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

[CaseySJ]

@CaseySJ, I wanted to check to see if what I did to enable GPU (RX 580) monitoring was the best way to go. I enabled CSM Support in bios, and then I used HWMonitorSMC2 and selected "Use the IOKit monitoring for GPUs", and voila! Now I have more info than ever (I also see my iGPU stats, since that's enabled). Are there any known alternatives to HWMonitorSMC2? iStat Menus and HWMonitor don't work for the GPU. (I only ask because it would be nice to have widgets in the taskbar, logging capabilities, etc.)

Also, is there any reason not to enable CSM support?

In my experience there is no problem with enabling CSM Support. It enhances support for some GPUs and, in particular, for eGPUs.

 

[CaseySJ]

Well... I had another "freeze" tonight. I tried switching the VBios jumper to low performance mode, and using OpenCore - didn't help. Only other thing I could try would be going to 1 monitor, but... tomorrow is my last day to return this video card for a refund and it wasn't cheap, so....if I can't reliably use it, it needs to go back. I don't know what else to do here, except return the card, switch to iGPU, and see if the problem goes away. If it does, try to find the best performing card that actually works reliably under Hackintosh and live with it.


I will say that I did some further testing this time, and the system does not fully lock up. Video/Audio continue to work. Not only that, but the screen is still updating visually. More so, I actually turned off my thunderbolt audio interface, and turned it back on, and it reconnected to the system and lit up, so what ever is happening isn't actually freezing the system.

It's really like the entire USB system just drops out of the computer, and everything else continues to work. I tried plugging my mouse and keyboard into other ports, they wouldn't come back up until I rebooted manually. I can't find anything in the logs either, and there is no kernel-panic. The mouse and keyboard just simply go dead. I don't get it, but it makes me wonder if it's even related to the RX 5700 XT - it might not be, but I don't know what else would be causing this issue.



Casey - I've attached my Clover config to this post for you. Can you please modify it so that I may use the iGPU on my i7-8700K please? Thank you!

-- Edit -- I forgot to mention, I don't have another Thunderbolt to DP adaptor, so I will have to use HDMI with iGPU.

Please try the attached config.plist after reentering serial numbers. HDMI port is enabled.

 

[xanderevo]

Hey @CaseySJ I have maybe isolated the popping issue, do you know what kind of thread is this?

 

[mazop89]

Hello @mazop89,

If the file is in your Downloads folder, simply type this in Terminal:

cd ~/Downloads
xattr -d com.apple.quarantine EFIClone-v4.sh

Also remember that TEST_SWITCH is set to Y, which puts the script in test mode. Once you have examined EFIClone.log from a test run and verified that it looks okay (i.e. only the files from Source EFI are being copied to Destination EFI), then change TEST_SWITCH to N.

View attachment 449357

Hi @CaseySJ , when I write the code that you sent to me (thanks!), i've got that answer that, I think, may highlight a problem :

xattr -d com.apple.quarantine EFIClone-v4.sh
xattr: EFIClone-v4.sh: No such xattr: com.apple.quarantine
flo@flos-iMac Downloads %

Is this normal ?

 

[RonnyHemlin]

Thunderbolt depends on:

• BIOS --> Thunderbolt Security --> No Security
• BIOS --> Thunderbolt --> GPIO3 Force Pwr --> Enabled
• And these two SSDTs that belong in the CLOVER/ACPI/patched folder:
  • SSDT-Z390-DESIGNARE-TB3HP-V4.aml
  • SSDT-DTPG.aml

Thanks man... will definetly try this ASAP.

 

[CaseySJ]

Hi @CaseySJ , when I write the code that you sent to me (thanks!), i've got that answer that, I think, may highlight a problem :

xattr -d com.apple.quarantine EFIClone-v4.sh
xattr: EFIClone-v4.sh: No such xattr: com.apple.quarantine
flo@flos-iMac Downloads %

Is this normal ?

Yes that’s actually good. You may use that file.

 

[jleahy2]

It’s interesting that you are using Thunderbolt audio like me. Maybe that’s the common thread here? I got more lockup’s today unfortunately.

