suspend/hibernate does not work in karmic on msi u100 netbook - regression

[Jorge Juan]

Suspend to ram stopped working after updating to karmic. Either selecting suspend in the shut down menu, closing the lid or running pm-suspend has the same effect: network is disconected, hd activity is like system is going to suspend and then network is connected again and the system keeps running: seems like a resume immediatly after suspend.

Screen does not blink or go black. Note that it is running in karmic's default kernel mode switching (KMS).

Seems to be similar to bug #426906, but in my case it do not work with 2.6.31-9 either.

I am attaching pm-suspend.log.

ProblemType: Bug
Architecture: i386
Date: Thu Sep 10 14:33:03 2009
DistroRelease: Ubuntu 9.10
Package: linux-image-2.6.31-10-generic 2.6.31-10.30
ProcEnviron:
 LANGUAGE=es_ES.UTF-8
 PATH=(custom, user)
 LANG=es_ES.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-10.30-generic
SourcePackage: linux
Tags: ubuntu-unr
Uname: Linux 2.6.31-10-generic i686

[Jorge Juan (jjchico)]

Definitively it should be a kernel issue since booting with my old Jaunty kernel solves the problem:

$ uname -a
Linux netbook 2.6.28-15-generic #49-Ubuntu SMP Tue Aug 18 18:40:08 UTC 2009 i686 GNU/Linux

with this kernel I do not have kernel mode switching as expected.

At the same time, a minor issue with brightness control has solved and USB devices works again (they stopped working with karmic as well), although these should be reported as a different bug.

[Jorge Juan (jjchico)]

My MSI U100 still does not suspend/hibernate after using ubuntu-boot PPA's. It just worked the first time I rebooted with ubuntu-boot/ppa, but stopped working after a new reboot. Enabling ubuntu-boot/staging did not update any packege and still does not work.

Additionally, booting now is much slower.

[Andy Whitcroft (apw)]

@Jorge -- can you confirm that this issue still occurs with the latest kernel, Ubuntu 2.6.31-10.32. Thanks!

[Jorge Juan (jjchico)]

Hi,

2009/9/15 Andy Whitcroft <email address hidden>:
> @Jorge -- can you confirm that this issue still occurs with the latest
> kernel, Ubuntu 2.6.31-10.32.  Thanks!
>
After yesterday's update (Sep. 16th), suspend was working for some
time. After 3 or 4 suspend cycles the system hanged. After that, the
system did not suspend anymore. After today's update (Sep. 17th)
suspend does not work either and the original behaviour described is
still present.

My current kernel is 2.6.31-10.34

I will try to provide further information.

jorge.

> --
> suspend/hibernate does not work in karmic on msi u100 netbook - regression
> https://bugs.launchpad.net/bugs/427327
> You received this bug notification because you are a direct subscriber
> of the bug.
>

--
Jorge Juan

[Jorge Juan (jjchico)]

AplayDevices:
 **** List of PLAYBACK Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC1200 Analog [ALC1200 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: jjchico 3171 F.... pulseaudio
CRDA: Error: command ['iw', 'reg', 'get'] failed with exit code 1: nl80211 not found.
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xffe00000 irq 16'
   Mixer name : 'Realtek ALC1200'
   Components : 'HDA:10ec0888,14620110,00100101'
   Controls : 21
   Simple ctrls : 13
DistroRelease: Ubuntu 9.10
HibernationDevice: RESUME=UUID=7845f726-99b3-47b3-b3ef-cf8b62639774
MachineType: MICRO-STAR INTERNATIONAL CO., LTD U-100
Package: linux (not installed)
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.31-10-generic root=UUID=7043558a-dade-4256-ac91-e165bae06e8a ro quiet splash
ProcEnviron:
 SHELL=/bin/bash
 PATH=(custom, no user)
 LANG=es_ES.UTF-8
 LANGUAGE=es_ES.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-10.34-generic
RelatedPackageVersions:
 linux-backports-modules-2.6.31-10-generic N/A
 linux-firmware 1.18
RfKill:

Tags: ubuntu-unr
Uname: Linux 2.6.31-10-generic i686
UserGroups:

WpaSupplicantLog:

dmi.bios.date: 07/16/2008
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 4.6.3
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: U-100
dmi.board.vendor: MICRO-STAR INTERNATIONAL CO., LTD
dmi.board.version: Ver.001
dmi.chassis.asset.tag: To Be Filled By O.E.M.
dmi.chassis.type: 3
dmi.chassis.vendor: MICRO-STAR INTERNATIONAL CO., LTD
dmi.chassis.version: Ver.001
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr4.6.3:bd07/16/2008:svnMICRO-STARINTERNATIONALCO.,LTD:pnU-100:pvrVer.001:rvnMICRO-STARINTERNATIONALCO.,LTD:rnU-100:rvrVer.001:cvnMICRO-STARINTERNATIONALCO.,LTD:ct3:cvrVer.001:
dmi.product.name: U-100
dmi.product.version: Ver.001
dmi.sys.vendor: MICRO-STAR INTERNATIONAL CO., LTD

[Jorge Juan (jjchico)]

Hi!

my system has started to suspend correctly after a couple of reboots without any change. Las time suspend was unsuccessful there was a kernel oop at boot up probably related to pulseaudio (which I apport reported). The last boot did not do oops and suspend works. It may be related.

