3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Christian Kujau]

Hi,

while trying to upgrade from 3.4.0 to 3.5.0-rc5 on this Powerbook G4 
(powerpc 32 bit), this happens during booting:

--------------
usb 2-1: new full-speed USB device number 4 using ohci_hcd
SCSI subsystem initialized
scsi0 : SBP-2 IEEE-1394
scsi1 : SBP-2 IEEE-1394
firewire_sbp2 fw1.0: logged in to LUN 0000 (0 retries)
scsi 0:0:0:0: Direct-Access     Ext Hard  Disk            PQ: 0 ANSI: 4

[25 seconds later]

BUG: soft lockup - CPU#0 stuck for 22s! [modprobe:695]
Modules linked in: sd_mod arc4 firewire_sbp2 scsi_mod b43 mac80211 cfg80211
irq event stamp: 56975532
hardirqs last  enabled at (56975531): <c053aea8>] _raw_spin_unlock_irqrestore+0x40/0x9c
hardirqs last disabled at (56975532): <c0011688>] reenable_mmu+0x40/0x98
softirqs last  enabled at (56966080): <c000f3f0>] call_do_softirq+0x14/0x24
softirqs last disabled at (56966073): <c000f3f0>] call_do_softirq+0x14/0x24
NIP: c053aeac LR: c053aea8 CTR: c009d510
REGS: eed1fd40 TRAP: 0901   Not tainted  (3.5.0-rc5)
MSR: 00009032 ME IR DR RI > CR: 24042484  XER: 00000000
TASK = eece5c00[695] 'modprobe' THREAD: eed1e000 #012GPR00: 012GPR08: 
NIP [c053aeac] _raw_spin_unlock_irqrestore+0x44/0x9c
LR [c053aea8] _raw_spin_unlock_irqrestore+0x40/0x9c
Call Trace:
[eed1fdf0] [c053aea8] _raw_spin_unlock_irqrestore+0x40/0x9c (unreliable)
[eed1fe00] [c005fbd8] lowest_in_progress+0xa8/0xcc
[eed1fe20] [c005fc5c] async_synchronize_cookie_domain+0x60/0x1f4
[eed1fe70] [c005fe3c] async_synchronize_full+0x3c/0x74
[eed1fe90] [c0083908] sys_init_module+0x174/0x1168
[eed1ff40] [c00117e8] ret_from_syscall+0x0/0x38
--- Exception: c01 at 0xff5d694#012[   36.339234]     LR = 0x100041c0
Instruction dump:
38630010 
73e08000 
--------------

This is what netconsole was able to receive. I've made a screenshot from 
another boot with the same vanilla 3.5.0-rc5 (today's mainline) and put it 
(along with some more details) up here: 

  http://nerdbynature.de/bits/3.5.0-rc5/soft_lockup/

Any ideas?

Thanks,
Christian.
-- 
BOFH excuse #436:

Daemon escaped from pentagram

Re: 3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Benjamin Herrenschmidt]

On Sun, 2012-07-01 at 00:18 -0700, Christian Kujau wrote:
> Hi,
> 
> while trying to upgrade from 3.4.0 to 3.5.0-rc5 on this Powerbook G4 
> (powerpc 32 bit), this happens during booting:
> 
> --------------
> usb 2-1: new full-speed USB device number 4 using ohci_hcd
> SCSI subsystem initialized
> scsi0 : SBP-2 IEEE-1394
> scsi1 : SBP-2 IEEE-1394
> firewire_sbp2 fw1.0: logged in to LUN 0000 (0 retries)
> scsi 0:0:0:0: Direct-Access     Ext Hard  Disk            PQ: 0 ANSI: 4

Interesting... I observed something roughly similar on a dual G4
the other day associated with a 30s to 1mn pause during boot. RCU
was complaining loudly.

In my case, it did continue booting normally, is that the case for you ?

Also if I compile the kernel without CONFIG_SMP, it did go away as well,
do you observe that too ?

I don't have a spare cycle to investigate this problem this week I'm
afraid, it might help if you could try a bisection though.

Cheers,
Ben.

Re: 3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Christian Kujau]

On Mon, 2 Jul 2012 at 14:50, Benjamin Herrenschmidt wrote:
> Interesting... I observed something roughly similar on a dual G4
> the other day associated with a 30s to 1mn pause during boot. RCU
> was complaining loudly.
> 
> In my case, it did continue booting normally, is that the case for you ?

No, in my case it stopped booting, though there was no "panic" message.

> Also if I compile the kernel without CONFIG_SMP, it did go away as well,
> do you observe that too ?

