Jump to content

iMac Pro X299 - live the future new with macOS 10.14 Mojave [Successful Build/Successful Guide]


KGP-iMacPro
 Share

142 posts in this topic

Recommended Posts

9 hours ago, vladie said:

Hi

Trying to run your config on i7-9800x, asus prime x299-a, rx vega 64.

Kernel starts to boot (from USB stick) and i see first screenshot for near 60 seconds, and then i see second screenshot (if i will wait more i see message 'still waiting for root device')

TSCAdjustReset was adopted.

 

UPDATE: seems that are only one USB port are seen by os x (i can insert usb stick into this port and boot installer). Tried kext-s from XHC-USB-Kext-Library, but no luck :( Keyboard/mouse not working, so i can't install macos in order to create working XHC-USB-Kext config with IoRegistryExplorer. I've also tried to boot high sierra installer (with usb port count patch enabled) - but result the same - no keyboard/mouse.

 

IMG_1132.JPG

IMG_1133.JPG

Hey can you share your EFI Folder?  I’m trying to get this running on the same setup as you but I can’t even get that far. 

Link to comment
Share on other sites

11 hours ago, Napka said:

@Vladie I had similar problems with latest BIOS v1602. Reverted to 1503 and everything’s good again.

 

 

My CPU i7-9800x is supported since v1602. But i will try to revert, thanks.

UPDATE: unfortunately, downgrading to v1503 have no effect. will try to buy usb hub today (i have one working usb port), connect usb flash+keyboard+mouse to it and install mac os.

Edited by vladie
Link to comment
Share on other sites

I've successfully installed mojave on asus x299-a prime + i7-9800x, using usb hub on single working usb port. Mojave didn't want to work with Asmedia 2142 ports, currently i have no idea how to make them active. Tried USBInjectAll+XHCI-unsupported, but no luck. Intel ports (except single working) also didn't work, but i see two USB 3.1 Bus intel controllers into USB device tree.

Edited by vladie
Link to comment
Share on other sites

30 minutes ago, vladie said:

I've successfully installed mojave on asus x299-a prime + i7-9800x, using usb hub on single working usb port. Mojave didn't want to work with Asmedia 2142 ports, currently i have no idea how to make them active. Tried USBInjectAll+XHCI-unsupported, but no luck. Intel ports (except single working) also didn't work, but i see two USB 3.1 Bus intel controllers into USB device tree.

 

Can you post your EFI folder?

Link to comment
Share on other sites

36 minutes ago, tio121 said:

 

Can you post your EFI folder?

 

it's just a efi folder from KGP github repository. you need to use usb hub, inserted into cyan usb connector (it's located above single usb-c connector), for connecting usb installer flash, keyboard and mouse.

Link to comment
Share on other sites

It's quite strange, but Asmedia ASM1074 driver is not loaded. I've installed KGP's ACPI patches (seems that they are matched X299-A prime, i've checked USB definitions), but this doesn't help. Is anyone know how to force this driver loading?

Screen Shot 2018-12-27 at 10.49.11 AM.png

Screen Shot 2018-12-27 at 10.53.18 AM.png

Link to comment
Share on other sites

I was ready to write this off as an incompatible CPU.  But vladie seems to be able to boot into OSX fine.  He's running the same CPU and same motherboard as I am.  

 

I'm using the EFI folder from github as-is with the addition of the TSC kext.

 

Any ideas?

Link to comment
Share on other sites

6 hours ago, tio121 said:

I was ready to write this off as an incompatible CPU.  But vladie seems to be able to boot into OSX fine.  He's running the same CPU and same motherboard as I am.  

 

I'm using the EFI folder from github as-is with the addition of the TSC kext.

 

Any ideas?

 

have you configured BIOS settings according to KGP instructions?

Link to comment
Share on other sites

18 hours ago, tio121 said:

Yes I followed his settings. 

 

Other people have this same issue as well with the new intel chips. Did you do anything different?

 

currently mojave boots without problems (i've installed it on SATA drive). video (rx vega 64) works without any setup, 4k support working. the only problem that i currently have - non working intel 200-series usb controller, seems that this is problem with bios (i cannot downgrade it because of new-generation cpu support). so i decided to buy other motherboard (deluxe version) and try again.

