> 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.