This PoweBook G4 is UP anyway and I'm always building w/o CONFIG_SMP.

> I don't have a spare cycle to investigate this problem this week I'm
> afraid, it might help if you could try a bisection though.

Yes, I've started a bisection already and will report back.

FYI, I've put a netconsole-log below from another boot (in the 
middle of a bisection), the instruction dump is slightly more complete.

Thanks!
Christian.

[   40.345973] BUG: soft lockup - CPU#0 stuck for 22s! [modprobe:691]
[   40.347737] Modules linked in: sd_mod arc4 firewire_sbp2 scsi_mod b43 mac80211 cfg80211
[   40.349443] irq event stamp: 65779272
[   40.351156] hardirqs last  enabled at (65779271): [<c052fae8>] _raw_spin_unlock_irqrestore+0x40/0x9c
[   40.352911] hardirqs last disabled at (65779272): [<c0011648>] reenable_mmu+0x40/0x98
[   40.354649] softirqs last  enabled at (65769864): [<c000f3c0>] call_do_softirq+0x14/0x24
[   40.356386] softirqs last disabled at (65769857): [<c000f3c0>] call_do_softirq+0x14/0x24
[   40.358098] NIP: c005ee64 LR: c005ee54 CTR: c009c578
[   40.359788] REGS: eece3d70 TRAP: 0901   Not tainted  (3.4.0-05609-gc80ddb5)
[   40.361517] MSR: 00009032 <EE,ME,IR,DR,RI>  CR: 24042482  XER: 00000000
[   40.363261] TASK = eec30b80[691] 'modprobe' THREAD: eece2000#012
[   40.363261] GPR00: c005ee54 eece3e20 eec30b80 00000000 00000002 c005edd0 00000000 00000000 #012
[   40.363261] GPR08: 00000000 eece2000 00000000 00000001 24042488 
[   40.368310] NIP [c005ee64] async_synchronize_cookie_domain+0x70/0x1f4
[   40.369978] LR [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4
[   40.371614] Call Trace:
[   40.373226] [eece3e20] [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4 (unreliable)
[   40.374884] [eece3e70] [c005f034] async_synchronize_full+0x3c/0x74
[   40.376524] [eece3e90] [c0082790] sys_init_module+0x178/0x1164
[   40.378151] [eece3f40] [c00117a8] ret_from_syscall+0x0/0x38
[   40.379783] --- Exception: c01 at 0xff5d694#012[   40.379783]     LR = 0x100041c0
[   40.382928] Instruction dump:
[   40.384489] 93a10044 419e0014 3d20c06f 8009d020 2f800000 419e017c 7fc3f378 4bfffed9 
[   40.386081] 7f9f1840 419d0060 7f9f1800 419e0050 <881bf008> 2f800000 419e0014 3d20c06f


-- 
BOFH excuse #241:

_Rosin_ core solder? But...

Re: 3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Paul E. McKenney]

On Sun, Jul 01, 2012 at 11:30:40PM -0700, Christian Kujau wrote:
> On Mon, 2 Jul 2012 at 14:50, Benjamin Herrenschmidt wrote:
> > Interesting... I observed something roughly similar on a dual G4
> > the other day associated with a 30s to 1mn pause during boot. RCU
> > was complaining loudly.
> > 
> > In my case, it did continue booting normally, is that the case for you ?
> 
> No, in my case it stopped booting, though there was no "panic" message.
> 
> > Also if I compile the kernel without CONFIG_SMP, it did go away as well,
> > do you observe that too ?
> 
> This PoweBook G4 is UP anyway and I'm always building w/o CONFIG_SMP.
> 
> > I don't have a spare cycle to investigate this problem this week I'm
> > afraid, it might help if you could try a bisection though.
> 
> Yes, I've started a bisection already and will report back.
> 
> FYI, I've put a netconsole-log below from another boot (in the 
> middle of a bisection), the instruction dump is slightly more complete.

Li Zhong posted a patch to fix this (async_synchronize_full() below):

	https://lkml.org/lkml/2012/7/2/32

This gets rid of this problem on my UP testing.  If it works for you,
please give Li Zhong a Tested-by.

