Page 6 of 44 FirstFirst ... 4567816 ... LastLast
Results 51 to 60 of 437

Thread: Toshiba L505D ACPI issue with 9.1/9.04

  1. #51
    Join Date
    Dec 2009
    Beans
    83

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Colin: I never use sleep mode, all of my normal starting-up/shutting-down works well, but I've never been a fan of sleep mode, so I don't have any good advice for troubleshooting it.

    Ivan: Lucid hasn't worked out very well for me, I probably won't try again until closer to release time.


    All: I think the very latest kernel may have a fix:

    commit e93166f10c741f247c7e172936811bad558b4135
    Author: Joerg Roedel <joerg.roedel@amd.com>
    Date: Mon Dec 21 15:51:23 2009 +0100

    x86/amd-iommu: Fix initialization failure panic

    commit 0f764806438d5576ac58898332e5dcf30bb8a679 upstream.

    The assumption that acpi_table_parse passes the return value
    of the hanlder function to the caller proved wrong
    recently. The return value of the handler function is
    totally ignored. This makes the initialization code for AMD
    IOMMU buggy in a way that could cause a kernel panic on
    initialization. This patch fixes the issue in the AMD IOMMU
    driver.

    Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>




    I haven't yet tried to compile my own, but I'd love to say that I think that may
    be our problem.

  2. #52
    Join Date
    Jun 2008
    Beans
    130

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by Linux_Lurker View Post
    Colin: I never use sleep mode, all of my normal starting-up/shutting-down works well, but I've never been a fan of sleep mode, so I don't have any good advice for troubleshooting it.

    Ivan: Lucid hasn't worked out very well for me, I probably won't try again until closer to release time.


    All: I think the very latest kernel may have a fix:

    commit e93166f10c741f247c7e172936811bad558b4135
    Author: Joerg Roedel <joerg.roedel@amd.com>
    Date: Mon Dec 21 15:51:23 2009 +0100

    x86/amd-iommu: Fix initialization failure panic

    commit 0f764806438d5576ac58898332e5dcf30bb8a679 upstream.

    The assumption that acpi_table_parse passes the return value
    of the hanlder function to the caller proved wrong
    recently. The return value of the handler function is
    totally ignored. This makes the initialization code for AMD
    IOMMU buggy in a way that could cause a kernel panic on
    initialization. This patch fixes the issue in the AMD IOMMU
    driver.

    Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>




    I haven't yet tried to compile my own, but I'd love to say that I think that may
    be our problem.
    When you say "latest kernel" do you mean Lucid Lynx or a Karmic 2.6.33-xx update?

    Colin

  3. #53
    Join Date
    Dec 2009
    Beans
    83

    Re: Toshiba L505D ACPI issue with 9.1/9.04


  4. #54
    Join Date
    Jun 2008
    Beans
    130

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by Linux_Lurker View Post
    I had downloaded and installed from here:

    http://kernel.ubuntu.com/~kernel-ppa...ine/v2.6.32.3/

    Is there an advantage to compiling from source on my own system (which I have never done)?

    Colin

  5. #55
    Join Date
    Dec 2009
    Beans
    83

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by colinwhipple View Post
    I had downloaded and installed from here:

    http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.32.3/
    Well, that would certainly be easier, wouldn't it?

    Quote Originally Posted by colinwhipple View Post
    Is there an advantage to compiling from source on my own system (which I have never done)?
    Not really(unless you have a good reason to), when you compile it yourself, it is missing any Ubuntu-specific packages that the Ubuntu devs included. It's not too hard, but it takes a long time, even on a fast machine, and you generally need to answer a whole bunch of questions that you may not know the answer to. I already started compiling(on my Phenom II 940BE desktop, laptops get HOT when you try to do something like that), but I also just DLed the Ubuntu version, I guess I'll try both.

    So, did that fix anything? Did you try it without any boot switches?

