ISSUE: Unable to handle kernel paging... Kernel 2.2.14 on COMPAQ with 24Mb

From: Frank Krauss (fmfkrauss@mindspring.com)
Date: Mon Feb 14 2000 - 18:13:47 EST


1. Summary
     Unable to handle kernel paging... 2.2.14 on COMPAQ with 24Mb

2. Description
     Upgrading System from 2.0.35 to 2.2.14 in order to support upgrading a
     Compaq Deskpro 66/M from the original 16Mb of Ram to 24Mb.
     If I do not specify the append = mem=24M option in Lilo, the 2.2.14
         Kernel runs fine but I only see 16Mb.
     If I specify the append = mem=24M option in Lilo, I then do see the
        full 24Mb but then I get various problems, especially this oops.
     I have also gotten, at various times, the following messages:
     1. Segmentation Fault
     2. Virtual Memory exhausted
     3. login: no shell: no such file or directory
     4. login: can't open cache 'etc/ld.so.cache'
     Please see additional information in Item X, Other Notes, below.

3. Keywords
     unknown

4. Kernel version (from /proc/version)
     Linux version 2.2.14 (root@frankmkrauss.new-york.ny.us)
          (gcc version 2.7.2.3) #1 Thu Feb 10 21:03:02 EST 2000

5. Output of oops
     ksymoops 2.3.3 on i486 2.0.35. Options used
          -v /usr/src/linux/vmlinux (specified)
          -K (specified)
          -L (specified)
          -o /lib/modules/2.2.14 (specified)
          -m /usr/src/linux/System.map (specified)

     No modules in ksyms, skipping objects
     Feb 11 16:36:31 frankmkrauss kernel: CPU: 486
     Feb 11 16:37:50 frankmkrauss kernel: Unable to handle kernel paging request at virtual address 003f3fb7
     Feb 11 16:37:50 frankmkrauss kernel: current->tss.cr3 = 002ba000, %cr3 = 002ba000
     Feb 11 16:37:50 frankmkrauss kernel: *pde = 00000000
     Feb 11 16:37:50 frankmkrauss kernel: Oops: 0002
     Feb 11 16:37:50 frankmkrauss kernel: CPU: 0
     Feb 11 16:37:50 frankmkrauss kernel: EIP: 0010:[<c011f0e3>]
     Using defaults from ksymoops -t elf32-i386 -a i386
     Feb 11 16:37:50 frankmkrauss kernel: EFLAGS: 00010206
     Feb 11 16:37:50 frankmkrauss kernel: eax: 003f3fb3 ebx: c0ffd108 ecx: 0000001d edx: 003f3fb7
     Feb 11 16:37:50 frankmkrauss kernel: esi: c17ff620 edi: c17ff620 ebp: c0ffdf90 esp: c000beb0
     Feb 11 16:37:50 frankmkrauss kernel: ds: 0018 es: 0018 ss: 0018
     Feb 11 16:37:50 frankmkrauss kernel: Process id (pid: 546, process nr: 25, stackpage=c000b000)
     Feb 11 16:37:50 frankmkrauss kernel: Stack: 00000282 00000015 c01f6bd4 c0ffd010 c000bed8 00000010 00000212 00000001
     Feb 11 16:37:50 frankmkrauss kernel: 00000015 00000000 c011f2d2 c17ff620 00000015 00000000 c03f6aec c03f6aa0
     Feb 11 16:37:50 frankmkrauss kernel: c03f8c20 c012f054 c17ff620 00000015 00000000 c03f6aec c03f6aa0 c03f8c20
     Feb 11 16:37:50 frankmkrauss kernel: Call Trace: [<c011f2d2>] [<c012f054>] [<c012a445>] [<c012a644>] [<c012a7a8>] [<c0123114>] [<c0123377>]
     Feb 11 16:37:50 frankmkrauss kernel: [<c0107aa4>]
     Feb 11 16:37:50 frankmkrauss kernel: Code: 89 02 8b 12 49 83 f9 ff 75 b3 c7 02 00 00 00 00 fa c7 45 08

>>EIP; c011f0e3 <kmem_cache_grow+2bb/394> <=====
     Trace; c011f2d2 <kmem_cache_alloc+ca/118>
     Trace; c012f054 <d_alloc+18/150>
     Trace; c012a445 <real_lookup+35/a0>
     Trace; c012a644 <lookup_dentry+10c/1ac>
     Trace; c012a7a8 <open_namei+6c/2ec>
     Trace; c0123114 <filp_open+44/f4>
     Trace; c0123377 <sys_open+33/8c>
     Trace; c0107aa4 <system_call+34/40>
     Code; c011f0e3 <kmem_cache_grow+2bb/394>
     00000000 <_EIP>:
     Code; c011f0e3 <kmem_cache_grow+2bb/394> <=====
        0: 89 02 movl %eax,(%edx) <=====
     Code; c011f0e5 <kmem_cache_grow+2bd/394>
        2: 8b 12 movl (%edx),%edx
     Code; c011f0e7 <kmem_cache_grow+2bf/394>
        4: 49 decl %ecx
     Code; c011f0e8 <kmem_cache_grow+2c0/394>
        5: 83 f9 ff cmpl $0xffffffff,%ecx
     Code; c011f0eb <kmem_cache_grow+2c3/394>
        8: 75 b3 jne ffffffbd <_EIP+0xffffffbd> c011f0a0 <kmem_cache_grow+278/394>
     Code; c011f0ed <kmem_cache_grow+2c5/394>
        a: c7 02 00 00 00 movl $0x0,(%edx)
     Code; c011f0f2 <kmem_cache_grow+2ca/394>
        f: 00
     Code; c011f0f3 <kmem_cache_grow+2cb/394>
       10: fa cli
     Code; c011f0f4 <kmem_cache_grow+2cc/394>
       11: c7 45 08 00 00 movl $0x0,0x8(%ebp)
     Code; c011f0f9 <kmem_cache_grow+2d1/394>
       16: 00 00