							Thanx, Paul

> Thanks!
> Christian.
> 
> [   40.345973] BUG: soft lockup - CPU#0 stuck for 22s! [modprobe:691]
> [   40.347737] Modules linked in: sd_mod arc4 firewire_sbp2 scsi_mod b43 mac80211 cfg80211
> [   40.349443] irq event stamp: 65779272
> [   40.351156] hardirqs last  enabled at (65779271): [<c052fae8>] _raw_spin_unlock_irqrestore+0x40/0x9c
> [   40.352911] hardirqs last disabled at (65779272): [<c0011648>] reenable_mmu+0x40/0x98
> [   40.354649] softirqs last  enabled at (65769864): [<c000f3c0>] call_do_softirq+0x14/0x24
> [   40.356386] softirqs last disabled at (65769857): [<c000f3c0>] call_do_softirq+0x14/0x24
> [   40.358098] NIP: c005ee64 LR: c005ee54 CTR: c009c578
> [   40.359788] REGS: eece3d70 TRAP: 0901   Not tainted  (3.4.0-05609-gc80ddb5)
> [   40.361517] MSR: 00009032 <EE,ME,IR,DR,RI>  CR: 24042482  XER: 00000000
> [   40.363261] TASK = eec30b80[691] 'modprobe' THREAD: eece2000#012
> [   40.363261] GPR00: c005ee54 eece3e20 eec30b80 00000000 00000002 c005edd0 00000000 00000000 #012
> [   40.363261] GPR08: 00000000 eece2000 00000000 00000001 24042488 
> [   40.368310] NIP [c005ee64] async_synchronize_cookie_domain+0x70/0x1f4
> [   40.369978] LR [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4
> [   40.371614] Call Trace:
> [   40.373226] [eece3e20] [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4 (unreliable)
> [   40.374884] [eece3e70] [c005f034] async_synchronize_full+0x3c/0x74
> [   40.376524] [eece3e90] [c0082790] sys_init_module+0x178/0x1164
> [   40.378151] [eece3f40] [c00117a8] ret_from_syscall+0x0/0x38
> [   40.379783] --- Exception: c01 at 0xff5d694#012[   40.379783]     LR = 0x100041c0
> [   40.382928] Instruction dump:
> [   40.384489] 93a10044 419e0014 3d20c06f 8009d020 2f800000 419e017c 7fc3f378 4bfffed9 
> [   40.386081] 7f9f1840 419d0060 7f9f1800 419e0050 <881bf008> 2f800000 419e0014 3d20c06f
> 
> 
> -- 
> BOFH excuse #241:
> 
> _Rosin_ core solder? But...
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: 3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Christian Kujau]

On Mon, 2 Jul 2012 at 09:45, Paul E. McKenney wrote:
> On Sun, Jul 01, 2012 at 11:30:40PM -0700, Christian Kujau wrote:
> > On Mon, 2 Jul 2012 at 14:50, Benjamin Herrenschmidt wrote:
> > > Interesting... I observed something roughly similar on a dual G4
> > > the other day associated with a 30s to 1mn pause during boot. RCU
> > > was complaining loudly.
> > > 
> > > In my case, it did continue booting normally, is that the case for you ?
> > 
> > No, in my case it stopped booting, though there was no "panic" message.
> > 
> > > Also if I compile the kernel without CONFIG_SMP, it did go away as well,
> > > do you observe that too ?
> > 
> > This PoweBook G4 is UP anyway and I'm always building w/o CONFIG_SMP.
> > 
> > > I don't have a spare cycle to investigate this problem this week I'm
> > > afraid, it might help if you could try a bisection though.
> > 
> > Yes, I've started a bisection already and will report back.
> > 
> > FYI, I've put a netconsole-log below from another boot (in the 
> > middle of a bisection), the instruction dump is slightly more complete.
> 
> Li Zhong posted a patch to fix this (async_synchronize_full() below):
> 
> 	https://lkml.org/lkml/2012/7/2/32
> 
> This gets rid of this problem on my UP testing.  If it works for you,
> please give Li Zhong a Tested-by.

Great! When applied to -rc5 Li's patch fixes it for my powerpc UP system 
as well.

  Tested-by: Christian Kujau <lists@nerdbynature.de>

Many thanks!
Christian.

