Hack freezes and reboots when waking from sleep since adding Sapphire RX 5500 XT

[gipsy]

Hi there,

 

I built my latest Hack a few months ago and everything was working well until I finally found a graphic card and added it a couple of weeks ago. Since then I am facing random system freezes+ either reboots or hard reboots and whenever the Hack goes to sleep and I wake it with a mouse click, the system will reboot. I did not face this issue when only the iGpu was used and I would appreciate help in solving this issue which is quite annoying

 

I'm using a headless framebuffer for CometLake (0300923E) and have added the DeviceID of the graphic card in the plist's DeviceProperties but it doesn't help:

 

USB ports are well configured so I don't think the issue comes from here as I had no issues before adding the card 

 

and of course I have no freezes or reboots issues under Windows.

 

After a reboot the system will sometime generate a panic report, here is the latest one:

 

panic(cpu 4 caller 0xffffff7f9c75ae05): "[3:0:0][PPLIB] Failed Power Play Resume. Shutting back down. TTL Error Message: {929795357422:[3:0:0] Error PSP: event_id=0x180214 event_info:type=3 hw_id=0 event_specific_tag=0x4 pData=0 data_size=0 psp_rap_is_supported failed

}{936369645061:[3:0:0] Error PSP: event_id=0xd0100203 event_info:type=3 hw_id=0 event_specific_tag=0x2 pData=0 data_size=0 psp_bootloader_load_sysdrv failed

}{936369677238:[3:0:0] TTL Event source_id=2, event_id=0x900c0401, event_info:type= 3,hw_id=0 : swip_client_id= 6, swip_event_id=5 : (SW_IP_CLIENT_ID__PSP,EVENT__POWER_UP)} --> {929795357422:[3:0:0] Error PSP: event_id=0x180214 event_info:type=3 hw_id=0 event_specific_tag=0x4 pData=0 data_size=0 psp_rap_is_supported failed

}{936369645061:[3:0:0] Error PSP: event_id=0xd0100203 event_info:type=3 hw_id=0 event_specific_tag=0x2 pData=0 data_size=0 psp_bootloader_load_sysdrv failed

}{936369677238:[3:0:0] TTL Event source_id=2, event_id=0x900c0401, event_info:type= 3,hw_id=0 : swip_client_id= 6, swip_event_id=5 : (SW_IP_CLIENT_ID__PSP,EVENT__POWER_UP)}.

" @AmdRadeonController.cpp:1987

Panicked task 0xffffff9511064670: 229 threads: pid 0: kernel_task

Backtrace (CPU 4), panicked thread: 0xffffff8b75eb7000, Frame : Return Address

0xffffffe0cebfb8f0 : 0xffffff800d485ffd

0xffffffe0cebfb940 : 0xffffff800d5e6035

0xffffffe0cebfb980 : 0xffffff800d5d5803

0xffffffe0cebfb9d0 : 0xffffff800d425a60

0xffffffe0cebfb9f0 : 0xffffff800d4863cd

0xffffffe0cebfbb10 : 0xffffff800d485b86

0xffffffe0cebfbb70 : 0xffffff800dd16409

0xffffffe0cebfbbe0 : 0xffffff7f9c75ae05

0xffffffe0cebfbc70 : 0xffffff7f9c71e50c

0xffffffe0cebfbcb0 : 0xffffff7f9c75a674

0xffffffe0cebfbce0 : 0xffffff7f9c73df73

0xffffffe0cebfbd00 : 0xffffff7f9c73df0f

0xffffffe0cebfbd20 : 0xffffff7f9c73dd5d

0xffffffe0cebfbd50 : 0xffffff7fa65b630b

0xffffffe0cebfbdf0 : 0xffffff7fa65b606c

0xffffffe0cebfbe20 : 0xffffff7fa65bd116

0xffffffe0cebfbed0 : 0xffffff800dc58c53

0xffffffe0cebfbf20 : 0xffffff800dc574ae

0xffffffe0cebfbf60 : 0xffffff800dc56ad7

0xffffffe0cebfbfa0 : 0xffffff800d42518e

      Kernel Extensions in backtrace:

         com.apple.iokit.IOGraphicsFamily(593.0)[212D9F00-4979-3ACB-9108-4BB19404BE85]@0xffffff7fa65aa000->0xffffff7fa65d8fff

