EVGA NVIDIA GTX 1080ti Hybrid (temporary until my RX 6600 arrives)
G.Skill Trident Z Neo RGB 192B (6 x 32GB) 2666MHz
WD Black SN7100 1TB PCIe 4.0
SanDisk 64GB UHS-I USB 3.2 (installer thumb drive)
Dynatron L29 Liquid Cooler (280W TDW)
Lian-Li Edge Gold 750W PSU
Lian-Li Odyessey X Black Ed XL-ATX Case

OpenCore 1.05
MacOS Ventura (latest)

EFI (TRx40 Xtreme).zip

A very interesting utility for AMD CPU to monitor all parameters you can monitor

To create a custom curves for CPU fan and auxiliary fans and many others interesting features

Great kudos to @Droga developer for his efforts to produce this shining apps 🙂

(Familiarity with ACPI is required)

Patching a DSDT is risky.

Do only if you know risks and how to recover it if something goes wrong!

Hi everyone, with the release of new bios for am5 platforms which allow pre-support for 8000 series CPUs we came across a bug that no longer started macos.

@fabiosun and I started by trying these new bios without success in starting macos, until I had the idea of injecting the old dsdt of the previous bios with true on the RebaseRegions quirks.

Subsequently I dedicated myself to patching new bios and removing new conditions that were added :

removing the entire CPVS section generates 30 errors in my case :

I started to correct these errors by removing conditions for loading devices or _PRW methods :

there are so many that it removes the if... string with the bracket { and at the end of the scope also remove a bracket } so as not to generate other errors.

and I also removed this whole part:

Attached in this post original dsdt of new asus b650e-i's bios f and the patched one 

(asus b650e-i)

Original - DSDT.aml.zip       

---------------------------                    

(asus b650e-i)

Patched - DSDT.aml.zip

---------------------------                    

(asus x670e gene)

Patched - DSDT.aml.zip

---------------------------         

(asus x670e hero)

Patched - DSDT.aml.zip

---------------------------

(MSI X670e Pro Carbon)

DSDT.aml.zip

---------------------------

(Asus ROG B650E-F)

DSDT.aml.zip

---------------------------

(ASUS ROG STRIX X670E-F GAMING WIFI)

DSDT.aml.zip

---------------------------

(ASUS X670E-I)

DSDT.aml.zip

----------------------------

https://github.com/DrogaBox/SMCAMDProcessor-personal

new feature in fast evolution from dev!

These below are ACPI patches useful to avoid ACPI errors with bios 1203

patches_OC.plist.zip

As always thanks @Corpnewt aka CorpGhost on AMD-OSX discord to do them for us 🙂

link

Configuration in detail:

Case ThermalTake Core X9

PSU Enermax 1700w platimax

Asus ROG CrossHair x870E Hero
192 96 Gb Corsair Vengenage 6400 Mhz 42x48Gb

2 6950xt 16 Gb GPU ( 🙂 )

2 KC3000 Nvme Kingston 2 Tb

2 Sabrent Rocket 4.0 1 Tb

2 SSD 4Tb each

2 HD 20 Tb each

I did an initial working EFI.

to support 6950xt OSX needs a spoof to 6900xt

One slot needs additional SSDT bridge

I did also a fast USB mapping , i have to refine it better

Attached my working EFI

Pay attention:

Intentionally i didnt build a MMIO whitelist Area but i have used DisableVAriableWrite quirk to boot without MMIO declaration

I do not advice to do the same , and you should build your proper MMIO area to whitelist

System works wonderfully and the only things i can see it is not working are ASMedia USB4 and WIFI because i do not want to disassamble ASUS motherboard plate , in a future maybe i will do

as last information i am using 1203 Asus Hero bios 

ASUS X870E Hero_Clover_tahoe.zip

ASUS-X870E Hero-OC.zip

~/Library/Application Support/discord/0.0.264/modules/discord_krisp

Everything works without environmental variables or libfakeintel

discord_krisp.node.zip

It is coming home! 🙂

EDIT:

ADDED an initial working EFI

to improve USB mapping

I am using old mapping done for Hero..it seems to work but i think it could be improved 🙂

ASUS X870E Creator WIFI.zip

Asus_X870E Proart_OC1.04.zip

Asus_X870E Proart_OC1.06.zip

Motherboard:

https://rog.asus.com/it/motherboards/rog-crosshair/rog-crosshair-x670e-hero-model/

