View previous topic :: View next topic |
Author |
Message |
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Sun Aug 26, 2018 4:47 am Post subject: IRQ disabled? |
|
|
I was working with certificates in a console window in Plasma when this popped up in the console window.
Code: |
Message from syslogd@9y84mj1 at Fri Aug 24 15:38:12 2018 ...
9y84mj1 kernel: [17131.992007] Disabling IRQ #16
|
Now the entire system is running SLOWLY. I can out-type the cursor in this window, for example. What is going on here? I never saw this before. I know a reboot can fix it, but why did it happen?
*EDIT*
It looks like IRQ16 is my video card. What the heck?
Code: |
CPU0 CPU1
0: 23679984 6277546 IO-APIC 2-edge timer
1: 17181 6307 IO-APIC 1-edge i8042
8: 7 4 IO-APIC 8-edge rtc0
9: 8958 701 IO-APIC 9-fasteoi acpi
12: 333471 169596 IO-APIC 12-edge i8042
16: 71854699 70645301 IO-APIC 16-fasteoi i915
17: 9 1 IO-APIC 17-fasteoi firewire_ohci
18: 0 0 IO-APIC 18-fasteoi mmc0, mmc1
19: 0 0 IO-APIC 19-fasteoi i801_smbus
20: 0 0 IO-APIC 20-fasteoi ehci_hcd:usb2, uhci_hcd:usb3, uhci_hcd:usb6
21: 0 0 IO-APIC 21-fasteoi uhci_hcd:usb4, uhci_hcd:usb7
22: 120 17 IO-APIC 22-fasteoi ehci_hcd:usb1, uhci_hcd:usb5, uhci_hcd:usb8
24: 0 0 DMAR-MSI 0-edge dmar0
25: 0 0 DMAR-MSI 2-edge dmar2
26: 476257 77552 PCI-MSI 512000-edge ahci[0000:00:1f.2]
27: 45 47 PCI-MSI 442368-edge snd_hda_intel:card0
28: 671304 317520 PCI-MSI 6291456-edge iwlwifi
29: 0 221 PCI-MSI 409600-edge enp0s25
NMI: 0 0 Non-maskable interrupts
LOC: 6536207 20697960 Local timer interrupts
SPU: 0 0 Spurious interrupts
PMI: 0 0 Performance monitoring interrupts
IWI: 1502354 1516057 IRQ work interrupts
RTR: 0 0 APIC ICR read retries
RES: 1128235 1410100 Rescheduling interrupts
CAL: 36028 35893 Function call interrupts
TLB: 34799 34727 TLB shootdowns
TRM: 0 0 Thermal event interrupts
THR: 0 0 Threshold APIC interrupts
MCE: 0 0 Machine check exceptions
MCP: 58 57 Machine check polls
ERR: 0
MIS: 0
PIN: 0 0 Posted-interrupt notification event
NPI: 0 0 Nested posted-interrupt event
PIW: 0 0 Posted-interrupt wakeup event
|
_________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Mon Sep 03, 2018 2:45 am Post subject: |
|
|
I am still not sure what caused this. It may have been due to a weak charge (battery getting low), but I verified that IRQ16 is indeed, the video card. When it disables, framerates drop into the sewer. Is there a way to tell the kernel to not disable my video card even when on low power? _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
Jaglover Watchman
Joined: 29 May 2005 Posts: 8291 Location: Saint Amant, Acadiana
|
|
Back to top |
|
|
P.Kosunen Guru
Joined: 21 Nov 2005 Posts: 309 Location: Finland
|
Posted: Mon Sep 03, 2018 11:18 am Post subject: |
|
|
Do you have "options i2c-i801 disable_features=0x10" set under /etc/modprobe.d/?
Code: | $ dmesg | grep -i smbus
[ 1.476403] i801_smbus 0000:00:1f.4: Interrupt disabled by user
[ 1.476544] i801_smbus 0000:00:1f.4: SMBus using polling |
|
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Mon Sep 03, 2018 10:05 pm Post subject: |
|
|
Unless it is set by default, I do not have it set. I have never had to mess with IRQ settings before.
Jag, I believe it is the kernel due to the message in my first post. I'm chilling there and then "9y84mj1 kernel: [17131.992007] Disabling IRQ #16" pops up. You know, it even says "kernel" sent the message, so if the kernel did not disable the IRQ as the message states, then I am thoroughly confused! _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
Jaglover Watchman
Joined: 29 May 2005 Posts: 8291 Location: Saint Amant, Acadiana
|
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Wed Sep 05, 2018 3:44 pm Post subject: |
|
|
Well, prior to using Debian and after that, Gentoo on this system, Windows never had the issue. This is indeed the first time it happened even in Linux, but only after a recent update. I imagine something changed to conserve power and it doesn't like doing that on my system. Either way, how do I troubleshoot or resolve this issue? _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
Jaglover Watchman
Joined: 29 May 2005 Posts: 8291 Location: Saint Amant, Acadiana
|
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Thu Sep 06, 2018 3:32 pm Post subject: |
|
|
This isn't hardware, I am 99.9% sure of that. I don't want to change my kernel for this, I just want to figure out what does it. If I need a debug kernel and all that, it isn't worth messing with. I actually use this system. Quite heavily, in fact.
