Contribute
Register

Lenovo G500 10.9.1: Almost done, just a little bit of help needed...

Status
Not open for further replies.
It worked! I have a full brightness range with quarter levels and everything! But it still reboots spontaneously, either just after waking up, or a few minutes later (but it always happens). There's some weird issue with MaciASL or my DSDT file I think, because I can't see the changes after I make them but they are there, but as far as I know, the _WAK patch is there. What else can be done about the reboots?
 
moonhack said:
It worked! I have a full brightness range with quarter levels and everything! But it still reboots spontaneously, either just after waking up, or a few minutes later (but it always happens). There's some weird issue with MaciASL or my DSDT file I think, because I can't see the changes after I make them but they are there, but as far as I know, the _WAK patch is there. What else can be done about the reboots?
Click to expand...

As far as weird issues with MaciASL. Realize that when you run MaciASL it loads DSDT *from ioreg*. If the DSDT that is in /Extra/dsdt.aml *matches* what is in ioreg, it pretends as if you're editing /Extra/dsdt.aml and when you save changes they go there (without using Save As). So, if you make some patches, then save, the first time they go to /Extra/dsdt.aml. If you quit MaciASL and run it again, it will again load the DSDT that you booted with (eg. the one captured in ioreg) and show "pre-edited DSDT". That DSDT is, of course, without the edits you made because that is just sitting on disk, not loaded into ioreg until the next reboot. If you want to avoid all this, just edit /Extra/dsdt.aml directly.

Your reboot problem is probably related to HPET or other power management... There are some patches in my repo you might try... see the ones in "System Patches" section.
 
I applied the HPET fix, nothing. I then added OS Check Fix, IRQ, SMBUS, RTC and MCHC as well, and still nothing. I even tried adding some patches on Clover that seem to be 'equivalent' and nothing. Apparently, it only reboots after a minute or two now, instead of instantly, but I've only tested 4 times. On the 4th time, it woke up by itself with zero brightness and it kept going back to sleep every 10 seconds without the backlight coming back. Then I put it to sleep manually, woke it up, and the backlight came back.

Edit: Nope, 5th try, rebooted as soon as I pressed the spacebar.
 
moonhack said:
I applied the HPET fix, nothing. I then added OS Check Fix, IRQ, SMBUS, RTC and MCHC as well, and still nothing. I even tried adding some patches on Clover that seem to be 'equivalent' and nothing. Apparently, it only reboots after a minute or two now, instead of instantly, but I've only tested 4 times. On the 4th time, it woke up by itself with zero brightness and it kept going back to sleep every 10 seconds without the backlight coming back. Then I put it to sleep manually, woke it up, and the backlight came back.

Edit: Nope, 5th try, rebooted as soon as I pressed the spacebar.
Click to expand...

Boot loader issue?
 
RehabMan said:
Boot loader issue?
Click to expand...
Could it be? I'm using Clover 2403 on UEFI mode. I haven't messed a lot with the configuration file other than hiding partitions and patching kexts... Keep in mind this only happens if I activate Sleep... if I don't, this is perfectly stable.

Any help would be appreciated...
 
moonhack said:
Could it be? I'm using Clover 2403 on UEFI mode. I haven't messed a lot with the configuration file other than hiding partitions and patching kexts...
Click to expand...

Clover builds are always a crapshoot. You never know what kind of bugs may be present in each build. It is always in a state of churn. FYI: Current clover build is v2428.

You might try Chameleon (Enoch branch) just as a test...
 
I updated Clover last night and decided to follow Apple's indications for calibrating the battery meter, so I didn't reboot into the new build until this morning. Battery lasted 5:20 switching between web use and non-sleep idling, pretty good since advertised battery time with Win8 is 5 hours. It took 2 hours to recharge completely this morning, and I noticed the battery cycle count is still at 0. Is there some DSDT or kext patch to report the correct amount (or does OS X log the cycles instead of the battery)? Well, so I just put it on sleep mode then woke it up around 5 minutes ago and it hasn't rebooted. It used to happen in a minute. Waiting patiently, maybe it was fixed after all...
 
moonhack said:
I updated Clover last night and decided to follow Apple's indications for calibrating the battery meter, so I didn't reboot into the new build until this morning. Battery lasted 5:20 switching between web use and non-sleep idling, pretty good since advertised battery time with Win8 is 5 hours. It took 2 hours to recharge completely this morning, and I noticed the battery cycle count is still at 0. Is there some DSDT or kext patch to report the correct amount (or does OS X log the cycles instead of the battery)? Well, so I just put it on sleep mode then woke it up around 5 minutes ago and it hasn't rebooted. It used to happen in a minute. Waiting patiently, maybe it was fixed after all...
Click to expand...

OS X does not track battery cycles. It is up to the battery hardware/DSDT. Think about it: battery cycles can happen when running a different OS, such as Windows, so it doesn't make sense for the OS to track this. Plus you may re-install the system, erasing any data the system might have collected about battery cycles.

For battery cycle count, you would need to determine where it is stored (in EC) and add that information to the table returned in _BIF (position 13).
 
Good news. After 12 different tries, no reboots on sleep now. It was Clover's fault after all. But it's not perfect yet. USB devices disconnect unsafely during sleep/when waking up which could cause problems because they aren't mounted back. Then, if, and only if, the lid is closed after choosing sleep and before it actually enters sleep, the brightness will be at zero when it wakes up, and will lock itself every few seconds. If you get to the desktop and put it to sleep again, wait until it enters sleep then wake it up, the backlight will be back. How can I fix these two, especially the second one?

I guess enabling cycle count will be postponed to "someday" together with HDMI audio. Unless you could guide me through the process. :mrgreen:
 
moonhack said:
Good news. After 12 different tries, no reboots on sleep now. It was Clover's fault after all. But it's not perfect yet. USB devices disconnect unsafely during sleep/when waking up which could cause problems because they aren't mounted back.
Click to expand...

Are you using GenericUSBXHCI? I have the same issue. I think it is a bug in that kext.

Then, if, and only if, the lid is closed after choosing sleep and before it actually enters sleep, the brightness will be at zero when it wakes up, and will lock itself every few seconds. If you get to the desktop and put it to sleep again, wait until it enters sleep then wake it up, the backlight will be back. How can I fix these two, especially the second one?
Click to expand...

I might have to try that scenario with my 4540s... Easy enough to bump the brightness up after that, provided your brightness keys are working.

I guess enabling cycle count will be postponed to "someday" together with HDMI audio. Unless you could guide me through the process. :mrgreen:
Click to expand...

It is a matter of guessing by looking at EC registers used by battery status and deciding if you can find the one related to cycle count. A little experimentation/dumping of registers never hurts either. You can output data to system.log from DSDT methods using ACPIDebug.kext. See: https://github.com/RehabMan/OS-X-ACPI-Debug
 
Status
Not open for further replies.
Copyright © 2010 - 2024 tonymacx86 LLC
Back
Top