[issue1077] Kernel lockup with SMP on -HEAD

Michael Kosarev sinknull at crater.dragonflybsd.org
Sat Jul 19 20:13:48 PDT 2008


New submission from Michael Kosarev <russiane39 at gmail.com>:

I've got a buddy @ Intel, and I've grabbed some platforms /
motherboards from him to test Dragonfly hardware compatilibilty, the
results shocked me - only one motherboard was compatible with
DragonFly, another 8 just hang kernel when I turn SMP on, /me tired
with ACPICA_IO turned on and turned off - nothing changes. The only
one motherboard, which works with DragonyFLY is rather old ASUS
workstation model, based on i975 chipset, another 8 motherboards was
from Intel,
5 motherboards from S5000 series, S3210SHLX, D975XBX2 and one SR1500
Intel platform. CPU's used in test - Core Quad Q6600 for LGA775
platform, Intel Xeon 5310 for LGA771 platform. All motherboards was
with latest bios firmware.
The hang I've noticed is extremely tricky to debug, I'm trying to find
problem for 4 days already, but can't find solution. Hang usually
occurs (as I could see visually) after detecting all devices, and
before (or maybe slightly after) calling /sbin/init, at least before
filesystems is mounted. I've talked with sephe about this issue, first
we thought it was bug of nata driver, but I've redone test with
Symbiosis SCSI controller, hang occurs absolutely at same place. All
my attempts at remote kgdb sessions failed, so the only meaningful
output from db> I'm sending to you. The problem with kgdb was
following:
kgdb -r /dev/cuaa0 kernel
Initiating remote session (or smthing like it)
Reply contains invalid hex digit 45
Session ended.
Feel free to contact me, so I can assist you in fixing that awful
issue. I should note, that DragonFly works nicely on all those systems
w/o SMP, and works only on one system with SMP.


db> ps
 pid      lwp  uid  ppid  pgrp  pflag  lflag stat  wmesg    wchan cmd
  1 cbdfafd0    0     0     0 000200 000000  1 1                  swapper
  0 c02f9c00    0     0     0 000200 000000  2 3 conifhk c02e0e60 swapper
oops, ran out of processes early!
cpu 0 tdrunqmask 10000000 curthread 0xc02fae78 reqflags 0012
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 28 0xc02fae78  -1 00210403 28/01/02 0xd269bce4        - ithread 1

