Hello, just upgraded from r245115 to r249810 and now box reliably panics on boot a second after mounting root. Trace: #9 0xffffffff8066a443 in calltrap () at /usr/src/head/sys/amd64/amd64/exception.S:228 #10 0xffffffff8051fe75 in propagate_priority (td=<value optimized out>) at /usr/src/head/sys/kern/subr_turnstile.c:277 #11 0xffffffff805206b2 in turnstile_wait (ts=<value optimized out>, owner=<value optimized out>, queue=<value optimized out>) at /usr/src/head/sys/kern/subr_turnstile.c:743 #12 0xffffffff804cb03e in __mtx_lock_sleep (c=0xffffffff80af5f48, tid=18446741874883377440, opts=<value optimized out>, file=0xfffffe000bf42468 "(z§\200ÿÿÿÿ", line=2542784) at /usr/src/head/sys/kern/kern_mutex.c:503 #13 0xffffffff804f02ca in softclock_call_cc (c=0xffffffff80abd090, cc=0xffffffff80af9200, direct=0) at /usr/src/head/sys/kern/kern_timeout.c:638 #14 0xffffffff804f0644 in softclock (arg=0xffffffff80af9200) at /usr/src/head/sys/kern/kern_timeout.c:802 #15 0xffffffff804b2fe8 in intr_event_execute_handlers ( p=<value optimized out>, ie=0xfffffe000bbea300) at /usr/src/head/sys/kern/kern_intr.c:1263 #16 0xffffffff804b39f8 in ithread_loop (arg=0xfffffe000ba78400) at /usr/src/head/sys/kern/kern_intr.c:1276 #17 0xffffffff804b0f55 in fork_exit ( callout=0xffffffff804b38d0 <ithread_loop>, arg=0xfffffe000ba78400, frame=0xffffff800026cc00) at /usr/src/head/sys/kern/kern_fork.c:991 #18 0xffffffff8066a97e in fork_trampoline () at /usr/src/head/sys/amd64/amd64/exception.S:602 The callout functions is the syscons one: (kgdb) fr 13 #13 0xffffffff804f02ca in softclock_call_cc (c=0xffffffff80abd090, cc=0xffffffff80af9200, direct=0) at /usr/src/head/sys/kern/kern_timeout.c:638 638 class->lc_lock(c_lock, sharedlock); (kgdb) p *c $2 = {c_links = {le = {le_next = 0xffffff800099d798, le_prev = 0xffffffff80af9310}, sle = {sle_next = 0xffffff800099d798}, tqe = {tqe_next = 0xffffff800099d798, tqe_prev = 0xffffffff80af9310}}, c_time = 67155431438, c_precision = 71582788, c_arg = 0xffffffff80abcac0, c_func = 0xffffffff803b9fa0 <scrn_timer>, c_lock = 0xffffffff80af5f30, c_flags = 130, c_cpu = 0} (kgdb) The lock is the correct one (syscons are Giant-locked): (kgdb) p *c->c_lock $3 = {lo_name = 0xffffffff80779329 "Giant", lo_flags = 17498112, lo_data = 1, lo_witness = 0x0} But __mtx_lock_sleep() founds a NULL instead of turnstile: (kgdb) fr 12 #12 0xffffffff804cb03e in __mtx_lock_sleep (c=0xffffffff80af5f48, tid=18446741874883377440, opts=<value optimized out>, file=0xfffffe000bf42468 "(z§\200ÿÿÿÿ", line=2542784) at /usr/src/head/sys/kern/kern_mutex.c:503 503 turnstile_wait(ts, mtx_owner(m), TS_EXCLUSIVE_QUEUE); (kgdb) p ts $4 = (struct turnstile *) 0x0 -- Totus tuus, Glebius.Received on Tue Apr 23 2013 - 18:25:13 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:36 UTC