Bug #1922

2.9 i386 mouse freeze, keyboard working, core dumped to leaf

Added by sgeorge.ml almost 4 years ago. Updated over 3 years ago.

Status:ClosedStart date:
Priority:NormalDue date:
Assignee:tuxillo% Done:

0%

Category:-
Target version:-

Description

Hi,

I have been getting mouse freezes on 2.9/x86 after the System runs
for around an hour.
There is nothing in the messages or the console in which 'startx' was
run that points to this.

This was happening I guess from 2-4 weeks.
I did the following

1) Change the window manager to see if it was WM specific - No it is
not, happens with fvwm as well as xfce
2) Followed the development source for around 4 weeks.
3) Test with GENERIC and GENERIC.MP KERNCONF file.

The result was same.

The mouse freezes and the mouse pointer gets stuck at the top right
corner of the screen.
Key board works.
Pressing Ctrl+Alt+F1 takes you to the console. Nothing there. Nothing
in /var/log/messages
At this point startx can be killed with Ctrl+c and then restarted
every thing is fine till around an hour when the freeze takes place

My kernel has the following sound modules complied into it

#Audio
device pcm
device snd_hda
device pnpbios

I have taken 2 core dumps and put it on leaf

1) leaf:/home/sgeorge/crash.mouse-crash.tar.bz2 - with an MP kernel
2) leaf:/home/sgeorge/crash/crash1.tar.bz2 - with GENERIC kernel

Currently I need to kill X every one hour or so to continue working.
Hope some interesting bug will be found and fixed :-)

thanks

--Siju

unnamed (1.24 KB) sgeorge.ml, 02/08/2011 05:29 AM

unnamed (3.17 KB) sgeorge.ml, 02/09/2011 02:25 PM

History

#1 Updated by ahuete.devel almost 4 years ago

Hi Siju,

I'm also experiencing trouble in latest master with psm(4). I'm
currently trying to track down this issue. In the mean time, you can
set sysctl debug.psm.loglevel to 9 and let me know the messages that
it spits out when the mouse get stuck.

Cheers,
Antonio Huete

2010/11/25 Siju George <>:
> Hi,
>
> I have been getting mouse freezes on 2.9/x86  after the System runs
> for around an hour.
> There is nothing in the messages or the console in which 'startx' was
> run that points to this.
>
> This was happening I guess from 2-4 weeks.
> I did the following
>
> 1) Change the window manager to see if it was WM specific - No it is
> not, happens with fvwm as well as xfce
> 2) Followed the development source for around 4 weeks.
> 3) Test with GENERIC and GENERIC.MP KERNCONF file.
>
> The result was same.
>
> The mouse freezes and the mouse pointer gets stuck at the top right
> corner of the screen.
> Key board works.
> Pressing Ctrl+Alt+F1 takes you to the console. Nothing there. Nothing
> in /var/log/messages
> At this point startx can be killed with Ctrl+c and then restarted
> every thing is fine till around an hour when the freeze takes place
>
> My kernel has the following sound modules complied into it
>
> #Audio
> device          pcm
> device          snd_hda
> device          pnpbios
>
> I have taken 2 core dumps and put it on leaf
>
> 1) leaf:/home/sgeorge/crash.mouse-crash.tar.bz2  - with an MP kernel
> 2) leaf:/home/sgeorge/crash/crash1.tar.bz2 - with GENERIC kernel
>
> Currently I need to kill X every one hour or so to continue working.
> Hope some interesting bug will be found and fixed :-)
>
> thanks
>
> --Siju
>

#2 Updated by sgeorge.ml almost 4 years ago

On Thu, Nov 25, 2010 at 2:44 PM, Antonio Huete Jimenez
<> wrote:
> Hi Siju,
>
> I'm also experiencing trouble in latest master with psm(4). I'm
> currently trying to track down this issue. In the mean time, you can
> set sysctl debug.psm.loglevel to 9 and let me know the messages that
> it spits out when the mouse get stuck.
>

Thanks Antonio :-)
As a surprise i have not had the mouse freeze after turning that
sysctl option but here is a sample of what it spits out.
Just wondering if it is interesting

