Re: IRQ: Nobody cared (2.6.19.1)

From: Chris Wilson
Date: Sat Jan 20 2007 - 10:10:22 EST


Hi all,

On Sat, 6 Jan 2007, Chris Wilson wrote:

Forwarded to lkml as suggested by Alan Stern. Please copy any replies to me, as I'm not on the list (too much traffic, sorry!).

Ping? It's been two weeks.

Cheers, Chris.

On Fri, 5 Jan 2007, Alan Stern wrote:
On Sat, 6 Jan 2007, Chris Wilson wrote:
> > I keep getting the following errors:
> > Jan 5 23:48:38 gcc kernel: irq 10: nobody cared (try booting with the
> "irqpoll" option)

> Jan 5 23:48:38 gcc kernel: handlers:

> Jan 5 23:48:38 gcc kernel: [<e0866a00>] (usb_hcd_irq+0x0/0x70 > [usbcore])
> Jan 5 23:48:38 gcc kernel: Disabling IRQ #10
> > There are no devices attached to that USB port, and it's the only device
> registered for IRQ 10.
> > This is a 2.6.19.1 kernel, last booted less than an hour ago. I had the
> same problem with 2.6.14.3 and older kernels, but less frequently.
> > Hardware is dual p3 coppermine, Gigabyte 6VXDC7 motherboard. Otherwise
> very stable, last up for 297 days (until I booted this kernel).

> /proc/interrupts:
> > CPU0 CPU1
> 0: 424892 412866 IO-APIC-edge timer
> 1: 2706 2034 IO-APIC-edge i8042
> 4: 5 1 IO-APIC-edge serial
> 5: 0 0 IO-APIC-fasteoi acpi
> 6: 5 0 IO-APIC-edge floppy
> 7: 0 0 IO-APIC-edge parport0
> 10: 75964 63749 IO-APIC-fasteoi uhci_hcd:usb1,
> uhci_hcd:usb2
> 12: 38217 29601 IO-APIC-edge i8042
> 14: 24424 14372 IO-APIC-edge ide0
> 15: 1 10 IO-APIC-edge ide1
> 16: 44129 1 IO-APIC-fasteoi eth0
> 17: 35 209490 IO-APIC-fasteoi eth1
> 18: 49348 50382 IO-APIC-fasteoi EMU10K1
> NMI: 0 0
> LOC: 837636 837635
> ERR: 0
> MIS: 0
> > Please let me know if I can provide any more information that might > help,
> or anything I can do to help fix this. I expect that the USB port is now
> useless until I reload the module.

This almost certainly is not caused by a problem in the USB hardware.
More likely some other device is using IRQ 10 and the kernel doesn't
realize it. In other words, it's a problem in IRQ assignment.

You can try booting with acpi=off on the boot command line, or acpi=noirq,
or noapic.

You can go ahead and report this on LKML; you don't have to subscribe to
the list in order to post on it. (That's what I do.) Include the dmesg
log showing the IRQ assignments during boot-up.

Alan Stern

Dmesg boot log attached. Any suggestions gratefully received.

It seems a bit drastic to disable a whole IRQ if it receives spurious interrupts that are not claimed by any driver. That could kill a machine if the IRQ is used for something critical like disks.

I'd rather not boot without ACPI if possible, as I don't want to lose power saving. I'm not sure about the negative consequences of booting with acpi=noirq or noapic, so I haven't tried that yet.

Cheers, Chris.
--
_ ___ __ _
/ __/ / ,__(_)_ | Chris Wilson <0000 at qwirx.com> - Cambs UK |
/ (_/ ,\/ _/ /_ \ | Security/C/C++/Java/Perl/SQL/HTML Developer |
\ _/_/_/_//_/___/ | We are GNU-free your mind-and your software |

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/