    PS: If you wanted to roll your own kernel, it'd go like:

    sudo apt-get install kernel-package fakeroot build-essential ncurses-dev
    Download source
    Extract tarball
    cd [kernel source directory]
    make oldconfig
    //answer a buttload of questions
    make menuconfig

    //this will take a loooooonnnnngggggg time
    CONCURRENCY_LEVEL=4 fakeroot make-kpkg --initrd --append-to-version=-custom kernel_image kernel_headers
    //4 is the number of cores on my desktop's CPU, change as needed for your machine

    cd ..; ls

    //install the kernel
    sudo dpkg -i [image.deb]; sudo dpkg -i [headers.deb]

  6. #56
    Join Date
    Jun 2008
    Beans
    130

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by Linux_Lurker View Post
    Well, that would certainly be easier, wouldn't it?

    So, did that fix anything? Did you try it without any boot switches?

    ....
    I tried the download of the prebuilt kernel, and got an error message about:

    ...kernel_thread_helper...

    I get that kind of message with your additional boot options and without any. With acpi=off I can boot it up.

    Colin

  7. #57
    Join Date
    Dec 2009
    Beans
    83

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by colinwhipple View Post
    I tried the download of the prebuilt kernel, and got an error message about:

    ...kernel_thread_helper...

    I get that kind of message with your additional boot options and without any. With acpi=off I can boot it up.

    Colin
    Same here. I only tried the Ubuntu pre-compiled kernel, I doubt the kernel I compiled would do any better. If you can hunt down a Ubuntu 2.6.31.11 kernel, that may help. AFAIK, that particular commit isn't included in the 2.6.31 kernel series yet, but after reading the .9/.10 changelogs, that same AMD kernel dev has been doing tonnes of commits around IOMMU, ACPI, etc...

  8. #58
    Join Date
    Jun 2008
    Beans
    130

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Quote Originally Posted by Linux_Lurker View Post
    Same here. I only tried the Ubuntu pre-compiled kernel, I doubt the kernel I compiled would do any better. If you can hunt down a Ubuntu 2.6.31.11 kernel, that may help. AFAIK, that particular commit isn't included in the 2.6.31 kernel series yet, but after reading the .9/.10 changelogs, that same AMD kernel dev has been doing tonnes of commits around IOMMU, ACPI, etc...
    No joy with 2.6.31.11. With no boot options it takes a long time to get to a text mode login. Then after logging in I do a "startx", and end up with a blank screen.

    With these boot options:

    acpi_os_name="Windows 2001" acpi_osi="Windows 2001" pci=noacpi pci=assign-busses

    It gets to the Gnome desktop, but WiFi doesn't work. A wired connection does work, which is what I am using now.

    Colin

  9. #59
    Join Date
    Dec 2009
    Beans
    83

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    I'm not sure if this kernel corresponds to any of the kernels that Colin tried, but, I enabled pre-release updates in software-sources, and there is a new kernel in the pipe called

    linux-headers-2.6.31.18-generic-pae(yours may not be pae, the pae is only installed if you have ~4gb or more of ram).

    I see the following patches from Ubuntu devs in the release notes:

    * ACPI / PCI: Fix NULL pointer dereference in acpi_get_pci_dev() (rev. 2)
    - LP: #480144

    * ACPI: Clarify resource conflict message

    * ACPI video: ignore buggy _BQC

    * ACPI video: work-around BIOS AML bug in _BQC
    - LP: #428910

    * [Upstream] acpi: video: Loosen strictness of video bus detection code
    - LP: #333386

    * usb: amd5536udc: fixed shared interrupt bug and warning oops
    - LP: #494633

    * x86/amd-iommu: attach devices to pre-allocated domains early
    - LP: #503430

    * x86/amd-iommu: un__init iommu_setup_msi
    - LP: #503430

    * x86, apic: Enable lapic nmi watchdog on AMD Family 11h
    - LP: #503430

    * x86/amd-iommu: Workaround for erratum 63
    - LP: #480144

    * x86/amd-iommu: Un__init function required on shutdown
    - LP: #480144

    * x86/amd-iommu: fix broken check in amd_iommu_flush_all_devices



    It looks like any number of those could fix the problem. I won't have time to try it this weekend, but it's just a matter of enabling pre-release updates in software-sources, and then running update-manager if anyone else wants to try.

  10. #60
    Join Date
    Jun 2008
    Beans
    130

    Re: Toshiba L505D ACPI issue with 9.1/9.04

    Lurker,

    Thanks, but it's the same-old. I need the boot options for it to work.

    Colin

Page 6 of 44 FirstFirst ... 4567816 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •