kern/38296: deadlock on p_stmutex?

[martin%duskware.de@localhost]

>Number:         38296
>Category:       kern
>Synopsis:       deadlock on p_stmutex?
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Mar 25 18:50:00 +0000 2008
>Originator:     Martin Husemann
>Release:        NetBSD 4.99.58
>Organization:
The NetBSD Foundation
>Environment:
System: NetBSD sunny-weather.duskware.de 4.99.58 NetBSD 4.99.58 (SUNNY) #74: 
Tue Mar 25 13:49:05 CET 2008 
martin%night-porter.duskware.de@localhost:/usr/src/sys/arch/sparc64/compile/SUNNY
 sparc64
Architecture: sparc64
Machine: sparc64
>Description:
Happened during a build.sh -j 4 on a two cpu machine:

Mutex error: mutex_vector_enter: spinout
                                        
lock address : 0x000000001847dd28 type     :               spin
shared holds :                  0 exclusive:                  1
shares wanted:                  0 exclusive:                  1
current cpu  :                  1 last held:                  0
current lwp  : 0x000000001845ab80 last held: 0x00000000183f68a0
last locked  : 0x000000000115cef0 unlocked : 0x00000000011520fc
initialized  : 0x00000000011413cc                              
owner field  : 0x00ff0f0000000000 wait/spin:                0/1
                                                               
panic: LOCKDEBUG
Stopped in pid 21782.1 (nbmake) at      netbsd:cpu_Debugger+0x4:        nop
db{1}>  bt                                                                 
lockdebug_abort1(1491328, 1491328, 12706f0, 12b5f80, 1, badcafe) at netbsd:lockd
ebug_abort1+0x7c                                                               
lockdebug_abort(1847dd28, 1410278, 12706f0, 12b5f80, 1847dd28, 1b) at netbsd:loc
kdebug_abort+0x7c                                                              
mutex_vector_enter(1847dd28, 1, 1845abf8, 2, 1, 1) at netbsd:mutex_vector_enter+
0x118                                                                          
statclock(e0017ed0, 184d7e00, 0, 3, 184d7cf0, 18) at netbsd:statclock+0xd8
statintr(e0017ed0, a, e0017ed0, badcafe, 1210880, badcafe) at netbsd:statintr+0x
10                                                                             
sparc_interrupt(1, 147d9f0, 2, 2, 4053c64c, 800) at netbsd:sparc_interrupt+0x29c
                                                                               
lockdebug_unlock_rd(1491368, a, 0, 2, 1, 3d) at netbsd:lockdebug_unlock_rd+0xc
lockdebug_locked(14845f0, 11648a0, 0, 1845ab80, 147d9f0, badcafe) at netbsd:lock
debug_locked+0x60                                                              
mutex_vector_enter(147d9f0, 12b8780, 18e, 12b87a0, 1845ab80, badcafe) at netbsd:
mutex_vector_enter+0x318                                                       
callout_schedule(147d9f0, 1f5, 1270800, 12ba800, 1845ab80, 1b) at netbsd:callout
_schedule+0x40                                                                 
sleepq_block(23, 1, 1845abf8, 2, 1, 1) at netbsd:sleepq_block+0x94
pollcommon(1845ab80, 184d7e00, 0, 3, 184d7cf0, 18) at netbsd:pollcommon+0x2b0
sys_poll(1845ab80, 184d7dc0, 184d7e00, badcafe, badcafe, badcafe) at netbsd:sys_
poll+0x5c                                                                      
syscall_plain(184d7ed0, 3, 4053c64c, 7, 4053c64c, 800) at netbsd:syscall_plain+0
x130                                                                           
?(40807490, 3, 1388, 222000, 1, 3d) at 0x10093e4

db{1}> mach cpu 0                               
db{0}> bt        
lwp_lock_retry(1845ab80, d02e8c0, 183c0ae0, 12ba800, 183f68a0, 0) at netbsd:lwp_
lock_retry+0x20                                                                
sched_pstats(1847dd00, 1845ab80, 147d9f0, 0, 1479400, 147ac00) at netbsd:sched_p
stats+0x23c                                                                    
callout_softclock(147d800, 183f68a0, 14586b0, 14566b0, 1459400, 1000000) at netb
sd:callout_softclock+0x340                                                     
softint_overlay(0, 0, ffffffffffffffff, 183b6000, 0, 183b9cf8) at netbsd:softint
_overlay+0x10c                                                                 
lwp_userret(183f68a0, 0, 183b9e00, 183f68a0, e0018000, 1) at netbsd:lwp_userret+
0x170                                                                          
syscall_plain(183b9ed0, 0, 4053c304, cb, 4053c304, 800) at netbsd:syscall_plain+
0x28c                                                                          
?(ffffffffffffffff, ffffffffffffaadc, 3, 0, badcafe, badcafe) at 0x10093e4

db{0}> x/i 0x00000000011520fc                            
netbsd:calcru+0x1c:     call            netbsd:mutex_vector_exit
db{0}> x/i 0x000000000115cef0                                   
netbsd:sched_pstats+0x90:       call            netbsd:mutex_vector_enter


>How-To-Repeat:
no idea

>Fix:
n/a