kern/52816: sl(4) cause panic

[Manuel.Bouyer%lip6.fr@localhost]

>Number:         52816
>Category:       kern
>Synopsis:       sl(4) cause panic
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Dec 13 16:10:00 +0000 2017
>Originator:     Manuel Bouyer
>Release:        NetBSD 8.99.9
>Organization:
>Environment:
System: NetBSD borneo.soc.lip6.fr 8.99.9 NetBSD 8.99.9 (XEN3_DOM0) amd64
Architecture: x86_64
Machine: amd64
>Description:
	traffic over a sl(4) interface over ucom(4) causes panic:
panic: kernel diagnostic assertion "mutex_owned(&tty_lock)" failed: file "/dsk/l1/misc/bouyer/HEAD/src/sys/dev/usb/ucom.c", line 1296
fatal breakpoint trap in supervisor mode
trap type 1 code 0 rip 0xffffffff802065f5 cs 0xe030 rflags 0x246 cr2 0x703e5d71e000 ilevel 0x6 rsp 0xffffa0003654fd20
curlwp 0xffffa00000e59420 pid 0.3 lowest kstack 0xffffa0003654c2c0
Stopped in pid 0.3 (system) at  ffffffff802065f5:       leave
?() at ffffffff802065f5
?() at ffffffff80207cce     
?() at ffffffff8066da37     
?() at ffffffff808788a5     
?() at ffffffff809967e5     
?() at ffffffff803c245a     
?() at ffffffff8091abbc     
?() at ffffffff8084e2cb

(this points to:
cpu_Debugger() sys/arch/amd64/amd64/db_interface.c:252
db_panic() sys/ddb/db_panic.c:65
vpanic()   sys/kern/subr_prf.c:342
ch_voltag_convert_in() sys/dev/scsipi/ch.c:1283 (obviously bogus)
ucomstart() sys/dev/usb/ucom.c:1296
slintr() sys/net/if_sl.c:869
softint_thread() sys/arch/amd64/compile/XEN3_DOM0/xen-ma/machine/cpu.h:55

This also fails under netbsd8/amd64 (non-Xen) but no stack trace as I was
running X11 at the time of the panic

>How-To-Repeat:
	setup a sl(4) over ucom(4), start ping.
>Fix:
	workaround: setup on a netbsd-7 host.