> In this case could you build a kernel with option MP_LOCKDEBUG and see if
it enters ddb?

Oh wait! I've a ddb when running with MP_LOCKDEBUG on real hardware! 
Previously it was on VMM.

so 'ps' show a huge list of sshd command:
PID     TID     PPID    UID     S       FLAGS   WAIT    COMMAND
68506   314235  39940   0       3       0x82    netio   sshd
56104   208367  39940   0       3       0x82    netio   sshd
...
77266   154542  66005   1000    3       0x83    netcon2 a.out
36210   35972   35940   0       2       0x82    netio   sshd
...


trace show: 

Debugger() at Debugger+0x9
__mp_lock() at __mp_lock+0x73
__mp_acquire_count() at __mp_acquire_count+0x2b
mi_switch() at mi_switch+0x21e
sleep_finish() at tsleep+0x154
soreceive() at socreceive+0x429
soo_read at soo_read+0x22
dofilereadv() at dofilereadv+0x204
sys_read() at sys_read+0x89
syscall() at syscall+0x27b
--- syscall (number 3) ---
end trace frame: 0x0, count: -11
0x145b662b0c5a

I can't copy/paste so I've typed this from pictures.

Reply via email to