Re: [Linux-usb-users] [REGRESSION] 2.6.23-rc1: uhci_hcd. irq 4:nobody cared

From: Alan Stern
Date: Sun Jul 29 2007 - 18:51:20 EST


On Sun, 29 Jul 2007, Mark Hindley wrote:

> On Sun, Jul 29, 2007 at 11:19:47AM -0400, Alan Stern wrote:
> > On Sun, 29 Jul 2007, Mark Hindley wrote:
> >
> > > Hi,
> > >
> > > I have just tried 2.6.23-rc1 on my Acer Aspire 1350.
> > >
> > > On boot I get the following error as the uhci_hcd module is loaded:
> > >
> > > Jul 28 18:23:20 mercury kernel: ACPI: PCI Interrupt 0000:00:10.0[A] -> Link [LNKA] -> GSI 4 (level, low) -> IRQ 4
> > > Jul 28 18:23:20 mercury kernel: uhci_hcd 0000:00:10.0: UHCI Host Controller
> > > Jul 28 18:23:20 mercury kernel: uhci_hcd 0000:00:10.0: new USB bus registered, assigned bus number 2
> > > Jul 28 18:23:20 mercury kernel: irq 4: nobody cared (try booting with the "irqpoll" option)
> >
> > Did it work okay with older kernels? What does /proc/interrupts say in
> > both 2.6.23-rc1 and in a working kernel?
>
> No boot error with 2.6.22.1.
>
> On 2.6.22.1:
> CPU0
> 0: 12312 XT-PIC-XT timer
> 1: 286 XT-PIC-XT i8042
> 2: 0 XT-PIC-XT cascade
> 4: 434 XT-PIC-XT uhci_hcd:usb2, via@pci:0000:01:00.0
> 5: 2000 XT-PIC-XT yenta, uhci_hcd:usb3, wifi0
> 6: 3 XT-PIC-XT floppy
> 7: 3 XT-PIC-XT parport0
> 8: 4 XT-PIC-XT rtc
> 9: 0 XT-PIC-XT uhci_hcd:usb4, VIA82XX-MODEM, VIA8233
> 10: 843 XT-PIC-XT acpi
> 11: 0 XT-PIC-XT ehci_hcd:usb1
> 12: 123 XT-PIC-XT i8042
> 14: 10951 XT-PIC-XT ide0
> 15: 53 XT-PIC-XT ide1
> NMI: 770
> LOC: 120091
> ERR: 0
> MIS: 0
>
> On 2.6.23-rc1:
> CPU0
> 0: 8616 XT-PIC-XT timer
> 1: 183 XT-PIC-XT i8042
> 2: 0 XT-PIC-XT cascade
> 4: 8233 XT-PIC-XT uhci_hcd:usb1, via@pci:0000:01:00.0
> 5: 4948 XT-PIC-XT uhci_hcd:usb2, yenta, wifi0
> 6: 3 XT-PIC-XT floppy
> 7: 22 XT-PIC-XT parport0
> 8: 4 XT-PIC-XT rtc
> 9: 0 XT-PIC-XT uhci_hcd:usb3, VIA82XX-MODEM, VIA8233
> 10: 854 XT-PIC-XT acpi
> 11: 0 XT-PIC-XT ehci_hcd:usb4
> 12: 123 XT-PIC-XT i8042
> 14: 12202 XT-PIC-XT ide0
> 15: 53 XT-PIC-XT ide1
> NMI: 809
> LOC: 150284
> ERR: 1
> MIS: 0

No significant differences. :-(

Well, about all I can suggest is bisection. And don't assume that this
problem was caused by a change to the USB stack. It could be a change
in the interrupt-handling core or a change to the driver for one of the
other devices sharing the IRQ.

It might even be caused by a change in the order the modules are
loaded, which has nothing to do with the kernel directly. In fact, it
might be worth checking this possibility first. For example, it's
clear that 2.6.22 loaded ehci-hcd before uhci-hcd whereas 2.6.23-rc1
did the reverse. And yenta vs. uhci-hcd is different.

Alan Stern

-
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/