            dependency: com.apple.iokit.IOPCIFamily(2.9)[897B72E0-B98F-30BA-8CB2-4E5E469CE4B6]@0xffffff800ffe7000->0xffffff8010011fff

         com.apple.kext.AMDRadeonX6000Framebuffer(4.0.7)[D447BB77-AFEE-3CF0-A35D-4EA43CA49E94]@0xffffff7f9c70d000->0xffffff7f9c998fff

            dependency: com.apple.AppleGraphicsDeviceControl(6.4.12)[F167C5E6-A75A-3C49-B8BC-F8EBE9B75B08]@0xffffff7fa4d8a000->0xffffff7fa4d8dfff

            dependency: com.apple.iokit.IOACPIFamily(1.4)[9AAF8737-B8CD-3A43-A654-69FD563A54CA]@0xffffff800fb40000->0xffffff800fb41fff

            dependency: com.apple.iokit.IOGraphicsFamily(593)[212D9F00-4979-3ACB-9108-4BB19404BE85]@0xffffff7fa65aa000->0xffffff7fa65d8fff

            dependency: com.apple.iokit.IOPCIFamily(2.9)[897B72E0-B98F-30BA-8CB2-4E5E469CE4B6]@0xffffff800ffe7000->0xffffff8010011fff

 

Process name corresponding to current thread (0xffffff8b75eb7000): kernel_task

Boot args: agdpmod=pikera chunklist-security-epoch=0 -chunklist-no-rev2-dev chunklist-security-epoch=0 -chunklist-no-rev2-dev

 

Mac OS version:

21D49

 

Kernel version:

Darwin Kernel Version 21.3.0: Wed Jan  5 21:37:58 PST 2022; root:xnu-8019.80.24~20/RELEASE_X86_64

Kernel UUID:

KernelCache slide: 0x000000000d200000

KernelCache base:  0xffffff800d400000

Kernel slide:      0x000000000d210000

Kernel text base:  0xffffff800d410000

__HIB  text base: 0xffffff800d300000

System model name: iMac20,2 (Mac-)

System shutdown begun: NO

Panic diags file available: YES (0x0)

Hibernation exit count: 0

 

System uptime in nanoseconds: 938489310659

Last Sleep:           absolute           base_tsc          base_nano

  Uptime  : 0x000000da82526fd1

  Sleep   : 0x000000d8899c5b8b 0x00000016f8bbb078 0x0000000000000000

  Wake    : 0x000000d88d797a93 0x00000000fc1953c8 0x000000d88b14b388

Zone info:

Foreign   : 0xffffff801307f000 - 0xffffff801308c000

Native    : 0xffffff8044418000 - 0xffffffa044418000

Readonly  : 0xffffff85110e4000 - 0xffffff86aaa78000

Metadata  : 0xffffffe17938c000 - 0xffffffe1996a0000

Bitmaps   : 0xffffffe1996a0000 - 0xffffffe1a56a0000

 

I would greatly appreciate help to fix this issue...

 

My specs:

ASRock Z490 Phantom Gaming ITX/TB3

Intel I9 10850k

Sapphire Pulse RX 5500 XT 8GB 

32 GB DDR4 Ram 2666MHz 

Opencore version: 0.7.7

SMBIOS iMac 20,2

 

Thanks a lot!!!!

 

 

 

Edited January 31, 2022 by gipsy

[Baio77]

ioreg please , use SSDT-USBW and Kext https://github.com/osy/USBWakeFixup 

[gipsy]

@Baio77

 

The hack reboots instead of waking up even with SSDT-USBW+kext sadly

Here is a copy of ioreg

 

Thanks!

ioreg.zip

[Baio77]

Post All EFI.

Insert in EFI\OC\ACPI this file and insert in configplist.
restart and ioreg.
Archivio.zip

 

You have TB3 ports and these are very annoying to sleep and wake.

Let's try to fine-tune the efi and then evaluate.

Edited February 1, 2022 by Baio77

[gipsy]

@Baio77

 

Thanks! will try these.

I know that Thunderbolt is a pain but I have fully deactivated the port in the bios for the time being

 

I'll report back

[gipsy]

Ok,

 

I replaced all aml files in ACPI folder with the ones you provided, made a clean snapshot of the plist saved and rebooted.

Hit reset nvram (twice to be sure) on OC picker and launched MacOS but it didn't load. I get to the Apple logo screen with the progress bar which loads very slowly and then hangs at around 40%.

