Gigabyte X299X - Catalina Support

[teosoft]

I have tried all BIOS released for this card and rest assured they all work. If it does not boot anymore, probably something went wrong during the update process, try again and check on info if the installed BIOS version is the correct one.
Also check that you haven't forgotten any settings.

Yes, it was my BIOS settings. Unfortunately, I was unable to upgrade to Big Sur from your latest EFI, modded per my previous message. However, I was able to upgrade to Big Sur 11.5.2, reverting to my previous EFI (also based on your work @LeleTuratti, thank you so much!), and using F3c.

Now, everything works in Bug Sur as it worked under Catalina.

I do hesitate to upgrade to Big Sur 11.6.3 though, as I need this box to work on my projects. Any advice, @LeleTuratti ? I've found that BIOS version doesn't matter as much as many people here tend to believe.

 

[LeleTuratti]

Yes, it was my BIOS settings. Unfortunately, I was unable to upgrade to Big Sur from your latest EFI, modded per my previous message. However, I was able to upgrade to Big Sur 11.5.2, reverting to my previous EFI (also based on your work @LeleTuratti, thank you so much!), and using F3c.

Oh ok, so Big Sur still doesn't work with the latest BIOS of this card.

My latest EFI should work perfectly with Big Sur, in fact now I am using it with the latest available version 11.6.3. The only change required is in UEFI > APFS: MinDate and MinVersion set to "0" (Zero).

 

[teosoft]

Oh ok, so Big Sur still doesn't work with the latest BIOS of this card.

My latest EFI should work perfectly with Big Sur, in fact now I am using it with the latest available version 11.6.3. The only change required is in UEFI > APFS: MinDate and MinVersion set to "0" (Zero).

I haven't tried the latest, which is F3j. I only tried F3i, so that I could reinstall F3c.
I already have UEFI -> APFS MinDate and MinVersion set to 0.

Now, @LeleTuratti I'd love to use your latest EFI, and I have a few of questions:

1. Which BIOS version I need to use for it to work? Which one do you use with it + 11.6.3?
2. Ocvalidate complains UEFI->Input->PointerSupport is set to Yes, but UEFI->Input->PointerSupportMode is not ASUS
3. Do I have to enable Resizable-BAR in BIOS? Does it actually "enhance GPU performance" as Gigabyte claims?

I think because of #2 the OpenCanopy sometimes crashes for me, if I try to select boot entries.

Thank you!

 

[gabevf]

Sorry, I thought Designare EX had Titan Ridge as a controller.

However I understand what the problem is. The SSDT you are using is made to enable Thunderbolt BUS, which is theoretically possible on the Alpine Ridge, but does not work on Gigabyte motherboards because it requires the "GPIO3 Force power" option enabled on the BIOS, but on Gigabyte motherboards this option was removed years ago.

I have personally tested an SSDT similar to yours two years ago on my previous GB X299 with the Alpine Ridge Thunderbolt PCI card, but the GPE method was not working properly and force injecting "GPIO3 Force power" with the SSDT caused Kernel Panic.
The only way to enable the Thunderbolt BUS therefore remained flashing the Thunderbolt chip but I excluded this option because in my case I didn't need it.

Add and enable this SSDT and don't forget to clean the NVRAM after adding it.

Thanks so much for this @LeleTuratti!
However, I couldn't get it to work either, here are my results:

• TB Boot Support ENABLED
  • could not boot with OWC Dock plugged in – does not get past the BIOS splash screen
  • was able to boot without OWC dock plugged in, but TB doesn't work. USB-C devices not recognized either
• TB Boot Support DISABLED
  • Unable to boot. Hangs at different parts of boot sequence. Seems like its NVMe related, but I can't tell. I have NVMeFix.kext, btw – not sure if that makes a difference.
  • Screenshots attached.

Any ideas?


Update 1: I was able to boot by disabling NVMeFix.kext, but USB-C does not work, and TB3 is still unstable.

Update 2: Turns out that Enabling Thunderbolt Boot Support disables USB-C functionality.


I don't mind keeping my system "native" (without the TB SSDT), but I would really like to fix the KP's during sleep. The KP's only occur after an extended period of sleep (like overnight). I don't know what's causing it, but I suspect its a power-up due to one of the TB devices waking up. I noticed that I'm able to wake up the machine with a mouse-click, for example if the mouse is connected to the OWC dock, even though I have the GPRW ACPI fix It would be nice if I could alter the GPRW ACPI fix to apply to TB devices as well, do you think that is possible? is it just removing the _PRW methods from the SSDT you sent?

 

[LeleTuratti]

Now, @LeleTuratti I'd love to use your latest EFI, and I have a few of questions:

1. Which BIOS version I need to use for it to work? Which one do you use with it + 11.6.3?
2. Ocvalidate complains UEFI->Input->PointerSupport is set to Yes, but UEFI->Input->PointerSupportMode is not ASUS
3. Do I have to enable Resizable-BAR in BIOS? Does it actually "enhance GPU performance" as Gigabyte claims?

