[Need help] Weird kernel panic

[_ML64]

Hello,

 

I'm actually not very new to hackintoshing, but I tried wrapping my head around this one and can't seem to get it right.

 

PC Specs:

 

i7-4770k (no vt-d)

GTX 780

4x4GB Corsair & Kingston RAM

256+512GB Samsung SSDs

Gigabyte GA-Z87M-D3H

No HDD

 

Using Clover.

 

Clover options:

Default (except verbose boot enabled, AICPUPM patch enabled, Kernel PM Patch enabled)

 

Clover config file untouched since after installing it.

 

Kexts injected: FakeSMC and NullCPUPM, also copied to InstallerPartition/S/L/E/

 

Now, please correct me if I'm wrong, but OSX on Haswell will _only_ use XCPM, am I correct? I've left Clover to do its own P-States generation, but in the end I always get a nasty kernel panic as per the below screenshot.

 

I managed to get a few snaps of the KP, but before doing anything else (dumping logs, if possible?) the system rebooted. I'll appreciate it if anyone can tell me how to disable reboot on panic.

 

Looks like it's either AppleSMC related, or XCPM.

 

Honestly, I'm pretty sure if I use Chameleon for this installation, it'd be a breeze. Strangely, my mileage with Clover has been terrible since forever. But I think it's time to move on and try facing Clover, and well this is what I get

 

I think I might get an answer to this from olarila, but for whatever reason my ISP seems to dislike that site (not blocked, just timeout, other ISPs work fine.)

 

Thanks!

 

[WaldMeister]

Hi,

 

 

 

AICPUPM patch enabled, Kernel PM Patch enabled

 

 

Kexts injected: FakeSMC and NullCPUPM, also copied to InstallerPartition/S/L/E/

 

Remove NullCPUPM and let Clover generate P-states and C-states if no SSDT for power management is present. 

 

NullCPUPM disables power management, AICPUPM and Kernel PM (XCPM mode) enables it by applying a patch to prevent kernel panics.

[_ML64]

Hi,

 

 

Remove NullCPUPM and let Clover generate P-states and C-states if no SSDT for power management is present. 

 

NullCPUPM disables power management, AICPUPM and Kernel PM (XCPM mode) enables it by applying a patch to prevent kernel panics.

 

Hi,

 

Unfortunately, that did not work. Tried it, and got the same KP. Dropped all OEM SSDTs on the 2nd attempt, too, still no-no.

 

Somehow the /S/L/E folder on the flash drive vanished when I plugged it into my Mac, I remembered it having around 200 kexts before!

 

<code>

ls /Volumes/YosemiteInstaller/System/Library/

-- result --

Caches CoreServices

----

</code>

 

(Renamed partiiton in this post to avoid confusion)

 

This probably isn't important, but why is GeForceSensors taking so long to "start accelerator"?

 

 

Thanks for the help!

[WaldMeister]

Geforcesensors is a kernel extension that will provide information about your card when using HWSensors. It loads before some kernel extensions, and is waiting for the accelerator to start. 

"The accelerator"  is the NVDAStartup and  NVDAGK100Hal, GeforceSensors has to wait for these to be loaded to read any values.

 

About the panic, could you remove the boot flag: -xcpm ?

[_ML64]

Geforcesensors is a kernel extension that will provide information about your card when using HWSensors. It loads before some kernel extensions, and is waiting for the accelerator to start. 

"The accelerator"  is the NVDAStartup and  NVDAGK100Hal, GeforceSensors has to wait for these to be loaded to read any values.

 

About the panic, could you remove the boot flag: -xcpm ?

 

Hi,

 

I'm pretty sure I've always left it out

[_ML64]

Created a new installation media, this time I've made sure not to do anything stupid with it. (copy kexts to the wrong place, etc)

config.plist attached, renamed to .txt. Same panic.

I'll try to generate a CPU SSDT using Piker's script, edit Clover's config.plist to something better than iMac14,2 Clover generated for me.

Contents of EFI/CLOVER/drivers64UEFI attached too.

config.txt