Contribute
Register

[Guide] Patching LAPTOP DSDT/SSDTs

Did a new USB.
Did a fresh install.
After install, kernel_task stays at 60%, well, I think I will let it settle down a little, because Spotlight might be indexing, etc...
Installed only FakeSMC and VoodooPS2.
But I think the issue is still there, it's been about 30min (and I have no data there, also, at SSD speeds Spotlight might index faster).
I will wait a little more, I don't trust Spotlight's efficiency.

On Apple discussions talk about Macbook Air's faulty temp sensor causing this issue, is there any way to report a fake temp to try?
I mean, if I can tell the OS the CPU is at 20°C for example...

No further ideas...
 
No further ideas...
I think that a solution is possible, I don't know...
I was thinking about DVD drive or the Intel WiFi+BT card (bluetooth is intermittent, sometimes appears to be working, the next boot it doesn't).
Also, my DSDT/SSDT includes a lot of ****, from 3x PEG devices (actually I think they are for multiple GPUs, at least on other computers), also 8 CPUs (OS X reports the last 4 as disabled), also I found NFC (???) and there are a lot of other things that don't know like PXSX or others.
I installed it using the same USB on a Clevo W550EU, patched DSDT (A LOT cleaner than mine) and everything is working perfectly, no kernel_task issue.
 
I think that a solution is possible, I don't know...
I was thinking about DVD drive or the Intel WiFi+BT card (bluetooth is intermittent, sometimes appears to be working, the next boot it doesn't).
Also, my DSDT/SSDT includes a lot of ****, from 3x PEG devices (actually I think they are for multiple GPUs, at least on other computers), also 8 CPUs (OS X reports the last 4 as disabled), also I found NFC (???) and there are a lot of other things that don't know like PXSX or others.
I installed it using the same USB on a Clevo W550EU, patched DSDT (A LOT cleaner than mine) and everything is working perfectly, no kernel_task issue.

I would start by looking at which kexts are loaded (give time for the kernel to unload kexts that didn't start).
Then one at a time, eliminate kexts that are suspicious or not essential until you find the culprit.
Also, the system logs may offer clues.
 
I would start by looking at which kexts are loaded (give time for the kernel to unload kexts that didn't start).
Then one at a time, eliminate kexts that are suspicious or not essential until you find the culprit.
Also, the system logs may offer clues.
Already did, only FakeSMC and it doesn't fix.

Uhm... I was thinking about the SD Card reader, I don't even know if it works, I don't even use it, I'm not sure if it would make any issue with this.
What do you think about it?
 
I don't think you understood. There is a while pile of kexts provided by Apple...
Hey!
I fixed it, there are a lot of _GPE that were messing my ACPI (WTF, why the hell there are GPEs for checking rotation lock?) I deleted all of them and now I have a smooth macOS, good battery life (actually reports about 1h with 24%, and I think it can be improved, because my actual SMBIOS makes the GPU work at 0'35GHz and the real lowest speed is 0'20GHz) and kernel_task issue is gone.
Thank you for all your time and help, I wish you the best!

Oh. Also... You should add a GPE remover on the laptop patches repo. Hahahaha
I learned A LOT of new things this last week, I really like to thank you and all this community :D
 
Hey!
I fixed it, there are a lot of _GPE that were messing my ACPI (WTF, why the hell there are GPEs for checking rotation lock?) I deleted all of them and now I have a smooth macOS, good battery life (actually reports about 1h with 24%, and I think it can be improved, because my actual SMBIOS makes the GPU work at 0'35GHz and the real lowest speed is 0'20GHz) and kernel_task issue is gone.
Thank you for all your time and help, I wish you the best!

Oh. Also... You should add a GPE remover on the laptop patches repo. Hahahaha
I learned A LOT of new things this last week, I really like to thank you and all this community :D

Removing all GPE methods (_Lxx) is probably a mistake...
Perhaps it is better to determine what is wrong with the methods instead and fix them... (use ACPIDebug.kext, and review the code, especially for multibyte EC access).
 
Removing all GPE methods (_Lxx) is probably a mistake...
Perhaps it is better to determine what is wrong with the methods instead and fix them... (use ACPIDebug.kext, and review the code, especially for multibyte EC access).
I'll check later, also... Do you know how to get rid of the screen garble at boot on 10.12? I've tried the El Capitan patches but no luck.
 
Back
Top