Problem with VIA Rhine II on eth0 - Suse

This is a discussion on Problem with VIA Rhine II on eth0 - Suse ; Every once in a while, for no apparent reason, my ethernet just stops. My /var/log/messages file shows: Jun 13 23:18:19 origin kernel: irq 209: nobody cared (try booting with the "irqpoll" option) Jun 13 23:18:19 origin kernel: [ ] __report_bad_irq+0x36/0x7d ...

+ Reply to Thread
Results 1 to 7 of 7

Thread: Problem with VIA Rhine II on eth0

  1. Problem with VIA Rhine II on eth0

    Every once in a while, for no apparent reason, my ethernet just stops.
    My /var/log/messages file shows:

    Jun 13 23:18:19 origin kernel: irq 209: nobody cared (try booting with
    the "irqpoll" option)
    Jun 13 23:18:19 origin kernel: [] __report_bad_irq+0x36/0x7d
    Jun 13 23:18:19 origin kernel: [] note_interrupt+0x1af/0x1eb
    Jun 13 23:18:19 origin kernel: [] handle_IRQ_event+0x23/0x49
    Jun 13 23:18:19 origin kernel: [] __do_IRQ+0xb9/0xee
    Jun 13 23:18:19 origin kernel: [] do_IRQ+0x71/0x83
    Jun 13 23:18:19 origin kernel: [] common_interrupt+0x1a/0x20
    Jun 13 23:18:19 origin kernel: []
    acpi_processor_idle+0x0/0x379 [processor]
    Jun 13 23:18:19 origin kernel: [] acpi_safe_halt+0x14/0x20
    [processor]
    Jun 13 23:18:19 origin kernel: []
    acpi_processor_idle+0x162/0x379 [processor]
    Jun 13 23:18:19 origin kernel: [] cpu_idle+0x9e/0xb8
    Jun 13 23:18:19 origin kernel: [] start_kernel+0x349/0x351
    Jun 13 23:18:19 origin kernel: [] unknown_bootoption+0x0/0x204
    Jun 13 23:18:19 origin kernel: handlers:
    Jun 13 23:18:19 origin kernel: [] (usb_hcd_irq+0x0/0x50 [usbcore])
    Jun 13 23:18:19 origin kernel: [] (rhine_interrupt+0x0/0x645
    [via_rhine])
    Jun 13 23:18:19 origin kernel: Disabling IRQ #209

    So two questions.

    1. What is the deal with the impossibly high IRQ number?
    2. What is the "irqpoll option," and how do I boot with it?

  2. Re: Problem with VIA Rhine II on eth0

    GateK33per wrote:

    > 1. What is the deal with the impossibly high IRQ number?


    It is APIC (!= ACPI ) the few years old feature that is used mainly for
    multiprocessor machines, it allows for up to 256 interrupts if computer
    hardware has support for it.

    > 2. What is the "irqpoll option," and how do I boot with it?


    If you have installed kernel sources than in:
    /usr/src/linux/Documentation
    file
    kernel-parameters.txt
    has all bout kernel paramateres.
    ....
    irqpoll [HW]
    When an interrupt is not handled search all handlers
    for it. Also check all handlers each timer
    interrupt. Intended to get systems with badly broken
    firmware running.
    ....
    Something that might help, but can make more troubles (not sure).
    On a first boot screen type
    irqpoll
    as test solution to see that other stuff will work.
    Than if everything works as before put that option in
    /boot/grub/menu.lst
    and see will that help.

    --
    Regards,
    Rajko.

  3. Re: Problem with VIA Rhine II on eth0

    Rajko M. wrote:
    > GateK33per wrote:
    >
    >> 1. What is the deal with the impossibly high IRQ number?

    >
    > It is APIC (!= ACPI ) the few years old feature that is used mainly for
    > multiprocessor machines, it allows for up to 256 interrupts if computer
    > hardware has support for it.
    >
    >> 2. What is the "irqpoll option," and how do I boot with it?

    >
    > If you have installed kernel sources than in:
    > /usr/src/linux/Documentation
    > file
    > kernel-parameters.txt
    > has all bout kernel paramateres.
    > ...
    > irqpoll [HW]
    > When an interrupt is not handled search all handlers
    > for it. Also check all handlers each timer
    > interrupt. Intended to get systems with badly broken
    > firmware running.
    > ...
    > Something that might help, but can make more troubles (not sure).
    > On a first boot screen type
    > irqpoll
    > as test solution to see that other stuff will work.
    > Than if everything works as before put that option in
    > /boot/grub/menu.lst
    > and see will that help.
    >

    I tried the irqpoll option, but it made things worse. The x-server hung
    at startup. Do you think there would be any point in disabling APIC? I
    do have an AMD 64 x2 processor, so technically this is a multi-processor
    machine.

  4. Re: Problem with VIA Rhine II on eth0

    GateKeeper wrote:

    > Rajko M. wrote:
    >> GateK33per wrote:
    >>
    >>> 1. What is the deal with the impossibly high IRQ number?

    >>
    >> It is APIC (!= ACPI ) the few years old feature that is used mainly for
    >> multiprocessor machines, it allows for up to 256 interrupts if computer
    >> hardware has support for it.
    >>
    >>> 2. What is the "irqpoll option," and how do I boot with it?

