Link to home
Start Free TrialLog in
Avatar of cwpak
cwpak

asked on

Kernel Bug at page_alloc.c:82


  I tried to install Redhat 7.1 booting
  from CDROM, and I got weird errors as following:

  Kernel command line: initrd=initrd.img lang=text devfs=
  nomount BOOT_IMAGE=vmlinuz
  Initializing CPU#0
  Detected 133.637 MHZ Processor
  Console: colour VGA+80x25
  Calibrating delay loop.. 266.24 BogoMIPS
  Kernel BUG at page_alloc.c:82!
  Invalid operand: 0000
  CPU: 0
  EIP: 0010:[<c01286ca>]
  EFLAGS: 00010286
  eax:0000001f ebx:c1000164 ecx:00000001 edx:c022c104
  esi:00000005 edi:c0272cc0 ebp:00000000 esp:c0241f70
  ds:0018  es:0018 ss:0018
  Process swapper cpid:0, stackpage=c0241000
  Stack: c01ee63b c01ee7d0 00000052 ...
         ... ...  ....................
         ............................. ffffffff
  Call Trace [<c01ee63b>] [<c01ee7d0>] [<c0105000>]
             [<c01e4459>] [<c0105000>] [<c0100191>]
  Code: 0f 0b 83 c4 0c 8b 43 08 85 c0 74 16 6a 54 68 d0
        e7 1e c0 68
  Kernel panic: Attempted to kill the idle task!
  In idle task - not syncing

  I have no idea what's going on.  Is this any
  hardware problem?  Could anyone let me know how
  to solve this problem?

  Thanks,
  cwp
 
ASKER CERTIFIED SOLUTION
Avatar of jlevie
jlevie

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of cwpak
cwpak

ASKER


  I have two PCs and I have installed RedHat 7.1
  with the same CD on the first machine successfully.
  I don't think this is CD image problem,
  but this is more like
  hardware problem as you mentioned.  Could you
  be more specific what components of hardware
  cause this problem and how to diagnose or solve them?

  Thanks,

  cwp
Avatar of cwpak

ASKER

 
  jlevie,

  I tried to swap memory chips on board, and
  that error's no longer shown.  I installed
  Redhat 7.1 OK finally.  However, I'm still
  trying to investigate why memory chips are
  bad.  While my machine was booting, memory
  test successfully performed, and I assumed
  that my memory chips were good enough to use.

  Do you have any comments on that?

  cwp
The memory test done by the BIOS during POST is typically far from exhaustive and in some cases is little more than just a memory sizing exercise. So it is entirely possible for the BIOS routine to "pass" memory that can't be used by the Linux kernel. I don't overly trust the self test of memory done by the BIOS, although you can trust it when it it reports less memory than you know to be installed. In that case a DIMM/SIMM is bad enough for the BIOS to detect and it certainly won't work with an operating system. I don't have a pointer to one handy, but there are real memory exercisers avaliable on the net that can do an exchaustive test of memory. Those are very useful when marginal memory is in the machine. Good enough to boot Linux, but not quite good enough for reliable operation.