Contribute
Register

Gigabyte Z490 Vision D (Thunderbolt 3) + i5-10400 + AMD RX 580

Apple implemented Thunderbolt and USB4 into their M1 chip... macOS currently supports Thunderbolt 3 Alpine Ridge and Titan Ridge... but we don’t know if it supports Thunderbolt 4 Maple Ridge...and even if it does, will the community need (or be able) to enable Tthunderbolt buss and hot plug in macOS on Maple Ridge (or will it just work out of the box?).

2021 will be interesting indeed.
Something else to note is that my 2020 13" MacBook Pro with four Thunderbolt ports and Ice Lake CPU uses Thunderbolt controller with device IDs:
  • 0x8A17 and 0x8A0D for the two NHI Controllers.
  • 0x8A1D, 0x8A1F, 0x8A21, 0x8A23 for the four Thunderbolt ports.
  • 0x8A13 for the USB 3 controller tied to those four USB-C ports.
However, Maple Ridge device ID is 0x1137 according to Linux source code.

What is interesting is that Apple has 3 versions of the Thunderbolt driver now:
  • Type3 for Alpine Ridge and Titan Ridge
  • Type4 for Ice Lake Thunderbolt
  • Type5 for Apple Silicon M1 Thunderbolt
Several questions come immediately to mind:
  • What is so different about Ice Lake Thunderbolt that a new driver class was created?
  • Does Type5 support Intel's Maple Ridge?
  • Is Apple's own M1 Thunderbolt based on Intel's Maple Ridge?
    • If so, then is Type5 a "universal" binary or is it ARM-only?
Screen Shot 2021-01-12 at 10.42.14 AM.png
Screen Shot 2021-01-12 at 10.26.29 AM.png
Screen Shot 2021-01-12 at 10.44.32 AM.png
 
I would be glad to upload the EFI folder from my X570 Taichi, which is already running macOS 11.2 beta and OpenCore 0.6.5 with the latest AMD 17H kernel patches. The system, however, lost its ability to go to sleep. Sleep did work during the late Catalina period, but hasn't been working with Big Sur. Might be due to my NZXT Kraken all-in-one liquid cooler that is connected to an internal USB 2 header, but I haven't spent much time chasing down the problem.

Because modern Macs feature Thunderbolt natively, my Hackintosh interest lies exclusively in systems where Thunderbolt can be made to work well. On the X570 Taichi, I'm using an ASRock Thunderbolt card connected to the Thunderbolt header. But I'm unable to flash this card and consequently my eGPU will not work. Basic Thunderbolt devices will connect, however, but hot plug has been iffy.
That's it. This is probably what's keeping my z490 from sleeping. I'm using the same NZXT Kraken liquid cooler.
 
If Type5 is universal, then Z590 should have a pretty good chance to have full TB4; if not, then maybe we can at least make TB3 work?

Something else to note is that my 2020 13" MacBook Pro with four Thunderbolt ports and Ice Lake CPU uses Thunderbolt controller with device IDs:
  • 0x8A17 and 0x8A0D for the two NHI Controllers.
  • 0x8A1D, 0x8A1F, 0x8A21, 0x8A23 for the four Thunderbolt ports.
  • 0x8A13 for the USB 3 controller tied to those four USB-C ports.
However, Maple Ridge device ID is 0x1137 according to Linux source code.

What is interesting is that Apple has 3 versions of the Thunderbolt driver now:
  • Type3 for Alpine Ridge and Titan Ridge
  • Type4 for Ice Lake Thunderbolt
  • Type5 for Apple Silicon M1 Thunderbolt
Several questions come immediately to mind:
  • What is so different about Ice Lake Thunderbolt that a new driver class was created?
  • Does Type5 support Intel's Maple Ridge?
  • Is Apple's own M1 Thunderbolt based on Intel's Maple Ridge?
    • If so, then is Type5 a "universal" binary or is it ARM-only?
View attachment 504773View attachment 504772View attachment 504774
 
Just to close the loop on the issue with Windows 10 and MacOS not waking up from sleep, because of the Fenvi T919 card.

My spec: i7-10700k, Radeon 5700 XT Gigabyte OC, Gigabyte Z490 Vision D (Bios F5a).

Replacing this card https://www.amazon.com/gp/product/B07WGS46PY/ with https://www.amazon.com/dp/B082X8MBMD/?tag=tonymacx86com-20 solved the issue for me and @dranose (if I understood it correctly)

Disclaimer: I'm not saying that https://www.amazon.com/gp/product/B07WGS46PY/ doesn't work. AFAIK many folks with the same hardware spec had no issues with this card.

