[Bug 219857] panic in scsi_cd code

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=219857

--- Comment #3 from Dexuan Cui <decui at microsoft.com> ---
Our test team sees a similar call-trace:

panic: sleepq_add: td 0xfffff8010581a000 to sleep on wchan 0xfffff8010f924048
with sleeping prohibited
cpuid = 1
time = 1527752111
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe00005815e0
vpanic() at vpanic+0x1a3/frame 0xfffffe0000581640
doadump() at doadump/frame 0xfffffe00005816c0
sleepq_add() at sleepq_add+0x378/frame 0xfffffe0000581710
_sleep() at _sleep+0x2c3/frame 0xfffffe00005817b0
cam_periph_runccb() at cam_periph_runccb+0x1ad/frame 0xfffffe0000581900
cdcheckmedia() at cdcheckmedia+0x8a/frame 0xfffffe0000581980
cdstrategy() at cdstrategy+0x55/frame 0xfffffe00005819b0
g_disk_start() at g_disk_start+0x48e/frame 0xfffffe0000581a20
g_io_schedule_down() at g_io_schedule_down+0x105/frame 0xfffffe0000581a60
g_down_procbody() at g_down_procbody+0x6d/frame 0xfffffe0000581a70
fork_exit() at fork_exit+0x84/frame 0xfffffe0000581ab0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0000581ab0
--- trap 0, rip = 0, rsp = 0, rbp = 0 ---
KDB: enter: panic
[ thread pid 13 tid 100113 ]
Stopped at      kdb_enter+0x3b: movq    $0,kdb_why
db> 


But our test team is just running some fio test with RAID (we run "kldload
geom_stripe). I'm not sure why our test is triggering cdstrategy() and
cdcheckmedia(), which seems to be related to CD? Our test team says the panic
happens to FreeBSD 10.3 and the latest HEAD code, meaning this should be a
longstanding issue...

Anyway, do we have a plan to fix the panic originally reported?

-- 
You are receiving this mail because:
You are the assignee for the bug.