    >>
    >> If you have installed kernel sources than in:
    >> /usr/src/linux/Documentation
    >> file
    >> kernel-parameters.txt
    >> has all bout kernel paramateres.
    >> ...
    >> irqpoll [HW]
    >> When an interrupt is not handled search all handlers
    >> for it. Also check all handlers each timer
    >> interrupt. Intended to get systems with badly broken
    >> firmware running.
    >> ...
    >> Something that might help, but can make more troubles (not sure).
    >> On a first boot screen type
    >> irqpoll
    >> as test solution to see that other stuff will work.
    >> Than if everything works as before put that option in
    >> /boot/grub/menu.lst
    >> and see will that help.
    >>

    > I tried the irqpoll option, but it made things worse. The x-server hung
    > at startup. Do you think there would be any point in disabling APIC? I
    > do have an AMD 64 x2 processor, so technically this is a multi-processor
    > machine.


    It real multiprocessor machine although both CPUs are in the same chip.

    It is cheap to try noapic option.
    Enter it the same way as irqpoll and see if it works.
    So far I know, the worst that can happen is that computer will not boot.

    --
    Regards,
    Rajko.

  5. Re: Problem with VIA Rhine II on eth0

    Rajko M. wrote:
    > GateKeeper wrote:
    >
    >> Rajko M. wrote:
    >>> GateK33per wrote:
    >>>
    >>>> 1. What is the deal with the impossibly high IRQ number?
    >>> It is APIC (!= ACPI ) the few years old feature that is used mainly for
    >>> multiprocessor machines, it allows for up to 256 interrupts if computer
    >>> hardware has support for it.
    >>>
    >>>> 2. What is the "irqpoll option," and how do I boot with it?
    >>> If you have installed kernel sources than in:
    >>> /usr/src/linux/Documentation
    >>> file
    >>> kernel-parameters.txt
    >>> has all bout kernel paramateres.
    >>> ...
    >>> irqpoll [HW]
    >>> When an interrupt is not handled search all handlers
    >>> for it. Also check all handlers each timer
    >>> interrupt. Intended to get systems with badly broken
    >>> firmware running.
    >>> ...
    >>> Something that might help, but can make more troubles (not sure).
    >>> On a first boot screen type
    >>> irqpoll
    >>> as test solution to see that other stuff will work.
    >>> Than if everything works as before put that option in
    >>> /boot/grub/menu.lst
    >>> and see will that help.
    >>>

    >> I tried the irqpoll option, but it made things worse. The x-server hung
    >> at startup. Do you think there would be any point in disabling APIC? I
    >> do have an AMD 64 x2 processor, so technically this is a multi-processor
    >> machine.

    >
    > It real multiprocessor machine although both CPUs are in the same chip.
    >
    > It is cheap to try noapic option.
    > Enter it the same way as irqpoll and see if it works.
    > So far I know, the worst that can happen is that computer will not boot.
    >

    Neither noapic nor irqpoll worked. Noapic had no effect, and irqpoll
    made the computer hang at reboot. So I tried UBUNTU. I had the same
    problem. Then I rebooted once with irqfixup. That solved the problem.
    Don't know if that will work with Suse.

  6. Re: Problem with VIA Rhine II on eth0

    Barry R Ford wrote:

    ....
    >>>>
    >>> I tried the irqpoll option, but it made things worse. The x-server hung
    >>> at startup. Do you think there would be any point in disabling APIC? I
    >>> do have an AMD 64 x2 processor, so technically this is a multi-processor
    >>> machine.

    >>
    >> It is a real multiprocessor machine although both CPUs are in the
    >> same chip.
    >>
    >> It is cheap to try noapic option.
    >> Enter it the same way as irqpoll and see if it works.
    >> So far I know, the worst that can happen is that computer will not boot.
    >>

    > Neither noapic nor irqpoll worked. Noapic had no effect, and irqpoll
    > made the computer hang at reboot. So I tried UBUNTU. I had the same
    > problem. Then I rebooted once with irqfixup. That solved the problem.
    > Don't know if that will work with Suse.


    Hi Barry,

    It is a kernel problem and that is the reason that Ubuntu failed in the same
    way. SUSE and Ubuntu use similar kernel versions, so solution for one can
    be for the other too.

    --
    Regards,
    Rajko.

  7. Re: Problem with VIA Rhine II on eth0

    Rajko M. wrote:
    > Barry R Ford wrote:
    >
    > ...
    >>>> I tried the irqpoll option, but it made things worse. The x-server hung
    >>>> at startup. Do you think there would be any point in disabling APIC? I
    >>>> do have an AMD 64 x2 processor, so technically this is a multi-processor
    >>>> machine.
    >>> It is a real multiprocessor machine although both CPUs are in the
    >>> same chip.
    >>>
    >>> It is cheap to try noapic option.
    >>> Enter it the same way as irqpoll and see if it works.
    >>> So far I know, the worst that can happen is that computer will not boot.
    >>>

    >> Neither noapic nor irqpoll worked. Noapic had no effect, and irqpoll
    >> made the computer hang at reboot. So I tried UBUNTU. I had the same
    >> problem. Then I rebooted once with irqfixup. That solved the problem.
    >> Don't know if that will work with Suse.

    >
    > Hi Barry,
    >
    > It is a kernel problem and that is the reason that Ubuntu failed in the same
    > way. SUSE and Ubuntu use similar kernel versions, so solution for one can
    > be for the other too.
    >

    The fix is still working with Ubuntu Feisty 7.04 for x86_64 (AMD64).

    Thanks again. I'll be on the Ubuntu newsgroup from now on.

+ Reply to Thread