Home » Mailing list archives » linux-kernel » 2007 » December » 9

Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41

Previous thread: Re: [PATCH 3/3] [UDP6]: Counter increment on BH mode by Herbert Xu on Monday, December 3, 2007 - 7:54 am. (6 messages)

Next thread: Re: x86: optimize page faults like all other achitectures and kill notifier cruft by Jan Beulich on Monday, December 3, 2007 - 9:31 am. (2 messages)
[view original message]
From: Toralf <toralf.foerster@...>
To: <linux-kernel@...>
Subject: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 3, 2007 - 9:19 am

Today I tried to hibernate my ThinkPad but got the following :


Dec  3 14:13:17 n22 ntpd[5504]: time reset +0.204868 s

Dec  3 14:14:06 n22 acpid: received event "ibm/hotkey HKEY 00000080 0000100=

c"

Dec  3 14:14:06 n22 acpid: notifying client 4693[0:0]

Dec  3 14:14:06 n22 acpid: executing action "/etc/acpi/default.sh ibm/hotke=

y HKEY 00000080 0000100c"

Dec  3 14:14:07 n22 swsusp: Marking nosave pages: 000000000009f000 - 000000=

0000100000

Dec  3 14:14:07 n22 swsusp: Basic memory bitmaps created

Dec  3 14:14:29 n22 c420 00000000=20

Dec  3 14:14:29 n22 dca29fb8 dca29f90 d30ad4c4 0851bf48 0851bf48 0851bfcb d=

ca29fb8 c01036f1=20

Dec  3 14:14:29 n22 Call Trace:

Dec  3 14:14:29 n22 [<c0135fe5>] refrigerator+0x55/0x80

Dec  3 14:14:29 n22 [<c0121b2a>] get_signal_to_deliver+0x3ca/0x3f0

Dec  3 14:14:29 n22 [<c01036f1>] do_notify_resume+0x91/0x6e0

Dec  3 14:14:29 n22 [<c0113c61>] update_curr+0x141/0x150

Dec  3 14:14:29 n22 [<c01c109d>] rb_erase+0x15d/0x280

Dec  3 14:14:29 n22 [<c01136f7>] update_stats_wait_end+0xa7/0xe0

Dec  3 14:14:29 n22 [<c028031b>] schedule+0x13b/0x2e0

Dec  3 14:14:29 n22 [<c01040c2>] work_notifysig+0x13/0x19

Dec  3 14:14:29 n22 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=

=3D=3D=3D=3D=3D

Dec  3 14:14:29 n22 linux-2.6.23  t 00000000     0 15474   9515

Dec  3 14:14:29 n22 c51c1f00 00200086 011c1fb8 00000000 00000000 00000000 0=

0000000 00000005=20

Dec  3 14:14:29 n22 c51c1fb8 c0120dc5 00000085 00000000 c012174c c51c1f18 0=

0000005 00000000=20

Dec  3 14:14:29 n22 00000085 00003c72 00000000 00000000 00000000 00000000 0=

0000000 00000000=20

Dec  3 14:14:29 n22 Call Trace:

Dec  3 14:14:29 n22 [<c0120dc5>] ptrace_stop+0x95/0xf0

Dec  3 14:14:29 n22 [<c012174c>] ptrace_notify+0x5c/0x70

Dec  3 14:14:29 n22 [<c0106871>] do_syscall_trace+0xc1/0x140

Dec  3 14:14:29 n22 [<c01040f1>] syscall_trace_entry+0x11/0x30

Dec  3 14:14:29 n22 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=

=3D=3D=3...
[ message continues ]
[view original message]
From: Daniel Drake <dsd@...>
To: Toralf Förster <toralf.foerster@...>
Cc: <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 3, 2007 - 9:49 am

And the outcome was what exactly? It didn't suspend, and hard-hung? Or 


Looks like your logger did not capture all the messages. Some important

messages at the top were likely lost as a result. dmesg will have them,

but you may have to increase the dmesg log buffer size in order to

capture them all.


Did you press alt+sysrq+T or did all this come about as a result of the

suspend?


Daniel

--
[view original message]
From: Toralf <toralf.foerster@...>
To: Daniel Drake <dsd@...>
Cc: <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 3, 2007 - 1:21 pm

Hhm,


in general it seems to be a regression of kernel 2.6.23 with suspend/hibern=

ate

if a user mode linux image is runnning.


