[SUCCESS] Gigabyte Designare Z390 (Thunderbolt 3) + i7-9700K + AMD RX 580

[CaseySJ]

Hi @CaseySJ and all,

Your test bench is amazing ! Love it !
As I see on the config.plist Clover configuration, there is a mistake on TunderboltConfig property compared to my previous post Here. This should be tested like that, as reminder, extracted properties are from iMac19,1.

We should also resolve XHC2 problem with SSP2 at 0x03 instead of 0x09 .. you should confirm under Windows that effectively we have Companion on XHC main USB controller by plugging (for each port) one USB3.1gen2, USB3.0 and another USB2.0 on USB-C/thunderbolt connector ! This last one should appear under XHC instead of XHC2.

Another remark compared to my config (also rMac), when a thunderbolt device is plugged-in, we should have an under-tree :

• IOThunderboltPort@1 and/or IOThunderboltPort@2 in case of first thunderbolt connector (appearing under DSB1
• IOThunderboltPort@3 and/or IOThunderboltPort@4 in case of second thunderbolt connector (appearing under DSB4)

I can't show how it look like under IOReg, currently I'm not at home with my eGPU thunderbolt device.

Again, Great test bench Set-up !

Hello @Elias64Fr

My plan for today is to start with Draft07 and systematically move up to Draft09, while also making changes to RWAK and RP05.PCI0 as you previously suggested.

I'll fix ThunderboltConfig right now.

Regarding XHC2, when we plug in a USB 2.0 device into top and bottom Thunderbolt ports, we see in macOS that they connect to XHC.HS08 and XHC.HS13. I can check this under Windows later today.

Two follow-ups:

1. Are you familiar with the bytes/format of pathcr? Are we using a valid pathcr?
2. Gigabyte's System SSDT-6 (attached) contains further ACPI definition of XHC.RHUB. Maybe it contains some relevant information?

 

[CaseySJ]

Voilà, is it good like that ?
View attachment 446759

Oui, c'est magnifique!

 

[scottkendall]

Yes, but we're first trying to see whether SSDT alone can accomplish our goal. FYI, we cannot flash Apple’s firmware as-is. We look for certain differences in the first 0x1000 bytes of the active partition and copy those bytes into the original Gigabyte firmware.

I read about flashing it a few months back I know its a some kind of doctored file and needs a hardware flasher. My buddy has a Hardware flasher I just need to make the 3/hr drive down to his house. Would like to see some progress made in that direction before I drive way down there to bang my head against the wall.

I will keep watching your progress.

 

[juantrix]

hi CaseySJ, how can I obtain a ACPIDebug Log like your?
I use ACPIDebug.kext for RMDT
thx

 

[Elias64Fr]

Hello @Elias64Fr

My plan for today is to start with Draft07 and systematically move up to Draft09, while also making changes to RWAK and RP05.PCI0 as you previously suggested.

I'll fix ThunderboltConfig right now.

Regarding XHC2, when we plug in a USB 2.0 device into top and bottom Thunderbolt ports, we see in macOS that they connect to XHC.HS08 and XHC.HS13. I can check this under Windows later today.

Two follow-ups:

1. Are you familiar with the bytes/format of pathcr? Are we using a valid pathcr?
2. Gigabyte's System SSDT-6 (attached) contains further ACPI definition of XHC.RHUB. Maybe it contains some relevant information?

Nice!

For answers:

1. Nope! I'm just using rMac data. It work on Alpine Ridge case why not for Titan Ridge!
2. I have updated my previous post. Effectively SSDT-6 show that HS08 and HS13 is defined as Type-C USB but we have another contradicting (for macOS set-up) data on SSDT-8. Let me know. How much have we thunderbolt related connector (USB Type A and C) on Designare board ?

 

[CaseySJ]

Nice !

For answers:

1. Nope ! just using rMac data. It work on Alpine Ridge case why not for Titan Ridge !
2. I have updated my previous post. Effectively SSDT-6 show that HS08 and HS13 is defined as Type-C USB but we have another contradicting (for macOS set-up) data on SSDT-8 .. Let me know .. How we have thunderbolt related connector (USB Type A and C) on Designare board ?

Not sure I understand: "How we have thunderbolt related connector (USB Type A and C) on Designare board ?" But the connectors look like this:

 

[CaseySJ]

hi CaseySJ, how can I obtain a ACPIDebug Log like your?
I use ACPIDebug.kext for RMDT
thx

We just need to use SSDTs that invoke ACPIDebug methods such as RMDT.P1, RMDT.P2, etc. This is done by the Thunderbolt SSDT that we're testing.

 

[Elias64Fr]

Not sure I understand: "How we have thunderbolt related connector (USB Type A and C) on Designare board ?" But the connectors look like this:

View attachment 446763

For futher information, on Asus Maximus desktop, I have the following set-up :

And Like I said on my first post Here, this board have only one Thunderbolt port that why there is only one thunderbolt ethernet appears on IOReg... But two connectors (one USB type A and one USB type C) are used with Thunderbolt device. That's an unknown set-up on Apple world but now it work !

 

[CaseySJ]

For futher information, on Asus Maximus desktop, I have the following set-up :

View attachment 446777

And Like I said on my first post Here, this board have only one Thunderbolt port that why there is only one thunderbolt ethernet appears on IOReg... But two connectors (one USB type A and one USB type C) are used with Thunderbolt device. That's an unknown set-up on Apple world but now it work !

Understood! Will do the Windows test with USB 3.1, USB 3.1 Gen 2, and USB 2.0.

 

[wessberg]

I just did a little more experimenting with NVRAM in Clover..
I can confirm what you are saying. It just isn't working. I think the time I thought it was kinda working I may have had a errant copy of EmuVariable in there and didn't realize it.

But native NVRAM seems to be working just fine in OpenCore.

It would seem that native NVRAM doesn't work with Clover, then. I wouldn't know where to start in terms of which files to keep and which files to drop from my EFI folder, so I'm waiting to see if the Catalina Mini-guide gets updated. @CaseySJ, do you have any insights here? I'd love to use the OpenCore guide to get native NVRAM working, but I'm new to hackintoshing, so I guess it's better to use Clover for now.