In the above pm-suspend.log you can see reports of unsuccessful suspends (first in the file where wake-up happens right after suspend) and successful ones (last in the file).

All in all I think suspend behaviour is unstable may be due to instabilities in other components of the system.

I'll report any change.

[Jorge Juan (jjchico)]

Bad news,

after a new reboot, suspend does not work again. Some times, right after entering the session I get XXX exited unexpectedly errors. Last time XXX was nautilus. I'll check if suspend fails only when these errors are present.

I am giving up until tomorrow.

[Jorge Juan (jjchico)]

I wiped out the whole disk and did a fresh install of Alpha 6 and an update. Suspend/hibernate behaviour is still erratic. Most of the time does not work. Sometimes after a reboot it works again. I'll wait for the next kernel.

[Jorge Juan (jjchico)]

Hi,

I have advanced quite a lot on finding the origin of the bug. It seems to be a bug in the uvcvideo module. From my dmesg:

At boot time, webcam is detected correctly: a Bison NB Pro:
...
[ 8.975626] Linux video capture interface: v2.00
[ 9.008304] uvcvideo: Found UVC 1.00 device BisonCam, NB Pro (5986:0203)
[ 9.024768] input: BisonCam, NB Pro as /devices/pci0000:00/0000:00:1d.7/usb1/1-5/1-5:1.0/input/input8
[ 9.025136] usbcore: registered new interface driver uvcvideo
[ 9.025150] USB Video Class driver (v0.1.0)
...

At suspend time, PM informs some devices (bison cam) failed to suspend, and it wakes up inmediatly:
...
[ 231.266406] PM: Entering mem sleep
[ 231.266427] Suspending console(s) (use no_console_suspend to debug)
[ 231.284163] uvcvideo: Failed to set UVC commit control : -108 (exp. 26).
[ 231.284178] uvcvideo 1-5:1.1: resume error -5
[ 231.284201] pm_op(): usb_dev_suspend+0x0/0x10 returns -108
[ 231.284211] PM: Device 1-5 failed to suspend: error -108
[ 231.284219] PM: Some devices failed to suspend
[ 231.338519] uvcvideo: Failed to set UVC commit control : -108 (exp. 26).
[ 231.338533] uvcvideo 1-5:1.1: resume error -5
[ 231.456330] PM: resume devices took 0.172 seconds
[ 231.456350] PM: Finishing wakeup.
[ 231.456354] Restarting tasks ... done.
...

I can (de)activate the web cam pressing FN+F6. When the cam is not active it does not show in lsusb (like unplugging a USB cam). I have tried some combinations and this is what happens:

Scenario 1: Camera is not attached
  - lsusb does not shows the camera (lsusb_cam_off.log)
  - Suspend works
  - USB storage works (I can use my USB stick)
  - The web cam (with cheese) does not work, as expected.

Scenario 2: Camera is attached (lsusb_cam_on.log, dmesg_cam_on_uvcvideo_on.log)
  - lsusb does show the camera
  - uvcvideo is loaded at boot time
  - Suspend does not work (dmesg summary above)
  - USB storage does not work (connecting a USB stick does nothing - no trace in dmesg)

Scenario 3: Camera is attached but uvcvideo is configured not to load at boot time (dmesg_cam_on_uvcvideo_off.log)
  - lsusb does show the camera
  - uvcvideo is not loaded (blacklisted in /etc/modprobe.d/)
  - Suspend works
  - USB storage works

So the workaround by now is to create a file /etc/modprobe.d/local_options that includes a line:
blacklist uvcvideo so that the webcam is disabled.

I change from S1 to S2 by pressing the webcam button (FN+F6).
I change from S2 to S1 by pressing the button and rebooting.

Once in one scenario, the netbook will remember the state from boot to boot.

I attach lsusb and dmesg output in all the cases.

[Jorge Juan (jjchico)]

I have reported a new (more accurate) bug that supersedes this one (#435352). I will continue further discussion with my self there :)

Stay tuned.