Need help with a bare metal Sequoia build

[henrikato]

I have spent the past few days attempting to get the installer for macOS Sequoia 15.8.1 to boot on my setup:

• CPU AMD Ryzen 9 7900X
• Mobo ASUS TUF Gaming B850M-Plus WiFi (BIOS ver. 1087)
• 2x 16GB ADATA XPG Lancer DDR5 at 6000MHz
• Revenger AMD RX 580 2048SP (BIOS was flashed to change its device id and identify itself as a standard RX 580 following this guide: https://www.tonymacx86.com/threads/...048sp-graphics-card-work-on-hackintosh.328087)
• WD Black SN770 512GB NVMe SSD
• Lexar NQ790 1TB NVMe SSD

The problem is no matter what I do, I never get past this message:
"Registered CoreAnalyticsHub functions with xnu.virtual bool CoreAnalyticsHub::start(IOService *)::229:CoreAnalyticsHub start completed"


By adding io=0x16 into my boot args the very last line that shows into the screen is:
"IOHIDProviderPropertyMerger[0x1000001ba]:robe fails"


I have tried many different configurations, disabling WiFi, Bluetooth, Onboard audio, IOMMU, no SSDTs loaded, SSDTTime-generated ones, a misc of boot arguments (npci=0x2000, -radvesa, -wegnoegpu) and many more tweaks into config.plist to the point I have truly forgotten every one of them. No matter what I do I get stuck at the exact same point.
I'm honestly lost here. I first hoped it would be an SSDT issue, then moved into USB, then CPU core count, nothing helped me.


I'm attaching my current EFI folder. Thanks in advance!

 

[Edhawk]

Where did you get the 2 x MmioWhitelist entries in your config.plist?
Did you follow this part of the OC troubleshooting guide when generating them?

Fixing KASLR slide values | OpenCore Install Guide

Does your system not require any AM5 ACPI patches?
If you post a copy of your System ACPI tables, someone may have a look to see which if any patches your system requires.

E

Post in thread 'DSDT patch'

Jul 12, 2026

In a further twist, AsRock B650E Steel Legend and X670E PG Lightning have been found to fit the patterns for MSI boards, while AsRock X870E Taichi Lite fits the "general" patterns … which are not so general, and just looking at the manufacturer may not be good enough.

Which leaves my "Universal Patches" in this unfortunate situation:

The Problem​

Modern DSDT tables (Intel and AMD) refer to configuration variables (G000 and so forth…) from "Operational Regions" in RAM, and then conditionally initialise devices accordingly. macOS operates according to an outdated version of...

• etorix

• 

You will need a copy of Miele's RTL812xLucy.kext for the Realtek 2.5G Ethernet port to work on your motherboard.

GitHub - Mieze/RTL812xLucy: A new macOS driver for the Realtek RTL812x family

A new macOS driver for the Realtek RTL812x family. Contribute to Mieze/RTL812xLucy development by creating an account on GitHub.

github.com

 

[henrikato]

I'm confident I must've have grabbed those MmioWhitelist entries from someone else's config.plist (that was showing the same symptom as I am).

I haven't read deep into KASLR but I will do later today after work.
I will also read the 'DSDT patch' post you added and see if I can figure something out.

I'm attaching the complete ACPI dump below, maybe someone can chime in to help.

Thanks!

 

[henrikato]

Quick update. I managed to boot into the installer by applying the three suggested patches on the 'DSDT patch' linked post.

I can format the disk just fine and start the graphical installer, however near the end of the installation the computer locks up (screen says "About 12 minutes remaining" but I know it's nowhere near a reliable estimate). No mouse movement, progress bar sits still.

On a second installation attempt I opened the installer logs and expanded to fill most of the screen to inspect whatever the installer is doing.
I added two screenshots from the logs below.

When the installation locked up I left it there for a solid 30 mins before restarting the PC.

OpenCore itself does not see any boot entries regarding any macOS preboot environment or the installed OS itself.
Just one entry for Windows, the macOS Installer on the USB drive itself and another entry to clear the NVRAM.

I will continue investigating, I'll probably have to keep digging into the DSDT itself to find for any more sections that need patching

 

[Edhawk]

OK, when your system DSDT.aml is checked in Hex Fiend, the 3 x AM5 Patches are all found. So that is all to the good.

Commonly the issue with a stalled installation at '12 minutes remaining' is a borked installation media/USB.

1. Did you create a Full installer 15GB+ in size on the USB drive, normally created on a real Mac or another Hack?
2. Did you create a 'Recovery' installer, which would be less than 1GB in size on the USB drive (normal Windows created installer)?

Whichever method you used, I would recommend wiping the USB and recreating the installer. Using the same installer will just put you back in the same stalled position as you currently find yourself.

I wouldn't mess with the OpenCore EFI anymore, not until you have macOS Sequoia installed. As doing so will just mess with your current booting EFI.

Have you temporarily removed the Lexar NVMe drive from the system, while installing macOS Sequoia? If not it might not be a bad idea to see if this helps, just for the installation phase. It can be reinstalled once you have Sequoia up and running. I am assuming this is your Windows drive and you are not using it for macOS.

 

[etorix]

Thanks! That's one more in my little collection.

And, unfortunately, the plot thickens… Your DSDT has the usual patterns from the general set #1, #2, #3… and then some more.

Short version: Here is an extended version of the "general" set with additional patches #4, #5, #6 and #7 for your DSDT. (It is quite possible that #7 serves no purpose, as the corresponding code should not be executed at the same time as the code under #5.)
I'll put the long version in the DSDT thread.

MMIO whitelist depend on installed memory, so you probably should follow the process of finding the right whitelist for your configuration.