If it helps, here is the content of my ACPI folder (with which MacOS boots)

 

[Baio77]

N1 :HW systems with CPUs 8th Gen and above on 300+ series chipsets need SSDT-AWAC + SSDT-EC_USBX + SSDT-PMCR.

Replace All ACPI Folder Integrate the new SSDTs into configplist.

Use OCAuxilaryTools to configure the configplist is the best application ever to manage EFI Hackintosh
ACPI.zip 

Read this ACPI guide https://github.com/5T33Z0/OC-Little-Translated

If you're still having trouble, I need all the EFI.

Edited February 1, 2022 by Baio77

[MaLd0n]

use efi folder from here and generate one smbios/serial

https://www.olarila.com/topic/5676-hackintosh-efi-folders-for-all-chipsets-clover-and-opencore/

disable igpu on bios/uefi

[gipsy]

2 hours ago, Baio77 said:

N1 :HW systems with CPUs 8th Gen and above on 300+ series chipsets need SSDT-AWAC + SSDT-EC_USBX + SSDT-PMCR.

Replace All ACPI Folder Integrate the new SSDTs into configplist.

Use OCAuxilaryTools to configure the configplist is the best application ever to manage EFI Hackintosh
ACPI.zip 

Read this ACPI guide https://github.com/5T33Z0/OC-Little-Translated

If you're still having trouble, I need all the EFI.

With the aml files in ACPI.zip, the hack managed to boot but I still have the issue waking up from sleep.

I have attached my EFI (with the aml files from ACPI.zip and without my ID's) for you to have a look.

 

Thanks a million

EFI.zip

[gipsy]

2 hours ago, MaLd0n said:

use efi folder from here and generate one smbios/serial

https://www.olarila.com/topic/5676-hackintosh-efi-folders-for-all-chipsets-clover-and-opencore/

disable igpu on bios/uefi

 

This EFI expects iMacPro 1,1 SMBIOS doesn't it? Can't I just replace the SMBIOS info there with iMac 20,2, my ID's and keep iGPU enabled?

I'll wait for Baio77's feedback on my EFI before trying what you suggest

 

 

[MaLd0n]

Just change smbios if need

[gipsy]

@Baio77 Good morning, did you manage to have a look at my EFI folder? 

 

@MaLd0n I used gensmbios to get my serial etc. to get SystemUUID, MLB, SystemProductName and SystemSerialNumber and got the ROM from my ethernet card.

When I look at your plist I see all this info in PlatformInfo>Generic which I will replace with my values but I also see info under PlatformInfo>Datahub and PlatformNVRAM 

 

Under Platforminfo>Datahub, I can get the BoardProduct from Hackintool but where do I get BoardRevision and SMCPlatform? Can I leave these values as such?

Under PlatformNVRAM, do I leave FirmwareFeatures and FirmwareFeatursMask as such?

 

Thanks!

[rthpjm]

I've had similar wake-from-sleep crashes. Monterey 12.2.1 was very stable for me, even wake from sleep was working!

 

When 12.3 came out, I made the mistake of upgrading before checking the forums. Ouch! 12.3 introduced numerous issues for me, most noticeable were graphics artefacts on the desktop background (which I later discovered was due to issues with high efficiency encoding of the Apple background images), and wake-from-sleep causing a kernel panic. It was SO bad I took the extreme measure of downgrading back to 12.2.1. All the problems went away! It's worth noting that I did not change my OpenCore version at all during the upgrade (argh!), and then the downgrade (phew!), therefore it would appear that the issues were with Apple's changes to the graphics kernel extensions.

 

Monterey 12.3.1 has fixed most of my issues, except the kernel crash on wake-from-sleep.

 

My config is:

ASUS TUF Gaming X570 Plus Wifi

AMD Ryzen 3900X

MSI AMD RX 5700 XT MECH OC

 

