Contribute
Register

High Sierra Random Freezing

Status
Not open for further replies.
Thireus said:
Which clearly states: "MTLCommandBufferErrorDomain" --> MTL = Metal.

I'm gonna launch the game again, this time I'm confident I'm gonna trigger a KP.
Click to expand...
Agreed - But that is a different error than you had before :

Thireus said:
last loaded kext at 244887126842148: com.apple.driver.usb.cdc.acm 5.0.0 (addr 0xffffff7f90f93000, size 57344)
last unloaded kext at 71698991526404: com.apple.driver.AppleXsanScheme 3 (addr 0xffffff7f90e50000, size 32768)
Click to expand...
 
Thireus said:
I had to restart the game 2 or 3 times until the hard freeze eventually happened. I think it's pretty safe to say Nvidia kexts are to blame.
Click to expand...
Maybe, but I still think that it is third party software related.

I guess that the way to find out would be to make a fresh installation on a different drive, install only the web drivers and your game then test.
 
P1LGRIM said:
Maybe, but I still think that it is third party software related.
Click to expand...

Yep, Nvidia kexts are third party drivers. ;)

Besides, I'm using NvidiaGraphicsFixup's patch https://github.com/lvs1974/NvidiaGraphicsFixup/commit/2548a94bce9fe38e855ca17f19bea05978938cd7, probably the patch doesn't fix everything and maybe introduces new instability issues.

Anyway, I'm glad I found a way to trigger these freezes myself rather than waiting 3-4 days of uptime for them to happen by themselves.

Edit: Nope, it's not related to the patch.
 
Last edited:
Do you have your iGPU enabled? I was experiencing freezing too until I followed some new steps, added some patches to my config.plist and enabled my iGPU, now no more freezing.
 
ashman70 said:
Do you have your iGPU enabled? I was experiencing freezing too until I followed some new steps, added some patches to my config.plist and enabled my iGPU, now no more freezing.
Click to expand...

No prob not. I have not done much of anything other then plug stuff in.
could you send me what you did?
 
Build 1 Titan Ridge TB3
Z370 AORUS Gaming 5 |  i9 9900K |  RX 6900 XT (16 GB)
Build 1 Titan Ridge TB3
Z370 AORUS Gaming 5 |  i9 9900K |  RX 6900 XT (16 GB)
3 days uptime without any KP!
 
jaraheel said:
3 days uptime without any KP!
Click to expand...

I just had a "random" crash again after a few days of uptime. I believe I finally have the answer about what's causing these crashes: Nvidia drivers.

I must be using some apps that take advantage of the GPU, or maybe I'm just have too many Windows at the same time. The random system hard freeze (KP) I just got today is exactly similar to the KP I was getting with the latest Nvidia drivers and NvidiaGraphicsFixup on 10.13.4 when playing Fortnite.

