I have a dual P3 system (via motherboard) with 1.5 gigs of RAM (with
highmem enabled in the kernel). Under heavy networking load, I get the
following error:
irq 26: nobody cared!
[<c01085aa>] __report_bad_irq+0x2a/0x90
[<c01086a0>] note_interrupt+0x70/0xb0
[<c01089e0>] do_IRQ+0x160/0x1a0
[<c0106a48>] common_interrupt+0x18/0x20
[<c0103e50>] default_idle+0x0/0x40
[<c0103e7c>] default_idle+0x2c/0x40
[<c0103f0b>] cpu_idle+0x3b/0x50
[<c040a920>] start_kernel+0x1c0/0x210
[<c040a370>] unknown_bootoption+0x0/0x160
handlers:
[<f8a25af0>] (usb_hcd_irq+0x0/0x70 [usbcore])
[<f8a25af0>] (usb_hcd_irq+0x0/0x70 [usbcore])
My USB2 devices become unusable.
If I do try and use /dev/scd0 (my usb2 DVD+RW, I get the following:
irq 18: nobody cared!
[<c01085aa>] __report_bad_irq+0x2a/0x90
[<c01086a0>] note_interrupt+0x70/0xb0
[<c01089e0>] do_IRQ+0x160/0x1a0
[<c0106a48>] common_interrupt+0x18/0x20
[<c0103e50>] default_idle+0x0/0x40
[<c0103e7c>] default_idle+0x2c/0x40
[<c0103f0b>] cpu_idle+0x3b/0x50
[<c040a920>] start_kernel+0x1c0/0x210
[<c040a370>] unknown_bootoption+0x0/0x160
handlers:
[<f8a25af0>] (usb_hcd_irq+0x0/0x70 [usbcore])
[<f9059b50>] (bttv_irq+0x0/0x380 [bttv])
At that point *none* of my USB devices are usable.
This is my /proc/interrupts
0: 879327 19 IO-APIC-edge timer
1: 6122 1 IO-APIC-edge i8042
2: 0 0 XT-PIC cascade
8: 0 1 IO-APIC-edge rtc
12: 32881 1 IO-APIC-edge i8042
14: 32788 0 IO-APIC-edge ide0
15: 22 0 IO-APIC-edge ide1
16: 67537 1 IO-APIC-level ohci_hcd, radeon@pci:0000:01:00.0
17: 192092 258917 IO-APIC-level ohci_hcd, Ensoniq AudioPCI, eth0
18: 295 1 IO-APIC-level ehci_hcd, bttv0
19: 4931 1 IO-APIC-level EMU10K1
26: 200000 0 IO-APIC-level uhci_hcd, uhci_hcd
And my lsmod:
I'm attaching the full dmesg output and thelsmod output. Please let me
know what I can do to get this working (and please CC me on the follow
ups).
Adam
On Mer, 2004-09-08 at 22:35, Adam K Kirchhoff wrote:
> I have a dual P3 system (via motherboard) with 1.5 gigs of RAM (with
> highmem enabled in the kernel). Under heavy networking load, I get the
> following error:
Try variously turning off acpi and the apic. If the routing tables are
still shot try the irqfixup patch I posted, it might well rescue your
box.
Alan Cox wrote:
>On Mer, 2004-09-08 at 22:35, Adam K Kirchhoff wrote:
>
>
>>I have a dual P3 system (via motherboard) with 1.5 gigs of RAM (with
>>highmem enabled in the kernel). Under heavy networking load, I get the
>>following error:
>>
>>
>
>Try variously turning off acpi and the apic. If the routing tables are
>still shot try the irqfixup patch I posted, it might well rescue your
>box.
>
>
>
Booting with 'noapic' seems to have solved the problem. At least my
attempts to replicate it so far (copying a large directory over nfs)
hasn't produced the same error.
Thanks for the help, Alan!
Adam
Adam K Kirchhoff wrote:
> Alan Cox wrote:
>
>> On Mer, 2004-09-08 at 22:35, Adam K Kirchhoff wrote:
>>
>>
>>> I have a dual P3 system (via motherboard) with 1.5 gigs of RAM (with
>>> highmem enabled in the kernel). Under heavy networking load, I get the
>>> following error:
>>>
>>
>>
>> Try variously turning off acpi and the apic. If the routing tables are
>> still shot try the irqfixup patch I posted, it might well rescue your
>> box.
>>
>>
>>
>
> Booting with 'noapic' seems to have solved the problem. At least my
> attempts to replicate it so far (copying a large directory over nfs)
> hasn't produced the same error.
>
> Thanks for the help, Alan!
>
> Adam
>
>
Actually, it looks like *enabling* ACPI (which had previously been
disabled) and enabling the APIC gets it working properly as well.
Adam