Contribute
Register

Laptop won't sleep/shutdown after a few hours of uptime

Status
Not open for further replies.
RehabMan said:
The WMI device is not used on OS X. Your deletions in patched files used on OS X will not affect the ACPI used on Windows.
Click to expand...

Not sure I understand what you mean. I'm trying to chase down a weird issue in OSX (single boot on this laptop), where sleep/shutdown locks up when the laptop has been running for 2 or 3 hours. Since I have no clue as to what the cause of the lockup is, I'm trying to clean up the DSDT of anything that seems like a good candidate to trigger a lockup. Since the WMI device gets a lot of attention from various EC queries, I figured it could help to get rid of it.
 
vbo said:
Not sure I understand what you mean. I'm trying to chase down a weird issue in OSX (single boot on this laptop), where sleep/shutdown locks up when the laptop has been running for 2 or 3 hours. Since I have no clue as to what the cause of the lockup is, I'm trying to clean up the DSDT of anything that seems like a good candidate to trigger a lockup. Since the WMI device gets a lot of attention from various EC queries, I figured it could help to get rid of it.
Click to expand...

I'm just saying that there is no harm in removing WMI as it is a Windows thing...

I do it as a matter of course for any hack I do...
 
RehabMan said:
I'm just saying that there is no harm in removing WMI as it is a Windows thing...

I do it as a matter of course for any hack I do...
Click to expand...

Oh I got you! I'm hoping it will solve my lockups.. Debugging something that only happens after a few hours is a pain! :)
 
vbo said:
Oh I got you! I'm hoping it will solve my lockups.. Debugging something that only happens after a few hours is a pain! :)
Click to expand...

Intermittent problems suck.
 
RehabMan said:
Intermittent problems suck.
Click to expand...
They sure do! Worked perfect all day across multiple sleep/wake cycles, until I left it run for 2 hours and 30 minutes without sleep and it locked up again..
Code: 
Nov  8 10:17:22 localhost kernel[0]: ACPIDebug: Version 0.1.2 starting
Nov  8 10:41:31 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: "_WAK exit"
Nov  8 10:58:28 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 11:36:59 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: "_WAK exit"
Nov  8 12:10:23 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 13:39:36 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: "_WAK exit"
[B]Nov  8 14:32:01 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 16:58:10 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 17:01:25 localhost kernel[0]: ACPIDebug: Version 0.1.2 starting[/B]

Back to the drawing board I guess.. I was thinking of dumping the EC inside _TTS so I can maybe spot a difference in its contents between the times were sleep works fine and the times it locks up..
Or maybe trying Clover's FixRegions.. I don't have any other symptoms of floating regions though so I don't know if that'll help. Doesn't hurt to add it in there I guess.
Any other ideas?
 
vbo said:
They sure do! Worked perfect all day across multiple sleep/wake cycles, until I left it run for 2 hours and 30 minutes without sleep and it locked up again..
Code: 
Nov  8 10:17:22 localhost kernel[0]: ACPIDebug: Version 0.1.2 starting
Nov  8 10:41:31 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 10:41:32 mba kernel[0]: ACPIDebug: "_WAK exit"
Nov  8 10:58:28 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 11:36:59 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 11:37:00 mba kernel[0]: ACPIDebug: "_WAK exit"
Nov  8 12:10:23 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 13:39:36 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: { "_PTS enter", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: "_PTS exit"
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: { "_WAK enter", 0x3, }
Nov  8 13:39:37 mba kernel[0]: ACPIDebug: "_WAK exit"
[B]Nov  8 14:32:01 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x0, }
Nov  8 16:58:10 mba kernel[0]: ACPIDebug: { "_TTS enter with state", 0x3, }
Nov  8 17:01:25 localhost kernel[0]: ACPIDebug: Version 0.1.2 starting[/B]

Back to the drawing board I guess.. I was thinking of dumping the EC inside _TTS so I can maybe spot a difference in its contents between the times were sleep works fine and the times it locks up..
Or maybe trying Clover's FixRegions.. I don't have any other symptoms of floating regions though so I don't know if that'll help. Doesn't hurt to add it in there I guess.
Any other ideas?
Click to expand...

