Resume from suspend leads to blank screen with intel driver on Samsung X20 (Intel GM915 Chipset)

Please change the graphic driver to intel instead of i810. Maybe this helps. Btw. the new Inlte driver is the default one in Gutsy for your card.

I changed the driver to "intel". The results are the same as with the i810 driver.
I've tried that with AIGLX activated and deactivated.
Actually I did not try to log in over network (ssh), but the symptoms on my notebook are exactly the same. Suspend works fine, but during resume the screen stays blank and the harddisk activity ends after max 10 to 20 seconds.
I'll try to log in later when I have more time.

The -i810 aspect of this issue (problems resuming from suspend, including crashes, corruption, and screen blanking) is already reported in other bug reports (127005, 98668, 83320, 57147, 28326, others...) Unfortunately, -i810 is no longer supported upstream and is considered deprecated, so these are not likely to be fixed.

The fact that the issue still occurs on -intel 2.1.1, however, is interesting. I'm moving this bug to -intel for further investigation.

Unfortunately, the Xorg.0.log shows no errors. There are some techniques for debugging X crashes https://wiki.ubuntu.com/DebuggingXorg if you'd be interested in investigating this further on the the -intel driver. A backtrace with gdb at the point of crash is needed to understand what codepath is generating this fault.

Can you test this deb? This has a fix for 133118, which is not identical to this bug, but fixes a similar symptom.

http://www2.eng.cam.ac.uk/~pcjc2/ubuntu/xserver-xorg-video-intel_2.1.1-0ubuntu10~pcjc2.2_i386.deb

If that doesn't help (and I'm skeptical, as the crashes fixed by that package appeared to occur on the way into suspend), can you try different settings in /etc/defaults/acpi-support

Coming back from resume with no video might indicate a need to alter some of the workarounds there, such as saving (and writing back) PCI config space, or saving the video state.

Options such as

SAVE_VBE_STATE=true
SAVE_VIDEO_PCI_STATE=true
POST_VIDEO=true

Might all be tried.

(NB: I need POST_VIDEO=true, but not the other two on 945GM. If your LCD fades / bleeds to a funny colour with any of these, pull the power. Mine does without POST_VIDEO, and it probably means the LCD is being fed a bad signal, you shouldn't leave it like that).

Thanks for the efforts, I'll try those things as soon as possible (lots of work at the moment).
I already tried different settings in /etc/default/acpi-support...
With the new intel driver, my system already managed to resume, but it took ages (at least for minutes). At each success I tried to log in over ssh, only to mention after a while that my screen was back.
The last few times I tried to resume, without parallel login over ssh, my screen stayed blank (waited over ten minutes).
Sorry, no xorg.logs or other logs available at the moment. But I'll try your tipps (xorg-debugging, new driver, acpi-support).
Perhaps it is only a fault on my old (always updated - since 5.10) system.
When I have the time (weekend?), I'll try a fresh gutsy installation, but before I have to free a partition.

I have the same problem with my Samsung X20 and tried the new deb file and the other settings: Problem remains. Not had the time for the debugging hints yet, sorry.

i am having the same problem on gutsy, trying for weeks, now it finally works somehow (network still does not resume)

i think what made the difference is commenting this:
file: /usr/share/acpi-support/screenblank
# if [ x$LOCK_SCREEN = xtrue ]; then
# su $user -c "(xscreensaver-command -lock)"
# fi

and /etc/default/acpi-support
#SAVE_VBE_STATE=true
# The file that we use to save the vbestate
#VBESTATE=/var/lib/acpi-support/vbestate

you may also try to disable compiz
good luck!

I had no problems with Suspend/Resume Gutsy. And unforunately your patch does not change anything in Feisty :-( Thanks for your suggestion anyway.

I've collected the reports from people commenting on bugs 138256 and 150109: https://wiki.ubuntu.com/X/Bugs/ScreenModeChange

I think there's more than two bugs represented here, but in many cases we don't have enough data. Please look at the above page and fill in the missing data.

I've also created a page summarizing how to analyze and work around these issues. It would be worthwhile for anyone experiencing these issues to review and try these steps, and then report their findings on the appropriate bug

https://wiki.ubuntu.com/X/Debugging#head-0b6e9c6b60fa07cda4013a1c51239f1c14c4f89d

I also own a Samsung X20 and have the same problems in Kubuntu Gutsy. With the current 2.6.22-Kernel resume only works from time to time, and if it works first the harddisk-LED is lit up for some minutes (the computer doesn't respond, the screen is black) and after a while the pc "awakes"...
None of the suggested things in here changed anything. I also think it might not be X-related because I get the same effect, when I suspend my laptop from the plain commandline (using "echo mem > /sys/power/state") without any xserver running!

Has anyone already found out more details? Currently I'm sticking to feisty's 2.6.20-kernel which works perfectly!

Same as c_t:
Samsung X20 XVM 1730 V
After uspend to ram the screen is black (turned off, same result with X running and "recovery mode" without X), some times the pc awakes later.
None of the suggested things worked
With the kernel parameter "acpi_sleep=s3_bios" the display remained black (turned on) but the pc freezed.
Resume works with feisty's kernel one time, the second time without reboot, the hard disk is dead (rejecting scsi io to dead device), but that seems to be a different bug.

@dafeu:
Solution for your dead hard-disk-problem can be found in bug 109762, see https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/109762/comments/32

For the gutsy-bug I've still not found any news...

