All of lore.kernel.org
 help / color / mirror / Atom feed
* 2.6.24-rt1 SSttrraannggee keyboard behaviour
@ 2008-02-01 15:06 Uli Brueggemann
  2008-02-01 16:20 ` Clark Williams
  0 siblings, 1 reply; 3+ messages in thread
From: Uli Brueggemann @ 2008-02-01 15:06 UTC (permalink / raw)
  To: RT

Hi,

I'm trying to use 2.6.24-rt1 in combination with BruteFIR (audio). So
far everything looks fine except:

After starting BruteFIR (which is doing its job) after a while the
keyboard responds like with double clicks. TTeerrrriibbllee. I cannot
enter a command anymore. The running BruteFIR itself still is running
stable.
 The system without starting BruteFIR keeps stable. BruteFIR itself is
working fine with older kernels. Also with e.g. 2.6.24-rc3-zen3. So I
guess the realtime patch is causing this.
BTW Brutefir is using realtime priorities 1 to 4 and it also uses SCHED_FIFO.

Any idea?

Uli

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 2.6.24-rt1 SSttrraannggee keyboard behaviour
  2008-02-01 15:06 2.6.24-rt1 SSttrraannggee keyboard behaviour Uli Brueggemann
@ 2008-02-01 16:20 ` Clark Williams
  2008-02-03 16:45   ` Uli Brueggemann
  0 siblings, 1 reply; 3+ messages in thread
From: Clark Williams @ 2008-02-01 16:20 UTC (permalink / raw)
  To: Uli Brueggemann; +Cc: RT

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Uli Brueggemann wrote:
> Hi,
> 
> I'm trying to use 2.6.24-rt1 in combination with BruteFIR (audio). So
> far everything looks fine except:
> 
> After starting BruteFIR (which is doing its job) after a while the
> keyboard responds like with double clicks. TTeerrrriibbllee. I cannot
> enter a command anymore. The running BruteFIR itself still is running
> stable.
>  The system without starting BruteFIR keeps stable. BruteFIR itself is
> working fine with older kernels. Also with e.g. 2.6.24-rc3-zen3. So I
> guess the realtime patch is causing this.
> BTW Brutefir is using realtime priorities 1 to 4 and it also uses SCHED_FIFO.
> 
> Any idea?
> 
> Uli

Uli,

I've seen this behavior before; on my system in addition to the doubled keystrokes,
cursor animations were jerky.

Are you running on a 64-bit kernel, is your clocksource hpet and is vsyscall64 on?

$ uname -m
x86_64
$ cat /sys/devices/system/clocksource/clocksource0/current_clocksource
hpet
$ cat /proc/sys/kernel/vsyscall64
1

If so, try putting this into your /etc/sysctl.conf and rebooting:

# vsyscall64 off
kernel.vsyscall64 = 0

You could try echoing "1" into /proc/sys/kernel/vsyscall64, but with the keyboard
doubling issue, it might be a problem :)

Clark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkejRsMACgkQHyuj/+TTEp14GgCg1PV1n23nDDJg8Bp6NTEXIUTQ
iukAoK56x+wQcZ8Z+YNea8glzIe41EpA
=ZKru
-----END PGP SIGNATURE-----

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 2.6.24-rt1 SSttrraannggee keyboard behaviour
  2008-02-01 16:20 ` Clark Williams
@ 2008-02-03 16:45   ` Uli Brueggemann
  0 siblings, 0 replies; 3+ messages in thread
From: Uli Brueggemann @ 2008-02-03 16:45 UTC (permalink / raw)
  To: RT

>
> I've seen this behavior before; on my system in addition to the doubled keystrokes,
> cursor animations were jerky.
>
> Are you running on a 64-bit kernel, is your clocksource hpet and is vsyscall64 on?
>
> $ uname -m
> x86_64
> $ cat /sys/devices/system/clocksource/clocksource0/current_clocksource
> hpet
> $ cat /proc/sys/kernel/vsyscall64
> 1
>
> If so, try putting this into your /etc/sysctl.conf and rebooting:
>
> # vsyscall64 off
> kernel.vsyscall64 = 0
>


Kernel is i686 (32bit), clocksource is tsc and vsyscall64 does not exist.
Only console, no X.

In the meantime I have further tried to detect a possible reason and I
guess I kow it now. But I do not know how to overcome:

The system is booted from USB memorystick. The keyboard is a USB
keyboard. And there is an IMON VFD display connected by USB. To use it
I run the lcdproc package.
It seems that the LCDd daemon (with imon_pad driver) is the reason. It
causes problems with the keyboard (as described in the topic) and also
with access to the memorystick.
If I use a PS/2 keyboard the keyboard problem does not occur. But
after a while the system still gives messages like

usb 5-3: reset high speed USB device using ehci_hcd and adress 3
usb 5-3: device not accepting adress 3, error -10

If I do not start LCDd then both keyboard and memorystick access keep stable.

So I suppose that updating the VFD display by LCDd via the USB
interface is the problem. But this is happening now the first time
with use of 2.6.24-rt1.
So avoiding LCDd helps but I like to use the VFD display. How to solve?

BTW I have noticed another effect: with a PS/2 keyboard the screen
switches off after some time of non-use (screensaver?). With an USB
keyboard the screen never switches off.

If someone can comment helpfully I would be happy.

Best

Uli

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2008-02-03 16:45 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-02-01 15:06 2.6.24-rt1 SSttrraannggee keyboard behaviour Uli Brueggemann
2008-02-01 16:20 ` Clark Williams
2008-02-03 16:45   ` Uli Brueggemann

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.