Suspend stopped working on Thinkpad X200s with 2.6.35-24

Bug #698097 reported by Björn Tillenius
48
This bug affects 7 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
Medium
Canonical Kernel Team
Maverick
Invalid
Medium
Seth Forshee

Bug Description

After updating the kernel to 2.6.35-24, suspending the computer doesn't work most of the time. The screen goes blank, but the sleep LED is blinking all the time, and it doesn't suspend. It usually works to suspend the first time after a boot, but the second time always fail. Sometimes the first time fails as well.

Booting up with 2.6.35-23 I can suspend many times without any problems.

ProblemType: Bug
DistroRelease: Ubuntu 10.10
Package: linux-image-2.6.35-24-generic 2.6.35-24.42
Regression: Yes
Reproducible: Yes
ProcVersionSignature: Ubuntu 2.6.35-24.42-generic 2.6.35.8
Uname: Linux 2.6.35-24-generic x86_64
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.23.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: CONEXANT Analog [CONEXANT Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: bjorn 2652 F.... pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xf2620000 irq 46'
   Mixer name : 'Conexant CX20561 (Hermosa)'
   Components : 'HDA:14f15051,17aa20ff,00100000'
   Controls : 15
   Simple ctrls : 8
Card29.Amixer.info:
 Card hw:29 'ThinkPadEC'/'ThinkPad Console Audio Control at EC reg 0x30, fw 7XHT21WW-1.03'
   Mixer name : 'ThinkPad EC 7XHT21WW-1.03'
   Components : ''
   Controls : 1
   Simple ctrls : 1
Card29.Amixer.values:
 Simple mixer control 'Console',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [on]
Date: Thu Jan 6 17:58:39 2011
MachineType: LENOVO 74695RG
ProcCmdLine: root=UUID=311d6aa9-f2f9-4a4c-8779-51580de83e03 ro quiet splash
ProcEnviron:
 LC_TIME=en_US.UTF-8
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
RelatedPackageVersions: linux-firmware 1.38
SourcePackage: linux
UserAsoundrc:
 pcm.btheadset {
         type bluetooth
         device 00:0D:3C:6B:34:39
         profile "auto"
 }
WpaSupplicantLog:

dmi.bios.date: 12/19/2008
dmi.bios.vendor: LENOVO
dmi.bios.version: 6DET38WW (2.02 )
dmi.board.name: 74695RG
dmi.board.vendor: LENOVO
dmi.board.version: Not Available
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnLENOVO:bvr6DET38WW(2.02):bd12/19/2008:svnLENOVO:pn74695RG:pvrThinkPadX200s:rvnLENOVO:rn74695RG:rvrNotAvailable:cvnLENOVO:ct10:cvrNotAvailable:
dmi.product.name: 74695RG
dmi.product.version: ThinkPad X200s
dmi.sys.vendor: LENOVO

Revision history for this message
Björn Tillenius (bjornt) wrote :
Changed in linux (Ubuntu):
assignee: nobody → Canonical Kernel Team (canonical-kernel-team)
importance: Undecided → High
Revision history for this message
rt_davies (rt-davies) wrote :

You may want to see the workaround at:

http://us.generation-nt.com/answer/2-6-36-2-regression-suspend-lenovo-x200s-broken-due-tpm-help-201488272.html

Essentially, disable the tpm_tis kernel module.

Revision history for this message
Russell Neches (ubuntu-vort) wrote :

I observed this problem as well, although I suspect the author of the original report was just lucky that their laptop successfully suspended the first time after a boot. It looks more like it happens at random. Sometimes it would crash on the first suspend attempt, and sometimes the fifth or sixth.

Upgrading to 2.6.35-25-generic broke suspend completely; the machine now immediately wakes up.

Come on, Canonical QA. The X200s is just about the most boring-vanilla-standard hardware you could ask for. That's why I bought it. Surely someone could test if suspend works before you push a new kernel?

Revision history for this message
Brian (x-brian) wrote :

With my Dell E6410 S3 suspend fails with 2.6.35-25-generic also:

$ uname -srvmo
Linux 2.6.35-25-generic #44-Ubuntu SMP Fri Jan 21 17:40:44 UTC 2011 x86_64 GNU/Linux

S3 suspend tries but immediately wakes.

Revision history for this message
Brian (x-brian) wrote :

S3 suspend worked fine the last time I tried it on Dell E6410 with 2.6.35-25-generic. Seems like an intermittent problem.

Revision history for this message
Andy Walker (walkeraj) wrote :

Ah, but this bug is about the X200. Posts about other laptops should be considered offtopic, unless this is a duplicate of another, more general bug, in which case this bug should be merged into that one. I have an x200, and suspend is still broken in 2.6.35-25

Revision history for this message
Greg Grossmeier (greg.grossmeier) wrote :

Right. On my x200s suspend is broken unless I remove the tpm_tis module. After that removal (and without rebooting) suspend works just fine.

The link in the first comment from rt_davies is completely correct and has much more information including the offending commit.

Kernel Team: What is the next step for this bug as there appears to be a pretty good amount of triage information?