Still waiting on feedback from Christopher, but setting to confirmed for now.

From comments from c_t and dafeu, this might be a kernel problem.

Hi All,

For those who experience this issue even when not running X, care to try some of the following:

echo 1 > /proc/sys/kernel/acpi_video_flags
/etc/acpi/sleep.sh force

if that doesn't work try:

echo 2 > /proc/sys/kernel/acpi_video_flags
/etc/acpi/sleep.sh force

and if that doesn't work try:

echo 3 > /proc/sys/kernel/acpi_video_flags
/etc/acpi/sleep.sh force

If none of those work, care to take a look at: https://wiki.ubuntu.com/DebuggingKernelSuspend .

Finally, I notice you are running Gutsy. Care to also test with a Hardy Alpha release? It contains an updated version of the kernel. You can download and try the new Hardy Heron Alpha release from http://cdimage.ubuntu.com/releases/hardy/ . You should be able to then test the new kernel via the LiveCD. Note you will only be able to test Suspend from the LiveCD, not Hibernate. If you can, please verify if this bug still exists or not and report back your results. General information regarding the release can also be found here: http://www.ubuntu.com/testing/ . Thanks.

Suspending from Hardy Alpha-4 LiveCD worked perfectly (actually my laptop suspended and woke up so fast like I had never seen it before!!).
I haven't tried the acpi_video_flags-thing, but since I'm running fine with feisty-kernel 2.6.20-16 for the moment I think I can wait for hardy to be released...

Since the latest Feisty kernel upgrade a few days ago resume works after a suspend to hdd. But not after a susped to ram.

But I am not sure if it was really this kernel upgrade or an earlier one. I haven't tried for a while because the change comments did not list anything which I though could fix this problem.

I'd becurious if it is possible using the Hardy kernel on Feisty. I dont like to use Hary on a production system as my laptop is.

correction: I mean Gutsy of course, NOT Feisty, sorry.

I'm having a similar problem with suspend on a gateway desktop system. It's proving difficult to pin down, even with all the suggestions mentioned on this thread, plus http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-advanced.html, plus https://wiki.ubuntu.com/DebuggingKernelSuspend. I'd like to help track it down if I can.

Does it work with Hardy Alpha 5 Live CD? It can be downloaded from http://cdimage.ubuntu.com/releases/hardy/alpha-5/ .

No, it doesn't. Same problem - no video output being sent to monitor, but caps lock toggles, etc.

Like c_t I tested the Alpha-4 LiveCD two weeks ago. And with that CD suspend worked really fine, I've never seen it so fast.
I have not yet tested Alpha-5.

@liquidporkgun
Please attach or post the following information:
/etc/X11/xorg.conf
/var/log/Xorg.0.log
/var/log/Xorg.0.log.old
output of lspci -vvnn
output of cat /proc/acpi/video/*/DOS
output of sudo cat /proc/acpi/dsdt

Sorry for the delay in response - only access this computer on the weekend.

I have attached the output of all but the last command to this file. Beginning of each command output is denoted by <<<<string, for example <<<</etc/X11/xorg.conf. End of output is denoted by >>>>. Hope that helps in weeding through it.

cat /proc/acpi/video/*/DOS did not output anything because /proc/acpi/video/ is empty.

cat /proc/acpi/dsdt produced binary output so I'm sending it in a separate file...

...and here is sudo cat /proc/acpi/dsdt

On a laptop with
00:00.0 Host bridge: Intel Corporation Mobile PM965/GM965/GL960 Memory Controller Hub (rev 0c)
running a daily update of Hardy alpha5
resume from suspend works fine. Before 2.6.24-11 I needed a manual switch to the console and back to X. (acpi-support's option for this had no effect)
Can't test hibernate, I don't have a swap partition.

@liquidporkgun
Thanks for the log but it is nearly unreadable. Could you please attach every file and output separately or at least attach them compressed with tar?
Btw. if I read your log right X is using the standard vesa driver and not the intel one.

The easiest way to save the output in a file is to pipe it.
e.g.: lspci -vvnn > lspciOutputFile.txt

I guess it is the best to mark this bug as fixed since the original reporter and other users have confirmed it.
Feel free to open it again if the issue still happens with i915 graphic card and Hardy Alpha 6 Live CD or later.

@liquidporkgun
Since you might have a different graphic card or a not recognized one I would create a new bug report and upload all the needed data.

I can confirm that this bug seems to be fixed in Hardy Beta with kernel 2.6.24-12-generic ...

After it worked nicely with kernel 2.6.24-12-generic the bug reappeared with one of the latest kernel-upgrades. Currently I'm running kernel 2.6.24-16-generic and resuming leads to a blank screen and the laptop doesn't respond to any key-press...

** Please reopen this bug! **

I'm running hardy-beta and I'm doing dayly upgrades. Hope this get's fixed again until the final Hardy-Release! I'm tired of sticking to 2.6.20-kernels...

I don't have any problems with kernel 2.6.24-16-generic. Suspend works fine.

Oh yeah, it seems like I messed something up. Actually I *do* have problems suspending with 2.6.24-16-generic unless I blacklist the ehci_hcd module. Otherwise I sometimes get a blank screen and also my usb-devices do not work! See bug #211572 for details. So no need to reopen this bug, since my issue seems to be a different one.

Also my lid-button completely stopped working with the latest kernels, see bug #217166 . Can you Christopher, as a X20 owner, confirm this?

No, my lid-button works fine, just tested with suspend to ram - worked.