>occurs.
>
>
>Its a Toshiba Satellite 1410-S173, currently running 2.6.7-rc2-mm2
>
>\__ Jason Munro
> \__ [email protected]
> \__ http://hastymail.sourceforge.net/
>
>
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to [email protected]
>More majordomo info at http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at http://www.tux.org/lkml/
>
>
>
--
<http://www.sun.com/software> <http://java.sun.com>
*Fernando Paredes
Identity & Collaboration
Sun Microsystems de M?xico
Prol. Reforma #600-110, Col. Santa Fe, M?xico D.F. 01210
Tel Dir: 52 + (55) 5258 6152
Fax: 52 + (55) 5258 6199*
Jason Munro wrote:
>>Not sure if I have exact problem like you do, but at least I have
>>something similar. Once in a while keyboard suddenly stopps working,
>>touchpad still work though (I have Toshiba Satellite Pro (centrino
>>based) laptop here).
I know that "me toos" are of limited use, but I have also been
experiencing this problem. I can give some specific details, and on
request am willing to work to produce additional diagnostics / test
patches etc.
I am using a Toshiba Satellite 2410-603 with a P4 M processor, and I
have experienced difficulties with Gentoo-patched 2.6.3 sources,
Gentoo-patched 2.6.5 sources and vanilla 2.6.6 sources, with
CONFIG_PREEMPT both on and off.
Of the three, 2.6.3 seems the least affected - although I do get
occasional keyboard lockups, if I use the mouse for a few seconds the
keyboard becomes re-enabled. On both the 2.6.5 and 2.6.6 kernels a
keyboard lockup seems permanent, although I haven't tried leaving it for
more than a minute or two.
I too can see warnings from atkbd.c in the kernel messages (on all three
kernel versions) reporting 'Unknown key pressed' and 'too many keys
pressed'.
I am well aware that Toshiba keyboards are buggy - under 2.4 kernels I
experienced multiple key event problems, so I think what's being asked
here on this thread is not that the keyboard driver be "fixed" as such,
but that, if possible, it is extended to work around the buggy hardware.
Thanks,
Sean
I just reinstalled Gentoo this week and the problem appeared! With
kernel 2.6.5- I didn't had this problem. I now have it with 2.6.6 and
2.6.7. I'm using a Toshiba Satellite A20 (Canadian version).
Sean Legassick wrote:
> Jason Munro wrote:
>
>>> Not sure if I have exact problem like you do, but at least I have
>>> something similar. Once in a while keyboard suddenly stopps working,
>>> touchpad still work though (I have Toshiba Satellite Pro (centrino
>>> based) laptop here).
>
>
> I know that "me toos" are of limited use, but I have also been
> experiencing this problem. I can give some specific details, and on
> request am willing to work to produce additional diagnostics / test
> patches etc.
>
> I am using a Toshiba Satellite 2410-603 with a P4 M processor, and I
> have experienced difficulties with Gentoo-patched 2.6.3 sources,
> Gentoo-patched 2.6.5 sources and vanilla 2.6.6 sources, with
> CONFIG_PREEMPT both on and off.
>
> Of the three, 2.6.3 seems the least affected - although I do get
> occasional keyboard lockups, if I use the mouse for a few seconds the
> keyboard becomes re-enabled. On both the 2.6.5 and 2.6.6 kernels a
> keyboard lockup seems permanent, although I haven't tried leaving it for
> more than a minute or two.
>
> I too can see warnings from atkbd.c in the kernel messages (on all three
> kernel versions) reporting 'Unknown key pressed' and 'too many keys
> pressed'.
>
> I am well aware that Toshiba keyboards are buggy - under 2.4 kernels I
> experienced multiple key event problems, so I think what's being asked
> here on this thread is not that the keyboard driver be "fixed" as such,
> but that, if possible, it is extended to work around the buggy hardware.
>
> Thanks,
>
> Sean
>
On Thu, Jun 17, 2004 at 06:37:27PM -0400, Gabriel Lavoie wrote:
> I just reinstalled Gentoo this week and the problem appeared! With
> kernel 2.6.5- I didn't had this problem. I now have it with 2.6.6 and
> 2.6.7. I'm using a Toshiba Satellite A20 (Canadian version).
If you could track down which change between 2.6.5 and 2.6.6 makes the
difference for you, that'd be very helpful.
--
Vojtech Pavlik
SuSE Labs, SuSE CR
On Saturday 19 of June 2004 20:36, Vojtech Pavlik wrote:
> On Thu, Jun 17, 2004 at 06:37:27PM -0400, Gabriel Lavoie wrote:
> > I just reinstalled Gentoo this week and the problem appeared! With
> > kernel 2.6.5- I didn't had this problem. I now have it with 2.6.6 and
> > 2.6.7. I'm using a Toshiba Satellite A20 (Canadian version).
>
> If you could track down which change between 2.6.5 and 2.6.6 makes the
> difference for you, that'd be very helpful.
I've been having such problems since approx. 2.6.5-rc2 (anyway, the release in
which the Prism54 driver appeared for the first time ;-)) on a Satellite
1400-103. However, I've found recently that it's sufficient to press the
left SHIFT after a lockup to "release" the keyboard. There may be some other
specific key or a combination of keys that helps on other Toshiba models, I
guess.
Yours,
rjw
--
Rafael J. Wysocki,
SiSK
[tel. (+48) 605 053 693]
----------------------------
For a successful technology, reality must take precedence over public
relations, for nature cannot be fooled.
-- Richard P. Feynman