kvm poweroff deadlock

kvm poweroff deadlock

[Zdenek Kabelac]

Hi

While testing latest  2.6.29-rc  inside qemu-kvm guest I've got this
deadlock while executing poweroff:

sd 0:0:0:0: [sda] Stopping disk

ACPI: Preparing to enter system sleep state S5

Disabling non-boot CPUs ...

BUG: spinlock lockup on CPU#0, halt/2389, ffff88001dc0c3c0

Pid: 2389, comm: halt Not tainted 2.6.28 #103

Call Trace:

 [<ffffffff803aa3f8>] _raw_spin_lock+0x158/0x170

 [<ffffffff80538924>] _spin_lock_irqsave+0x64/0x90

 [<ffffffff80259268>] ? __queue_work+0x18/0x90

 [<ffffffff80259268>] __queue_work+0x18/0x90

 [<ffffffff8025936d>] queue_work_on+0x4d/0x60

 [<ffffffff80286ca7>] __stop_machine+0xe7/0x140

 [<ffffffff80523930>] _cpu_down+0x120/0x2f0

 [<ffffffff802469da>] disable_nonboot_cpus+0x8a/0x120

 [<ffffffff802571d1>] kernel_power_off+0x21/0x50

 [<ffffffff802574b0>] sys_reboot+0x110/0x230

 [<ffffffff8023c11f>] ? finish_task_switch+0x5f/0x120

 [<ffffffff8023c0c0>] ? finish_task_switch+0x0/0x120

 [<ffffffff8053540b>] ? thread_return+0x3d/0x702

 [<ffffffff80254509>] ? sys_kill+0xd9/0x200

 [<ffffffff802544ad>] ? sys_kill+0x7d/0x200

 [<ffffffff805380dc>] ? lockdep_sys_exit_thunk+0x35/0x67

 [<ffffffff80538066>] ? trace_hardirqs_on_thunk+0x3a/0x3f

 [<ffffffff80231a48>] sysenter_dispatch+0x7/0x2c

 [<ffffffff80538066>] ? trace_hardirqs_on_thunk+0x3a/0x3f



Zdenek

Re: kvm poweroff deadlock

[Zdenek Kabelac]

2009/1/6 Zdenek Kabelac <zdenek.kabelac@gmail.com>:
> Hi
>
> While testing latest  2.6.29-rc  inside qemu-kvm guest I've got this
> deadlock while executing poweroff:
>
> sd 0:0:0:0: [sda] Stopping disk
>
> ACPI: Preparing to enter system sleep state S5
>

Hi

This might be related to my previous post - but today with
commit: ede6f5aea054d3fb67c78857f7abdee602302043

I'm getting this oops inside qemu guest while doing poweroff:

ACPI: Preparing to enter system sleep state S5

Disabling non-boot CPUs ...

BUG: unable to handle kernel paging request at 000077ffe1adad5f

IP: [<ffffffff802592d4>] queue_work_on+0x44/0x60

PGD 0

Oops: 0000 [#1] SMP

last sysfs file:
/sys/devices/pci0000:00/0000:00:01.1/host1/target1:0:0/1:0:0:0/block/sdc/removable

CPU 0

Modules linked in: nfs lockd nfs_acl auth_rpcgss sunrpc autofs4 ipv6
dm_crypt crypto_blkcipher crypto_algapi dm_mod loop rtc_cmos i2c_piix4
uhci_hcd rtc_core rtc_lib 8139cp psmouse serio_raw evdev floppy
i2c_core mii usbcore button

