[SAMSUNG ELECTRONICS CO., LTD. NC10] suspend/resume failure

Hi Fabien,

Are you able to reproduce this? ie Does it always fail after a few suspend/resume cycles? If you could also please include any information as to the circumstances leading up to this failure, for example did your battery run out? Did any messages get logged or printed? If it is reproducible could you try to suspend from VT1, press ctrl-alt-F1, login, and run pm-suspend there. If you could test and report back here that would be helpful. Thanks in advance.

On my nc10 this also happens. I tried your advice an ran pm-suspend from VT1. I got the KernelOops after the 3rd suspend, after the 8th suspend and after the 13th suspend. It happens totally random. But I noticed the hdd lamp is permanently glowing if the system cannot wake up.

I added some screen shots I made after a failed wakeup. I did the suspend by closing the laptop. After the wakeup the mouse-pointer was visible and movable, however the login screen did not show up. I was able to open the virtual terminal, but was not able to log in. Instead a lot of error messages I don't understand showed up. After about 2 minutes the error messages did stop and I tried to log in but was not successful (figure 2) and another error message appeared.

figure 2. Sorry for uncompressed picture in previous post.

it reoccurred several times since i filed this bug.

The situation is almost always the same. I have some music playing in rhythmbox, wifi (~zero traffic), and i close the lid. It suspends quite well, then resume starts quite fine, the music restarts to where it stopped, but only for a few seconds, then things turn weird. Either X crashes, or i get a corrupted display, or all the apps are getting confused. On two occasions, i got the see error messages from the kernel. They are the same as in the screenshot from Tobias in comment #4.

Failed the Suspend/Resume test script [1] at TEST 4 but i previously passed the 1st 6 tests so it looks random.

Moving status to confirmed as Tobias is also impacted.

[1]: https://lists.ubuntu.com/archives/ubuntu-devel-announce/2009-March/000556.html)

I've got the same problem. Brand new NC10 fails randomly with after resume. kernel 2.6.28-11.38.

Hard disk LED glows constantly when it fails. When I switch to text console, it prints out ATA1 errors:
ata1.00: status { DRDY }
ata1: SRST failed (errno=-16)

Then after a while, ext3 complains about I/O errors (unable to read inode block). Hard disk is spinning normally, but no seek noises can be heard.

Today that same ATA error happened after normal boot. Maybe this has something to do with the resume problems. This time the system recovered after a while. Now it's running Jaunty kernel 2.6.28-11.39.

The problem appeared when Gnome desktop was starting, after WLAN association. Dmesg attached.

Same problem here. Brand new NC10 - I've been going to suspend manually, but this time shut lid. On resume, I ended up with an ATA problem and had to force a shutdown.

Same problem here. As mentioned someone before the computer is kind of stuck and the kernel is reporting I/O errors..

Regards, guschora.

The problem can be reproduced without suspending. Disk I/O during power changes triggers the problem.

1. Start heavy disk I/O in the background: sudo dd if=/dev/sda of=/dev/null bs=131072

2. Start vmstat to monitor disk I/O: vmstat 1

3. Plug/unplug the AC adapter a couple of times.

4. Vmstat shows that the amount of disk I/O drops to zero.

5. The previously reported ATA error messages appear to kernel log, But this time the system recovers after ata1 soft reset.

When the AC adapter is not unplugged, no problems appear during disk I/O.

I investigated this issue on my machine and somehow it seems that it can be worked around (at least it worked for me the workaround suppressed the error so far) using a hidden BIOS setting by changing the SATA mode from "Compatible" to "Enhanced". I posted some details on how to activate it at: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/340014/comments/7

Disclaimer: This is an undocumented setting, use it at your own risk!

Do not enable AHCI, this will prevent GRUB from loading.

I did the same - and haven't had the freeze since, but I hadn't equated the
two actions.

To get the hidden menu put the num-lock(fn+f11) and scroll-lock(fn+f12) on
while your in the BIOS, and then scroll to the side. A secret menu labelled
'intel' will appear.

On Thu, Apr 16, 2009 at 12:32 AM, Thilo-Alexander Ginkel
<email address hidden>wrote:

> I investigated this issue on my machine and somehow it seems that it can
> be worked around (at least it worked for me the workaround suppressed
> the error so far) using a hidden BIOS setting by changing the SATA mode
> from "Compatible" to "Enhanced". I posted some details on how to
> activate it at:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/340014/comments/7
>
> Disclaimer: This is an undocumented setting, use it at your own risk!
>
> Do not enable AHCI, this will prevent GRUB from loading.
>
> --
> [SAMSUNG ELECTRONICS CO., LTD. NC10] suspend/resume failure
> https://bugs.launchpad.net/bugs/344183
> You received this bug notification because you are a direct subscriber
> of the bug.
>

--
Luke Whitmore

Cannot confirm this. When enabling your suggested BIOS setting the crash happens again after a few suspends.

Most likely there is another variable in the equation: Probably the BIOS revision. My NC10 is running revision 04CA.

Mine is running the same revision and it still crashes after a few suspends.

Am Donnerstag, 16. April 2009 07:58:02 schrieb Thilo-Alexander Ginkel:
> Most likely there is another variable in the equation: Probably the BIOS
> revision. My NC10 is running revision 04CA.

Leann, how can we move forward with this bug?
Any more info I can provide?