panic: vm_page_insert: page already inserted

Ok, now another panic that doesn't involve my code.  Simply loading
the bktr module causes this:
login: bktr_mem: memory holder loaded
bktr0: <BrookTree 878> mem 0xe7005000-0xe7005fff irq 11 at device 11.0 on pci0
bktr0: [GIANT-LOCKED]
smbus0: <System Management Bus> on bktr0
iicbb0: <I2C bit-banging driver> on bktr0
iicbus0: <Philips I2C bus> on iicbb0 master-only
panic: vm_page_insert: page already inserted
cpuid = 0
KDB: enter: panic
[thread pid 662 tid 100090 ]
Stopped at      kdb_enter+0x2b: nop     
db> tr
Tracing pid 662 tid 100090 td 0xc3916510
kdb_enter(c0694489) at kdb_enter+0x2b
panic(c06abef1,0,0,c1655fb8,d1d30a90) at panic+0x127
vm_page_insert(c1655fb8,c075a440,11eb1,0) at vm_page_insert+0x36
contigmalloc2(c1655fb8,8,0,1,c1655fb8) at contigmalloc2+0x106
contigmalloc(8000,c06ce020,1,0,ffffffff) at contigmalloc+0x95
get_bktr_mem(0,8000,11,36e109e,36e109e) at get_bktr_mem+0x20
common_bktr_attach(c3959000,0,36e109e,11,0) at common_bktr_attach+0x34
bktr_attach(c366d200) at bktr_attach+0x37a
device_attach(c366d200,c366d200,c366d200,a,c3670300) at device_attach+0x58
device_probe_and_attach(c366d200,c366d200,c3670300) at device_probe_and_attach+0xe0
pci_driver_added(c3661880,c3ac85b4) at pci_driver_added+0xd1
devclass_add_driver(c35ad840,c3ac85b4,c3949900,c3ac8594,c390fc20) at devclass_add_driver+0xb7
driver_module_handler(c3949900,0,c3ac85a0,c06fad00,c069362f) at driver_module_handler+0x59
module_register_init(c3ac8594) at module_register_init+0x52
linker_file_sysinit(c39acc00,c39acc00,1,c39acc00,c390fc20) at linker_file_sysinit+0x7d
linker_load_file(c390fc20,d1d30c44,400,0,c378ec00) at linker_load_file+0xce
linker_load_module(0,c378ec00,0,0,d1d30c70) at linker_load_module+0xb7
kldload(c3916510,d1d30d04,c,c3916510,d1d30d30) at kldload+0xcb
syscall(3b,3b,3b,0,bfbfe8e8) at syscall+0x27e
Xint0x80_syscall() at Xint0x80_syscall+0x1f
--- syscall (304, FreeBSD ELF32, kldload), eip = 0x280baf83, esp = 0xbfbfe85c, ebp = 0xbfbfe898 ---
db> 

I thought it was reproducable, but I haven't been able to reproduce
it, though it happened to be before..  and unluckily the dump failed..
Dumping 375 MB (2 chunks)
  chunk 0: 1MB (159 pages) ... ok
  chunk 1: 375MB (95984 pages) 359 343 327 311 295 279 263 247 231 215 199 183 167 151ad0: FAILURE - out of memory in start
 ... fail

** DUMP FAILED (ERROR 12) **

Though w/ the new ata DMA dumping code, it went a bit faster..

-- 
  John-Mark Gurney				Voice: +1 415 225 5579

     "All that I will do, has been done, All that I have, has not."