Bug 135124

Summary: [lor] lor vfs_bio/ufs_dirhash
Product: Base System Reporter: randy
Component: amd64Assignee: freebsd-amd64 (Nobody) <amd64>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: Unspecified   
Hardware: Any   
OS: Any   

Description randy 2009-06-01 01:20:00 UTC
lock order reversal:
 1st 0xffffff8052aba050 bufwait (bufwait) @ /usr/src/sys/kern/vfs_bio.c:2556
 2nd 0xffffff0004f53000 dirhash (dirhash) @ /usr/src/sys/ufs/ufs/ufs_dirhash.c:275
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2a
_witness_debugger() at _witness_debugger+0x2e
witness_checkorder() at witness_checkorder+0x81e
_sx_xlock() at _sx_xlock+0x55
ufsdirhash_acquire() at ufsdirhash_acquire+0x33
ufsdirhash_add() at ufsdirhash_add+0x19
ufs_direnter() at ufs_direnter+0x88b
ufs_makeinode() at ufs_makeinode+0x31c
VOP_CREATE_APV() at VOP_CREATE_APV+0xb1
vn_open_cred() at vn_open_cred+0x3fd
kern_openat() at kern_openat+0x159
syscall() at syscall+0x1dd
Xfast_syscall() at Xfast_syscall+0xd0
--- syscall (5, FreeBSD ELF64, open), rip = 0x800c0bfcc, rsp = 0x7fffffffe6f8, rbp = 0x7fffffffe860 ---

i think this may be different than 261
<http://sources.zabbadoz.net/freebsd/lor/261.html>
Comment 1 John Baldwin freebsd_committer freebsd_triage 2009-06-01 14:25:25 UTC
It is not, it is the same exact LOR.

-- 
John Baldwin
Comment 2 John Baldwin freebsd_committer freebsd_triage 2009-06-01 14:25:38 UTC
State Changed
From-To: open->closed

This LOR is a duplicate of earlier reports.  It cannot deadlock as the 
dirhash lock is acquired on a new object before it is visible to other 
threads.
Comment 3 randy 2009-06-01 15:30:21 UTC
>> i think this may be different than 261
>> <http://sources.zabbadoz.net/freebsd/lor/261.html>
> It is not, it is the same exact LOR.

apologies

randy
Comment 4 John Baldwin freebsd_committer freebsd_triage 2009-06-01 15:40:06 UTC
On Monday 01 June 2009 10:30:21 am Randy Bush wrote:
> >> i think this may be different than 261
> >> <http://sources.zabbadoz.net/freebsd/lor/261.html>
> > It is not, it is the same exact LOR.
> 
> apologies

No worries. :)

-- 
John Baldwin