 

ps. try my clover:

 

https://www.dropbox.com/s/9fwikz8rz85joby/CLOVER.tar.bz2?dl=0

 

UPDATE: i've found a reason - why my USB doesn't work. on initialization, AppleUSBXHCI reports next:

 

XHCI@(null): AppleUSBXHCI::start: hardware exception occured

 

Seems that this can be some incompatibility with new CPU :(

 

UPDATE2: yessss!!!! XHCI is working. adding npci=0x3000 to kernel parameters solved this problem.

UPDATE3: Areca 1883i card works plug'n'play, Intel X540-T2 modified to be a small tree 10gbe too.

Bluetooth works plug'n'play with iogear bluetooth usb stick. 

Edited by vladie
Link to comment
Share on other sites

Hi all,

 

First of all, I have to say that reading this topic (and its predecessor for HS) feels like heaven when you think about renewing your good-old-x58-based hackintosh like me :).

Long story short, I am currently looking at the ASRock X299 Taichi XE manual, and I am not 100% sure that this motherboard has its MSR 0xE2 register unlocked:

Could someone confirm that?

 

(Sorry if this has already been written somewhere else, but looking at the forum posts, I found several member having this mobo in their signature, but no clear confirmation is it is running great for this matter...)

Edited by MinusZwei
Link to comment
Share on other sites

On 1/2/2019 at 8:40 PM, tio121 said:

Even with your EFI folder I get the same issue.

 

Can you check your bios settings if you have anything different than the optimized defaults and KGP's settings?

 

have you updated BIOS to latest version ?

reset bios settings to defaults and try again to setup it according to KGP manual

Link to comment
Share on other sites

  • 3 weeks later...

Working USB port limit patches for 10.14.3 and 10.14.4 Beta1

 

Thanks to @ydeng, there are now working USB port limit patches for 10.14.3

 

kext     : com.apple.iokit.IOUSBHostFamily
Find     : 83FB0F0F 871604 00
Replace  : 83FB3F0F 871604 00
Name     : USB Port Limit Patch 1 ©ydeng

 

kext     : com.apple.driver.usb.AppleUSBXHCI
Find     : 83FB0F0F 838F0400 00
Replace  : 83FB3F0F 838F0400 00
Name     : USB Port Limit Patch 2 ©ydeng

 

kext     : com.apple.iokit.IOUSBHostFamily
Find     : 83E30FD3 E34109DF
Replace  : 83E33FD3 E34109DF
Name     : USB Port Limit Patch 3 ©ydeng

 

And here are the respective USB port limit patches for 10.14.4 Beta1 provided by @PMHeart

 

kext     : com.apple.iokit.IOUSBHostFamily
Find     : 83FB0F0F 870B0400 00
Replace  : 83FB3F0F 870B0400 00
Name     : USB Port Limit Patch 1 ©PMHeart

 

kext     : com.apple.driver.usb.AppleUSBXHCI
Find     : 4183FF0F 0F839404 0000
Replace  : 4183FF3F 0F839404 0000
Name     : USB Port Limit Patch 2 ©PMHeart

 

kext     : com.apple.iokit.IOUSBHostFamily
Find     : 83E30FD3 E34109DF
Replace  : 83E33FD3 E34109DF
Name     : USB Port Limit Patch 3 ©PMHeart

 

The port limit patches should be used in line with the fully implemented XHC USB kexts.

 

However, it is still recommended to disable the port limit patches and use truncated 15-port kexts instead. This might avoid buffer overruns and will make your system more vanilla with respect to future macOS updates.

 

Enjoy and have fun,

 

kgp.png.6db60e059a828b390e59a30e00e676b4.png

  • Thanks 2
Link to comment
Share on other sites

  • 2 weeks later...