Nov 25 17:37:36 dfly-vmsrv kernel: psmintr: 28 02 ff 00 0a 00
Nov 25 17:37:36 dfly-vmsrv kernel: psmintr: 28 01 ff 00 0a 00
Nov 25 17:37:38 dfly-vmsrv kernel: psm0: lost interrupt?
Nov 25 17:37:44 dfly-vmsrv last message repeated 6 times
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 04 02 00 08 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 06 06 00 04 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 07 06 00 03 00
Nov 25 17:37:45 dfly-vmsrv kernel: psm0: lost interrupt?
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 09 07 00 01 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 0e 0a 00 01 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 0d 09 00 00 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 11 09 00 03 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 0e 05 00 00 00
Nov 25 17:37:45 dfly-vmsrv kernel: psmintr: 08 11 05 00 00 00

thanks for looking into this :-)

--Siju

#3 Updated by sgeorge.ml almost 4 years ago

On Thu, Nov 25, 2010 at 2:44 PM, Antonio Huete Jimenez
<> wrote:
>
> Hi Siju,
>
> I'm also experiencing trouble in latest master with psm(4). I'm
> currently trying to track down this issue. In the mean time, you can
> set sysctl debug.psm.loglevel to 9 and let me know the messages that
> it spits out when the mouse get stuck.
>

Hi Antonio, this is it

Nov 26 10:38:26 dfly-vmsrv kernel: psmintr: 28 00 ff 00 06 00
Nov 26 10:38:26 dfly-vmsrv kernel: psmintr: 28 00 ff 00 08 00
Nov 26 10:38:28 dfly-vmsrv kernel: psm0: lost interrupt?
Nov 26 10:38:54 dfly-vmsrv last message repeated 26 times
Nov 26 10:38:54 dfly-vmsrv login: ROOT LOGIN (root) ON ttyv0
Nov 26 10:38:55 dfly-vmsrv kernel: psm0: lost interrupt?
Nov 26 10:39:18 dfly-vmsrv last message repeated 23 times

thanks

--Siju

#4 Updated by ahuete.devel almost 4 years ago

Hi Siju,

When your mouse freezes again, can you please check in 'vmstat -i'
whether psm0 interrupts are increased when you move the mouse or stay
the same all the time?

PD: I can't use my mouse on SMP kernel at all in latest master, as psm
misses all the interrupts. In UP kernel I have no problems.

Cheers,
Antonio Huete

2010/11/26 Siju George <>:
> On Thu, Nov 25, 2010 at 2:44 PM, Antonio Huete Jimenez
> <> wrote:
>>
>> Hi Siju,
>>
>> I'm also experiencing trouble in latest master with psm(4). I'm
>> currently trying to track down this issue. In the mean time, you can
>> set sysctl debug.psm.loglevel to 9 and let me know the messages that
>> it spits out when the mouse get stuck.
>>
>
> Hi Antonio, this is it
>
> Nov 26 10:38:26 dfly-vmsrv kernel: psmintr: 28 00 ff 00 06 00
> Nov 26 10:38:26 dfly-vmsrv kernel: psmintr: 28 00 ff 00 08 00
> Nov 26 10:38:28 dfly-vmsrv kernel: psm0: lost interrupt?
> Nov 26 10:38:54 dfly-vmsrv last message repeated 26 times
> Nov 26 10:38:54 dfly-vmsrv login: ROOT LOGIN (root) ON ttyv0
> Nov 26 10:38:55 dfly-vmsrv kernel: psm0: lost interrupt?
> Nov 26 10:39:18 dfly-vmsrv last message repeated 23 times
>
> thanks
>
> --Siju
>

#5 Updated by sgeorge.ml almost 4 years ago

On Fri, Nov 26, 2010 at 5:14 PM, Antonio Huete Jimenez
<> wrote:
> Hi Siju,
>
> When your mouse freezes again, can you please check in 'vmstat -i'
>

Will do that on monday because it is my office machine :-)

>whether psm0 interrupts are increased when you move the mouse or stay
> the same all the time?
>

When I move the mouse during normal operations like switching pages in
fvwm or scrolling a page.

