kernel: attempt to access beyond end of device

From: Norman Hooper (normanh_at_ossafrica.com)
Date: 05/06/05

  • Next message: Balazs Javor: "Managing shares and authentication for a small network" 
    To: debian-user@lists.debian.org
Date: Fri, 06 May 2005 12:51:07 +0200

    Hi there,

    This has me a bit stumped, and Google isn't coming up with much.

    I have a box with an Intel SRCS16 SATA controller. It uses the megaraid
    module. At the moment it's running a standard 2.6.7-1-686-smp kernel.
    I've built a custom kernel based on the .config from the 2.6.7-1 kernel,
    the only difference being max addressable RAM: I changed it from 4GB to
    64GB; and changed my CPU to Pentium4/Xeon/etc.

    When I boot the custom kernel it comes up fine, but after a while, it
    starts with:
            kernel: attempt to access beyond end of device
            sda2: rw=0, want=1218969608, limit=1146494790
    (sda2 is the /home partition)

    The "want" value differs depending on which file has been requested. The
    "limit" value remains the same.

    Why does changing the value for max addressable RAM / CPU have this
    effect? It's the same megaraid module, and the two kernels use identical
    initrd images to load it. Am I barking up the wrong tree?

    I've tried starting with a standard 2.6.8-2-686-smp kernel too, but the
    same thing happened. Here's a diff of my config files:

    /boot:$ diff config-2.6.8-2005042901 config-2.6.8-2-686-smp
    70c70
    < # CONFIG_M686 is not set 

    ---
> CONFIG_M686=y
74c74
< CONFIG_MPENTIUM4=y
---
> # CONFIG_MPENTIUM4 is not set
88c88
< CONFIG_X86_L1_CACHE_SHIFT=7
---
> CONFIG_X86_L1_CACHE_SHIFT=5
89a90
> CONFIG_X86_PPRO_FENCE=y
95d95
< CONFIG_X86_INTEL_USERCOPY=y
119,120c119,120
< # CONFIG_HIGHMEM4G is not set
< CONFIG_HIGHMEM64G=y
---
> CONFIG_HIGHMEM4G=y
> # CONFIG_HIGHMEM64G is not set
122d121
< CONFIG_X86_PAE=y
1973a1973
> CONFIG_VIDEO_MEYE=m
Hints, tips, clues, pointers all welcome.
Cheers,
Norman.
-- 
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
  • Next message: Balazs Javor: "Managing shares and authentication for a small network"

    Relevant Pages

    • Re: SuSE: migrating tot linux software RAID?
      ... > third machine with a megaraid that crashed because of this kernel ... >> The reason that you had data loss at all... ... > I had backups and successfully restored the system after a full install. ...
      (alt.os.linux.suse)
    • Poor I/O Performance with MegaRaid SATA 150-4; bug or feature?
      ... one of which occupied by a LSI MegaRaid SATA 150-4. ... and I'm running kernel 2.6.11.6. ... I could test megaraid and megaraid2, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • RE: Not able to load megaraid module after kernel upgrade from 2.4.9-e2 to 2.4.9-e25
      ... Grab the latest megaraid driver and compile it for your kernel, ... Not able to load megaraid module after kernel upgrade from ...
      (Linux-Kernel)
    • RE: megaraid on opteron w/ 8G RAM
      ... I'm continuing the work on this system; Dan or I will try to get a serial ... The driver stops after sending the first enquiry command to the card; ... > I'm using the megaraid 1.19.6 driver on RedHat Work Station ... > I also tried the megaraid2 module and that kernel panics ...
      (Linux-Kernel)
    • Re: stupid dell boxes and kernel recompile with scsi perc 4/di
      ... > All I want to do is recompile a kernel but for some reason I JUST CANT ... If your / is on an AMI Megaraid-connected SCSI disk, ... you need to build SCSI support, SCSI disk support, support for the ... > where you need the megaraid driver loaded before kernel loads. ...
      (comp.os.linux.misc)