FixRegions can't hurt. From the sounds of it it is almost more likely to be some sort of memory corruption issue caused by bad kexts/drivers/etc. So you might review what you have installed and if possible eliminate kexts one-by-one. Also, you could compare against other version of OS X such as Mavericks...
 
RehabMan said:
FixRegions can't hurt. From the sounds of it it is almost more likely to be some sort of memory corruption issue caused by bad kexts/drivers/etc. So you might review what you have installed and if possible eliminate kexts one-by-one. Also, you could compare against other version of OS X such as Mavericks...
Click to expand...

Guess where I got these bad boys :)
Code: 
ls /Volumes/EFI/EFI/CLOVER/kexts/10.10/
ACPIBacklight.kext       ACPIDebug.kext           FakeSMC.kext
ACPIBatteryManager.kext  CodecCommander.kext      VoodooPS2Controller.kext
Everything else is vanilla except an AppleHDA injector in SLE.

I'll give FixRegions a go and pull out my old Mavericks install to try to reproduce the issue if it still doesn't work. This would be much less of a hassle if I didn't have to wait 2+ hours before testing :)
Thanks for the help RehabMan, I appreciate it.
 
vbo said:
Guess where I got these bad boys :)
Code: 
ls /Volumes/EFI/EFI/CLOVER/kexts/10.10/
ACPIBacklight.kext       ACPIDebug.kext           FakeSMC.kext
ACPIBatteryManager.kext  CodecCommander.kext      VoodooPS2Controller.kext
Everything else is vanilla except an AppleHDA injector in SLE.

I'll give FixRegions a go and pull out my old Mavericks install to try to reproduce the issue if it still doesn't work. This would be much less of a hassle if I didn't have to wait 2+ hours before testing :)
Thanks for the help RehabMan, I appreciate it.
Click to expand...

Any config.plist kext patches?

Are you using native WiFi (not USB)?
 
RehabMan said:
Any config.plist kext patches?
Click to expand...
Sure, but nothing out of the ordinary:
https://github.com/vbourachot/Dell-XPS13-9333-DSDT-Patch/blob/master/config.plist#L119
  • AppleHDA
  • Trim
  • 2nd stage boot graphics fix for 10.10
  • Brcm4360 patches for wifi
  • I just added the handoff patch from the-darkvoid (issue predates this addition though, and the patch is probably not even applied since I don't enable bluetooth via BTFirmwareUploader so the kext is probably not in kernelcache. I just added it thinking someone following the guide I posted would ask for it at some point).

RehabMan said:
Are you using native WiFi (not USB)?
Click to expand...
Yes PCi-E wifi - BCM94352Z, which is BCM4352 in M2/NGFF format.
 
vbo said:
Sure, but nothing out of the ordinary:
https://github.com/vbourachot/Dell-XPS13-9333-DSDT-Patch/blob/master/config.plist#L119
  • AppleHDA
  • Trim
  • 2nd stage boot graphics fix for 10.10
  • Brcm4360 patches for wifi
  • I just added the handoff patch from the-darkvoid (issue predates this addition though, and the patch is probably not even applied since I don't enable bluetooth via BTFirmwareUploader so the kext is probably not in kernelcache. I just added it thinking someone following the guide I posted would ask for it at some point).


Yes PCi-E wifi - BCM94352Z, which is BCM4352 in M2/NGFF format.
Click to expand...

Seems pretty clean. Try without Trim for a while... Some SSDs have issues with Trim on OS X. My desktop Mushkin Chronos Deluxe 256GB)... I use without Trim as it slows boot down considerably. Check that you have current firmware for the SSD too.
 
Status
Not open for further replies.
Copyright © 2010 - 2024 tonymacx86 LLC
Back
Top