First steps with a Sapphire Radeon VII under macOS 10.14.3 (18D42) and 10.14.3 SU (18D109)

 

IMG_3107.jpg.f6fa56cf8628e4152680f4e5f948e483.jpg

 

Today, I encountered a unique opportunity to perform some dirty first steps and tests with a Sapphire Radeon VII under macOS 10.14.3 (18D42) and macOS 10.14.3 SU (18D109) on a friend's system here in Berlin. 

 

IMG_3095.jpg.b90fdfc6cdc535085d0171277b728640.jpg

1.) The first disappointing discovery independent from macOS was that this GPU for now apparently seems not compatible with UEFI. The BIOS enforced "CSM" by default otherwise the ASUS Prime X299 Deluxe system would not have booted at all. 

2.) Once under macOS 10.14.3, the Radeon VII was rather recognised by macOS like a Vega Frontier at first place, with only one DP1.4 port actually natively working. 

 

1119083164_Screenshot2019-02-08at13_59_03.png.7f8c501032f0654b03655586189c8e7c.png

 

3.) After applying just some cosmetics to SSDT-X299-Vega64.aml and implementing the latter in line with Whatevergreen.kext, the Radeon VII appeared under "About this Mac" such:

 

226320375_Screenshot2019-02-08at15_49_11.png.f464a495b12bcbcda1c21b2a54bb4f09.png

 

Thanks to SSDT-X299-Radeon-VII.aml (which is basically SSDT-X299-Vega64.aml), Radeon VII and Radeon VII HD-Audio PCI drivers also correctly popped-up in Section "PCI" of Apple's system Report as properly implemented and fully loaded:

 

889179259_Screenshot2019-02-08at15_48_16.thumb.png.20d83a5df74dc33788e955119a13e1c5.png

 

831755848_Screenshot2019-02-08at15_48_29.thumb.png.4c1ac9e00f6bef412ecc46ea668bf479.png

 

In combination with Whatevergreen.kext, the three DP1.4 ports and the single HDMI port of the Sapphire Radeon VII also have been properly implemented in form of just four Radeon frame buffers + the respective HDAU device. 

 

1238642935_Screenshot2019-02-08at15_46_58.thumb.png.a95439f2795a5a9effabcf0563bd63a0.png

 

All four ports were fully functional, including display HotPlug and multi-monitor support. Also sleep/wake worked absolutely flawless. 

The UI worked fluid without an issues or lag and also video playback fully worked as expected. 

Now the bad news at least for macOS 10.14.3 (18D42) and macOS 10.14.3 SU (18D109):

All benchmark tools available, i.e. Cinebench OpenGL, Geekbench OpenGL and Metal, LuXMark OpenGL or Heaven and Valley were not able to find any GPU supporting OpenGL or Metal. 

Thus, unfortunately under the above mentioned macOS versions, the Radeon VII seems yet not to support OpenGL or Metal so far.

Thought that this recent findings also might be interesting for others .. :wink_anim:

 

kgp.png.9dda268624d64006e8b3283a27edfc0f.png

 

 

Edited by KGP-iMacPro
  • Like 1
  • Thanks 2
Link to comment
Share on other sites

Sapphire Radeon VII now fully UEFI compatible!

 

IMG_3107.jpg.42c6486a90582fcddad04b61d46a9408.jpg

 

At first place I should highlight some important correction. My first tests and results concerning the Radeon VII did not base on a MSI but Sapphire Radeon VII. I simply confused the brand my friend employed on his system. I therefore already correct my initial lapsus in my former post above up to my best knowledge..  image.gif.9ab16978bd6883677c6071f9890c66a8.gif:lol:

This morning, @DSM2  pointed me to some approach recently discussed in the German Hackintosh forum: 

@Mork vom Ork recently 
provided a link for a windows tool developed by ASRock, which implements a hybrid firmware on the ASRock Radeon VII. Unfortunately, ASRX7_16G190201.exe did not work in my case with the Sapphire Radeon VII.

However, @CMMChris
 extracted the VBIOS from ASRX7_16G190201.exe by means of a hex editor and provided the resulting BIOS ROM radeon-vii-uefi.rom in his respective post in the German Hackintosh Forum. 