> PD: I can't use my mouse on SMP kernel at all in latest master, as psm
> misses all the interrupts. In UP kernel I have no problems.
>

Since it was difficult to use the mouse properly i switched to a USB
mouse which I left because of

http://leaf.dragonflybsd.org/mailarchive/users/2010-05/msg00132.html
http://leaf.dragonflybsd.org/mailarchive/users/2010-05/msg00152.html
http://leaf.dragonflybsd.org/mailarchive/users/2010-05/msg00166.html

I find that it is working OK now except for some latence in movement some times.

I will re attach the PS/2 mouse on monday and give you te results you asked

Thanks again for looking into this :-)

--Siju

#6 Updated by sgeorge.ml almost 4 years ago

On Fri, Nov 26, 2010 at 5:17 PM, Antonio Huete Jimenez (via DragonFly
issue tracker) <> wrote:
>
> Antonio Huete Jimenez <> added the comment:
>
> Hi Siju,
>
> When your mouse freezes again, can you please check in 'vmstat -i'
>

Here is it :-)

interrupt total rate
clk 1362927 278
rl1/rl2 122969 25
sio0 0 0
uhci0/ehci0/rl0 2 0
fdc0 4 0
ppc0 1 0
acpi0 0 0
pcm0/uhci3 191856 39
uhci1/atapci1 32994 6
psm0 100516 20
ata0 35 0
uhci2/ata1 0 0
irq19 4881 0
irq21 28302 5
swi_siopoll 0 0
swi_cambio 0 0
swi_vm 0 0
swi_mp_taskq/swi_taskq 0 0
Total 1844487 376

thanks :-D

-Siju

#7 Updated by tuxillo almost 4 years ago

Hi Siju,

Can you please retry with latest master? I was having issues with my PS/2 also
until Sephe committed some fixes for the IO APIC work.

Cheers,
Antonio Huete

#8 Updated by sgeorge.ml almost 4 years ago

On Sat, Jan 15, 2011 at 2:43 PM, Antonio Huete Jimenez (via DragonFly
issue tracker) <> wrote:
>
> Antonio Huete Jimenez <> added the comment:
>
> Hi Siju,
>
> Can you please retry with latest master? I was having issues with my PS/2 also
> until Sephe committed some fixes for the IO APIC work.
>

Ok Antonio I will do that as soon as I reach office on Monday.
Hope the New changes for big memory is quite stable?
I am running x86 so I wonder what will be the effects since i guess
x86 can use only 3GB of RAM anyways.

Thanks

--Siju

#9 Updated by tuxillo almost 4 years ago

Hi Siju,

Did you have a chance to try it? It would be really nice to see if it worked or not.

Cheers,
Antonio Huete

#10 Updated by sgeorge.ml almost 4 years ago

Oh,

I am sorry I was not able to test it.
Sick and at home.

Will do it tomorrow as I get to office.

Thanks and sorry for the delay :-(

--Siju

On Tue, Feb 8, 2011 at 3:52 AM, Antonio Huete Jimenez (via DragonFly issue
tracker) <> wrote:

>
> Antonio Huete Jimenez <> added the comment:
>
> Hi Siju,
>
> Did you have a chance to try it? It would be really nice to see if it
> worked or not.
>
> Cheers,
> Antonio Huete
>
> ----------
> assignedto: -> tuxillo
> nosy: +tuxillo
>
> _____________________________________________________
> DragonFly issue tracker <>
> <http://bugs.dragonflybsd.org/issue1922>
> _____________________________________________________
>

#11 Updated by sgeorge.ml almost 4 years ago

Hi Antonio,

I have run the system for more than 1 hr using

DragonFly dfly-vmsrv.hifxnx.local 2.9-DEVELOPMENT DragonFly
v2.9.1.531.g21e43-DEVELOPMENT #0: Tue Jan 25 07:44:27 IST 2011

and so far there is no freezed but for psm log level 9 it gives this

Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 02 00 ff 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fd 03 00 fd 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 03 00 fc 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fd 03 00 ff 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 02 00 00 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 02 00 00 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 01 00 ff 00
Feb 9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 01 00 fb 00
Feb 9 14:19:18 dfly-vmsrv kernel: psmintr: 18 ff 00 00 f1 00
Feb 9 14:19:19 dfly-vmsrv kernel: psm0: lost interrupt?
Feb 9 14:19:37 dfly-vmsrv last message repeated 18 times
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 02 00 fe 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 07 00 00 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 07 00 01 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 05 00 01 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 03 00 04 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 03 00 04 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 02 00 02 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 02 00 02 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 00 00 01 00
Feb 9 14:19:37 dfly-vmsrv kernel: psmintr: 28 01 ff 00 02 00
Feb 9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 fd 00 02 00
Feb 9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 fb 00 01 00
Feb 9 14:19:38 dfly-vmsrv kernel: psmintr: 28 03 fa 00 02 00
Feb 9 14:19:38 dfly-vmsrv kernel: psmintr: 28 03 f9 00 02 00
Feb 9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 f9 00 01 00

Is that OK?