CPU:
AMD Ryzen 7950X

https://www.amd.com/en/direct-buy/5676499200/it

128 Gb DDR5:

2kit of Kingston Fury Beast DDR5 RGB 64GB (2x32GB) 5200MT/s DDR5 CL40 DIMM

https://www.amazon.it/gp/product/B0B1QNPM2T/

I will keep the CPU heatsink and some of the disks that I have now

This system loses in multicore to my TRX40 but performs better in other areas

I hope to see you still developing our historical thread on the TRX40, I certainly will

a touched TrexIntosh 🙂 🙂

SysReport.zip

EFI_ASUS_ROG_x670E_HERO_Clover5149.zip

EFI_ASUS_ROG_x670E_HERO_OC_098.zip

EFI-ASUS_x670E_HERO_Bios2124-fabiosun.zip

AMD Clover Vanilla patches Github:

https://github.com/AMD-OSX/AMD_Vanilla/tree/clover

I have converted only patches mandatory for my chipset and cpu..

you can follow the same schema in nomenclature definitions

By the way ,some definitions in clover 5137 are not described well and I think must be update their nomenclature 

Also MAtchOs need some check and I hope devs will solve a very annoying bug for AMD users 🙂

MAtchOS is not perfect by now in my definition of this plist..but was useful to boot with a clean and also to update from 11.5 beta 4

I used the attached patch plist to install from scratch and then to update

Only GoldFish64 patch has some definition problem in CLoverBootloader due Empty Find value (in verbose it prints out a not destructive error)

I have also attached two debug log to see problem I have said before and also others job to do.

1)installation debug (GetOS Empty all patches applied.....😇

2) after installation is completed....GetOS detection works as it must do

Thanks to patches-manager and creator 🙂

debug_pre inst_inst.zip

opencore-2021-07-01-161751.txt.zip

for lryzen friend (17h19h)

pay attention on patch 43..I commented limit there 😉

thanks @iGPUfor Goldfish64 2F tip! 😉

------

UPDATE 18-07-2021

https://www.macos86.it/topic/4913-amd-clover-bootloader-monterey-beta-1-installation-patches/?do=findComment&comment=116586

Clover-Kernel_Patches_Quirks.plist.zip

This is the discussion thread that references this post:
 

Feel free to discuss, share excitement, errors and whatnot! Let's bring AMD hackintoshes on par with their Intel counterparts!

The topic has been recently documented in a more exhaustive way in the OpenCore configuration pdf, but nowadays it has assumed more and more value to have a Hackintosh at its maximum potential also in light of the new CPU architectures in use.

By default  the quirk is not active and for many systems it is not essential to make your hack work and start.

In December 2019 with the release of the third generation AMD ThreadRipper CPUs (sTRX4), many users, despite themselves, were forced to take an interest in the functioning of the quirk in question.

At that time the OpenCore documentation on the subject was not very exhaustive also because it was a quirk useful only for old systems.

With the sometimes cryptic help of the OpenCore developers (to whom thanks are always due for their work) they started to take the first steps to understand how to use it to debug on a platform like the one described above ( sTRX4) in conjunction with the MMIO Whitelist functionality.

link

By activating the quirk and only using a debug version of OpenCore with the Misc / Debug / Target option set to 67, it is possible, even with a machine that does not start, a debug in which the MMIO areas of your PC are highlighted (number of areas and values may vary):

