Jump to content

Asus G750JX sporadic reboots


8 posts in this topic

Recommended Posts

Hi again my friends...!!!

I'm working in a hack for a friend.

His laptop is an Asus G750JX running Big Sur 11.1.

Intel Core i7 4700HQ

Graphics: GeForce GTX770m

 

For some reason, laptop haves sporadic reboots.

I believe my patch work is right, but...

Can anyone see my SenMe report file and tell me if it have any error?
Thanks in advance.

 

PD. BootLoader is OC.

Send me MacBook-Pro-de-Antonio.zip

Edited by kali2000

There's quite a lot that looks wrong in your setup. You also omitted to posts detailed system's specs so :thumbsdown_anim:...

 

Patched DSDT table:

  • other than the nVidia dGPU (POP2.GFX0), iGPU HDMI audio (HDAU), audio (HDEF), wireless (ARPT) or USB power (EHC1/EHC2/XHC) properties injection, is there anything else you patched?
  • You've mixed the old-fashioned DTGP-based _DSM methods with more modern Return (Package ()) ones. Latter is better/simpler when former requires DTGP method to be added if not already present. All is fine though.
  • some of the injected nVidia properties are incorrect: invalid connector-type for con0 (02040000) and very invalid vram size (FFFFFFFF, imagine the decimal value of that !!!).
  • many of the nVidia properties you inject are not required. In fact, did you encounter issues with the dGPU without injecting properties?
  • final debug Store line of code at bottom of DSDT is clearly unnecessary but somehow indicative of your patching process...

 

OC config:

  • ACPI:
    • use of patched DSDT arguably unnecessary depending on your answers to above questions.
    • use of SSDT-PNLF definitely unnecessary. It's for iGPU and you're running on nVidia dGPU; get rid of it.
    • you use SSDT-XOSI but you've not implemented the binary patch that gores with it; it's therefore useless...
    • you've not implemented patches that rename EHCx to EH0x. DSDT shows you have USB2.0 controllers but these do not show in IOReg; not necessarily an issue though, I've experienced this on a Toshiba laptop when it runs in UEFI mode (only the XHC USB3 controller is active).
  • Device Properties:
    • All is wrong there...
    • for audio device @1b: device-id injection is incorrect and unnecessary anyway. This entire device entry is actually not required since you use redundant boot arg alcid=3 and you also injected that layout in the patched DSDT! Delete this entry or inject the correct properties if you want it listed in SysProfiler->PCI.
    • for iGPU device @2: wrong layout id for a HD4600 laptop (you inject the desktop layout) and entire device entry is not required since you run on the dGPU. Nothing shows in IOReg... Delete this entry.
    • for dGPU device @1,0: several incorrect properties are incorrect and/or invalid (Family (???)/device-id (???)/NVCAP (string instead of hex data)). Why inject properties here anyway when you already have a patched DSDT to that effect. Choose one or the other, not both, that's totally messy especially when done incorrectly. Delete this entry.
  • Kernel:
    • Quirks: AppleCpuPmCfgLock unnecessary. It's the equivalent of Clover's AICPUPM patch and applies to Sandy Bridge/Ivy Bridge platforms. Get rid of it. No impact of having it though.
  • NVRAM:
    • boot-args:
      • alcid=3: redundant with property injected in patched DSDT. Causes no harm though...
      • nvda_drv=1: this is for nVidia Web Driver. No such thing since Mojave so get rid of this.
      • -applbklbeta: what is that? Get rid of that.
      • kext-dev-mode=1: old and defunct boot arg applicable only to OS X Yosemite; get rid of this, it's totally useless.
  • SMBIOS:
    • you picked up MacBookPro11,5, (quad-core Haswell platform with AMD graphics) when MacBookPro11,3 would be much better (quad-core Haswell platform with nVidia Kepler graphics).
    • looks like you deliberately hid serial numbers, MLB, ROM, etc. If so, Ok; if not, make sure these are properly inserted; you may use OpenCore Configurator app to that effect.

 

Read up the documentation at Dortania:

https://dortania.github.io/OpenCore-Install-Guide/

 

Good luck.

Ok, so you've added a couple of specs:

  1. i7-4700HQ, which I had gathered through one of the debug files.
  2. GeForce GTX 770M, which differs from GTX 780M injected in DSDT and OC config.

IOReg shows id 10de:11e0 which would confirm GTX 770M and so would 10de:0e0b which is GK106 HDMI audio controller. It's important to note that:

  • GeForce GTX 770M, being based on Kepler GK106, is likely to suffer from Kepler driver's bug (an OS X/macOS issue never resolved by Apple) which causes glitches/corruption.
  • GeForce GTX 780M, based on Kepler GK104, would not be subject to driver's bug.

If laptop suffers from the GK106-related bug, you may want to fallback to HD4600 iGPU which would be done by 1) enabling Optimus in BIOS, 2) disabling the nVidia dGPU through a patched SSDT and 3) injecting adequate HD4600 properties + SMBIOS.

@Hervé Thanks a lot of for your reply and your help...

I have modified my files following your instructions.

Please take a look at my new ACPI folder and my config.plist to check my changes and tell me if they are correct.

Once again, thank you very much for your help.

 

PD. My DSDT has some other patches... battery, FNkeys, Brightness keys ...

 

Archivo comprimido.zip

Edited by kali2000
3 hours ago, Hervé said:

Ok, so you've added a couple of specs:

  1. i7-4700HQ, which I had gathered through one of the debug files.
  2. GeForce GTX 770M, which differs from GTX 780M injected in DSDT and OC config.

IOReg shows id 10de:11e0 which would confirm GTX 770M and so would 10de:0e0b which is GK106 HDMI audio controller. It's important to note that:

  • GeForce GTX 770M, being based on Kepler GK106, is likely to suffer from Kepler driver's bug (an OS X/macOS issue never resolved by Apple) which causes glitches/corruption.
  • GeForce GTX 780M, based on Kepler GK104, would not be subject to driver's bug.

If laptop suffers from the GK106-related bug, you may want to fallback to HD4600 iGPU which would be done by 1) enabling Optimus in BIOS, 2) disabling the nVidia dGPU through a patched SSDT and 3) injecting adequate HD4600 properties + SMBIOS.

For now, and if you give the green light to my new files, I'm going to follow up with nVidia and put the computer through graphical rendering work to test its performance and check if it reboots.
If continue the issue, I will have to use HD4600.
Thank you.

You've not followed all the recommendations I made and you somehow managed to make your DSDT even worse for the Nvidia properties injection. Revert to the bare minimum and refrain from making things complex.

  • 5 months later...
×
×
  • Create New...