It certainly could be. I'm running a Focusrite Clarett 8Pre thunderbolt audio interface. I didn't see it on the officially supported list on this thread, but, so far, these issues aside, it's worked without issue using a legit Apple thunderbolt adaptor.

Please try the attached config.plist after reentering serial numbers. HDMI port is enabled.

Thank you Casey. I've just removed the RX 5700 XT, and switched over to iGPU with your config. I'll report back with updates.

 

[Elias64Fr]

@Elias64Fr,

Attached are three LSPCI dumps:

• DarwinDumper_3.0.4_With_TB3_Apple_ICM_Controlled.zip
  • This is generated using standard SSDT that runs in ICM mode.
  • In this case, TB3 devices operate successfully and can hot-plug successfully.
• DarwinDumper_3.0.4_With_TB3_TbtOnPch_1.zip
  • This is generated using our SSDT-TbtOnPch.
  • TB3 device was connected, but not functional.
  • No TNODE/TBUS.
• DarwinDumper_3.0.4_With_TB3_TbtOnPch_2.zip
  • This is generated using our SSDT-TbtOnPch.
  • TB3 device was connected, but not functional.
  • TNODE/TBUS activated.

Hi @CaseySJ,

Thanks for send files !

I called each setting CONF01 to CONF3, in order to files, for the following clarifications (MELD 3.19.2 (r5) app used Here) :

Spoiler: Compare CONF02 (left side) to CONF03 (right side)

We can see on first screenshot that RP05 have same registers configuration for both CONF.
On second screenshot, there is some issue with UP18, UP19 and UP1A data on CONF03, all these registers which define current and sub-devices are all null value. Probably due to Uncorrectable error tagged on Device Status register.
On last screenshot, we have confirmation that there is no sub devices (DSB0..) appearing on this files (don't know why you have all TNode/TBus populated with that ! ).

 

[atz1995]

Hey
@CaseySJ,

I built my first hackintosh to your specs (except for an i9-9900), and it worked perfectly until an issue with the Thunderbolt 3/USB-c which stopped working. Upon trying to fix things and restarting, I am getting this...

Apologies in advance as I am very much a beginner. Any help would be greatly appreciated.

I would hate to have to wipe the boot drive and start over again, if possible to avoid that.

-- Austin

 

[Elias64Fr]

@Elias64Fr,

Attached are three LSPCI dumps:

• DarwinDumper_3.0.4_With_TB3_Apple_ICM_Controlled.zip
  • This is generated using standard SSDT that runs in ICM mode.
  • In this case, TB3 devices operate successfully and can hot-plug successfully.
• DarwinDumper_3.0.4_With_TB3_TbtOnPch_1.zip
  • This is generated using our SSDT-TbtOnPch.
  • TB3 device was connected, but not functional.
  • No TNODE/TBUS.
• DarwinDumper_3.0.4_With_TB3_TbtOnPch_2.zip
  • This is generated using our SSDT-TbtOnPch.
  • TB3 device was connected, but not functional.
  • TNODE/TBUS activated.

@CaseySJ
and for CONF02 to CONF01 compare :

Spoiler: compare CONF02 (left side) to CONF01 (right side)

On first screenshot, there is tagged Correctable error on Device Status register of CONF02 (don't know why this is not corrected .. in fact this conf doesn't work).

On UpStreamBridge (second screenshot), differencies are on CLKPM (Clock PM) and CommClk (Common Clock) bits disabled on Link Control register of CONF02.

On DownStreamBridge0 (third screenshot), CONF02 has this device on D3 state (inactive state). There is another difference on CommClk bit configuration !

On last screenshot, optionally for our need, we can see that CONF02 Cache Line Size is not defined on XHC2 controller. Like CommClk is not same.

Conclusion: some work is required to correctly setting all theses registers ! This might help us for Thunderbolt full tree populating.

Another ACPIDebugLog required for our progressing work.. at begin and end of ICMB method

Spoiler: ACPIDebug log code

            DBG2 ("NH00- Read VID/DID =", NH00)
            DBG2 ("XH00- Read VID/DID =", XH00)
            DBG2 ("RP18 =", RP18)
            DBG2 ("RP19 =", RP19)
            ...
            until RP28
            ...
            same for other sub-devices