Bug 802085 - [abrt] kernel: irq 18: nobody cared (try booting with the "irqpoll" option)
[abrt] kernel: irq 18: nobody cared (try booting with the "irqpoll" option)
Status: CLOSED ERRATA
Product: Fedora
Classification: Fedora
Component: kernel (Show other bugs)
16
x86_64 Unspecified
unspecified Severity unspecified
: ---
: ---
Assigned To: Kernel Maintainer List
Fedora Extras Quality Assurance
abrt_hash:ca3961d2e0102b5f26f5af1935c...
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2012-03-10 18:12 EST by Tommi R.
Modified: 2012-07-12 11:46 EDT (History)
6 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2012-07-12 11:46:20 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)
File: smolt_data (5.09 KB, text/plain)
2012-03-10 18:12 EST, Tommi R.
no flags Details
/proc/interrupts (1.68 KB, text/plain)
2012-03-14 11:02 EDT, Tommi R.
no flags Details
lspci -vvv output (32.06 KB, text/plain)
2012-03-14 11:07 EDT, Tommi R.
no flags Details

  None (edit)
Description Tommi R. 2012-03-10 18:12:00 EST
libreport version: 2.0.8
abrt_version:   2.0.7
cmdline:        BOOT_IMAGE=/vmlinuz-3.2.9-1.fc16.x86_64 root=UUID=b7e0b37a-64b6-43c5-91f6-a64e8a97e48b ro rd.md=0 rd.lvm=0 rd.dm=0 quiet SYSFONT=latarcyrheb-sun16 rhgb KEYTABLE=fi rd.luks=0 LANG=en_US.UTF-8
comment:        This happens with my new Asus E45M1-M Pro mother board (CPU on board), one PCI slot has network interface card (using driver 8139too) and  Interrupt: pin A routed to IRQ 18. Usually this will happen 5-15 times / day (24 hours). But sometimes there can be more than 24 hours without this. I tried with suggested irqpoll boot parameter, but it seems to not help much.  I also tried different network interface card, but no luck. Also using the other PCI slot does not help, but then problematic IRQ is 19.
event_log:      2012-03-11-01:04:19> Smolt profile successfully saved
kernel:         3.2.9-1.fc16.x86_64
reason:         irq 18: nobody cared (try booting with the "irqpoll" option)
time:           Sat 10 Mar 2012 11:12:25 PM EET

smolt_data:     Text file, 5210 bytes

backtrace:
:irq 18: nobody cared (try booting with the "irqpoll" option)
:Pid: 0, comm: swapper/0 Not tainted 3.2.9-1.fc16.x86_64 #1
:Call Trace:
: <IRQ>  [<ffffffff810e11ad>] __report_bad_irq+0x3d/0xe0
: [<ffffffff810e146d>] note_interrupt+0x16d/0x220
: [<ffffffff810dec39>] handle_irq_event_percpu+0xa9/0x220
: [<ffffffff8101b953>] ? native_sched_clock+0x13/0x80
: [<ffffffff810dedf4>] handle_irq_event+0x44/0x70
: [<ffffffff810e1edf>] handle_fasteoi_irq+0x5f/0xf0
: [<ffffffff81016226>] handle_irq+0x46/0xb0
: [<ffffffff815ed5da>] do_IRQ+0x5a/0xe0
: [<ffffffff815e2f2e>] common_interrupt+0x6e/0x6e
: <EOI>  [<ffffffff81338e8b>] ? arch_local_irq_enable+0x8/0xd
: [<ffffffff81096e7a>] ? sched_clock_idle_wakeup_event+0x1a/0x20
: [<ffffffff81339bd2>] acpi_idle_enter_simple+0xd3/0x115
: [<ffffffff81493671>] cpuidle_idle_call+0xc1/0x280
: [<ffffffff8101322a>] cpu_idle+0xca/0x120
: [<ffffffff815bffce>] rest_init+0x72/0x74
: [<ffffffff81aebbfe>] start_kernel+0x3ba/0x3c5
: [<ffffffff81aeb347>] x86_64_start_reservations+0x132/0x136
: [<ffffffff81aeb140>] ? early_idt_handlers+0x140/0x140
: [<ffffffff81aeb44d>] x86_64_start_kernel+0x102/0x111
Comment 1 Tommi R. 2012-03-10 18:12:05 EST
Created attachment 569137 [details]
File: smolt_data
Comment 2 Tommi R. 2012-03-14 10:28:23 EDT
More information to this: after updating to newer kernel:
Linux anjo 3.2.9-2.fc16.x86_64 #1 SMP Mon Mar 5 20:55:39 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux
this behavior went away.
But there are very often messages in /var/log/messages like:
Mar 14 16:05:59 anjo kernel: [88737.586287] Disabling IRQ 18
Mar 14 16:05:59 anjo kernel: [88737.596030] Polling IRQ 18
Mar 14 16:06:00 anjo kernel: [88738.596262] Reenabling IRQ 18