Now I no longer use the Gigabyte X299X Designare, however:

1. I was using the latest BIOS F3j with MacOS Catalina. I don't know if there are any problems with the latest versions of Big Sur with the F3c BIOS, but I don't think so.
2. It is not relevant.
3. My EFI has Resizable-BAR enabled, so you have to activate it in the BIOS as well, otherwise you will have a KP. Unfortunately only the RX 6000 series benefits from this option in MacOS, so you won't get a real performance boost. Alternatively you have to disable the Resizable-BAR in the Config.

 

[LeleTuratti]

I don't mind keeping my system "native" (without the TB SSDT), but I would really like to fix the KP's during sleep. The KP's only occur after an extended period of sleep (like overnight). I don't know what's causing it, but I suspect its a power-up due to one of the TB devices waking up. I noticed that I'm able to wake up the machine with a mouse-click, for example if the mouse is connected to the OWC dock, even though I have the GPRW ACPI fix It would be nice if I could alter the GPRW ACPI fix to apply to TB devices as well, do you think that is possible? is it just removing the _PRW methods from the SSDT you sent?

I believe there are some configuration problems in your EFI folder.
If you upload your EFI and Ioreg I try to check as soon as I have a moment.
What MacOS version do you use?
Which BIOS version?
Which OC version?

 

[gabevf]

I believe there are some configuration problems in your EFI folder.
If you upload your EFI and Ioreg I try to check as soon as I have a moment.
What MacOS version do you use?
Which BIOS version?
Which OC version?

Here are the requested files, plus a few other things.

Info:
macOS: 12.1
BIOS: F7e
OC: 0.7.7

I got the F7e BIOS directly from GA technical support when I was having issues with my multi-GPU setup, and this one is meant to enable the Resizable BAR option.

 

[LeleTuratti]

Hi @gabevf, I checked your EFI and removed all patches and SSDTs that are not needed and enabled some essential quirks. Sleep should now work.
I have enabled ResizeAppleGpuBars, remember to enable it on the BIOS as well.
Remember to perform an NVRAM reset on startup (Press the spacebar at the OC screen to see the auxiliary options).

Ps. I also noticed that the USB port mapping is incomplete and possibly incorrect, this could have consequences with Sleep and Thunderbolt.

Let me know if everything works.

 

[gabevf]

Hi @gabevf, I checked your EFI and removed all patches and SSDTs that are not needed and enabled some essential quirks. Sleep should now work.
I have enabled ResizeAppleGpuBars, remember to enable it on the BIOS as well.
Remember to perform an NVRAM reset on startup (Press the spacebar at the OC screen to see the auxiliary options).

Ps. I also noticed that the USB port mapping is incomplete and possibly incorrect, this could have consequences with Sleep and Thunderbolt.

Let me know if everything works.

Hi @LeleTuratti thanks for this!

I still need to test out sleep and report back, but the TB SSDT didn't work at all. TB devices are not visible (even after cold boot + NVRAM reset), and the system is noticeably stuttery, especially right after booting – I'll disable the SSDT for now.

Questions:

• Looks like you've removed all my Device Properties pci mappings. Can I add them back for aesthetic purposes, or do I risk causing things to break?
• Under Kernel you've changed the Arch from 'Any' to 'x86_64'. What's the Difference?
• Can you explain the reason for removing HybernationFixup.kext?
• What's the benefit of adding "j160" under Misc > Security ?
• Any particular motive for toggling EnableVectorAcceleration ?

Most importantly, what do you mean by USB mapping being incomplete?

 

[LeleTuratti]

still need to test out sleep and report back, but the TB SSDT didn't work at all. TB devices are not visible (even after cold boot + NVRAM reset), and the system is noticeably stuttery, especially right after booting – I'll disable the SSDT for now.

Without a proper SSDT for Thunderbolt, sleep can't work if you have devices connected.
Give it one last try with KGP's SSDT, I used it for a long time with the Alpine Ridge years ago.

Looks like you've removed all my Device Properties pci mappings. Can I add them back for aesthetic purposes, or do I risk causing things to break?

Device Properties has a purely cosmetic function, only those concerning the audio are necessary to inject the layout-id.
You used software to generate them automatically and it added every single ACPI path, but this doesn't make much sense and many names are wrong. I advise you to use OpenCore Configurator to manually add through the drop-down menu below only those that interest you, such as the graphics cards or other information you want to view on System Info.

Under Kernel you've changed the Arch from 'Any' to 'x86_64'. What's the Difference?

MacOs is now 64-bit only.

Can you explain the reason for removing HybernationFixup.kext?

Your card doesn't need it.

For other questions you can find more in-depth information by searching the web.

Most importantly, what do you mean by USB mapping being incomplete?

What is the mapping of the XHC2 device due to? In your IOreg it is not present. It could be the cause of some of your instability issues when adding the SSDT-TB.
The rest of the mapping is only for the rear USB ports and one internal I suppose. If you don't use the front USB ports on the case this is fine.