Thinkpad x60 doesn't recover from suspend or hibernate

Bug #59851 reported by Mark A. Hershberger
12
Affects Status Importance Assigned to Milestone
linux-source-2.6.17 (Ubuntu)
Fix Released
Medium
Ben Collins

Bug Description

Trying to suspend or hibernate on my x60 (after an upgrade to edgy from dapper, where it worked) fails.

That is, it appears to suspend (the machine shuts down and the suspend LED flashes) but it won't come back up.

I experience the same thing with hibernation.

Tags: edgy hibernate
Revision history for this message
Matthew Garrett (mjg59) wrote :

Can you try booting with the

acpi_sleep=s3_bios

argument and see if you get any output on resume?

Revision history for this message
Mark A. Hershberger (hexmode) wrote :

That and "pci=routeirq" seems to be almost it.

Everything seems to come back up, but I get errors from sdhci similar to #44959.

That is, everything is the same, though my messages stop right at "Unexpected auto CMD12 error".

X seems to be up and the virtual consoles, but everything that tries to access the disk seems to hang forever. For instance, I can type a username at a console login prompt, but it hangs instead of giving a password prompt. Similar for the GDM login.

Revision history for this message
Matthew Garrett (mjg59) wrote :

What happens if you don't use "pci=routeirq"? Thanks for the test, it sounds like the hard drive isn't being resumed correctly. Any chance you can provide the output of lspci?

Revision history for this message
Mark A. Hershberger (hexmode) wrote :

If I don't use pci=routeirq, I don't get as far. It looks like the system resumes part way, but nothing is displayed.

I tried following #44959's suggestion of unloading sdhci before suspend and I got the following:
hda_intel: azx_get_response timeout, switching to single_cmd mode.

According to a google search, the problem appears to be the snd_hda_intel module.

Revision history for this message
Mark A. Hershberger (hexmode) wrote :

Actually, it does appear that if I leave out pci=routeirq and then remove the snd_hda_intel and sdhci/mmc_core modules, suspend gets to the same point.

In any case, unloading snd_hda_intel doesn't appear to get me the rest of the way.

Clue here: http://www.thinkwiki.org/wiki/Problems_with_SATA_and_Linux#Hang_on_resume_from_suspend_to_RAM

Revision history for this message
Mark A. Hershberger (hexmode) wrote :

Ahh... changelog for linux-image-2.6.17 has the following:

  * Disable CONFIG_SCSI_SATA_ACPI until it's forward ported.

Revision history for this message
Mark A. Hershberger (hexmode) wrote :

And, the above comment was made in May.

In June, the patch was forward-ported:

http://cvs.fedora.redhat.com/viewcvs/rpms/kernel/FC-5/linux-2.6-console-suspend.patch?rev=1.1&view=markup

Would it be possible to get that patch added?

Revision history for this message
Matthew Garrett (mjg59) wrote : Re: [Bug 59851] Re: Thinkpad x60 doesn't recover from suspend or hibernate

Ok, that should be integrated. Ben's looking at that now.

--
Matthew Garrett | <email address hidden>

Revision history for this message
Ben Collins (ben-collins) wrote :
Changed in linux-source-2.6.17:
assignee: nobody → ben-collins
status: Unconfirmed → Fix Committed
Changed in linux-source-2.6.17:
status: Fix Committed → Fix Released
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.