The period how often this happens is now typically less than 1 minute!
And disabling IRQ message is emergency message, as default to wall everybody, so I needed to change this behavior to get emergency messages to go only one file.

However, everything seems to work, even also the external network, and interrupt of this network interface card pin A is routed to IRQ 18!

MB has the latest bios, I updated it already about 2 weeks ago from original, but not seeing any difference.
Comment 3 Josh Boyer 2012-03-14 10:40:27 EDT
(In reply to comment #2)
> More information to this: after updating to newer kernel:
> Linux anjo 3.2.9-2.fc16.x86_64 #1 SMP Mon Mar 5 20:55:39 UTC 2012 x86_64 x86_64
> x86_64 GNU/Linux
> this behavior went away.
> But there are very often messages in /var/log/messages like:
> Mar 14 16:05:59 anjo kernel: [88737.586287] Disabling IRQ 18
> Mar 14 16:05:59 anjo kernel: [88737.596030] Polling IRQ 18
> Mar 14 16:06:00 anjo kernel: [88738.596262] Reenabling IRQ 18

These messages will be gone in the kernel that is in updates-testing.
Comment 4 Tommi R. 2012-03-14 11:02:51 EDT
Created attachment 570007 [details]
/proc/interrupts
Comment 5 Tommi R. 2012-03-14 11:07:44 EDT
Created attachment 570010 [details]
lspci -vvv output

Right now I can not boot computer and test the later kernel from testing reposity, but I can attach these two files if they can be helpful to investigate this. I'm not sure if this is duplicate of other bug, I saw similar ones.
Comment 6 Dave Jones 2012-03-22 13:16:42 EDT
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.
Comment 7 Dave Jones 2012-03-22 13:18:35 EDT
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.
Comment 8 Dave Jones 2012-03-22 13:26:59 EDT
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.
Comment 9 Paweł Brodacki 2012-07-08 06:51:51 EDT
Hello,

It seems that I'm hitting the same bug running F17 

Linux hex 3.4.4-3.fc17.x86_64 #1 SMP Tue Jun 26 20:54:56 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux

I have a Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06) in one of PCI slots of Asus E45M1-M PRO motherboard and see plentiful 
IRQ 19 might be stuck.  Polling
messages in dmesg output. Also the card seems to be periodically loosing link. I access the machine remotely and after I get disconnected and re-connect I see 
8139too 0000:06:01.0: p5p1: link up, 100Mbps, full-duplex, lpa 0x41E1
message in dmesg.

This bug is tagged with NEEDINFO, but I do not see any information requests. What can I do to help you to squash this bug?
Comment 10 Dave Jones 2012-07-12 11:41:45 EDT
Pawel, That sounds like a different bug. Please file a separate report for it.
Comment 11 Dave Jones 2012-07-12 11:46:20 EDT
If you can still reproduce this in 3.4, please reopen. We believe this should be fixed with the current updates.

Note You need to log in before you can comment on or make changes to this bug.