Thanks @CaseySJ for the great tutorial and config.
How does men know which version is right?

I am planning in buying this card:


But how do I know if this is the correct one?
 
Hello @daddaa,

One or both of your external drives is formatted as ExFAT, and the kernel panic is most likely occurring when reading or writing to those drives.

Questions:
  • How are these external drives connected?
    • Are they mounted in an external USB enclosure and connected via USB cable?
    • Are they mounted inside a multi-bay drive array?
    • If the two drives are connected to two different USB ports, which ports are they connected to?
    • Or are they connected to a USB hub?
    • Are the two drives self-powered or powered through the USB cable itself?
    • Are the two drives regular (mechanical) hard drives or SSDs?
  • If you copy the Adobe After Effects file(s) to an internal SSD temporarily (and disconnecting both of the external drives), can you perform all the functions on them successfully without encountering a system crash?
  • Is it necessary to format your drives as ExFAT or can they be formatted as either HFS+ or APFS?
    • Do not reformat the drives at this time.
Hi @CaseySJ,
thanks for your quick reply!

Normally, I also would suspect the drive. Maybe the example with the external drive was a bit misleading, as the freezing also occours without any external drive and opening the After Effects project sitting on the internal nvme SSD (which btw is a Samsung 970 Evo plus 500gb). -> So this answers your second bullet point :)

Here you find the panic report from the crash without any external drives:
Code:
Panic(CPU 0, time 87550693709): NMIPI for unresponsive processor: TLB flush timeout, TLB state:0x0
RAX: 0x0000000009000000, RBX: 0x000000000001fbe4, RCX: 0xffffff800a80ceb8, RDX: 0x0000000009000000
RSP: 0xffffff8384f30e80, RBP: 0xffffff8384f30e80, RSI: 0x0000000000000000, RDI: 0xffffff800a80ceb8
R8:  0x0000000000000002, R9:  0x0000000004104d00, R10: 0x000000000000001f, R11: 0x0000000000000000
R12: 0x0000000000000000, R13: 0xffffff800a80ceb8, R14: 0xffffff8009f36e70, R15: 0xffffff8055c1d310
RFL: 0x0000000000000046, RIP: 0xffffff800a03cd32, CS:  0x0000000000000008, SS:  0x0000000000000000
Backtrace (CPU 0), Frame : Return Address
0xffffff8384f30d30 : 0xffffff800a043e3b mach_kernel : _interrupt + 0x12b
0xffffff8384f30d80 : 0xffffff8009ebed47 mach_kernel : _return_to_iret + 0x157
0xffffff8384f30e80 : 0xffffff8009f634f4 mach_kernel : _timer_queue_expire_with_options + 0x2a4
0xffffff8384f30f40 : 0xffffff800a031847 mach_kernel : _timer_intr + 0x87
0xffffff8384f30f80 : 0xffffff800a043f3a mach_kernel : _interrupt + 0x22a
0xffffff8384f30fd0 : 0xffffff8009ebebed mach_kernel : _hndl_allintrs + 0x11d

BSD process name corresponding to current thread: After Effects
Boot args: keepsyms=1 dart=0 debug=0x100 agdpmod=pikera alcid=11 shikigva=80

Mac OS version:
19H15

Kernel version:
Darwin Kernel Version 19.6.0: Thu Oct 29 22:56:45 PDT 2020; root:xnu-6153.141.2.2~1/RELEASE_X86_64
Kernel UUID: 9B5A7191-5B84-3990-8710-D9BD9273A8E5
Kernel slide:     0x0000000009c00000
Kernel text base: 0xffffff8009e00000
__HIB  text base: 0xffffff8009d00000
System model name: iMac19,1 (Mac-AA95B1DDAB278B95)
System shutdown begun: NO
Panic diags file available: YES (0x0)

