Kernel BUG in JFS 2.6.0-test4

After 2 days of uptime, I got the BUG below. I didn't notice until a couple of hours later, when I noticed that I a had a non-functioning and non-killable postgresql process, and a lilo install that refused to finish or die. This is with a stock 2.6.0-test4 + the Aug 26th reiser4 patches, but the problem (at least on the surface) doesn't appear to be related to the reiser4 stuff. (I'm CCing that list just in case somebody there thinks otherwise).

-Tupshin

Aug 28 15:38:32 bastard kernel: BUG at fs/jfs/jfs_dmap.c:468 assert(blkno + nblocks <= bmp->db_mapsize)
Aug 28 15:38:32 bastard kernel: kernel BUG at fs/jfs/jfs_dmap.c:468!
Aug 28 15:38:32 bastard kernel: invalid operand: 0000 [#1]
Aug 28 15:38:32 bastard kernel: CPU:    0
Aug 28 15:38:32 bastard kernel: EIP:    0060:[<c02526d3>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
Aug 28 15:38:32 bastard kernel: EFLAGS: 00010282
Aug 28 15:38:32 bastard kernel: eax: 0000004b   ebx: 00000064   ecx: c053b544   edx: c0483ad8
Aug 28 15:38:32 bastard kernel: esi: 00726174   edi: 00000000   ebp: f03d8000   esp: f7d41e9c
Aug 28 15:38:32 bastard kernel: ds: 007b   es: 007b   ss: 0068
Aug 28 15:38:32 bastard kernel: Stack: c043595b c0435b5c 000001d4 c0449000 f03e024c c1652e08 d893bd3c 00001000
Aug 28 15:38:32 bastard kernel: c013da09 f03e01c0 00000004 d893bd3c 00001000 c0155d82 c1652e08 000000d0
Aug 28 15:38:32 bastard kernel: d893bd3c c0156dbe 00000000 00000000 f7dbf60c f03d8000 c1652e08 c01575c8
Aug 28 15:38:32 bastard kernel: Call Trace:
Aug 28 15:38:32 bastard kernel: [<c013da09>] __set_page_dirty_nobuffers+0xa9/0xe0
Aug 28 15:38:32 bastard kernel: [<c0155d82>] mark_buffer_dirty+0x32/0x50
Aug 28 15:38:32 bastard kernel: [<c0156dbe>] __block_commit_write+0x8e/0x90
Aug 28 15:38:32 bastard kernel: [<c01575c8>] block_commit_write+0x28/0x30
Aug 28 15:38:32 bastard kernel: [<c0263ef0>] txFreeMap+0x200/0x2f0
Aug 28 15:38:32 bastard kernel: [<c0263a94>] txUpdateMap+0x174/0x240
Aug 28 15:38:32 bastard kernel: [<c0261e96>] txEnd+0xb6/0x120
Aug 28 15:38:32 bastard kernel: [<c0264220>] txLazyCommit+0x20/0xd0
Aug 28 15:38:32 bastard kernel: [<c026436a>] jfs_lazycommit+0x9a/0x1c0
Aug 28 15:38:32 bastard kernel: [<c011a7a0>] default_wake_function+0x0/0x30
Aug 28 15:38:32 bastard kernel: [<c010b172>] ret_from_fork+0x6/0x14
Aug 28 15:38:32 bastard kernel: [<c011a7a0>] default_wake_function+0x0/0x30
Aug 28 15:38:32 bastard kernel: [<c02642d0>] jfs_lazycommit+0x0/0x1c0
Aug 28 15:38:32 bastard kernel: [<c0109235>] kernel_thread_helper+0x5/0x10
Aug 28 15:38:32 bastard kernel: Code: 0f 0b d4 01 5c 5b 43 c0 8b 84 24 94 00 00 00 8b 94 24 94 00

Trace; c013da09 <__set_page_dirty_nobuffers+a9/e0> Trace; c0155d82 <mark_buffer_dirty+32/50> Trace; c0156dbe <__block_commit_write+8e/90> Trace; c01575c8 <block_commit_write+28/30> Trace; c0263ef0 <txFreeMap+200/2f0>
Trace; c0263a94 <txUpdateMap+174/240> Trace; c0261e96 <txEnd+b6/120>
Trace; c0264220 <txLazyCommit+20/d0> Trace; c026436a <jfs_lazycommit+9a/1c0> Trace; c011a7a0 <default_wake_function+0/30> Trace; c010b172 <ret_from_fork+6/14> Trace; c011a7a0 <default_wake_function+0/30> Trace; c02642d0 <jfs_lazycommit+0/1c0> Trace; c0109235 <kernel_thread_helper+5/10>

Code; c02526d3 <dbUpdatePMap+83/500>
00000000 <_EIP>:

Code;  c02526d3 <dbUpdatePMap+83/500>   <=====
  0:   0f 0b                     ud2a      <=====
Code;  c02526d5 <dbUpdatePMap+85/500>
  2:   d4 01                     aam    $0x1
Code;  c02526d7 <dbUpdatePMap+87/500>
  4:   5c                        pop    %esp
Code;  c02526d8 <dbUpdatePMap+88/500>
  5:   5b                        pop    %ebx
Code;  c02526d9 <dbUpdatePMap+89/500>
  6:   43                        inc    %ebx
Code;  c02526da <dbUpdatePMap+8a/500>
  7:   c0 8b 84 24 94 00 00      rorb   $0x0,0x942484(%ebx)
Code;  c02526e1 <dbUpdatePMap+91/500>
  e:   00 8b 94 24 94 00         add    %cl,0x942494(%ebx)