08:808 00:001 OCABC: MMIO devirt 0xE2100000 (0x81 pages, 0x8000000000000001) skip 0
08:809 00:001 OCABC: MMIO devirt 0xE3180000 (0x81 pages, 0x8000000000000001) skip 0
08:810 00:001 OCABC: MMIO devirt 0xEF100000 (0x181 pages, 0x8000000000000001) skip 0
08:811 00:001 OCABC: MMIO devirt 0xFA180000 (0x81 pages, 0x8000000000000001) skip 0
08:812 00:001 OCABC: MMIO devirt 0xFA300000 (0x100 pages, 0x8000000000000001) skip 0
08:813 00:001 OCABC: MMIO devirt 0xFEA00000 (0x100 pages, 0x8000000000000001) skip 0
08:814 00:001 OCABC: MMIO devirt 0xFEC00000 (0x1 pages, 0x8000000000000001) skip 0
08:816 00:001 OCABC: MMIO devirt 0xFEC10000 (0x1 pages, 0x8000000000000001) skip 0
08:817 00:001 OCABC: MMIO devirt 0xFED00000 (0x1 pages, 0x8000000000000001) skip 0
08:818 00:001 OCABC: MMIO devirt 0xFED40000 (0x5 pages, 0x8000000000000001) skip 0
08:819 00:001 OCABC: MMIO devirt 0xFED80000 (0x10 pages, 0x8000000000000001) skip 0
08:820 00:001 OCABC: MMIO devirt 0xFEDC2000 (0xE pages, 0x8000000000000001) skip 0
08:821 00:001 OCABC: MMIO devirt 0xFEDD4000 (0x2 pages, 0x8000000000000001) skip 0
08:822 00:001 OCABC: MMIO devirt 0xFEE00000 (0x100 pages, 0x8000000000000001) skip 0
08:823 00:001 OCABC: MMIO devirt 0xFF000000 (0x1000 pages, 0x8000000000000001) skip 0
08:825 00:001 OCABC: MMIO devirt 0x4040000000 (0x10400 pages, 0x8000000000000001) skip 0
08:826 00:001 OCABC: MMIO devirt 0x8BB0000000 (0x10400 pages, 0x8000000000000001) skip 0
08:827 00:001 OCABC: MMIO devirt 0x8BE0000000 (0x10400 pages, 0x8000000000000001) skip 0
08:828 00:001 OCABC: MMIO devirt 0xD750000000 (0x10400 pages, 0x8000000000000001) skip 0

It is of particular importance that these areas can be different even under conditions of the same BIOS and manufacturer, but only by swapping options in the BIOS or filling the nvme and pciexpress slots differently.

Note:

a common mistake is to take this list from any configuration and put it in your config.plist

It must always be taken and calculated in the system in use and rechecked if you change options in the BIOS or add hardware!

Activating the quirk involves having devirtualized memory areas and in some cases freed several megabytes of memory (typically from 64 to 256 Mb) which allows some problematic systems to start without having the notorious error at boot (memory allocation errors), and this right in the initial stages of the boot.

skip 0 means devirtualized area and if during its internal operations UEFI bios needed that area, not having it at its disposal, malfunctions or Kernel Panic would be created.

On the indications of the Opencore devs we concentrated on the last 4 areas which are the most substantial at the level of pages to possibly devirtualize and with 16 combinations (4!) It was established that the Kernel patches in use at the time were malfunctioning for sTRX4 systems. Let's now neglect this statement which turned out to be not exactly correct, but then it was practically a fact that led many sTRX4 users with the desire to use OSX on their machines to explore the fascinating world of ProXmox virtualisation.

Now the tests made at the time were to report the last 4 areas to full use of OSX (skip 1) in various combinations (16).

How?

By adding these areas in the Booter / MmioWhitelist section of our config.plist. To do this we are helped by the scientific calculator or the excellent app in the download area HackCheck (visible in the photo) for download it click below:

https://www.macos86.it/files/file/95-hack-check/

now let's just focus on the area:

08:828 00:001 OCABC: MMIO devirt 0xD750000000 (0x10400 pages, 0x8000000000000001) skip 0

and take the hexadecimal value 0xD750000000, this will be converted into a number as shown in the figure below:

and must be reported precisely in the appropriate area of the config.plist by activating it:

After saving the config.plist and restarting we will have this situation in our next debug log:

08:808 00:001 OCABC: MMIO devirt 0xE2100000 (0x81 pages, 0x8000000000000001) skip 0
08:809 00:001 OCABC: MMIO devirt 0xE3180000 (0x81 pages, 0x8000000000000001) skip 0
08:810 00:001 OCABC: MMIO devirt 0xEF100000 (0x181 pages, 0x8000000000000001) skip 0
08:811 00:001 OCABC: MMIO devirt 0xFA180000 (0x81 pages, 0x8000000000000001) skip 0
08:812 00:001 OCABC: MMIO devirt 0xFA300000 (0x100 pages, 0x8000000000000001) skip 0
08:813 00:001 OCABC: MMIO devirt 0xFEA00000 (0x100 pages, 0x8000000000000001) skip 0
08:814 00:001 OCABC: MMIO devirt 0xFEC00000 (0x1 pages, 0x8000000000000001) skip 0
08:816 00:001 OCABC: MMIO devirt 0xFEC10000 (0x1 pages, 0x8000000000000001) skip 0
08:817 00:001 OCABC: MMIO devirt 0xFED00000 (0x1 pages, 0x8000000000000001) skip 0
08:818 00:001 OCABC: MMIO devirt 0xFED40000 (0x5 pages, 0x8000000000000001) skip 0
08:819 00:001 OCABC: MMIO devirt 0xFED80000 (0x10 pages, 0x8000000000000001) skip 0
08:820 00:001 OCABC: MMIO devirt 0xFEDC2000 (0xE pages, 0x8000000000000001) skip 0
08:821 00:001 OCABC: MMIO devirt 0xFEDD4000 (0x2 pages, 0x8000000000000001) skip 0
08:822 00:001 OCABC: MMIO devirt 0xFEE00000 (0x100 pages, 0x8000000000000001) skip 0
08:823 00:001 OCABC: MMIO devirt 0xFF000000 (0x1000 pages, 0x8000000000000001) skip 0
08:825 00:001 OCABC: MMIO devirt 0x4040000000 (0x10400 pages, 0x8000000000000001) skip 0
08:826 00:001 OCABC: MMIO devirt 0x8BB0000000 (0x10400 pages, 0x8000000000000001) skip 0
08:827 00:001 OCABC: MMIO devirt 0x8BE0000000 (0x10400 pages, 0x8000000000000001) skip 0
08:828 00:001 OCABC: MMIO devirt 0xD750000000 (0x10400 pages, 0x8000000000000001) skip 1

So the area we converted and inserted into the config by activating YES is now listed in the config as skip 1.

What does it mean?

It means that now this area is again available to UEFI Bios as if we hadn't activated the quirk..but only this MMIO area.

If in the lucky hypothesis that our system starts anyway and arrives at login, it is possible to proceed to insert and activate all the other areas in MMIOWhitelist until we find the one that is harmful in our system.

What is a useful goal for the user who needs to use this procedure?

Get back as many skip 1s as possible!

If I activate the DevirtualizeMMIO quirk and whitelist all areas by activating them, what happens?

First option:

- system starts properly!

Well you wasted a lot of time because you don't need this procedure! having all areas declared and activated in the config plist section Booter / MmmioWhitelist is like leaving the DevirtualiseMMIO quirk set to OFF.

Second option:

- system does not start properly!

Very well that's why you activated the quirk, if it were possible for us to keep everything unchanged we would not have activated the quirk.

So, with holy patience, they try to convert all the values one by one and insert them in MMIOWhitelist (skip 1) until they find those that cannot be inserted and that block your system.

At this point, if your ideas aren't more tangled up than when you started reading, you should ask yourself:

why do this operation of re-making available to UEFI Firmware as many areas as possible?

Simple, even if your PC starts and seems to work correctly only by activating the DevirtualiseMMIO quirk, during daily operation you could (certainly have) KP, Reboot problems or simply malfunctions of the Nvram and the sleep / wake cycle.

So it's a good thing to waste some time re-assigning them to free use of UEFI Bios until you find or find the ones that are impossible to reassign!

you can also find useful tip here:

https://www.macos86.it/topic/3307-trx40-bare-metal-vanilla-patches-yes-it-worksbutproxmox-is-better/?do=findComment&comment=85469

inspired by the works of@Gengik84 And@foskvs i tried to edit a ssdt basic only for the generations mentioned above.

this ssdt for intel 600/700 series contains :

• ssdt plug alt 
• ssdt ec
• ssdt usbx
• ssdt pmc
• ssdt awac
• fix to turn off EC0 if found in acpi origin
• Fix to turn off unsupported igpu (kext WhateverGreen is required)
• usbw fix for 1 touch wake up (need kext USBWakeFixup.kext)

this ssdt for AMD AM5 contains  :

• ssdt cpur 
• ssdt ec
• ssdt usbx
• fix to turn off EC0 if found in acpi origin
• Fix to turn off unsupported igpu (kext WhateverGreen is required)
• ubw fix for various usb controllers for 1 touch wake up (need kext USBWakeFixup.kext and on asus mb it works fine, on other mb it may not work on 1 touch)

I thanks @fabiosun And @tomnic for the tests carried out and for their patience 🙂 

SSDT-Basic-600-700-AM5.aml.zip

for those who try it and want to report bugs or malfunctions, come forward and I will do my best to fix it.

It seems @Pavobooted with big Sur in A vanilla mode

some problems to solve but for him it is booting now 🙂

for me no...