After saving the original Firware of the Sapphire Radeon VII with Winflash.exe (admin permissions) of AMD/ATI ATIFlash 2.84 under Windows 10, I subsequently flashed the Sapphire Radeon VII with radeon-vii-uefi.rom without any issues.


Subsequently I disabled CSM in the BIOS of my ASUS Prime X299 Deluxe and successfully booted my system with the Sapphire Radeon VII and new ASRock hybrid firmware..

Legacy only problem of Radeon VII fully removed! :headbang:

I guess that the latter approach should work for any Radeon VII.

Enjoy and have fun,

 

kgp.png.38801e7b9990e3319f124a9f10559513.png

  • Like 2
Link to comment
Share on other sites

21 hours ago, Balamut said:

Still having issues with OpenCL and Metal?

 

The entire correct, native VII macOS implementation is still missing.. At present the VII is recognised and implemented Vega like, without hardware acceleration.   

  • Like 1
Link to comment
Share on other sites

KGP, 

 

Fantastic write up and guide! I just upgraded from a dual e5 xeon hackintosh running Mojave to skylake-x

 

My system is as follows

EVGA x299 Micro

I7-7800X

8gb 2800 g.skill ddr4 (more to come)

Powercolor Vega56 8gb

250gb Samsung Evo 970 NVMe

 

Ive used your respective EFI folder and created new serial numbers for smbios. Ive read the guide a couple times maybe I'm missing something but I cant even boot at all just hangs at End RandomSeed.

 

Ive searched google for answers on this and have tried everything suggested. So I must be missing something from your write up. I did reenable the MSR patch as far as ive read evga doesnt unlock it. Any ideas or suggestions would be greatly appreciated. 

 

Thanks Epnk82

20190215_220116.jpg

Link to comment
Share on other sites

Did you implement a properly adopted TSCAdjustReset.kext under /EFI/CLOVER/kexts/Others? Maybe you also need boot flag npci=0x2000?

 

Disable MSR lock in BIOS settings! If you do not disable MSR lock in BIOS settings, you must check “KernelPm” and enable  xcpm core scope kernel patch in config.plist for being able to boot your system. With disabled MSR lock, “KernelPm” and xcpm core scope kernel patch can remain disabled. 

 

Everything detailed above clearly outlined in my guide and absolutely nevessary for succesful system boot! 

 

Good luck,

 

KGP

  • Like 1
Link to comment
Share on other sites

On 2/8/2019 at 8:25 PM, KGP-iMacPro said:

Today, I encountered a unique opportunity to perform some dirty first steps and tests with a Sapphire Radeon VII under macOS 10.14.3 (18D42) and macOS 10.14.3 SU (18D109) on a friend's system here in Berlin.

Did you try the newest beta yet? I heard they added the device-id.

 

The latest MacBook Pro also has a Vega 20 card option... Could you fake it as that, possibly?

Link to comment
Share on other sites

KGP, thanks for the great and detailed guide :graduated:! It's made the process of getting Asus x299 Prime Deluxe rig up and running quickly and painlessly.

 

I am having an issue getting your SSDTs to load though. I've recompiled them, checked the code, all seems fine. Example is my Sata controller still reads "Generic AHCI Controller" in system report. Wondering if you could offer a clue as to where to look? I do have your ACPI edits enabled in my Clover config.plist.

Link to comment
Share on other sites

9 hours ago, J Lamp said:

KGP, thanks for the great and detailed guide :graduated:! It's made the process of getting Asus x299 Prime Deluxe rig up and running quickly and painlessly.

 

I am having an issue getting your SSDTs to load though. I've recompiled them, checked the code, all seems fine. Example is my Sata controller still reads "Generic AHCI Controller" in system report. Wondering if you could offer a clue as to where to look? I do have your ACPI edits enabled in my Clover config.plist.

 

Based on the info you provide above I am not able to help at all. Please read beginning of section C.) and provide respective information. 

 

https://github.com/KGP/X299-System-SSDTs/blob/master/Distribution/SSDT-X299-SAT1.aml should work OoB as long you also implement https://github.com/KGP/X299-System-SSDTs/blob/master/Distribution/SSDT-DTPG.aml

 

Edited by KGP-iMacPro
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...