I did just do a new search with a few words changed and apparently this is an old issue. It's all over the place. Doubt it is my hardware due to this. Will search more when time permits. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
tholin Apprentice
Joined: 04 Oct 2008 Posts: 204
|
Posted: Thu Sep 06, 2018 5:17 pm Post subject: |
|
|
The_Great_Sephiroth wrote: | I did just do a new search with a few words changed and apparently this is an old issue. It's all over the place. Doubt it is my hardware due to this. Will search more when time permits. |
Most of those problems look like the same thing. Some device is broken and have a stuck interrupt. The kernel will notice this ("nobody cared") and disable interrupts for that device and instead poll it a few times per second. There was an old Asmedia PCIe-PCI bridge that was infamous for having this problem. |
|
Back to top |
|
|
CooSee Veteran
Joined: 20 Nov 2004 Posts: 1468 Location: Earth
|
Posted: Thu Sep 06, 2018 5:18 pm Post subject: |
|
|
did you try with any other live-cd, if this still exist ? _________________ " Die Realität ist eine Illusion, die durch Mangel an ehrlicher Kommunikation entsteht "
---
" Der Mensch ist von Natur aus neugierig, was am Ende übrig bleibt ist die Gier " |
|
Back to top |
|
|
Ant P. Watchman
Joined: 18 Apr 2009 Posts: 6920
|
Posted: Fri Sep 07, 2018 5:56 am Post subject: |
|
|
This isn't hardware failure. The intel graphics driver isn't great, it's been this way for a long time. One of the workarounds they added many years ago is a hang detector that forces it into dumb framebuffer mode instead of crashing entirely, that's what's happening in this case. No idea what the cause is, but rebooting will fix it. Or maybe power-cycling the hardware - try suspending?
Sometimes the cure is worse than the disease, though; this used to be a daily occurrence on my netbook around 2011-12, but I barely noticed when it happened because I run Openbox and the GPU was already slow. Running KDE's compositor without 3D acceleration is going to sting. |
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Sat Sep 08, 2018 11:18 pm Post subject: |
|
|
I believe Ant hit the nail on the head. This was indeed my graphics card. After I got hibernation working I was hibernating it or suspending it all week between client locations. Basically, I'd suspend it, drive to wherever, wake it up, work, and repeat. At the end of the day I'd suspend it so I could pick up where I left off Monday. I started it up one Monday and after about ten minutes it did this. My guess is that the driver had been "up" for too long and screwed up. Will try this again to see if it is reproducible. If it is, I can narrow it down to the driver not liking to be loaded all week. It has not done this since, but I am also back to shutting down every day. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Sat Sep 15, 2018 2:13 am Post subject: |
|
|
Just following up here. It happened AGAIN. I was in Plasma and I only had VirtualBox open (DOS 6.22 w/WFW3.11) and Chromium. Suddenly DOS got slow (DOS was NEVER slow) and when I went back to Plasma, slow as heck. I shutdown the VM and rebooted and when the desktop disappeared and it went to the shell, I saw it had disabled IRQ #16 again.
This is most likely a kernel issue at this point. I did NOT have this issue in 4.14.52, but it is rampant in 4.14.65. I imported the old configuration with "make oldconfig" as always, so it is an identical configuration. Not sure what it is, but something in the new kernel is flaky. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
tholin Apprentice
Joined: 04 Oct 2008 Posts: 204
|
Posted: Sat Sep 15, 2018 8:33 am Post subject: |
|
|
The_Great_Sephiroth wrote: | I did NOT have this issue in 4.14.52, but it is rampant in 4.14.65. |
It should be easy to git-bisect then. My bet would be on this patch in 4.14.58.
https://patchwork.freedesktop.org/patch/229559/
It changes how the kernel handles interrupts of hotplug events (hotplugging of monitors). That patch only apply to some older IGPUs. What hardware do you have anyway? |
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Fri Sep 21, 2018 2:06 am Post subject: |
|
|
It is an older Dell latitude E6400 system off the top of my head. Core2Duo, 8GB RAM, 1TB SATA2 disk, DVD-RAM drive, smart-chip reader, the whole nine yards. Next time I use it I will post the specifics. I use the i965 driver. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
tholin Apprentice
Joined: 04 Oct 2008 Posts: 204
|
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Fri Sep 21, 2018 8:59 pm Post subject: |
|
|
I'm not hot-plugging though. I rarely use a second monitor these days and I rely on the one on the laptop itself 99% of the time. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
Maitreya Guru
Joined: 11 Jan 2006 Posts: 445
|
Posted: Sat Sep 22, 2018 3:14 pm Post subject: |
|
|
The_Great_Sephiroth wrote: | I'm not hot-plugging though. I rarely use a second monitor these days and I rely on the one on the laptop itself 99% of the time. |
I think some close/open lid actions are implemented as (un)plug |
|
Back to top |
|
|
The_Great_Sephiroth Veteran
Joined: 03 Oct 2014 Posts: 1602 Location: Fayetteville, NC, USA
|
Posted: Sun Sep 23, 2018 5:06 pm Post subject: |
|
|
That could possibly be it, except the most recent occurrence happened without any lid action going on. It simply disabled the IRQ while I was working with a virtual machine. _________________ Ever picture systemd as what runs "The Borg"? |
|
Back to top |
|
|
|