Find the answer to your Linux question:
Results 1 to 4 of 4
Hej! I have just started using Linux one week ago. I installed Debian from a "liveCD", the current kernel is 2.6.18.6 and works. There are some things that don't work ...
  1. #1
    Just Joined!
    Join Date
    May 2008
    Posts
    3

    New linux user: help with compiling kernel

    Hej! I have just started using Linux one week ago. I installed Debian from a "liveCD", the current kernel is 2.6.18.6 and works. There are some things that don't work so well yet, so I would like to install a new kernel. (For example, the inbuilt Atheros wireless adapter doesn't work and the kernel doesn't notice that my laptop is from Toshiba.)

    I downloaded the newest kernel sources from The Linux Kernel Archives, and i've followed the instructions to compile them and install them. It was a steep learning curve and I made lots of mistakes in the beginning but I think now I know reasonably well what I'm doing.

    However, I don't get the new kernel to boot. After entering the LUKS passphrase to the encrypted LVM drive of my hard disk, the message is "Key slot 0 unlocked", then "Volume group ... not found", then something like "yaird - success, the new root is in the place of the old root", and finally "Kernel panic - not syncing".

    I think it's because I made a mistake with the options in the "make gconfig" step. Or because the yaird program fails to include the information necessary about what to do with the encrypted LVM module after it has become unlocked. Or maybe the encryption algorithm isn't present. I don't know where to look for further clues, can someone help me?

    I'm attaching the .config file that I used when compiling the new kernel (I had to zip it to attach it in this forum). I've tried to be specific about what is in my laptop, but maybe there's something that I have switched off without realising that it is needed for the ne kernel to properly load? I had once deactivated the dm-crypt module, and the yaird program told me that it is missing, so I have fixed that. But something is still wrong.

    Also, is there any way I can do something like "dmesg" on the new kernel that is not loading? I cannot get a prompt because of the kernel panic, so I cannot type it anywhere.

    Thanks for help! (I'm away for a few days but I'll reply as soon as I can.)
    Attached Files Attached Files

  2. #2
    Just Joined!
    Join Date
    May 2008
    Posts
    37
    sounds like you didn't include your hard drive's drivers

  3. #3
    Just Joined!
    Join Date
    May 2008
    Posts
    3
    @proprietarysucks: How do I know which drivers to include? I'm attaching the dmesg of the currently running kernel, does it contain that information? (If not, where do I find it?)

    I was quite confident that I had got the SCSI/disk part working because, just before the LUKS passphrase prompt, it says that it's got /sda /sda1 /sda2 /sda5 or something like that. (And it read the initrd image in the first place, so I thought the hard disk driver parts must be ok then.) But I really don't know...

    dmesg of the working 2.6.18.6 kernel attached.
    Attached Files Attached Files

  4. #4
    Just Joined!
    Join Date
    May 2008
    Posts
    3
    More information on my problem (I haven't changed anything yet, just tried to read most of the messages that appear on the screen while I attempt to boot my new kernel image...).

    I get a chance to scroll up through the bootup messages when the prompt of LUKS passphrase appears. Unfortunately I cannot see all the way to the top, but I can see up to somewhere in the middle of the many lines that go like follows, there are a few of them in both my working 2.6.18.6 kernel and the new one I built.
    Code:
    ACPI: PCI interrupt link [LNKA] (IRQs 10 11) *0, disabled
    I don't know what the kernel messages of the new kernel are above this point, so it is possible that I am missing error messages or that some things are loaded there which I don't know about in this post. (And my spelling may not be exactly the same as in the message because I have to copy them by hand.)

    So, the differences that I can see are:

    • Code:
      ACPI: PCI Interrupt 0000.02.06.0[A] --> GSI 23 (level low) -> IRQ 177
      This got IRQ 23 in my new kernel
    • Code:
      Initialising cryptographic API
      I can't see this message in the new kernel, but maybe it is above the ACPI section and I cannot scroll there to see it?
    • Code:
      RAMDISK driver initialised
      I don't see this either.
    • Code:
      TCP bic registered
      Don't see this in the new kernel either.
    • Code:
      NET: Registered protocol family ...
      My old kernel registered 1 17 8 and 20, my new kernel registers 1 17 10 and 15
    • Code:
      Using IPI no shortcuts mode
      My new kernel uses shortcuts mode.
    • Code:
      Time: tsc clocksource has been installed.
      Cannot see this either.
    • Code:
      ACPI: PCI interrupt 0000.00.13.0[A] --> GSI 19 (level low) -> IRQ 185
      That line also got IRQ 19 in the new kernel, and a few more ACPI: PCI messages that look similar all get 19 instead of 185.
    • Code:
      ohci_hcd 0000:00:13.1: new USB bus registered, assigned bus number 1
      now gets bus number 2, similarly the numbers of the ehci_hcd and another ohci_hcd USB buses are changed
    • Code:
      [ohci_hcd 0000:00;13;0: irq 185, io mem ...
      these lines now all have irq 19 instead of 185
    • Code:
      ohci_hcd 0000:00:13;0: wakeup
      I don't see this in my new kernel
    • Code:
      sata_sil 0000:00:12:0 version 2.0
      I don't see this either
    • Code:
      ACPI: PCI interrupt 0000:00:12.0(A) --> GSI 22 (level low) -> IRQ 193
      This now gets IRQ 22
    • Code:
      ata1: SATA max UDMA/100 cmd 0xF8836080 ctl 0xF883608A bmdma 0xF8836000 irq 193
      This line has a bit different format and the information in it is now
      Code:
      max UDMA/100 mmio m512@0xc000400 tf 0xc0004080 irq 22
    • in the similar line about ata2, the tf number is 0xc00040c0
    • Code:
      ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 0)
      My new kernel has SStatus 0 here
    • Code:
      ata1.00: ATA-7, max UDMA/100 ...
      The new kernel identifies this as my hard disk TOSHIBA MK6034GSX AH101A max UDMA/100
    • The new kernel prints some extra information about this hard disk:
      Code:
      scsi 0.0.0.0 direct-access ATA Toshiba MK6034GS AH10 PQ:0 ANSI:5
    • Code:
      device-mapper: ioctl: 4.7.0-ioctl
      The version number in the new kernel is 4.13.0
    • Code:
      ACPI: PCI interrupt 0000.00.14.1(A) --> GSI 16 level low -> IRQ 201
      This now gets irq 16
    • Code:
      ATTIXP... ide0: BM-DMA at 0x8460-0x8467 BIOS settings hda:pio, hdb:pio
      The new kernel spells is PIO.

    In the dmesg file that I attached before, the LUKS prompt comes between the lines device-mapper and kjournald starting. As far as I can see, the differences that are listed above are the only differences that appear on the screen when I boot my working kernel or my new kernel.

    Can an expert point me to the problem that I have and suggest some things that I could try out to fix this? Help is very appreciated, thank you!

Posting Permissions

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