Code: 
Anonymous UUID:       3016C5F0-081C-315F-2D7E-D03EF4FFBBA3

Mon Apr  9 10:00:05 2018

*** Panic Report ***
Machine-check capabilities: 0x0000000000000c14
 family: 6 model: 85 stepping: 4 microcode: 33554499
 signature: 0x50654
 Intel(R) Core(TM) i9-7980XE CPU @ 2.60GHz
 20 error-reporting banks
Processor 18: IA32_MCG_STATUS: 0x0000000000000005
 IA32_MC0_STATUS(0x401): 0xb200000000070005
Processor 19: IA32_MCG_STATUS: 0x0000000000000005
 IA32_MC0_STATUS(0x401): 0xb200000000070005
mp_kdp_enter(): 68652367871, 1, 36 TIMED-OUT WAITING FOR NMI-ACK, PROCEEDING
panic(cpu 26 caller 0xffffff8010b895da): "Machine Check at 0xffffff8010b99f69, registers:\n" "CR0: 0x000000008001003b, CR2: 0x000000010c013208, CR3: 0x00000000261e3000, CR4: 0x00000000003627e0\n" "RAX: 0x0000000000000020, RBX: 0xffffff80113ef7c0, RCX: 0x0000000000000001, RDX: 0x0000000000000000\n" "RSP: 0xffffffce53963ee0, RBP: 0xffffffce53963f10, RSI: 0x0000000000000002, RDI: 0xffffff801120b7f8\n" "R8:  0x0000000000000024, R9:  0x0000000000000001, R10: 0x0000754c0ccac240, R11: 0xffffff80113e9c80\n" "R12: 0x0000000000000020, R13: 0x0000000000000000, R14: 0x0000754c0ccac3a5, R15: 0x000000000000001a\n" "RFL: 0x0000000000000046, RIP: 0xffffff8010b99f69, CS:  0x0000000000000008, SS:  0x0000000000000010\n" "Error code: 0x0000000000000000\n"@/BuildRoot/Library/Caches/com.apple.xbs/Sources/xnu/xnu-4570.51.1/osfmk/i386/trap_native.c:168
Backtrace (CPU 26), Frame : Return Address
0xffffff8010959320 : 0xffffff8010a6e166 mach_kernel : _handle_debugger_trap + 0x4c6
0xffffff8010959370 : 0xffffff8010b96714 mach_kernel : _kdp_i386_trap + 0x114
0xffffff80109593b0 : 0xffffff8010b88a00 mach_kernel : _kernel_trap + 0x4d0
0xffffff8010959430 : 0xffffff8010a20180 mach_kernel : _return_from_trap + 0xe0
0xffffff8010959450 : 0xffffff8010a6dbdc mach_kernel : _panic_trap_to_debugger + 0x21c
0xffffff8010959580 : 0xffffff8010a6d99c mach_kernel : _panic + 0x5c
0xffffff80109595e0 : 0xffffff8010b895da mach_kernel : _panic_64 + 0xda
0xffffff80109596d0 : 0xffffff8010a209ff mach_kernel : _hndl_machine_check + 0xf
0xffffffce53963f10 : 0xffffff8010b8a80c mach_kernel : _machine_idle + 0x1dc
0xffffffce53963f30 : 0xffffff8010a8c442 mach_kernel : _processor_idle + 0xd2
0xffffffce53963f80 : 0xffffff8010a8ca10 mach_kernel : _idle_thread + 0x20
0xffffffce53963fa0 : 0xffffff8010a1f4f7 mach_kernel : _call_continuation + 0x17

BSD process name corresponding to current thread: kernel_task
Boot args: -v darkwake=0 keepsyms=1 debug=0x100

Mac OS version:
17E199

Kernel version:
Darwin Kernel Version 17.5.0: Mon Mar  5 22:24:32 PST 2018; root:xnu-4570.51.1~1/RELEASE_X86_64
Kernel UUID: 1B55340B-0B14-3026-8A47-1E139DB63DA3
Kernel slide:     0x0000000010800000
Kernel text base: 0xffffff8010a00000
__HIB  text base: 0xffffff8010900000
System model name: iMacPro1,1 (Mac-7BA5B2D9E42DDD94)

This KP above was not triggered by Fortnite this time, but just after using my hackintosh for several days (with many Finder/TextEdit Windows and a few apps open). Fortnite that is intensively using the GPU triggers these hard freezes above almost instantly after the game is launched or after 1min in-game, while a normal use of my hackintosh would only trigger these GPU-related hard freezes once in a while (every 2-4 days uptime).

Previous on 10.13.3 and 378.10.10.10.25.106 drivers, the random KP I was getting were traced as the following in the logs, like @rasa.acharya, @giftmusic, @eiriasemrys, @ophelia, @juantrix:

Code: 
Double fault at 0xffffff8***19a8

But now on 10.13.4 and 387.10.10.10.30.103 they trace as:

Code: 
TIMED-OUT WAITING FOR NMI-ACK, PROCEEDING

The stack trace is just dumb in my opinion and can't locate where exactly things started to go wrong, which clearly is (to me) linked to NVidia kexts.

Anyway... I don't have a solution for these. I'm just going to swap my Nvidia card for an AMD one I have to see if I can pass my 4 days uptime record. If I can, it will be a confirmation of the above and I will eventually think about buying a new AMD card.

EDIT: I swapped my Nvidia card for AMD card. The same exact KP happened when I was connected through Remote Desktop and played a video remotely. Which means the issue is not Nvidia related.

I was able to locate the real cause of the issue: AVX offsets.

My AVX offset was set way too high: at -5. I had to reduce it even more to -8. Now I can play Fortnite with my Nvidia card without major issues (still some Metal crashes from time to time, but no KP at all) and play videos through RDP without triggering the hard freeze (KP). Let's see if that holds a few weeks.
 
Last edited:
Status
Not open for further replies.
Copyright © 2010 - 2024 tonymacx86 LLC
Back
Top