6. Example Program
     N/A - from what I see of this problem, it appears that when I login to
          my second Id, I then get this problem. The first Login appears
          to normally work O.K.

7. Environment

     7.1 Software
          System Information
             System name Linux
             Node name frankmkrauss.new-york.ny.us
             Release 2.2.14
             Version #1 Thu Feb 10 21:03:02 EST 2000
             Machine i486
          Product Information Current Future
             Kernel modules 2.1.121
             Gnu C 2.7.2.3
             Binutils 2.9.1.0.4
             Linux C Library 5.4.46
             Dynamic Linker (ld.so) 1.9.9
             Linux C++ Library 27.2.8
             Procps 1.2.7 2.0.3
             Procinfo 0.11 17
             Util-linux (Mount) 2.10
             Net-tools 1.33 1.52
             PSMISC 16
             Patch 2.5
             Kbd 0.94
             Sh-utils 1.16
             Sysklogd 1.3-3 1.3-31
             RPM 2.5.2.col
             Pcmcia 3.0.3 3.0.14
             PPP 2.2 2.3.10
             Ksymoops 2.3.3

     7.2 Processor information (from /proc/cpuinfo)
          processor: 0
          vendor_id: unknown
          cpu family: 4
          model: 0
          model name: unknown
          stepping: unknown
          fdiv_bug: no
          hlt_bug: no
          sep_bug: no
          f00f_bug: no
          coma_bug: no
          fpu: yes
          fpu_exception: yes
          cpuid level: -1
          wp: yes
          flags:
          bogomips: 33.18

     7.3 Module information (from /proc/modules)
          nls_iso8859-1 2024 0 (unused)
          nls_cp437 3548 0 (unused)
          sr_mod 15652 0 (unused)
          scsi_mod 47896 1 [sr_mod]
          slhc 4160 0 (unused)

     7.4 SSCI information (from /proc/scsi/scsi)
          attached devices: none

     7.5 Other Information
          A. Output of FREE immediately after Boot WITHOUT append statement
                              total used free shared buffers cached
               Mem: 14700 9200 5500 7964 540 4240
            -/+ buffers/cache: 4420 10280
Swap: 16568 0 16568
B. Output of FREE immediately after Boot WITH append statement
             total used free shared buffers cached
Mem: 22932 9616 13316 7960 536 4432
-/+ buffers/cache: 4648 18284
Swap: 16568 0 16568

X. Other Notes
     History of this problem
     I have been running the 2.0.35 Kernel from the Caldera Open Linux
     1.3 Distribution successfully for the past 8 Months with 16Mb.
     About two Months ago I first attempted to upgrade my Compaq by
     adding on 8Mb more. Without the append statement, the new Storage
     was not reconized. With the append statement specified, I got all
     different types of oops. I was told by everyone on the Boards that
     unless I upgraded to the latest Kernel, no Developer would look at
     the problem, since the 2.0.35 Kernel was so old. That is the
     reason that I'm upgrading to the latest Production Kernel.
 
     Conserning the status of the Various Components mentioned in the
     Documentation/Changes file.
     I'm in the process of updating them, one at a time, to the correct
     levels. If some one tells me that my problem is caused by one of
     the Products in Item 7.1 above not being at the correct level, I
     will naturally consentrate on upgrading that one even though I
     believe that the problem is in the Kernel rather than a Product.

     Conserning my Virtual Memory (Swap Space)
     I do not know if the following is of any importance but I will
        include the information anyway.
     When I installed my System iniatially, since I had at that time 16Mb
     of Ram, I defined a 16Mb Swap Partition. Since I added the
     additional 8Mb of Ram I have NOT increased the Swap Space for the
     following reasons:
     1. I did not see any Doc that says that the Swap space has to be
          the same size as the Ram.
     2. I'm extremely tight on Disk space and need every byte possible.
     3. I assumed that with more Ram I would be doing even less Swapping
          then before.

     Uniqeness of my Problem
     A large part of my problem appears to be the fact that I'm running
     on a COMPAQ PC. It seems, from what I have heard on the Boards, that
     they use some type of non-standard or Proprietary BIOS which is a
     big cause of my problems. I think that someone with a strong
     knowledge of the Compaq way of doing things may be needed in order
     to solve this problem.

Though I'm a new person in the Linux/Unix world, I have been for many years
a System Programmer in the IBM Main Frame area and so I can follow any
directions that some one would give me in order to debug this Problem.

Any help in solving this problem would be greatly appreciated.

Yours truly,

Frank Krauss
fmfkrauss@mindspring.com

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:28 EST