[hardy] suspend to ram doesn't work on intel i965 chipset mobo

Bug #191613 reported by nyékhelyi gábor
4
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Confirmed
Medium
Ben Collins

Bug Description

In previous Hardy alphas suspend worked well. But now the same bug occurs as in Gutsy.

When I select suspend, the system goes to suspend, black screen is coming, the computer goes to power off, then the system immediately wakes up, then the login screen appears.

I have an Abit AB9 (i965 chipset) mobo, nVIDIA 7600gs videocard with nvidia binary driver, and running daily updated Hardy (currently : the 2.6.24-7-generic kernel). I don't use compiz.

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

Thank you for taking the time to report this bug and helping to make Ubuntu better.

Please attach the following information (as _separate_ attachments):

Output of 'sudo lspci -vvnn'

Output of 'sudo dmidecode'

Try to hibernate/suspend and then restart the system and attach /var/log/kern.log.0

Tarball of /proc/acpi directory

Thanks!

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
description: updated
Changed in linux:
assignee: nobody → ubuntu-kernel-acpi
importance: Undecided → Medium
status: Incomplete → Confirmed
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Now with 2.6.24-11-generic still doesn't work.

Changed in linux:
status: Confirmed → Triaged
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Updated to 2.6.24-12-generic...still doesn't work.
Any chaces that Hardy final will suspend correctly on my machine? (on Archlinux it works seamlessly, so it is possible)
I'd be glad to help you with this bug, even testing patches.

Revision history for this message
Ben Collins (ben-collins) wrote :

Please do not include kern.log. Instead, try a suspend/resume cycle and send the output of dmesg, since it is much more useful. In your kern.log, I see no remnants of a suspend attempt (note that your two files before and after attempt are empty. So in this case, the kernel has performed no actions, unless you just mistakenly attached the wrong file.

Revision history for this message
Ben Collins (ben-collins) wrote :

In the above comment I meant that your two kern.log files were the same, not empty.

Changed in linux:
assignee: ubuntu-kernel-acpi → ben-collins
status: Triaged → Incomplete
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

I attach the before-suspend and after-suspend dmesg outputs. After suspend the computer still wakes up immediately.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
Greg Grossmeier (greg.grossmeier) wrote :

Many sources are now showing that using the 64bit version of Ubuntu and adding "ec_intr=0" before "ro splash" in the /boot/grub/menu.lst file.

Before it looks something like this:
kernel /boot/vmlinuz-2.6.22-14-generic root=/dev/sda1 ro quiet

Change it to look similar to this:
kernel /boot/vmlinuz-2.6.22-14-generic root=/dev/sda1 ec_intr=0 ro quiet

In other words, don't change anything but adding "ec_intr=0" after the root=/dev/sdaX

For more information about this fix see this forum post: http://ubuntuforums.org/showthread.php?t=609722&page=3

The system they are referencing is the System76 Darter Ultra 2, which is based on the i965 chipset.

Changed in linux:
status: Incomplete → Confirmed
Revision history for this message
Greg Grossmeier (greg.grossmeier) wrote :

Also, here are related bugs (they are referring to bad power management information, but the cause seems to be the same)
bug 147560
bug 139701

Ben, I'll leave it to you to decide if this should be marked duplicate of one of those (or which of those should be a master for this issue. Paul Sladen has commented on bug 147560 because he was given a problem laptop from System76).

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

I tried your suggestion...doesn't work.

I've modified my menu.lst:
kernel /boot/vmlinuz-2.6.24-12-generic root=UUID=410c3a15-68c7-4672-bc2b-f18b2d06d5d8 ec_intr=0 ro splash

I've reattached the dmesg outputs before, and after the suspend.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

I use 32bit Hardy, not 64bit

Changed in linux:
status: Confirmed → In Progress
Revision history for this message
Ben Collins (ben-collins) wrote :

The original report shows that this occured with the nvidia binary driver installed. Can we see about getting this reproduced using the open nv xorg driver instead?

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

I couldn't make a dmesg log after suspend, because system didn't come up after the suspend (monitor stayed power off), but when computer went power off, it immediately waked up again according to the fans.

Revision history for this message
Ben Collins (ben-collins) wrote :

Ok, please blacklist the vboxdrv module, reboot and try a suspend/resume. Trying to alleviate the possibility that this is a bug in something other than the stock kernel.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Completely uninstalled virtualbox (don't actually use it now). The bug still occurs.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Tried to trigger a suspend from root terminal (with the same results, maybe the messages are useful):

# /etc/acpi/sleep.sh sleep
ifdown: interface eth0 not configured
 * Shutting down ALSA... [ OK ]
 * Saving the system clock
/etc/acpi/resume.d/17-video-restore.sh: line 5: /var/lib/acpi-support/vbestate: No such file or directory
Function not supported
 * Setting the system clock
Ignoring unknown interface eth0=eth0.
 * Setting up ALSA... [ OK ]
FATAL: Module acpi_sbs not found.
FATAL: Module acpi_sbs not found.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Updated to the new kernel (2.6.24-15-generic), suspend still doesn't work.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

(2.6.24-16-generic) new kernel, and the old bug.
So i965 guys won't have working suspend in Hardy?!

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

"Wake up by USB device" and "Wake up by LAN" was turned on in the BIOS, turned off them, and suspend works. So the bug should be searched that way. In windows, the suspend works well with those wake up settings above.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

With the new 2.6.22-17 kernel, my computer goes to sleep, but doesn't wake up.
Only I recive is blank unpowered monitor, and the HDD led keeps lighting.

I'm using Hardy with all updates enabled, and daily updated.

Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

Same error with the new 2.6.22-18 kernel. Working nice with 2.6.22-16.
What could cause the problem in the newer kernels?

Changed in linux:
status: In Progress → Confirmed
Revision history for this message
nyékhelyi gábor (n0gabor) wrote :

2.6.22-19 kernel solved the problem, suspend works again!
Thanks.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Thanks for testing and the update. I'm glad to hear the 2.6.24-19 kernel is working for you. Based on your last comment it would seem that fixes that were applied for bug 226279 has resolved this issue for you as well. I'm going to go ahead and mark this a duplicate of that bug just so this gets closed at the same time as bug 226279. Thanks!

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

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