 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8003a4  -1 00200000 12/01/00 0xff807d30        - idle_0
 32 0xc03037c0  -1 00010400 12/01/01 0xcbdaad88        - ithread emerg
 32 0xc02fad64  -1 00210400 28/01/00 0xcbdadd88        - ithread 0
 32 0xc030e6d4  -1 00010400 14/02/01 0xcbdb0d18        - softclock 0
 32 0xc02ff6b4  -1 00210400 14/01/00 0xcbdbcd88        - ithread 68
 32 0xc03064c4  -1 00000400 12/01/01 0xcbdbfd88        - dsched 0
 32 0xc03055a0  -1 00000400 12/01/01 0xcfc5ad88        - usched 0
 32 0xd24d8ac0   1 00000600 10/01/01 0xd2470d34        - swapper
 32 0xc0311ec0  -1 00040400 12/02/01 0xd2473cc4 waitport netisr_cpu 0
 32 0xc0310d20  -1 00040400 12/02/00 0xd247fcb8 waitport ifnet 0
 32 0xd24d8bb0  -1 00000e80 12/03/01 0xd24ebc8c ccb_scan xpt_thrd
 32 0xc02ff5a0  -1 00210400 14/02/01 0xd24eece4        - ithread 67
 32 0xd24d8ca0  -1 00000680 12/03/01 0xd24f1c9c    tqthr taskqueue 0
 32 0xc02ff7c8  -1 00210400 14/02/01 0xd24fdce4        - ithread 69
 32 0xc02fb718  -1 00210400 28/01/00 0xd2642d88        - ithread 9
 32 0xc02fb940  -1 00210400 28/01/00 0xd2665d88        - ithread 11
 32 0xc02fb82c  -1 00210400 28/02/01 0xd2688ce4        - ithread 10
 32 0xc02fb3dc  -1 00210400 28/02/01 0xd2697ce4        - ithread 6
 32 0xc02fae78  -1 00210403 28/01/02 0xd269bce4        - ithread 1
 32 0xc02ff264  -1 00210400 14/01/00 0xd269ed88        - ithread 64
 32 0xc02fb1b4  -1 00210400 28/01/00 0xd27e2d88        - ithread 4
 32 0xd24d9330  -1 00000680 12/02/01 0xd2af5c80    rwait random
 32 0xc0337fa0  -1 00040400 12/02/01 0xd2afccc4 waitport udp_thread 0
 32 0xc0337120  -1 00040400 12/02/01 0xd2dcbcc0 waitport tcp_thread 0
 32 0xd24d9510  -1 00040600 12/02/01 0xd2dd7cc0 waitport rtable_cpu 0
 32 0xc02f9d20   0 00000080 12/02/01 0xc0377ca0  conifhk swapper
cpu 1 tdrunqmask 00000000 curthread 0xff8083a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8083a4  -1 00200001 12/01/00 0xff80fd30        - idle_1
 32 0xc030e7d0  -1 00010400 14/02/01 0xcbdb3d18        - softclock 1
 32 0xc03065b4  -1 00000400 12/01/01 0xcfc51d88        - dsched 1
 32 0xc0305690  -1 00000400 12/01/01 0xcfc5dd88        - usched 1
 32 0xc0311fa8  -1 00040400 12/02/01 0xd2476cc4 waitport netisr_cpu 1
 32 0xc0310e08  -1 00040400 12/02/00 0xd24e2cb8 waitport ifnet 1