With previous kernel versions (tested gentoo-sources-2.6.22-r9 and=20

gentoo-sources-2.6.21-r4) I've no problem to either suspend or to hibernate=

 the

ThinkPad T41.


However with gentoo-sources-2.6.23-r3 the system didn't neither

suspend nor hibernate. The good news is that the system doesn't hang :-)


=2D-=20

MfG/Sincerely


Toralf F=F6rster

pgp finger print: 7B1A 07F4 EC82 0F90 D4C2 8936 872A E508 7DB6 9DA3
[view original message]
From: Rafael J. Wysocki <rjw@...>
To: Toralf <toralf.foerster@...>
Cc: Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 3, 2007 - 5:54 pm

Please file a bug report at

http://bugzilla.kernel.org/enter_bug.cgi?product=Power%20Management

against Hibernation/Suspend and add my address to the report's CC list.


Thanks,

Rafael

--
[view original message]
From: Pavel Machek <pavel@...>
To: Rafael J. Wysocki <rjw@...>
Cc: Toralf <toralf.foerster@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Sunday, December 9, 2007 - 1:41 pm

I smell refrigerator problem; can you suspend if you SIGSTOP your uml,

first?


--

(english) http://www.livejournal.com/~pavelmachek

(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

--
[view original message]
From: Toralf <toralf.foerster@...>
To: Pavel Machek <pavel@...>
Cc: Rafael J. Wysocki <rjw@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 10, 2007 - 4:41 am

No, I cannot suspend my host system if I send SIGSTOP  to the UML process f=

irst.


=2D-=20

MfG/Sincerely


Toralf F=F6rster

pgp finger print: 7B1A 07F4 EC82 0F90 D4C2 8936 872A E508 7DB6 9DA3
[view original message]
From: Rafael J. Wysocki <rjw@...>
To: Pavel Machek <pavel@...>
Cc: Toralf <toralf.foerster@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Sunday, December 9, 2007 - 5:31 pm

That might be, but the issue is apparently fixed in 2.6.24-rc and I'm

struggling to figure out how the !@#$%^ (please see

http://bugzilla.kernel.org/show_bug.cgi?id=9501).


Greetings,

Rafael

--
[view original message]
From: Pavel Machek <pavel@...>
To: Rafael J. Wysocki <rjw@...>
Cc: Toralf <toralf.foerster@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Sunday, December 9, 2007 - 5:21 pm

No idea... I'm glad its fixed :-). Maybe someone fixed uninterruptible

sleep somewhere in ptrace...

								Pavel

--

(english) http://www.livejournal.com/~pavelmachek

(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

--
[view original message]
From: Rafael J. Wysocki <rjw@...>
To: Pavel Machek <pavel@...>
Cc: Toralf <toralf.foerster@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Sunday, December 9, 2007 - 5:50 pm

Yeah, but we should find a -stable fix, if possible.


Greetings,

Rafael

--
[view original message]
From: Pavel Machek <pavel@...>
To: Rafael J. Wysocki <rjw@...>
Cc: Toralf <toralf.foerster@...>, Daniel Drake <dsd@...>, <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Sunday, December 9, 2007 - 5:37 pm

I do not think this bug is severe enough to warrant stable fix. It

does not eat data, and it should be easy to SIGSTOP and retry suspend.


								Pavel

--

(english) http://www.livejournal.com/~pavelmachek

(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

--
[view original message]
From: Toralf Förster <toralf.foerster@...>
To: Daniel Drake <dsd@...>
Cc: <linux-kernel@...>
Subject: Re: hibernation issue with kernel 2.6.23-gentoo-r3 at T41
Date: Monday, December 3, 2007 - 10:12 am

Ah, sorry - the system came back and  I could continue my work.

However I couldn't hibernate my system (I tried it 2 additional times)

as long as

my user mode linux image was running (2.6.23 UML kernel). Within UML I

ran cupsd to which firefox

was trying to make a connection. Firefox itself was waiting for a

response of the cupsd server,

the cupsd itself was generating the ssl -files under /etc/cupsd/ssl as

I pressed <Fn>+<F12> to hibernate the system, and b/c the scrren was

black and came back some secs later I had have a look into

/var/log/message and copied the outout into my first mail.


--
Previous thread: Re: [PATCH 3/3] [UDP6]: Counter increment on BH mode by Herbert Xu on Monday, December 3, 2007 - 7:54 am. (6 messages)

Next thread: Re: x86: optimize page faults like all other achitectures and kill notifier cruft by Jan Beulich on Monday, December 3, 2007 - 9:31 am. (2 messages)