some domU preempt kernel oops

some domU preempt kernel oops

[Ted Kaczmarek]

dom0 does not have preempt, 
domU preempt enabled
FC4 dom0

Is this even a valid setup in Xen?

If not ignore the rest of this post :-)

changeset:   7366:43677823b395
tag:         tip
user:        kaf24@firebug.cl.cam.ac.uk
date:        Thu Oct 13 06:12:11 2005 +0100
summary:     Fix client dependency on libxenstore in Makefile.


xen conole output

stop_this_cpu disable_local_APIC

smp_send_stop disable_local_APIC

smp_send_stop disable_local_APIC

stop_this_cpu disable_local_APIC

stop_this_cpu disable_local_APIC

smp_send_stop disable_local_APIC


1st attempt to bring online domu

register_blkdev: cannot get major 8 for sd
<1>Unable to handle kernel NULL pointer dereferencexen_blk: can't get
major 8 with name sd
raid5: automatically using best checksumming function: pIII_sse
 at virtual address 00000118
 printing eip:
c02536c8
*pde = ma 00000000 pa 55555000
Oops: 0002 [#1]
PREEMPT SMP
Modules linked in:
CPU:    0
EIP:    0061:[<c02536c8>]    Not tainted VLI
EFLAGS: 00010286   (2.6.12-xenU)
EIP is at blk_start_queue+0x8/0x60
eax: 00000000   ebx: 00000000   ecx: c043b060   edx: 00000000
esi: c046e004   edi: c0519dc3   ebp: c0055f9c   esp: c0055f28
ds: 007b   es: 007b   ss: 0069
Process xenwatch (pid: 10, threadinfo=c0055000 task=c004f590)
Stack: c025d512 c046e000 c025f79e 00000000 c046e000 c02602ac c046e000
00000000
       00000801 00000000 00000200 c046e000 c0333f65 c0055f78 c03419b3
c0333b55
       c0055f70 00000000 00000200 00b00000 00000000 c02601c0 c0519d60
c0055000
Call Trace:
 [<c025d512>] xenbus_dev_ok+0x32/0x60
 [<c025f79e>] kick_pending_request_queues+0x1e/0x40
 [<c02602ac>] watch_for_status+0xec/0x140
 [<c02601c0>] watch_for_status+0x0/0x140
 [<c025db27>] xenwatch_thread+0x87/0x140
 [<c0138020>] autoremove_wake_function+0x0/0x60
 [<c025daa0>] xenwatch_thread+0x0/0x140
 [<c0137a97>] kthread+0xc7/0xd0
 [<c01379d0>] kthread+0x0/0xd0
 [<c010791d>] kernel_thread_helper+0x5/0x18
Code: 00 00 00 8b 44 24 04 8b 48 6c ff e1 8d b4 26 00 00 00 00 81 44 24
04 b0 00 00 00 e9 53 27 00 00 8d 76 00 53 83 ec 04 8b 5c 24 0c <f0> 0f
ba b3 18 01 00 00 02 f0 0f ba ab 18 01 00 00 06 19 c0 85
 <6>note: xenwatch[10] exited with preempt_count 2
   pIII_sse  :  2427.200 MB/sec
raid5: using function: pIII_sse (2427.200 MB/sec)


This one happened when I still had the domU's /root mounted on the domO.

