Loading ...
Sorry, an error occurred while loading the content.

Help with OpenSlug-OpenDebian

Expand Messages
  • jack_2006b
    Please answer to this question. Recently I upgrade my slug with 64MB (2 chips) and install OpenSlug and the OpenDebian with Apex as the bootloader. But I
    Message 1 of 6 , Oct 30, 2005
    • 0 Attachment
      Please answer to this question. Recently I upgrade my slug with 64MB
      (2 chips) and install OpenSlug and the OpenDebian with Apex as the
      bootloader. But I notice in the message log the following entry:

      Oct 29 09:56:49 localhost kernel: Pid: 224, comm:
      fsck.ext3
      Oct 29 09:56:49 localhost kernel: CPU: 0
      Oct 29 09:56:49 localhost kernel: pc : [<c0115078>] lr :
      [<c003bc60>] Not tainted
      Oct 29 09:56:49 localhost kernel: sp : c3979cb0 ip : ffffbd24 fp :
      c3979ccc
      Oct 29 09:56:49 localhost kernel: r10: c3979cd0 r9 : c01b12c0 r8 :
      c01c8834
      Oct 29 09:56:49 localhost kernel: r7 : 00000000 r6 : 60000013 r5 :
      c01c9270 r4 : c03b1918
      Oct 29 09:56:49 localhost kernel: r3 : 00000000 r2 : 00000001 r1 :
      c03b199c r0 : ffffffff
      Oct 29 09:56:49 localhost kernel: Flags: nZCv IRQs on FIQs on
      Mode SVC_32 Segment user
      Oct 29 09:56:49 localhost kernel: Control: 39FF Table: 03A18000
      DAC: 00000015
      Oct 29 09:56:49 localhost kernel: Function entered at [<c001d800>]
      from [<c001db74>]
      Oct 29 09:56:49 localhost kernel: r5 = 0000001A r4 = C01C2B20
      Oct 29 09:56:49 localhost kernel: Function entered at [<c001daf0>]
      from [<c001dbf8>]
      Oct 29 09:56:49 localhost kernel: r6 = 04000000 r5 = 0000001F r4
      = C01C2B20
      Oct 29 09:56:49 localhost kernel: Function entered at [<c001dbac>]
      from [<c001c940>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c0114f9c>]
      from [<c003bf30>]
      Oct 29 09:56:49 localhost kernel: r6 = C0114F9C r5 = 00000100 r4
      = C3978000
      Oct 29 09:56:49 localhost kernel: Function entered at [<c003bdb4>]
      from [<c00381f8>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c003819c>]
      from [<c001dcac>]
      Oct 29 09:56:49 localhost kernel: r8 = C3979D5C r7 = 00000000 r6
      = 00000020 r5 = 0000001F
      Oct 29 09:56:49 localhost kernel: r4 = C01C26DC
      Oct 29 09:56:49 localhost kernel: Function entered at [<c001dbac>]
      from [<c001c940>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c004f1d0>]
      from [<c004eda0>]
      Oct 29 09:56:49 localhost kernel: r8 = C0215540 r7 = 00001000 r6
      = 00F781A0 r5 = 00F781A1
      Oct 29 09:56:49 localhost kernel: r4 = C0280144
      Oct 29 09:56:49 localhost kernel: Function entered at [<c004ebec>]
      from [<c0051804>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c0051648>]
      from [<c005196c>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c00518d8>]
      from [<c006b958>]
      Oct 29 09:56:49 localhost kernel: r8 = C3978000 r7 = 00053878 r6
      = C393BF00 r5 = 00000000
      Oct 29 09:56:49 localhost kernel: r4 = 00008000
      Oct 29 09:56:49 localhost kernel: Function entered at [<c006b890>]
      from [<c006bc44>]
      Oct 29 09:56:49 localhost kernel: Function entered at [<c006bbf8>]
      from [<c001ccc0>]
      Oct 29 09:56:49 localhost kernel: r8 = C001CE44 r7 = 00000003 r6
      = 00031470 r5 = 7819A000
      Oct 29 09:56:49 localhost kernel: r4 = 0000000F
      Oct 29 09:56:49 localhost kernel: irq26: nobody cared

      Is this a memory problem? Bad kernel, bad Apex configuration or is
      normal? Thanks with any help.
    • John Bowler
      From: jack_2006b ... I can t remember how these things get output in the log but I think the traceback corresponds to the kernel warning at the end (i.e. the
      Message 2 of 6 , Oct 30, 2005
      • 0 Attachment
        From: jack_2006b
        >Oct 29 09:56:49 localhost kernel: irq26: nobody cared

        I can't remember how these things get output in the log but I think the
        traceback corresponds to the kernel warning at the end (i.e. the one
        I quoted above.)

