[Bug 275] New: Consistant oops on dual athlon boards

[bugzilla-daemon@oss.sgi.com]

http://oss.sgi.com/bugzilla/show_bug.cgi?id=275

           Summary: Consistant oops on dual athlon boards
           Product: Linux XFS
           Version: unspecified
          Platform: AMD
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: Medium
         Component: XFS kernel code
        AssignedTo: xfs-master@oss.sgi.com
        ReportedBy: xfs@djc.f2o.org


Over the past few months, I've been able to consistantly crash XFS-enabled
kernels with a 35Gb rsync(or any high disk I/O) to a dual AMD server. Will try
to get a backtrace ASAP. This is on a stock 2.4.21 kernel with XFS 1.3-pre6

Unable to handle kernel paging request at virtual address 593fe098
c013049b
*pde = 00000000
Oops: 0000
CPU:    1
EIP:    0010:[<c013049b>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010007
eax: 20202020   ebx: 20202020   ecx: d8bf6000   edx: c15b9800
esi: 00000029   edi: 00000054   ebp: dffea268   esp: decb9e6c
ds: 0018   es: 0018   ss: 0018
Process sendmail (pid: 672, stackpage=decb9000)
Stack: dffea270 dffea278 dffea268 00000246 000001f0 c158d440 c013117b dffea268
       c15b9800 000001f0 fffffff4 c158d4ac c158c1c0 c158d440 c0130597 dffea268
       000001f0 c014ecbc dffea268 000001f0 00000007 fffffff4 c158d4ac c158c1c0
 [<c013117b>] __kmem_cache_alloc+0x5b/0x130 [kernel]
 [<c0130597>] kmem_cache_alloc+0x17/0x20 [kernel]
 [<c014ecbc>] d_alloc+0x1c/0x1b0 [kernel]
 [<c0144b0f>] real_lookup+0xaf/0x140 [kernel]
 [<c01452a8>] link_path_walk+0x5a8/0x6c0 [kernel]
 [<c01453d6>] path_walk+0x16/0x20 [kernel]
 [<c01455c9>] path_lookup+0x39/0x40 [kernel]
 [<c0145ad0>] open_namei+0x70/0x570 [kernel]
 [<c01388de>] filp_open+0x3e/0x70 [kernel]
 [<c0138cc3>] sys_open+0x53/0xc0 [kernel]
 [<c010765f>] system_call+0x33/0x38 [kernel]
Code: 8b 44 81 18 0f af 5d 18 89 41 14 03 59 0c 40 74 33 8b 4c 24

>>EIP; c013049b <kmem_cache_alloc_batch+4b/e0>   <=====
Code;  c013049b <kmem_cache_alloc_batch+4b/e0>
00000000 <_EIP>:
Code;  c013049b <kmem_cache_alloc_batch+4b/e0>   <=====
   0:   8b 44 81 18               mov    0x18(%ecx,%eax,4),%eax   <=====
Code;  c013049f <kmem_cache_alloc_batch+4f/e0>
   4:   0f af 5d 18               imul   0x18(%ebp),%ebx
Code;  c01304a3 <kmem_cache_alloc_batch+53/e0>
   8:   89 41 14                  mov    %eax,0x14(%ecx)
Code;  c01304a6 <kmem_cache_alloc_batch+56/e0>
   b:   03 59 0c                  add    0xc(%ecx),%ebx
Code;  c01304a9 <kmem_cache_alloc_batch+59/e0>
   e:   40                        inc    %eax
Code;  c01304aa <kmem_cache_alloc_batch+5a/e0>
   f:   74 33                     je     44 <_EIP+0x44> c01304df
<kmem_cache_alloc_batch+8f/e0>
Code;  c01304ac <kmem_cache_alloc_batch+5c/e0>
  11:   8b 4c 24 00               mov    0x0(%esp,1),%ecx


1 warning and 1 error issued.  Results may not be reliable.
root@shell:/home/djc> uname -a
Linux shell 2.4.21-xfs-djc #4 SMP Mon Aug 18 13:55:43 CDT 2003 i686 unknown
root@shell:/home/djc> gcc -v
Reading specs from /usr/local/lib/gcc-lib/i686-pc-linux-gnu/3.3.1/specs
Configured with: ./configure --host=i686-pc-linux-gnu --enable-threads=posix
--with-cpu=i686 --disable-libgcj
Thread model: posix
gcc version 3.3.1



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.