Lock order reversals

Adriaan de Groot adridg at cs.kun.nl
Fri Feb 27 15:56:34 PST 2004

Hash: SHA1

Attached find the dump of a lock order reversal I found on my console just 
now. It might have happened while I was compiling and ripping CDs at the same 
time. Anyway .. is it useful to post these?They do suggest the _possibility_ 
of deadlock.

I also recently managed to panic my box with an, um, non-sleepable mutex 
discombobulation. Forget what exactly, but it was while trying to mount a USB 
pendrive. Next time .. 

- -- 
pub  1024D/FEA2A3FE 2002-06-18 Adriaan de Groot <groot at kde.org>
                     Would you like a freem?
Version: GnuPG v1.2.4 (FreeBSD)

-------------- next part --------------
lock order reversal
 1st 0xffffff0003a929a0 vm object (vm object) @ /mnt/sys/CURRENT/src/sys/vm/swap
 2nd 0xffffffff807dcca0 swap_pager swhash (swap_pager swhash) @ /mnt/sys/CURRENT
 3rd 0xffffff003e0e8700 vm object (vm object) @ /mnt/sys/CURRENT/src/sys/vm/uma_
Stack backtrace:
backtrace() at backtrace+0x17
witness_checkorder() at witness_checkorder+0x678
_mtx_lock_flags() at _mtx_lock_flags+0x80
obj_alloc() at obj_alloc+0x41
slab_zalloc() at slab_zalloc+0xa2
uma_zone_slab() at uma_zone_slab+0xd5
uma_zalloc_internal() at uma_zalloc_internal+0x3c
uma_zalloc_arg() at uma_zalloc_arg+0x3a6
swp_pager_meta_build() at swp_pager_meta_build+0x14b
swap_pager_putpages() at swap_pager_putpages+0x2f3
default_pager_putpages() at default_pager_putpages+0xa
vm_pageout_flush() at vm_pageout_flush+0x15e
vm_pageout_clean() at vm_pageout_clean+0x292
vm_pageout_scan() at vm_pageout_scan+0x65f
vm_pageout() at vm_pageout+0x377
fork_exit() at fork_exit+0xd9
fork_trampoline() at fork_trampoline+0xe
--- trap 0, rip = 0, rsp = 0xffffffffa9976d00, rbp = 0 ---

More information about the freebsd-amd64 mailing list