Pid: 4168, comm: halt Not tainted 2.6.28 #104

RIP: 0010:[<ffffffff802592d4>]  [<ffffffff802592d4>] queue_work_on+0x44/0x60

RSP: 0018:ffff88001e47dd68  EFLAGS: 00010246

RAX: 000077ffe1adad5f RBX: ffff88001dd7e040 RCX: ffff88001dd7e040

RDX: 0000000000000000 RSI: ffff88001dd7e040 RDI: 0000000000000000

RBP: ffff88001e47dd68 R08: 0000000000000000 R09: ffffffff806ce8b8

R10: 0000000000000003 R11: 0000000000000000 R12: 0000000000000000

R13: 0000000000000000 R14: ffffffff806ce8b8 R15: 0000000000000001

FS:  0000000000000000(0000) GS:ffffffff8090d040(0063) knlGS:00000000f7e626b0

CS:  0010 DS: 002b ES: 002b CR0: 000000008005003b

CR2: 000077ffe1adad5f CR3: 000000001df5c000 CR4: 00000000000006e0

DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000

DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400

Process halt (pid: 4168, threadinfo ffff88001e47c000, task ffff88001e91a210)

Stack:

 ffff88001e47dd98 ffffffff80286607 0000000000000010 0000000000000001

 00000000ffffffea ffffffff806ce8b8 ffff88001e47ddf8 ffffffff80524620

 0000000000000010 0000000000000001 000000000000000f 0000001d4321fedc

Call Trace:

 [<ffffffff80286607>] __stop_machine+0xe7/0x140

 [<ffffffff80524620>] _cpu_down+0x120/0x2f0

 [<ffffffff802468fa>] disable_nonboot_cpus+0x8a/0x120

 [<ffffffff80257141>] kernel_power_off+0x21/0x50

 [<ffffffff80257420>] sys_reboot+0x110/0x230

 [<ffffffff8023c03f>] ? finish_task_switch+0x5f/0x110

 [<ffffffff8023bfe0>] ? finish_task_switch+0x0/0x110

 [<ffffffff8053618b>] ? thread_return+0x3d/0x702

 [<ffffffff80254439>] ? sys_kill+0xd9/0x200

 [<ffffffff802543dd>] ? sys_kill+0x7d/0x200

 [<ffffffff8020d1a9>] ? retint_swapgs+0xe/0x13

 [<ffffffff8026f18a>] ? trace_hardirqs_on_caller+0x16a/0x1d0

 [<ffffffff80538e0e>] ? trace_hardirqs_on_thunk+0x3a/0x3f

 [<ffffffff802319c8>] sysenter_dispatch+0x7/0x2c

 [<ffffffff80538e0e>] ? trace_hardirqs_on_thunk+0x3a/0x3f

Code: 1f 84 00 00 00 00 00 48 8d 41 08 48 39 41 08 75 2c 44 8b 46 20
48 8b 06 45 85 c0 48 f7 d0 0f 45 3d 02 56 47 00 48 89 ce 48 63 d7 <48>
8b 3c d0 e8 e3 fe ff ff ba 01 00 00 00 c9 89 d0 c3 0f 0b eb

RIP  [<ffffffff802592d4>] queue_work_on+0x44/0x60

 RSP <ffff88001e47dd68>

CR2: 000077ffe1adad5f

---[ end trace 862d0eb04a8b6d35 ]---

PM: Removing info for No Bus:vcs1

PM: Removing info for No Bus:vcsa1


Zdenek

Re: kvm poweroff deadlock

[Zdenek Kabelac]

2009/1/7 Zdenek Kabelac <zdenek.kabelac@gmail.com>:
> 2009/1/6 Zdenek Kabelac <zdenek.kabelac@gmail.com>:
>> Hi
>>
>> While testing latest  2.6.29-rc  inside qemu-kvm guest I've got this
>> deadlock while executing poweroff:
>>
>> sd 0:0:0:0: [sda] Stopping disk
>>
>> ACPI: Preparing to enter system sleep state S5
>>
>
> Hi
>
> This might be related to my previous post - but today with
> commit: ede6f5aea054d3fb67c78857f7abdee602302043
>
> I'm getting this oops inside qemu guest while doing poweroff:
>
> ACPI: Preparing to enter system sleep state S5
>
> Disabling non-boot CPUs ...
>
> BUG: unable to handle kernel paging request at 000077ffe1adad5f
>
> IP: [<ffffffff802592d4>] queue_work_on+0x44/0x60
>

My problem seems to be fixed by this patch:


http://marc.info/?l=linux-kernel&m=123131978225429&w=2

Zdenek