I'm using 2.4.19-pre10, and got an oops. I sometimes also get a lockup, and have no idea why.
This oops happened when reading my lkml mailbox. Ksymoops output:
Warning (compare_maps): ksyms_base symbol vmalloc_to_page_R__ver_vmalloc_to_page not found in System.map. Ignoring ksyms_base entry
Unable to handle kernel NULL pointer dereference at virtual address 00000000
c013f35e
*pde = 00000000
Oops: 0002
CPU: 0
EIP: 0010:[<c013f35e>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010286
eax: c09abb94 ebx: c09abb7c ecx: 00000000 edx: c25927f0
esi: c09abb64 edi: c9c606a0 ebp: 00000f53 esp: c1237f60
ds: 0018 es: 0018 ss: 0018
Process kswapd (pid: 5, stackpage=c1237000)
Stack: 0000000c 000001d0 00000020 00000006 c013f64b 000048ba c0129a51 00000006
000001d0 00000006 000001d0 c028cfd4 00000000 c028cfd4 c0129a9c 00000020
c028cfd4 00000001 c1236000 c0129b33 c028cf20 00000000 c1236249 0008e000
Call Trace: [<c013f64b>] [<c0129a51>] [<c0129a9c>] [<c0129b33>] [<c0129b8e>]
[<c0129c9d>] [<c0106efc>]
Code: 89 11 89 46 30 89 40 04 8b 46 4c 85 c0 74 13 8b 40 14 85 c0
>>EIP; c013f35e <prune_dcache+9e/138> <=====
>>eax; c09abb94 <_end+69cfb0/c52641c>
>>ebx; c09abb7c <_end+69cf98/c52641c>
>>edx; c25927f0 <_end+2283c0c/c52641c>
>>esi; c09abb64 <_end+69cf80/c52641c>
>>edi; c9c606a0 <_end+9951abc/c52641c>
>>ebp; 00000f53 Before first symbol
>>esp; c1237f60 <_end+f2937c/c52641c>
Trace; c013f64b <shrink_dcache_memory+1b/34>
Trace; c0129a51 <shrink_caches+69/80>
Trace; c0129a9c <try_to_free_pages+34/54>
Trace; c0129b33 <kswapd_balance_pgdat+43/8c>
Trace; c0129b8e <kswapd_balance+12/28>
Trace; c0129c9d <kswapd+99/bc>
Trace; c0106efc <kernel_thread+28/38>
Code; c013f35e <prune_dcache+9e/138>
00000000 <_EIP>:
Code; c013f35e <prune_dcache+9e/138> <=====
0: 89 11 mov %edx,(%ecx) <=====
Code; c013f360 <prune_dcache+a0/138>
2: 89 46 30 mov %eax,0x30(%esi)
Code; c013f363 <prune_dcache+a3/138>
5: 89 40 04 mov %eax,0x4(%eax)
Code; c013f366 <prune_dcache+a6/138>
8: 8b 46 4c mov 0x4c(%esi),%eax
Code; c013f369 <prune_dcache+a9/138>
b: 85 c0 test %eax,%eax
Code; c013f36b <prune_dcache+ab/138>
d: 74 13 je 22 <_EIP+0x22> c013f380 <prune_dcache+c0/138>
Code; c013f36d <prune_dcache+ad/138>
f: 8b 40 14 mov 0x14(%eax),%eax
Code; c013f370 <prune_dcache+b0/138>
12: 85 c0 test %eax,%eax
After that I got a few times "kernel BUG at dcache.c:345"
Kurt