> > [   40.345973] BUG: soft lockup - CPU#0 stuck for 22s! [modprobe:691]
> > [   40.347737] Modules linked in: sd_mod arc4 firewire_sbp2 scsi_mod b43 mac80211 cfg80211
> > [   40.349443] irq event stamp: 65779272
> > [   40.351156] hardirqs last  enabled at (65779271): [<c052fae8>] _raw_spin_unlock_irqrestore+0x40/0x9c
> > [   40.352911] hardirqs last disabled at (65779272): [<c0011648>] reenable_mmu+0x40/0x98
> > [   40.354649] softirqs last  enabled at (65769864): [<c000f3c0>] call_do_softirq+0x14/0x24
> > [   40.356386] softirqs last disabled at (65769857): [<c000f3c0>] call_do_softirq+0x14/0x24
> > [   40.358098] NIP: c005ee64 LR: c005ee54 CTR: c009c578
> > [   40.359788] REGS: eece3d70 TRAP: 0901   Not tainted  (3.4.0-05609-gc80ddb5)
> > [   40.361517] MSR: 00009032 <EE,ME,IR,DR,RI>  CR: 24042482  XER: 00000000
> > [   40.363261] TASK = eec30b80[691] 'modprobe' THREAD: eece2000#012
> > [   40.363261] GPR00: c005ee54 eece3e20 eec30b80 00000000 00000002 c005edd0 00000000 00000000 #012
> > [   40.363261] GPR08: 00000000 eece2000 00000000 00000001 24042488 
> > [   40.368310] NIP [c005ee64] async_synchronize_cookie_domain+0x70/0x1f4
> > [   40.369978] LR [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4
> > [   40.371614] Call Trace:
> > [   40.373226] [eece3e20] [c005ee54] async_synchronize_cookie_domain+0x60/0x1f4 (unreliable)
> > [   40.374884] [eece3e70] [c005f034] async_synchronize_full+0x3c/0x74
> > [   40.376524] [eece3e90] [c0082790] sys_init_module+0x178/0x1164
> > [   40.378151] [eece3f40] [c00117a8] ret_from_syscall+0x0/0x38
> > [   40.379783] --- Exception: c01 at 0xff5d694#012[   40.379783]     LR = 0x100041c0
> > [   40.382928] Instruction dump:
> > [   40.384489] 93a10044 419e0014 3d20c06f 8009d020 2f800000 419e017c 7fc3f378 4bfffed9 
> > [   40.386081] 7f9f1840 419d0060 7f9f1800 419e0050 <881bf008> 2f800000 419e0014 3d20c06f

-- 
BOFH excuse #280:

Traceroute says that there is a routing problem in the backbone.  It's not our problem.

Re: 3.5.0-rc5: BUG: soft lockup - CPU#0 stuck for 22s

[Christian Kujau]

On Mon, 2 Jul 2012 at 14:50, Benjamin Herrenschmidt wrote:
> > while trying to upgrade from 3.4.0 to 3.5.0-rc5 on this Powerbook G4 
> > (powerpc 32 bit), this happens during booting:
> > 
> > --------------
> > usb 2-1: new full-speed USB device number 4 using ohci_hcd
> > SCSI subsystem initialized
> > scsi0 : SBP-2 IEEE-1394
> > scsi1 : SBP-2 IEEE-1394
> > firewire_sbp2 fw1.0: logged in to LUN 0000 (0 retries)
> > scsi 0:0:0:0: Direct-Access     Ext Hard  Disk            PQ: 0 ANSI: 4
> 
> Interesting... I observed something roughly similar on a dual G4
> the other day associated with a 30s to 1mn pause during boot. RCU
> was complaining loudly.
> 
> In my case, it did continue booting normally, is that the case for you ?
> 
> Also if I compile the kernel without CONFIG_SMP, it did go away as well,
> do you observe that too ?
> 
> I don't have a spare cycle to investigate this problem this week I'm
> afraid, it might help if you could try a bisection though.

OK, after a git-bisect, the following commit has been identified as bad:

---------------------------
a7a20d103994fd760766e6c9d494daa569cbfe06 is the first bad commit
commit a7a20d103994fd760766e6c9d494daa569cbfe06
Author: Dan Williams <dan.j.williams@intel.com>
Date:   Thu Mar 22 17:05:11 2012 -0700

    [SCSI] sd: limit the scope of the async probe domain
---------------------------

Unfortunately it's not possible to "git revert" just this single commit:

------------
$ git revert a7a20d103994fd760766e6c9d494daa569cbfe06
error: could not revert a7a20d1... [SCSI] sd: limit the scope of the async probe domain
hint: after resolving the conflicts, mark the corrected paths  
hint: with 'git add <paths>' or 'git rm <paths>'
hint: and commit the result with 'git commit'
------------

I _guess_ this is probably due to:

------------
commit ea80dadec7a06889562b478cf0b87afbe62b7ac8
Author: James Bottomley <JBottomley@Parallels.com>
Date:   Wed Jun 6 14:54:13 2012 +0900

    [SCSI] Fix sd_probe_domain config problem
------------

The whole git-bisect log, dmesg, .config and so on is here:

  http://nerdbynature.de/bits/3.5.0-rc5/soft_lockup/

Anyone got an idea how to go from here?

Thanks,
Christian.
-- 
BOFH excuse #380:

Operators killed when huge stack of backup tapes fell over.