Unable to handle kernel NULL pointer dereference at virtual address
00000000
 printing eip:
c02436e5
*pde = ma 00000000 pa 55555000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    0
EIP:    0061:[<c02436e5>]    Not tainted VLI
EFLAGS: 00010246   (2.6.12-xenU)
EIP is at visual_init+0x65/0x190
eax: 00000000   ebx: c0532a00   ecx: 00000000   edx: 00000000
esi: c0532a00   edi: 00000000   ebp: 00000000   esp: c0029e80
ds: 007b   es: 007b   ss: 0069
Process init (pid: 1, threadinfo=c0029000 task=c002aa40)
Stack: c0029000 c0304cca c03a9000 c0235bff 00000000 c0532a00 c0532b54
c02438af
       c0532a00 00000000 00000001 ffffffed c03a9000 00000000 00000000
00400000
       c0246998 00000000 ffffffed c057be20 00000001 c0236a47 c03a9000
c057be20
Call Trace:
 [<c0304cca>] _spin_unlock_irqrestore+0x5a/0x60
 [<c03a9000>] inflate_codes+0x60/0x480
 [<c0235bff>] init_dev+0x2af/0x600
 [<c02438af>] vc_allocate+0x9f/0x150
 [<c03a9000>] inflate_codes+0x60/0x480
 [<c0246998>] con_open+0x48/0xb0
 [<c0236a47>] tty_open+0x117/0x300
 [<c03a9000>] inflate_codes+0x60/0x480
 [<c03a9000>] inflate_codes+0x60/0x480
 [<c0236930>] tty_open+0x0/0x300
 [<c01711a6>] chrdev_open+0xd6/0x1d0
 [<c0165f1a>] dentry_open+0xfa/0x270
 [<c0165e0c>] filp_open+0x5c/0x70
 [<c0166138>] get_unused_fd+0xa8/0xd0
 [<c0166289>] sys_open+0x49/0x90
 [<c0109b45>] syscall_call+0x7/0xb
Code: 01 00 00 8b 46 08 ff 4e 14 a8 08 0f 85 1f 01 00 00 8b 15 ec 1f 40
c0 89 53 28 8b 04 bd 40 31 40 c0 85 c0 74 03 89 43 28 8b 53 28 <8b> 32
85 f6 74 3a 89 34 24 e8 8d a7 ef ff 85 c0 0f 84 ad 00 00
 <0>Kernel panic - not syncing: Attempted to kill init!

This happens on all following attempts

Unable to handle kernel NULL pointer dereference at virtual address
00000000
 printing eip:
c02436e5
*pde = ma 00000000 pa 55555000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    0
EIP:    0061:[<c02436e5>]    Not tainted VLI
EFLAGS: 00010246   (2.6.12-xenU)
EIP is at visual_init+0x65/0x190
eax: 00000000   ebx: c03c9e00   ecx: 00000000   edx: 00000000
esi: c03c9e00   edi: 00000000   ebp: 00000000   esp: c0029e80
ds: 007b   es: 007b   ss: 0069
Process init (pid: 1, threadinfo=c0029000 task=c002aa40)
Stack: 00000000 c0081cfc c0541000 c0235bff 00000000 c03c9e00 c03c9f54
c02438af
       c03c9e00 00000000 00000001 ffffffed c0541000 00000000 00000000
00400000
       c0246998 00000000 ffffffed c053e880 00000001 c0236a47 c0541000
c053e880
Call Trace:
 [<c0235bff>] init_dev+0x2af/0x600
 [<c02438af>] vc_allocate+0x9f/0x150
 [<c0246998>] con_open+0x48/0xb0
 [<c0236a47>] tty_open+0x117/0x300
 [<c0236930>] tty_open+0x0/0x300
 [<c03bf4ac>] af_unix_init+0x7c/0x80
 [<c01711a6>] chrdev_open+0xd6/0x1d0
 [<c03bf4ac>] af_unix_init+0x7c/0x80
 [<c03bf4ac>] af_unix_init+0x7c/0x80
 [<c0165f1a>] dentry_open+0xfa/0x270
 [<c03bf4ac>] af_unix_init+0x7c/0x80
 [<c0165e0c>] filp_open+0x5c/0x70
 [<c0166138>] get_unused_fd+0xa8/0xd0
 [<c0166289>] sys_open+0x49/0x90
 [<c0109b45>] syscall_call+0x7/0xb
Code: 01 00 00 8b 46 08 ff 4e 14 a8 08 0f 85 1f 01 00 00 8b 15 ec 1f 40
c0 89 53 28 8b 04 bd 40 31 40 c0 85 c0 74 03 89 43 28 8b 53 28 <8b> 32
85 f6 74 3a 89 34 24 e8 8d a7 ef ff 85 c0 0f 84 ad 00 00
 <0>Kernel panic - not syncing: Attempted to kill init!



Keir knows where to get objdumps if needed.

Regards,
Ted