 32 0xd24d8d90  -1 00000680 12/03/01 0xd24f4c9c    tqthr taskqueue 1
 32 0xc0338088  -1 00040400 12/02/01 0xd2affcc4 waitport udp_thread 1
 32 0xc0337208  -1 00040400 12/02/01 0xd2dcecc0 waitport tcp_thread 1
 32 0xd44a9700  -1 00040600 12/02/01 0xd2ddacc0 waitport rtable_cpu 1
cpu 2 tdrunqmask 00000000 curthread 0xff8103a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8103a4  -1 00200001 12/01/00 0xff817d30        - idle_2
 32 0xc030e8cc  -1 00010400 14/02/01 0xcbdb6d18        - softclock 2
 32 0xc03066a4  -1 00000400 12/01/01 0xcfc54d88        - dsched 2
 32 0xc0305780  -1 00000400 12/01/01 0xcfc82d88        - usched 2
 32 0xc0312090  -1 00040400 12/02/01 0xd2479cc4 waitport netisr_cpu 2
 32 0xc0310ef0  -1 00040400 12/02/00 0xd24e5cb8 waitport ifnet 2
 32 0xd24d8e80  -1 00000680 12/03/01 0xd24f7c9c    tqthr taskqueue 2
 32 0xc0338170  -1 00040400 12/02/01 0xd2dc5cc4 waitport udp_thread 2
 32 0xc03372f0  -1 00040400 12/02/01 0xd2dd1cc0 waitport tcp_thread 2
 32 0xd44e9700  -1 00040600 12/02/01 0xd2dddcc0 waitport rtable_cpu 2
cpu 3 tdrunqmask 00000000 curthread 0xff8183a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8183a4  -1 00200001 12/01/00 0xff81fd30        - idle_3
 32 0xc030e9c8  -1 00010400 12/01/01 0xcbdb9d88        - softclock 3
 32 0xc0306794  -1 00000400 12/01/01 0xcfc57d88        - dsched 3
 32 0xc0305870  -1 00000400 12/01/01 0xcfc85d88        - usched 3
 32 0xc0312178  -1 00040400 12/02/01 0xd247ccc4 waitport netisr_cpu 3
 32 0xc0310fd8  -1 00040400 12/02/00 0xd24e8cb8 waitport ifnet 3
 32 0xd24d8f70  -1 00000680 12/03/01 0xd24fac9c    tqthr taskqueue 3
 32 0xc0338258  -1 00040400 12/02/01 0xd2dc8cc4 waitport udp_thread 3
 32 0xc03373d8  -1 00040400 12/02/01 0xd2dd4cc0 waitport tcp_thread 3
 32 0xd4529700  -1 00040600 12/02/01 0xd382bcc0 waitport rtable_cpu 3
CURCPU 0 CURTHREAD 0xc02fae78 (-1)
db>
 pid      lwp  uid  ppid  pgrp  pflag  lflag stat  wmesg    wchan cmd
  1 cbdfafd0    0     0     0 000200 000000  1 1                  swapper
  0 c02f9c00    0     0     0 000200 000000  2 3 conifhk c02e0e60 swapper
oops, ran out of processes early!
cpu 0 tdrunqmask 10000000 curthread 0xc02fae78 reqflags 0012
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 28 0xc02fae78  -1 00210403 28/01/02 0xd269bce4        - ithread 1