You creators users could try 🙂

ops only you @iGPU

others users are Asrock

🙂

fabiosun's EFI and ioreg

System Description:

Ploddles's EFI and ioreg

System Description:

Arrakis's EFI and ioreg

System Description:

mSedek's EFI and ioreg

System Description:

Rox67er's EFI and ioreg

System Description:

UPDATE 17-07-2021:

EFI-OC-1.0.5.zip

EFI-5162.zip

Candle benchmark for DaVinci Resolve Studio

Thread is open also for unsupported GPU in OSX like 6900 xt, 6800 Xt, 6800 (we hope in a near support for them) but also it is possible to post results for Nvidia GPU like RTX series that will never be supported by OSX (Sob 🙂)

I will start with mine GPU, you can see in this video

On YouTube a simple video showing initial Davinci resolve setup (OpenGL / Metal or Cuda if it is available)

If you like you can post your results using a screenshot for frame rate in different environment you use

It is welcome any new GPU benchmark (AMD rx 6900 xt and similar 🙂 🙂 )

You must play inside Color page and take note only there of your maximal FPS

Attached useful file for this benchmark

Candle benchmark.zip

Trying to get my MMIO Whitelist info for my X870 MSI. When I use the Opencore Release OC 1.0.3 I get to the Opencore boot menu with no issue. Starts to boot but then boot loop as there is no MMIO Whitelist included in the config. So when I use OC 1.0.3 Opencore.efi debug I get the boot menu as in the attached pic, but then a boot loop. I never get to the Opencore boot menu. Any idea?

https://forum.amd-osx.com/threads/ryzen-threadripper-trx50.4776/

trx50-config-oc-log.zip

Ryzen 9950x 16-core

ASUS ProArt x670e

Corsair 192GB 4x48GB (CMK192GX5M4B5200C38)

FENVi T919 (Pulled from Trx50 build)

I never did get around to updating the Trx50 build beyond 14.4.1.

I would like to clone my Trx50 setup but perhaps starting fresh is ideal. Thoughts?

My config-OC-M5A.plist is attached for your reference.

config-OC-M5A.plist.zip

Hello Fabio. Been trying to adopt your EFI for my computer for a day already with no success. Could you please help me with that? I really like clover, it's been always easy in a past and I hoped that it is still be like that but can't figured what cause hang during installation. All bios settings as it should and I removed TB aml from EFI since I don't have it on my motherboard. rest should be same I think but still not working out.

Only About this Hack got wrong bootloader at its home page, but its System Report showed correct Clover version.

config-5156.plist.zip

For this program the installation is straightforward: just mount the downloaded dmg, drag the app into your Applications folder in macOS drive. Then you only need to overwrite libiomp5.dylib, libmkl_avx512.dylib and libmkl_core.dylib with my patched versions into Mathematica.app/Contents/SystemFiles/Libraries/MacOSX-x86-64  

I include a little benchmark to test the functionality of the app, bench.nb, just unzip it and open it in Mathematica then run it placing your cursor before the word "Needs" in the worksheet and choose the menu function Evaluation -> Evaluate cells.

Feel free to post some results, here are mine 😉

Note: as always, native libs of this specific app version are patched to redirect the problematic functions intel fastmemset and intel fastmemcpy to the fastest compatible similar functions for our AMD hackintoshes, also fooling the Genuine Intel CPU check with a trivial function which gives always true as the returning value: no more need to inject code, no more need to use environmental variables or old prepatched libs.

libiomp5.dylib.zip libmkl_avx512.dylib.zip libmkl_core.dylib.zip bench.nb.zip

I recently built a new PC with specs similar to yours: an AMD Ryzen Threadripper 7970x, ASUS Pro WS TRX50 Sage WiFi motherboard, 128GB of DDR5 RAM, and an AMD RX6950XT GPU. I'm interested in installing MacOS on this system. I've attempted to follow your EFI installation guide, but unfortunately, it hasn't been successful. I've also exported the Mmiowhitelist as per your instructions in your post, but the system seems to be stuck at the initial boot stage. I would greatly appreciate your advice on resolving these issues. Thank you very much.

EFI-7970x.zip

Recodesign it:

sudo codesign --force --deep --sign - /Library/Application\ Support/Waves/Modules/MklWaves.framework/Versions/A/libmkl_waves.dylib

If you mix them with v12 or inferior versions plugins you'll need further patches.

libmkl_waves.dylib.zip