Changed in linux (Ubuntu):
status: New → Confirmed
Revision history for this message
Russell Neches (ubuntu-vort) wrote :

Blacklisting tpm_tis allows the machine to enter S3 about 75% of the time, but now we're just back to the original problem. It still intermittently crashes while attempting to enter S3.

Andy Whitcroft (apw)
tags: added: kernel-key
Changed in linux (Ubuntu Maverick):
status: New → Confirmed
Revision history for this message
Andy Whitcroft (apw) wrote :

There are two aspects here, fixing this in Maverick and ensure it is fixed in Natty when that releases. Could those of you who are able could you either test a Natty kernel on your Maverick system or test a current Natty daily ISO. We have had specific fixes for both suspend/resume and for tpm related issues for Lenovo systems in general. This will help us ensure Natty is fixed in this regard, and will help us identify any fixes for Maverick. Please report any testing back here.

Changed in linux (Ubuntu):
status: Confirmed → Incomplete
Changed in linux (Ubuntu Maverick):
importance: Undecided → Medium
Changed in linux (Ubuntu):
importance: High → Medium
Revision history for this message
sabby (sabby) wrote :

I will comment here since my original bug report was marked as a duplicate of this one. I did test with 2.6.37.12 and 2.6.38.4 backport for lucid and it did fix the problem with suspend to ram for me.

Revision history for this message
Jeremy Foshee (jeremyfoshee) wrote :

Bjorn,
     Can you verify sabby's findings above?

Thanks!

~JFo

Changed in linux (Ubuntu):
status: Incomplete → Confirmed
status: Confirmed → Incomplete
Revision history for this message
Björn Tillenius (bjornt) wrote : Re: [Bug 698097] Re: Suspend stopped working on Thinkpad X200s with 2.6.35-24

On Tue, May 24, 2011 at 06:20:15PM -0000, Jeremy Foshee wrote:
> Bjorn,
> Can you verify sabby's findings above?

I'm running Natty now, with 2.6.38-8, and it seems to work fine now.

--
Björn Tillenius | https://launchpad.net/~bjornt

Revision history for this message
Seth Forshee (sforshee) wrote :

I'm closing the development task based on the feedback that this is fixed in natty.

This sounds similar to bug #705845, which is noted as being fixed in 2.6.35-27.48. Please test to see if suspend still fails in the latest maverick kernels. Thanks!

Changed in linux (Ubuntu Maverick):
assignee: nobody → Seth Forshee (sforshee)
status: Confirmed → Incomplete
Changed in linux (Ubuntu):
status: Incomplete → Fix Released
Revision history for this message
sabby (sabby) wrote :

I don't think we should close it, yes using the lucid backport of natty kernel fix the problem but that's not a real fix. The problem is still in the latest 2.6.32.31 kernel in lucid. As my original bug report stated this started with kernel 2.6.32-28, so a diff from 28-27 should show what caused the problem and can be fix in a later lucid kernel update.

Revision history for this message
Seth Forshee (sforshee) wrote :

@sabby, the task I closed corresponds to the most recent version of the ubuntu kernel, in which the problem has been fixed. For lucid, what we should do is to apply the fix from bug #705845 to the lucid kernel and test it. I'll start a build and post it here once it's done. If that fixes the bug, what I'll probably do is open a lucid task on bug #705845 and mark this bug as a duplicate of that one.

It would still be helpful if anyone is able to test the latest maverick kernel to see if the problem still exists there.

Revision history for this message
sabby (sabby) wrote :

@Seth, I will try the fixed lucid kernel once you post it. As far as the latest maverick kernel, I could try it on lucid if that's any help.

Revision history for this message
Seth Forshee (sforshee) wrote :

Test build posted to:

http://people.canonical.com/~sforshee/lp698097/linux-2.6.32-32.63~lp698097v201105311500/

Please report back here with your results. Thanks!

Revision history for this message
sabby (sabby) wrote :

Well I want to start with a little apology... After doing some more test and stumbling by chance on my real problem it turns out the system did not suspend because flgrx module were not compiled for that kernel. Somehow, dkms no longer generate module automatically when installing a new kernel. Will have to see why later. So I went back and tried all the latest kernel of 2.6.38, 2.6.37, 2.6.35 and 2.6.32 with the fglrx module manually compiled and the system suspends without any problem. Now this brings up the fact that there should really have been some sort of message in the log somewhere showing the gpu hanging was causing the system to not suspend but I couldn't find anything. Will look later what happened to dkms. Sorry to have wasted people's time, never thought missing the fglrx module could have caused that. I seem to think that it wasn't like that at one point, i.e. you could still suspend without the flgrx kernel module, but maybe I'm wrong. So for me this issue is resolved. Perhaps the other people can verify if that was the problem, not sure if there's equivalent module shoud be generated by dkms for intel/nvidia gpu.

Revision history for this message
Seth Forshee (sforshee) wrote :

sabby, no problem.

Can anyone confirm whether or not this problem still exists in lucid/maverick?

Revision history for this message
Seth Forshee (sforshee) wrote :

Closing Maverick task as Invalid due to lack of response.

Changed in linux (Ubuntu Maverick):
status: Incomplete → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.