System uptime in nanoseconds: 87550834991
panic(cpu 12 caller 0xffffff800a01f51b): "IPI timeout, unresponsive CPU bitmap: 0x1, NMIPI acks: 0x0, now: 0x1, deadline: 87550692276, pre-NMIPI time: 0x14626d8c35, current: 0x14626fbb1e, global: 0"@/AppleInternal/BuildRoot/Library/Caches/com.apple.xbs/Sources/xnu/xnu-6153.141.2.2/osfmk/x86_64/pmap.c:2909
Backtrace (CPU 12), Frame : Return Address
0xffffffa3d5bfb8d0 : 0xffffff8009f1868d mach_kernel : _handle_debugger_trap + 0x49d
0xffffffa3d5bfb920 : 0xffffff800a052ab5 mach_kernel : _kdp_i386_trap + 0x155
0xffffffa3d5bfb960 : 0xffffff800a04463e mach_kernel : _kernel_trap + 0x4ee
0xffffffa3d5bfb9b0 : 0xffffff8009ebea40 mach_kernel : _return_from_trap + 0xe0
0xffffffa3d5bfb9d0 : 0xffffff8009f17d57 mach_kernel : _DebuggerTrapWithState + 0x17
0xffffffa3d5bfbad0 : 0xffffff8009f18147 mach_kernel : _panic_trap_to_debugger + 0x227
0xffffffa3d5bfbb20 : 0xffffff800a6bf2bc mach_kernel : _panic + 0x54
0xffffffa3d5bfbb90 : 0xffffff800a01f51b mach_kernel : _pmap_flush_tlbs + 0x83b
0xffffffa3d5bfbc40 : 0xffffff800a025212 mach_kernel : _pmap_remove_range_options + 0x3a2
0xffffffa3d5bfbd20 : 0xffffff800a025fa6 mach_kernel : _pmap_remove_options + 0x1f6
0xffffffa3d5bfbd80 : 0xffffff8009fb1c45 mach_kernel : _vm_map_destroy + 0x8d5
0xffffffa3d5bfbed0 : 0xffffff8009fb960e mach_kernel : _vm_map_remove + 0x5e
0xffffffa3d5bfbf00 : 0xffffff7f8c3a606d com.apple.kec.pthread : __bsdthread_terminate + 0xa4
0xffffffa3d5bfbf40 : 0xffffff800a582437 mach_kernel : _unix_syscall64 + 0x287
0xffffffa3d5bfbfa0 : 0xffffff8009ebf206 mach_kernel : _hndl_unix_scall64 + 0x16
      Kernel Extensions in backtrace:
         com.apple.kec.pthread(1.0)[02A1EFC8-F978-38C0-9441-52EFA6C4016A]@0xffffff7f8c3a5000->0xffffff7f8c3aefff

BSD process name corresponding to current thread: After Effects
Next steps to try:
- I ordered a WD SN750 drive and will clone macOS to that to rule out the Samsung SSD
- Do a Memtest to rule out issues with the RAM (Crucial Ballistix 2x 16gb, 3200mhz)
 
Hi @CaseySJ,
thanks for your quick reply!

Normally, I also would suspect the drive. Maybe the example with the external drive was a bit misleading, as the freezing also occours without any external drive and opening the After Effects project sitting on the internal nvme SSD (which btw is a Samsung 970 Evo plus 500gb). -> So this answers your second bullet point :)

Here you find the panic report from the crash without any external drives:
...
This kernel panic is squarely due to Adobe After Effects. Could it be attributed to a particular plug-in? Does After Effects support plug-ins? Do you have a real Mac on which to test the same use case with the same file(s)?

Next steps to try:
- I ordered a WD SN750 drive and will clone macOS to that to rule out the Samsung SSD
Ouch, the WD SN750 is on our list of problematic drives. Recommended brands are:
  • Sabrent Rocket NVMe
  • Silicon Power NVMe
  • ADATA NVMe
  • Many others also qualify, but I have personal experience with the above three
- Do a Memtest to rule out issues with the RAM (Crucial Ballistix 2x 16gb, 3200mhz)
If the kernel panic only occurs while using Adobe After Effects, then memory might not be the issue. However, this is still a good thing to do.
 
This kernel panic is squarely due to Adobe After Effects. Could it be attributed to a particular plug-in? Does After Effects support plug-ins? Do you have a real Mac on which to test the same use case with the same file(s)?
I'm using a template with some imported presets. That shouldn't be anything special I hope. And yes, I have a real Mac. On my Macbook Pro the project works without any issue from an external drive (except the Macbook is of course way slower than the Hackintosh).

If the kernel panic only occurs while using Adobe After Effects, then memory might not be the issue. However, this is still a good thing to do.
It occured once without opening AE, but unfortunately didn't keep the panic report. Let's see what will turn out from the check.

Ouch, the WD SN750 is on our list of problematic drives. Recommended brands are:

Ooops :oops: Thanks for the hint! Just read about the EVO plus being problematic and didn't investigate any further in panic of getting an alternative. Then probably I'll use the WD for the Windows side of the machine.
 
Back
Top