HP DC7700 vs. Linux - Networking

This is a discussion on HP DC7700 vs. Linux - Networking ; Anyone know what HP screwed up on the DC7700 series computers that prevents Linux from coming up unless ACPI is completely shut off? -- |---------------------------------------/----------------------------------| | Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below | | ...

+ Reply to Thread
Results 1 to 5 of 5

Thread: HP DC7700 vs. Linux

  1. HP DC7700 vs. Linux

    Anyone know what HP screwed up on the DC7700 series computers that prevents
    Linux from coming up unless ACPI is completely shut off?

    --
    |---------------------------------------/----------------------------------|
    | Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
    | first name lower case at ipal.net / spamtrap-2007-08-14-1311@ipal.net |
    |------------------------------------/-------------------------------------|

  2. Re: HP DC7700 vs. Linux

    phil-news-nospam@ipal.net wrote:
    > Anyone know what HP screwed up on the DC7700 series computers that prevents
    > Linux from coming up unless ACPI is completely shut off?


    No, but the output of dmesg will give you some instructions on commands to run and where to send the
    output. Keeping the ACPI code working with the brain-dead BIOS's in new machines is a full-time
    activity. Your output will help them fix Linux to work with that machine.

    Larry

  3. Re: HP DC7700 vs. Linux

    On Tue, 14 Aug 2007 22:25:17 GMT Larry Finger wrote:
    | phil-news-nospam@ipal.net wrote:
    |> Anyone know what HP screwed up on the DC7700 series computers that prevents
    |> Linux from coming up unless ACPI is completely shut off?
    |
    | No, but the output of dmesg will give you some instructions on commands to run and where to send the
    | output. Keeping the ACPI code working with the brain-dead BIOS's in new machines is a full-time
    | activity. Your output will help them fix Linux to work with that machine.

    In what system will dmesg tell me where to send its output? On normal
    machines, dmesg gives me recent kernel printk output, but nothing about
    where to send stuff. And I assume what they want is the output with
    ACPI off ... since I can't run dmesg with it on since the kernel does
    not make it all the way up to start userland.

    --
    |---------------------------------------/----------------------------------|
    | Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
    | first name lower case at ipal.net / spamtrap-2007-08-14-2159@ipal.net |
    |------------------------------------/-------------------------------------|

  4. Re: HP DC7700 vs. Linux

    phil-news-nospam@ipal.net wrote:
    > On Tue, 14 Aug 2007 22:25:17 GMT Larry Finger wrote:
    > | phil-news-nospam@ipal.net wrote:
    > |> Anyone know what HP screwed up on the DC7700 series computers that prevents
    > |> Linux from coming up unless ACPI is completely shut off?
    > |
    > | No, but the output of dmesg will give you some instructions on commands to run and where to send the
    > | output. Keeping the ACPI code working with the brain-dead BIOS's in new machines is a full-time
    > | activity. Your output will help them fix Linux to work with that machine.
    >
    > In what system will dmesg tell me where to send its output? On normal
    > machines, dmesg gives me recent kernel printk output, but nothing about
    > where to send stuff. And I assume what they want is the output with
    > ACPI off ... since I can't run dmesg with it on since the kernel does
    > not make it all the way up to start userland.


    Your original message implied that your computer did come up with ACPI off. In that case, you would
    get dmesg output. On one system that I have been helping debug a problem, the dmesg output includes
    the following:

    ACPI: System BIOS is requesting _OSI(Linux)
    ACPI: Please test with "acpi_osi=!Linux"
    Please send dmidecode to linux-acpi@vger.kernel.org

    If your ACPI is broken, it should be issuing this, or similar, messages.

    Larry

  5. Re: HP DC7700 vs. Linux

    On Wed, 15 Aug 2007 04:28:05 GMT Larry Finger wrote:
    | phil-news-nospam@ipal.net wrote:
    |> On Tue, 14 Aug 2007 22:25:17 GMT Larry Finger wrote:
    |> | phil-news-nospam@ipal.net wrote:
    |> |> Anyone know what HP screwed up on the DC7700 series computers that prevents
    |> |> Linux from coming up unless ACPI is completely shut off?
    |> |
    |> | No, but the output of dmesg will give you some instructions on commands to run and where to send the
    |> | output. Keeping the ACPI code working with the brain-dead BIOS's in new machines is a full-time
    |> | activity. Your output will help them fix Linux to work with that machine.
    |>
    |> In what system will dmesg tell me where to send its output? On normal
    |> machines, dmesg gives me recent kernel printk output, but nothing about
    |> where to send stuff. And I assume what they want is the output with
    |> ACPI off ... since I can't run dmesg with it on since the kernel does
    |> not make it all the way up to start userland.
    |
    | Your original message implied that your computer did come up with ACPI off. In that case, you would
    | get dmesg output. On one system that I have been helping debug a problem, the dmesg output includes
    | the following:
    |
    | ACPI: System BIOS is requesting _OSI(Linux)
    | ACPI: Please test with "acpi_osi=!Linux"
    | Please send dmidecode to linux-acpi@vger.kernel.org
    |
    | If your ACPI is broken, it should be issuing this, or similar, messages.

    I rebuilt a kernel without ACPI because I will be running it via PXE without
    an interactive prompt. Would it still work that way? I should have a chance
    to test it today, anyway.

    --
    |---------------------------------------/----------------------------------|
    | Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
    | first name lower case at ipal.net / spamtrap-2007-08-15-0707@ipal.net |
    |------------------------------------/-------------------------------------|

+ Reply to Thread