I will run with the latest kernel for a day and tell you if I see anything
wrong

Thanks a million :-)

--Siju

On Tue, Feb 8, 2011 at 3:52 AM, Antonio Huete Jimenez (via DragonFly issue
tracker) <> wrote:

>
> Antonio Huete Jimenez <> added the comment:
>
> Hi Siju,
>
> Did you have a chance to try it? It would be really nice to see if it
> worked or not.
>
> Cheers,
> Antonio Huete
>
> ----------
> assignedto: -> tuxillo
> nosy: +tuxillo
>
> _____________________________________________________
> DragonFly issue tracker <>
> <http://bugs.dragonflybsd.org/issue1922>
> _____________________________________________________
>

#12 Updated by sgeorge.ml almost 4 years ago

Hi,

I have been running

2.9-DEVELOPMENT DragonFly v2.9.1.664.g744f6-DEVELOPMENT #2: Thu Feb 10
11:31:15 IST 2011

with no probs!

Thanks a million :-)

--Siju

On Wed, Feb 9, 2011 at 7:52 PM, Siju George <> wrote:
> Hi Antonio,
>
> I have run the system for more than 1 hr using
>
> DragonFly dfly-vmsrv.hifxnx.local 2.9-DEVELOPMENT DragonFly
> v2.9.1.531.g21e43-DEVELOPMENT #0: Tue Jan 25 07:44:27 IST 2011
>
> and so far there is no freezed but for psm log level 9 it gives this
>
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 02 00 ff 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fd 03 00 fd 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 03 00 fc 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fd 03 00 ff 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 fe 02 00 00 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 02 00 00 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 01 00 ff 00
> Feb  9 14:19:17 dfly-vmsrv kernel: psmintr: 18 ff 01 00 fb 00
> Feb  9 14:19:18 dfly-vmsrv kernel: psmintr: 18 ff 00 00 f1 00
> Feb  9 14:19:19 dfly-vmsrv kernel: psm0: lost interrupt?
> Feb  9 14:19:37 dfly-vmsrv last message repeated 18 times
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 02 00 fe 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 07 00 00 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 07 00 01 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 02 05 00 01 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 03 00 04 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 03 00 04 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 02 00 02 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 00 02 00 02 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 08 01 00 00 01 00
> Feb  9 14:19:37 dfly-vmsrv kernel: psmintr: 28 01 ff 00 02 00
> Feb  9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 fd 00 02 00
> Feb  9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 fb 00 01 00
> Feb  9 14:19:38 dfly-vmsrv kernel: psmintr: 28 03 fa 00 02 00
> Feb  9 14:19:38 dfly-vmsrv kernel: psmintr: 28 03 f9 00 02 00
> Feb  9 14:19:38 dfly-vmsrv kernel: psmintr: 28 02 f9 00 01 00
>
>
> Is that OK?
>
> I will run with the latest kernel for a day and tell you if I see anything
> wrong
>
> Thanks a million :-)
>
> --Siju
>
>
>
> On Tue, Feb 8, 2011 at 3:52 AM, Antonio Huete Jimenez (via DragonFly issue
> tracker) <> wrote:
>>
>> Antonio Huete Jimenez <> added the comment:
>>
>> Hi Siju,
>>
>> Did you have a chance to try it? It would be really nice to see if it
>> worked or not.
>>
>> Cheers,
>> Antonio Huete
>>
>> ----------
>> assignedto:  -> tuxillo
>> nosy: +tuxillo
>>
>> _____________________________________________________
>> DragonFly issue tracker <>
>> <http://bugs.dragonflybsd.org/issue1922>
>> _____________________________________________________
>
>

Also available in: Atom PDF