You are not logged in.

#1 2019-08-23 17:12:50

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Kernel Panic Oops 11 smp_timer

I'm getting a kernel oops 11 followed by kernel panic Fatal Exception in Interrupt.

hrtimer_interrupt

apic_timer_interrupt
smp_apic_timer_interrupt
sched_clock_cpu

I'm also getting random bsod's on Windows.  Maybe the same problem.

I did a memtest some time ago and no problems.

When I unplug the unit from the wall the CMOS clock is lost, so it might be the CMOS battery, but I  don't think that would prevent the computer from running once it gets plugged back in.  I've changed out hardware to see if that's the problem but don't think so.

I am running stock kernel, but when I PXE boot linux-rt it seems to be stable, but that's if it boots.  Maybe if I try a lts it won't crash.

Any ideas?

Last edited by nomorewindows (2019-08-23 17:14:46)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#2 2019-08-23 17:44:55

loqs
Member
Registered: 2014-03-06
Posts: 17,196

Re: Kernel Panic Oops 11 smp_timer

What is the full text of the panic?

Offline

#3 2019-08-23 18:32:06

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

loqs wrote:

What is the full text of the panic?

Is there a way for me to grab it?  I'm not seeing all of it on the sceen anyhow.


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#4 2019-08-23 19:02:52

loqs
Member
Registered: 2014-03-06
Posts: 17,196

Re: Kernel Panic Oops 11 smp_timer

Shift+PageUp but may well not be available after an Oops.
I was expecting the Oops to contain register contents,  a backtrace,  function name or address
(some may well be lost off screen and being inside an interrupt handler might explain some more being lost was hoping there was more to go on)
If nothing is in the journal possibly link to a screenshot.

Last edited by loqs (2019-08-23 19:03:15)

Offline

#5 2019-08-24 05:44:54

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

loqs wrote:

Shift+PageUp but may well not be available after an Oops.
I was expecting the Oops to contain register contents,  a backtrace,  function name or address
(some may well be lost off screen and being inside an interrupt handler might explain some more being lost was hoping there was more to go on)
If nothing is in the journal possibly link to a screenshot.

I caught function names on the screen in the first post...but yeah...doubt I'm seeing all of it.
It would possibly give a clue about what is going on.

Something about smp, hrtimer, and apic.  But deciding what is having a problem...

Hrtimer is high resolution timer

Apic Advanced programmable Interrupt Controller...possibly an IRQ conflict with something but hasn't been a problem before...but something similar is trying to be said in the random BSODs on windows one time it's a network driver, next time it's a file system driver, next time...

smp is tied to multiple processors and sometimes tied to apic...so don't know what is going on there.

Maybe it's having a problem managing the swapping between processors for some reason.

Changing the power supply might be the next thing to try.  Pulled just about all the hardware out and still a problem.

Last edited by nomorewindows (2019-08-24 05:53:31)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#6 2019-08-24 17:30:05

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

It locked on a PXE boot, so scratch that.  And it was a rt kernel, not stock, so kernel is out (most likely and because it crashes under windows just the same).

Didn't get any messages this time (on the wrong tty).

Memory test would lock up in the same place if I used smp mode, but would run in single processor only and pass the memory.

Last edited by nomorewindows (2019-08-24 17:43:29)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#7 2019-09-05 19:49:01

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

It probably wouldn't help to give some help to show the BSODs (from bluescreenview) since they are so random, as Linux seems to give at least a consistent oops messages.

I thought the machine was running for a few days without problem but in fact it rebooted into Windows and stayed at the login screen and then when I finally log in it tells me I BSODed.  Oh wait...5 BSODs in a single day...whoa.

Last edited by nomorewindows (2019-09-05 19:50:44)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#8 2019-09-18 17:15:45

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

Running an older kernel I'm getting some

NOHZ: local_softirq_pending 80

Wonder if this is the same thing?  Instead of a kernel panic/oops I'm just getting these instead.

But sometimes the machine runs for days without crashing on recent stock kernel.

While I ran pacman it would stop in the middle of downloading a package and then when it recovered, it printed above message.

Last edited by nomorewindows (2019-09-18 19:20:30)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#9 2019-09-18 20:46:39

seth
Member
Registered: 2012-09-03
Posts: 49,992

Re: Kernel Panic Oops 11 smp_timer

Context? But from your OP it looks related.
This is what NOHZ is and how you can deactivate it: https://stackoverflow.com/questions/977 … nux-kernel

Might be CPU powerstate related => lscpu?

Offline

#10 2019-09-20 19:24:30

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

seth wrote:

Context? But from your OP it looks related.
This is what NOHZ is and how you can deactivate it: https://stackoverflow.com/questions/977 … nux-kernel

Might be CPU powerstate related => lscpu?