OpenCore 0.7.9 using SMC set to MacPro7,1

 

panic(cpu 4 caller 0xffffff7f9c90cc5b): "[12:0:0][PPLIB] Failed Power Play Resume. Shutting back down. TTL Error Message: {} --> {79474988709:[12:0:0] Error SW_IP_CLIENT_ID__SMU: event_id=0xc0140403 event_info:type=3 hw_id=0 event_specific_tag=0x50 pData=0 data_size=0 %s %s %d %s}{79475011392:[12:0:0] Error SW_IP_CLIENT_ID__SMU: event_id=0x18020c event_info:type=3 hw_id=0 event_specific_tag=0x1 pData=0 data_size=0 %s %s %d %s}.
" @AmdRadeonController.cpp:1998
Panicked task 0xffffff904f1ad670: 261 threads: pid 0: kernel_task
Backtrace (CPU 4), panicked thread: 0xffffff904ff32000, Frame : Return Address
0xffffffd066d7b7f0 : 0xffffff800d683e2d mach_kernel : _handle_debugger_trap + 0x41d
0xffffffd066d7b840 : 0xffffff800d7e3cb6 mach_kernel : _kdp_i386_trap + 0x116
0xffffffd066d7b880 : 0xffffff800d7d350d mach_kernel : _kernel_trap + 0x51d
0xffffffd066d7b8d0 : 0xffffff800d623a60 mach_kernel : _return_from_trap + 0xe0
0xffffffd066d7b8f0 : 0xffffff800d6841fd mach_kernel : _DebuggerTrapWithState + 0xad
0xffffffd066d7ba10 : 0xffffff800d6839b6 mach_kernel : _panic_trap_to_debugger + 0x2b6
0xffffffd066d7ba70 : 0xffffff800df164bf mach_kernel : _panic + 0x84
0xffffffd066d7bb60 : 0xffffff7f9c90cc5b com.apple.kext.AMDRadeonX6000Framebuffer : __ZN34AMDRadeonX6000_AmdRadeonController10doGPUPanicEPKcz + 0x1b9
0xffffffd066d7bc70 : 0xffffff7f9c8d0100 com.apple.kext.AMDRadeonX6000Framebuffer : __ZN33AMDRadeonX6000_AmdPowerPlayHelper4wakeEv + 0xc0
0xffffffd066d7bcb0 : 0xffffff7f9c90c494 com.apple.kext.AMDRadeonX6000Framebuffer : __ZN34AMDRadeonX6000_AmdRadeonController4wakeEv + 0xa4
0xffffffd066d7bce0 : 0xffffff7f9c8efb8d com.apple.kext.AMDRadeonX6000Framebuffer : __ZN35AMDRadeonX6000_AmdRadeonFramebuffer6doWakeEv + 0x5b
0xffffffd066d7bd00 : 0xffffff7f9c8efb29 com.apple.kext.AMDRadeonX6000Framebuffer : __ZN35AMDRadeonX6000_AmdRadeonFramebuffer19setSystemPowerStateENS_15AmdFbPowerStateE + 0x7f
0xffffffd066d7bd20 : 0xffffff7f9c8ef8cb com.apple.kext.AMDRadeonX6000Framebuffer : __ZN35AMDRadeonX6000_AmdRadeonFramebuffer18doPowerStateChangeENS_15AmdFbPowerStateE + 0x43
0xffffffd066d7bd50 : 0xffffff7fa674b21b com.apple.iokit.IOGraphicsFamily : __ZN13IOFramebuffer14checkPowerWorkEj + 0x27d
0xffffffd066d7bdf0 : 0xffffff7fa674af80 com.apple.iokit.IOGraphicsFamily : __ZN14IOFBController14checkPowerWorkEj + 0xa0
0xffffffd066d7be20 : 0xffffff7fa675209c com.apple.iokit.IOGraphicsFamily : __ZN13IOFramebuffer10systemWorkEP8OSObjectP22IOInterruptEventSourcei + 0x124
0xffffffd066d7bed0 : 0xffffff800de47454 mach_kernel : __ZN22IOInterruptEventSource12checkForWorkEv + 0x114
0xffffffd066d7bf20 : 0xffffff800de45d0e mach_kernel : __ZN10IOWorkLoop15runEventSourcesEv + 0x13e
0xffffffd066d7bf60 : 0xffffff800de45337 mach_kernel : __ZN10IOWorkLoop10threadMainEv + 0x37
0xffffffd066d7bfa0 : 0xffffff800d62318e mach_kernel : _call_continuation + 0x2e
      Kernel Extensions in backtrace:
         com.apple.iokit.IOGraphicsFamily(594.0)[3ECB2385-311A-374E-9C0F-65639BA26A05]@0xffffff7fa673f000->0xffffff7fa676dfff
            dependency: com.apple.iokit.IOPCIFamily(2.9)[A436E92C-DE10-3718-AEF4-ED2A788A466A]@0xffffff8010203000->0xffffff801022efff
         com.apple.kext.AMDRadeonX6000Framebuffer(4.0.8)[BF08CA07-2A84-3890-89C1-DBDEC2A33F28]@0xffffff7f9c8bf000->0xffffff7f9cb46fff
            dependency: com.apple.AppleGraphicsDeviceControl(6.5.7)[B9E606A5-2F2C-3F4C-A2DE-0D3BF7E114A5]@0xffffff7fa4f29000->0xffffff7fa4f2cfff
            dependency: com.apple.iokit.IOACPIFamily(1.4)[CF250EA9-AE5C-3215-BCF5-763DB70A0228]@0xffffff800fd60000->0xffffff800fd61fff
            dependency: com.apple.iokit.IOGraphicsFamily(594)[3ECB2385-311A-374E-9C0F-65639BA26A05]@0xffffff7fa673f000->0xffffff7fa676dfff
            dependency: com.apple.iokit.IOPCIFamily(2.9)[A436E92C-DE10-3718-AEF4-ED2A788A466A]@0xffffff8010203000->0xffffff801022efff