        Try running without IRQ debugging (i.e. with noirqdebug on the kernel
        command line). IRQ debugging is useful when there are problems, but if
        there aren't any you are likely to see spurious warnings. Here's an
        explanation from l-a-k:

        From: linux-arm-kernel-bounces@... On Behalf Of
        Russell King - ARM Linux
        Sent: Monday, October 24, 2005 3:30 PM
        Subject: Re: ide0: unexpected interrupt (pxa255 and at91)
        >On Mon, Oct 24, 2005 at 05:16:38PM -0500, Bill Gatliff wrote:
        >> I have two different boards, one pxa255 and one at91rm9200, both of
        >> which have nonremovable CF ATA cards. Same 2.6.12 kernel runs on each
        >> (a few different config settings, obviously).
        >
        >At a guess, you're seeing the classic problems if hooking an interrupt
        >output which is designed for level-based signalling to an edge-based
        >input.
        >
        >You have two options:
        >1. Drop interrupts and risk devices stopping working afterwards.
        >2. Gain extra spurious interrupts which the driver may ignore.
        >
        >We opt for (2) because that's the more reliable option - at the
        >expense of occasionally creating unexpected interrupts. The IDE
        >subsystem is unfortunately particularly noisy when it comes to
        >these.

        I don't know whether this applies to the USB on the NSLU2 PCI
        controller, but that irq26 is the IRQ from the USB which is on
        the PCI bus and it is a spurious interrupt so far as we have
        ever been able to see.

        John Bowler <jbowler@...>
      • jack_2006b
        Thanks for the info, but I new with OpenDebian, maybe is a simple answer, but where is the Kernel Command line where I need to write the noirqdebug? Sorry.
        Message 3 of 6 , Oct 30, 2005
        • 0 Attachment
          Thanks for the info, but I new with OpenDebian, maybe is a simple
          answer, but where is the Kernel Command line where I need to write
          the noirqdebug? Sorry.

          Thanks with any help


          --- In nslu2-linux@yahoogroups.com, John Bowler <jbowler@a...> wrote:
          >
          > From: jack_2006b
          > >Oct 29 09:56:49 localhost kernel: irq26: nobody cared
          >
          > I can't remember how these things get output in the log but I
          think the
          > traceback corresponds to the kernel warning at the end (i.e. the
          one
          > I quoted above.)
          >
          > Try running without IRQ debugging (i.e. with noirqdebug on the
          kernel
          > command line). IRQ debugging is useful when there are problems,
          but if
          > there aren't any you are likely to see spurious warnings. Here's
          an
          > explanation from l-a-k:
          >
          > From: linux-arm-kernel-bounces@l... On Behalf Of
          > Russell King - ARM Linux
          > Sent: Monday, October 24, 2005 3:30 PM
          > Subject: Re: ide0: unexpected interrupt (pxa255 and at91)
          > >On Mon, Oct 24, 2005 at 05:16:38PM -0500, Bill Gatliff wrote:
          > >> I have two different boards, one pxa255 and one at91rm9200,
          both of
          > >> which have nonremovable CF ATA cards. Same 2.6.12 kernel runs
          on each
          > >> (a few different config settings, obviously).
          > >
          > >At a guess, you're seeing the classic problems if hooking an
          interrupt
          > >output which is designed for level-based signalling to an edge-
          based
          > >input.
          > >
          > >You have two options:
          > >1. Drop interrupts and risk devices stopping working afterwards.
          > >2. Gain extra spurious interrupts which the driver may ignore.
          > >
          > >We opt for (2) because that's the more reliable option - at the
          > >expense of occasionally creating unexpected interrupts. The IDE
          > >subsystem is unfortunately particularly noisy when it comes to
          > >these.
          >
          > I don't know whether this applies to the USB on the NSLU2 PCI
          > controller, but that irq26 is the IRQ from the USB which is on
          > the PCI bus and it is a spurious interrupt so far as we have
          > ever been able to see.
          >
          > John Bowler <jbowler@a...>
          >
        • John Bowler
          From: jack_2006b ... The one from APEX - i.e. it needs to be in the arguments passed to the kernel by APEX. A standard NSLU2 kernel build includes the option
          Message 4 of 6 , Oct 30, 2005
          • 0 Attachment
            From: jack_2006b
            >Thanks for the info, but I new with OpenDebian, maybe is a simple
            >answer, but where is the Kernel Command line where I need to write
            >the noirqdebug? Sorry.

            The one from APEX - i.e. it needs to be in the arguments passed to
            the kernel by APEX. A standard NSLU2 kernel build includes the option
            on the default command line, so I assume that the APEX boot process
            overrides this, but I may be wrong as I don't know much about APEX...

            John Bowler <jbowler@...>
          • Isthar
            Where can I read more about adding RAM to the slug? Thanks, Eddie
            Message 5 of 6 , Oct 31, 2005
            • 0 Attachment
              Where can I read more about adding RAM to the slug?

              Thanks,
              Eddie

              --- In nslu2-linux@yahoogroups.com, "jack_2006b" <jack@j...> wrote:
              >
              > Please answer to this question. Recently I upgrade my slug with 64MB
              > (2 chips) and install OpenSlug and the OpenDebian with Apex as the
              > bootloader. But I notice in the message log the following entry:
            • Marcel Nijenhof
              ... http://www.nslu2-linux.org/wiki/HowTo/FattenYourSlug Marceln
              Message 6 of 6 , Nov 1, 2005
              • 0 Attachment
                On Tue, 2005-11-01 at 01:55 +0000, Isthar wrote:

                > Where can I read more about adding RAM to the slug?


                http://www.nslu2-linux.org/wiki/HowTo/FattenYourSlug

                Marceln
              Your message has been successfully submitted and would be delivered to recipients shortly.