Bug 459426

Summary: msi wind u100: pm-suspend fails
Product: [Fedora] Fedora Reporter: Need Real Name <lsof>
Component: hal-infoAssignee: David Zeuthen <davidz>
Status: CLOSED CURRENTRELEASE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: medium    
Version: 10CC: hdegoede, lsof, mclasen, opensource, richard
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-05-07 21:01:59 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
pm-suspend.log none

Description Need Real Name 2008-08-18 19:12:16 UTC
Created attachment 314490 [details]
pm-suspend.log

pm-hibernate works fine, but pm-suspends fails.

I don't know how to debug this, so here is the log.

I notice it seems to suspend okay, but pushing a key does nothing (probably by design).
Pushing the power button seems to do something (the hard disk light flickers a bit), but the display does nothing.

Comment 1 Need Real Name 2008-08-18 19:20:18 UTC
The box remains unpingable throughout.

Comment 2 Till Maas 2008-08-18 19:43:05 UTC
Please post the output of:

lshal | egrep "(system.hardware.(product|vendor|version)|system.firmware.version|power_management.quirk)"

Here you can find some hints about what you can do to debug this, but I am not sure, that this may fix the problem, it could also be a kernel problem:

http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-index.html

Comment 3 Need Real Name 2008-08-18 20:02:40 UTC
lshal output:
  system.firmware.version = '4.6.3'  (string)
  system.hardware.product = 'U-100'  (string)
  system.hardware.vendor = 'MICRO-STAR INTERNATIONAL CO., LTD'  (string)
  system.hardware.version = 'Ver.001'  (string)

Comment 4 Need Real Name 2008-08-18 20:18:32 UTC
Hm, the box is actually resumed, but the screen is still asleep, even with --quirk-dpms-on

Comment 5 Need Real Name 2008-08-18 20:22:39 UTC
--quirk-dpms-suspend too. I have no more quirks to try.

Comment 6 Need Real Name 2008-08-18 20:24:50 UTC
Aha! pm-suspend --quirk-vbe-post works!
Only downside is the screen goes white on resume :)

Comment 7 Need Real Name 2008-08-18 20:25:37 UTC
How can I make this permanent?

Comment 8 Till Maas 2008-08-18 20:32:33 UTC
You need to edit an xml file, probably:
20-video-quirk-pm-misc.fdi

And you need to send a patch upstream like it is described here:
http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-report.html

Comment 9 Need Real Name 2008-08-18 20:44:26 UTC
Okay. For info, two quirks are needed:
 pm-suspend --quirk-vbe-post --quirk-dpms-on
The white screen problem goes away.

I can't install gcc on this box at the moment, so can't send upstream.

Comment 10 Till Maas 2008-08-18 22:53:53 UTC
I sent a patch upstream. I hope it appears in the archives soon:
http://lists.freedesktop.org/archives/hal/2008-August/thread.html

Setting to ASSIGNED because triaging is done.

Comment 11 Need Real Name 2008-08-23 15:34:35 UTC
Thanks. btw will the "MSI" prefix expand to the whole string?

Comment 12 Till Maas 2008-08-23 17:59:19 UTC
(In reply to comment #11)
> Thanks. btw will the "MSI" prefix expand to the whole string?

No, it will not. But it is not used. system.hardware.vendor is matched with the case insensitive prefix "micro-star".

Comment 13 Need Real Name 2008-08-24 19:04:47 UTC
There is a second part to this bug (but I will open a new bug if you want):

pm-suspend only works once. On the second _resume_, I just get a blank screen.

This is also noted here:
 http://forums.msiwind.net/default-msiwind/ubuntu-can-suspend-more-than-once-t854.html

Comment 14 Till Maas 2008-08-29 20:24:18 UTC
(In reply to comment #13)
> There is a second part to this bug (but I will open a new bug if you want):
> 
> pm-suspend only works once. On the second _resume_, I just get a blank screen.

If you cannot fix it with any of the know quirks, it is probably a bug in the kernel and therefore you should create a new bug with the component "kernel" to make sure the kernel guys notice this bug.

Comment 15 Till Maas 2008-10-30 09:02:38 UTC
If you are running a 2.6.27 kernel and the i915, it should probably work without any quirks. You can test it then with:

echo mem > /sys/power/state

Please report back once you have tested this, please report back.

Comment 16 Need Real Name 2008-10-30 22:14:02 UTC
2.6.27.4-47.rc3.fc10.i686

No it doesn't work. It seems to come back, but the screen stays dark.
I suspect it comes back because ctrl+alt+delete works and the disk does something.

2.6.27.4-58.fc10.i686 doesn't boot at all (switchroot failed: no such file or directory, booting has failed). So can't test that.

Comment 17 Need Real Name 2008-10-30 22:17:26 UTC
Bug 469267 entered for the second issue.

Comment 18 Bug Zapper 2008-11-26 02:48:42 UTC
This bug appears to have been reported against 'rawhide' during the Fedora 10 development cycle.
Changing version to '10'.

More information and reason for this action is here:
http://fedoraproject.org/wiki/BugZappers/HouseKeeping

Comment 19 Need Real Name 2008-12-07 13:44:13 UTC
(In reply to comment #15)
> If you are running a 2.6.27 kernel and the i915, it should probably work
> without any quirks. You can test it then with:
> 
> echo mem > /sys/power/state
> 
> Please report back once you have tested this, please report back.

Works perfectly in F10.

Comment 20 Hans de Goede 2009-05-07 21:01:59 UTC
Closing per comment #19