Process name corresponding to current thread (0xffffff904ff32000): kernel_task
Boot args: 

Mac OS version:
21E258

Kernel version:
Darwin Kernel Version 21.4.0: Fri Mar 18 00:45:05 PDT 2022; root:xnu-8020.101.4~15/RELEASE_X86_64
Kernel UUID: B6F8637B-0844-355F-8C82-60FA06149384
KernelCache slide: 0x000000000d400000
KernelCache base:  0xffffff800d600000
Kernel slide:      0x000000000d410000
Kernel text base:  0xffffff800d610000
__HIB  text base: 0xffffff800d500000
System model name: MacPro7,1 (redacted)
System shutdown begun: NO
Panic diags file available: YES (0x0)
Hibernation exit count: 0

System uptime in nanoseconds: 79475483968
Last Sleep:           absolute           base_tsc          base_nano
  Uptime  : 0x00000012811ba7d5
  Sleep   : 0x00000010bc8adf69 0x0000001a562de0d5 0x0000000000000000
  Wake    : 0x00000010e1da6f0c 0x0000000136c7fd44 0x00000010dabcf0c7
Compressor Info: 0% of compressed pages limit (OK) and 0% of segments limit (OK) with 0 swapfiles and OK swap space
Zone info:
  Foreign : 0xffffff801e3ff000 - 0xffffff801e40d000
  Native  : 0xffffff804f0d8000 - 0xffffffa04f0d8000
  Readonly: 0xffffff851bda4000 - 0xffffff86b573d000
  Metadata: 0xfffffff8fe525000 - 0xfffffff91e832000
  Bitmaps : 0xfffffff91e832000 - 0xfffffff92a832000

 

I think we'll need to wait and see if there is any improvement with 12.4.0. I have seen other MacBook users complaining of similar issues when using Navi graphics cards (even 5700 XTs in eGPU enclosures). There are other forum threads that have similar issues with wake-from-sleep kernel panics citing Power Play Resume failures going back to Mac OS 10.15 (Catalina, 10.15.4 in particular seems to have been affected):

• https://discussions.apple.com/thread/251339239
• https://forums.macrumors.com/threads/16-mbp-gpu-panics.2221004/

Some of these older threads that reference MacBook Pros say the issue was resolved by a motherboard replacement. My issue is unlikely to be hardware because 12.2.1 was stable for me, and 12.3.x has not been with no changes to hardware (and no changes to the OpenCore boot loader).

 

It seems to that the AMDRadeonX6000Framebuffer.kext has been troublesome for a while. Worse than that, Apple engineering seem to "fix it", but then re-introduce the same issue as a regression as they are moving through the OS upgrade cycle. I have my fingers crossed that the Apple Engineering teams will see some of these threads and feedback/crash reports and be motivated to address the issue. I can live with the issue by disabling sleep and disabling power nap. I now choose to put the display to sleep (rather than putting the whole computer to sleep) whilst away from the keyboard for short periods, and choosing to shutdown fully if walking away for a long period (saves electricity any way, better for the planet and my wallet!)! Since my rig is booting from an NVMe SSD it's not a big problem since the boot time is acceptable, but it would be nice to get wake-from-sleep working again!