My machine won't boot with anything plugged into the USB ports.

This is actually the /var/log/kern.log from booting with 2.6.21 
yesterday, but I got exactly the same booting up 2.6.22 with the USB 
headphones plugged in. This doesn't happen with the vanilla 2.6.18 from 
etch.

Aug  9 11:44:37 localhost kernel:  =======================
Aug  9 11:44:37 localhost kernel: handlers:
Aug  9 11:44:37 localhost kernel: [<c01f2046>] (acpi_irq+0x0/0x14)
Aug  9 11:44:37 localhost kernel: irq 9: nobody cared (try booting with 
the "irqpoll" option)
Aug  9 11:44:37 localhost kernel:  [<c01426d6>] __report_bad_irq+0x36/0x75
Aug  9 11:44:37 localhost kernel:  [<c01428a0>] note_interrupt+0x18b/0x1bb
Aug  9 11:44:37 localhost kernel:  [<c0141e5d>] thread_simple_irq+0x53/0x5f
Aug  9 11:44:37 localhost kernel:  [<c01424d2>] do_irqd+0xd4/0x282
Aug  9 11:44:37 localhost kernel:  [<c01423fe>] do_irqd+0x0/0x282
Aug  9 11:44:37 localhost kernel:  [<c012e48c>] kthread+0xb4/0xdc
Aug  9 11:44:37 localhost kernel:  [<c012e3d8>] kthread+0x0/0xdc
Aug  9 11:44:37 localhost kernel:  [<c0105037>] 
kernel_thread_helper+0x7/0x10
Aug  9 11:44:37 localhost kernel:  =======================

ad infinitum ... (at least until I take the USB device out)

This also happens similarly when I have attempted to boot with a USB 
wifi dongle in too.

I don't appear to be able to find any logs from this morning's failed 
boot apart from the less-than-useful /var/log/dmesg

I'll try rebooting with irqpoll option as suggested:
"""
The "irqpoll" option is a kernel arguement so it needs to be appended to 
the kernel line of grub. So it should look something like:

kernel /boot/vmlinux-2.6.22-8-generic root=UUID=stuffhere ro 
resume=UUID=stuffhere irqpoll
"""

 From reading it seems that acpi=off may also be worth trying as this 
could be a conflict between uhci-hcd and an ACPI interrupt. I have also 
read reports of noirqdebug being useful. However, irqpoll and some of 
the other options are known to do things like disable CD drives and mess 
up keyboards, so none of these suggested workarounds look much like fixes.

It appears to be a common problem with 2.6 series kernels on older 
hardware like mine.

$ cat /proc/interrupts
            CPU0
   0:    1534942    XT-PIC-XT        timer
   1:       7131    XT-PIC-XT        i8042
   2:          0    XT-PIC-XT        cascade
   6:          2    XT-PIC-XT        floppy
   7:          0    XT-PIC-XT        parport0
   8:          1    XT-PIC-XT        rtc
   9:          1    XT-PIC-XT        acpi
  10:          0    XT-PIC-XT        uhci_hcd:usb1
  11:     147762    XT-PIC-XT        FM801, eth0
  12:      57999    XT-PIC-XT        i8042
  14:      28973    XT-PIC-XT        ide0
  15:      59658    XT-PIC-XT        ide1
NMI:          0
LOC:          0
ERR:          0
MIS:          0

I'd be interested to hear from anyone else experiencing irq conflicts 
and lock-ups like this. It would be even cooler to fix it once and for 
all. Suggestions, of course, welcome. :)

cheers,

tim
/|\

_______________________________________________
64studio-devel mailing list
[email protected]
http://lists.64studio.com/mailman/listinfo/64studio-devel

Reply via email to