 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8003a4  -1 00200000 12/01/00 0xff807d30        - idle_0
 32 0xc03037c0  -1 00010400 12/01/01 0xcbdaad88        - ithread emerg
 32 0xc02fad64  -1 00210400 28/01/00 0xcbdadd88        - ithread 0
 32 0xc030e6d4  -1 00010400 14/02/01 0xcbdb0d18        - softclock 0
 32 0xc02ff6b4  -1 00210400 14/01/00 0xcbdbcd88        - ithread 68
 32 0xc03064c4  -1 00000400 12/01/01 0xcbdbfd88        - dsched 0
 32 0xc03055a0  -1 00000400 12/01/01 0xcfc5ad88        - usched 0
 32 0xd24d8ac0   1 00000600 10/01/01 0xd2470d34        - swapper
 32 0xc0311ec0  -1 00040400 12/02/01 0xd2473cc4 waitport netisr_cpu 0
 32 0xc0310d20  -1 00040400 12/02/00 0xd247fcb8 waitport ifnet 0
 32 0xd24d8bb0  -1 00000e80 12/03/01 0xd24ebc8c ccb_scan xpt_thrd
 32 0xc02ff5a0  -1 00210400 14/02/01 0xd24eece4        - ithread 67
 32 0xd24d8ca0  -1 00000680 12/03/01 0xd24f1c9c    tqthr taskqueue 0
 32 0xc02ff7c8  -1 00210400 14/02/01 0xd24fdce4        - ithread 69
 32 0xc02fb718  -1 00210400 28/01/00 0xd2642d88        - ithread 9
 32 0xc02fb940  -1 00210400 28/01/00 0xd2665d88        - ithread 11
 32 0xc02fb82c  -1 00210400 28/02/01 0xd2688ce4        - ithread 10
 32 0xc02fb3dc  -1 00210400 28/02/01 0xd2697ce4        - ithread 6
 32 0xc02fae78  -1 00210403 28/01/02 0xd269bce4        - ithread 1
 32 0xc02ff264  -1 00210400 14/01/00 0xd269ed88        - ithread 64
 32 0xc02fb1b4  -1 00210400 28/01/00 0xd27e2d88        - ithread 4
 32 0xd24d9330  -1 00000680 12/02/01 0xd2af5c80    rwait random
 32 0xc0337fa0  -1 00040400 12/02/01 0xd2afccc4 waitport udp_thread 0
 32 0xc0337120  -1 00040400 12/02/01 0xd2dcbcc0 waitport tcp_thread 0
 32 0xd24d9510  -1 00040600 12/02/01 0xd2dd7cc0 waitport rtable_cpu 0
 32 0xc02f9d20   0 00000080 12/02/01 0xc0377ca0  conifhk swapper
cpu 1 tdrunqmask 00000000 curthread 0xff8083a4 reqflags 0000
cpu 1 tdrunqmask 00000000 curthread 0xff8083a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
--More--
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8083a4  -1 00200001 12/01/00 0xff80fd30        - idle_1
 32 0xc030e7d0  -1 00010400 14/02/01 0xcbdb3d18        - softclock 1
 32 0xc03065b4  -1 00000400 12/01/01 0xcfc51d88        - dsched 1
 32 0xc0305690  -1 00000400 12/01/01 0xcfc5dd88        - usched 1
 32 0xc0311fa8  -1 00040400 12/02/01 0xd2476cc4 waitport netisr_cpu 1
 32 0xc0310e08  -1 00040400 12/02/00 0xd24e2cb8 waitport ifnet 1
 32 0xd24d8d90  -1 00000680 12/03/01 0xd24f4c9c    tqthr taskqueue 1
 32 0xc0338088  -1 00040400 12/02/01 0xd2affcc4 waitport udp_thread 1
 32 0xc0337208  -1 00040400 12/02/01 0xd2dcecc0 waitport tcp_thread 1
 32 0xd44a9700  -1 00040600 12/02/01 0xd2ddacc0 waitport rtable_cpu 1
cpu 2 tdrunqmask 00000000 curthread 0xff8103a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
--More--
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8103a4  -1 00200001 12/01/00 0xff817d30        - idle_2
 32 0xc030e8cc  -1 00010400 14/02/01 0xcbdb6d18        - softclock 2
 32 0xc03066a4  -1 00000400 12/01/01 0xcfc54d88        - dsched 2
 32 0xc0305780  -1 00000400 12/01/01 0xcfc82d88        - usched 2
 32 0xc0312090  -1 00040400 12/02/01 0xd2479cc4 waitport netisr_cpu 2
 32 0xc0310ef0  -1 00040400 12/02/00 0xd24e5cb8 waitport ifnet 2
 32 0xd24d8e80  -1 00000680 12/03/01 0xd24f7c9c    tqthr taskqueue 2
 32 0xc0338170  -1 00040400 12/02/01 0xd2dc5cc4 waitport udp_thread 2
 32 0xc03372f0  -1 00040400 12/02/01 0xd2dd1cc0 waitport tcp_thread 2
 32 0xd44e9700  -1 00040600 12/02/01 0xd2dddcc0 waitport rtable_cpu 2
cpu 3 tdrunqmask 00000000 curthread 0xff8183a4 reqflags 0000
    INCOMING IPIQS:
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
--More--
 tdq     thread pid    flags pri/cs/mp        sp    wmesg comm
 32 0xff8183a4  -1 00200001 12/01/00 0xff81fd30        - idle_3
 32 0xc030e9c8  -1 00010400 12/01/01 0xcbdb9d88        - softclock 3
 32 0xc0306794  -1 00000400 12/01/01 0xcfc57d88        - dsched 3
 32 0xc0305870  -1 00000400 12/01/01 0xcfc85d88        - usched 3
 32 0xc0312178  -1 00040400 12/02/01 0xd247ccc4 waitport netisr_cpu 3
 32 0xc0310fd8  -1 00040400 12/02/00 0xd24e8cb8 waitport ifnet 3
 32 0xd24d8f70  -1 00000680 12/03/01 0xd24fac9c    tqthr taskqueue 3
 32 0xc0338258  -1 00040400 12/02/01 0xd2dc8cc4 waitport udp_thread 3
 32 0xc03373d8  -1 00040400 12/02/01 0xd2dd4cc0 waitport tcp_thread 3
 32 0xd4529700  -1 00040600 12/02/01 0xd382bcc0 waitport rtable_cpu 3
CURCPU 0 CURTHREAD 0xc02fae78 (-1)

_____________________________________________________
DragonFly issue tracker <bugs at lists.dragonflybsd.org>
<https://bugs.dragonflybsd.org/issue1077>
_____________________________________________________





More information about the Bugs mailing list