Maybe something similar: https://bbs.archlinux.org/viewtopic.php?id=143460

The old kernel I'm running right now is almost as old as linux-lts.

lscpu

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              2
On-line CPU(s) list: 0,1
Thread(s) per core:  1
Core(s) per socket:  2
Socket(s):           1
NUMA node(s):        1
Vendor ID:           GenuineIntel
CPU family:          15
Model:               6
Model name:          Intel(R) Pentium(R) D CPU 2.80GHz
Stepping:            4
CPU MHz:             2400.000
CPU max MHz:         2800.0000
CPU min MHz:         2400.0000
BogoMIPS:            5603.15
L1d cache:           16K
L2 cache:            2048K
NUMA node0 CPU(s):   0,1
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc pebs bts nopl cpuid pni dtes64 monitor ds_cpl est cid cx16 xtpr pdcm lahf_lm pti

Maybe I need to get to a more recent kernel for spectre and meltdown status.

The fact that it crashes on Windows also seems to point to some hardware problem (memory/disk/power supply).

Last edited by nomorewindows (2019-09-20 19:41:42)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#11 2019-09-20 21:31:42

seth
Member
Registered: 2012-09-03
Posts: 49,992

Re: Kernel Panic Oops 11 smp_timer

That's rather the skylake generation, isn't?
Anyway: https://wiki.archlinux.org/index.php/In … ete_freeze

Offline

#12 2019-09-21 05:54:32

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

seth wrote:

That's rather the skylake generation, isn't?
Anyway: https://wiki.archlinux.org/index.php/In … ete_freeze

It's 10 years older, it's a dual core pentium 4 probably some of the first dual core processors.  Baytrail is a graphics board?

NOHZ is the tickless timer, but the oops is in smp_timer, hrtimer, apic so it might be the same timer that is switching between cpus?  But it seems to have been done away with in the more recent kernels. 

Seems there used to be a tool called symoops that would decipher oops but still how do you capture them...don't remember much about using symoops, probably never used it.  But apparently it's not available anymore.

Maybe if I can somehow figure out what the irqs are doing.  Hmmm...there are some irq tuning tools in the repos.  IRQbalance, tuna, rtirq, but of course that wouldn't help whatever goes on in windows, and may still oops in linux.


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#13 2019-09-21 07:12:34

seth
Member
Registered: 2012-09-03
Posts: 49,992

Re: Kernel Panic Oops 11 smp_timer

Oh, that's a netburst CPU… no, they don't have PM states ;-)
Baytrail is a codename for a CPU generation.

You're also getting BSODs on windows, so given the HW age, we might just be dealing w/ a physical issue. Did you try some like *really* old kernel (live Ubuntu pre-2010 or so)?

Offline

#14 2019-09-22 05:38:34

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

I found another pentium d machine, and I put what dmidecode said the memory would take.  But I power it up and it says bad memory.  So I run it with the memory that came out of the problem box.  It complains about bad memory.  I just randomly pull one of the chips and make an exact spec replacement.  It fires up, The problem box said it would take 4G, but it only takes 3G.  Possible the chipset is the same and this other box only takes 3G too.  It seems the memory chip maybe bad. Have to play around with it some more.  Possibly IRQ problems related to bad memory, but the tests never come up bad. 

Another machine only fails if I do the memory check in SMP mode.  Pretty sure the one chip is bad on it, it locks up and beeps indicating a memory problem, even after running awhile.  Once it hits that hole it stops.

Last edited by nomorewindows (2019-09-22 05:38:53)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#15 2019-09-22 16:58:41

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

I kept trying with extra memory I had and finally got the machine to take the full 4G of ram.  It threw some fits with the memory out of the older box and sometimes weird what it did take.  Makes me wonder if I was having trouble with the memory all along and now it was just really having a fit.  Not sure though.


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#16 2019-09-24 17:59:37

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

So right now both boxes have 4x1GB sticks that BIOS accepts, and both of them are showing only 3.5GB available even in 64-bit arch.  3.5GB is all that would be available with 32-bit, but also with 64-bit.

But I know before I tried putting all 4GB in the troublesome box and it had stability issues in both windows and linux.

So there might have been something wrong with one of the memory sticks some time ago, but I don't remember which ones they were.  But I know that instead of all 4G, I was running 3G, because I wasn't even getting 3.5GB to work without crashing.


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#17 2019-09-28 18:59:35

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

Had a oops on the second machine while doing a disk operation involving a NTFS partition from the first machine.  May have been IRQ maybe not. 

Didn't get any screenshot as I was in GUI Desktop. 

So maybe it's something with the disk or just ... not sure.

So now both machines have oopsed even when booting PXE/diskless.

Last edited by nomorewindows (2019-09-28 19:02:15)


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

#18 2019-10-14 03:22:19

nomorewindows
Member
Registered: 2010-04-03
Posts: 3,362

Re: Kernel Panic Oops 11 smp_timer

I put Windows 7 back together and it kept bsoding.  Then I put Windows XP back on, and it's sleeping (ACPI) just fine now. 
It's going to be a video card, or a network card, that I took out.
And older Linux kernel went for days without any noise.
Was beginning to think it was hard drive possibly also.


I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.

Offline

Board footer

Powered by FluxBB