Partially resolved ! I'm stuck trying to install macOS Sequoia on GA-AB350-Gaming, Ryzen 7 2700X, Radeon RX 550

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
Hello Ryzentosher,

I'm having major trouble installing macOS Sequoia on my system. Here are my hardware specifications:

  • Motherboard: Gigabyte GA-AB350-Gaming
  • Processor: Ryzen 7 2700X
  • Graphics Card: Radeon RX 550/560 Series (4 GB, Polaris 21)
I understand that macOS Sequoia is still in beta, and I realize that certain adjustments are needed to make it run smoothly on all systems. Still, I'd really appreciate any support in getting the installation to boot successfully without getting stuck in an annoying boot loop.

Does anyone have tips or experience on how to get the system to boot stably? Unfortunately, my phone's video quality isn't great, but I hope someone might still have an idea on how to fix the issue.

Error-Loop


Thanks in advance for any help!
 
Last edited:

Shaneee

The AMD Guy
Staff member
Administrator
Joined
Mar 13, 2020
Messages
2,341
Try with this minimal EFI. I'm sure that board has Above 4G as an option in the BIOS. Make sure that's enabled. If you still encounter the same error remake your installer EFI as shown in the Dortania guide and use this EFI still.
 

Attachments

  • EFI.zip
    6.8 MB · Views: 11

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
Which WiFi/Bluetooth card/dongle are you using? I noticed a couple of Bluetooth related 'issues/errors' in the Verbose text.
 

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
Here is another EFI for you to try with your Gigabyte GA-AB350 system. It has been created using the following:
  1. OpenCore 1.0.2
  2. Latest AMD Kernel Patches
  3. Latest essential kexts
  4. MacPro7,1 SMBIOS.
No allowance for any WiFi or Bluetooth cards has been made.
No sensor or other extraneous kexts have been included.
Assumed that the RX 550 has a Baffin chip, not a Lexa chip and is natively supported in macOS.
Assumed you haven't enabled Above 4G Decoding in your bios, so I have included the npci=0x3000 boot argument in the config.plist. Delete this boot argument if the opposite is true.

The EFI should work with Big Sur or any newer macOS release, including Sequoia.
 

Attachments

  • EFI-spakk.zip
    9.2 MB · Views: 10

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
Vielen Dank, Edhawk, für das Hochladen des EFI-Ordners.

Ich bin gerade online gegangen und lade gerade eine neue Installations-App für macOS Sequoia herunter. Mir bleiben noch etwa 20 Minuten, um den Download abzuschließen. Gestern vermuteten wir, dass die aktuelle Installationsversion von macOS Sequoia, die ich verwende, beim Download beschädigt worden sein könnte, was die Ausführung der Installation verhinderte. Sobald der Download abgeschlossen ist, werde ich es mit Ihrem EFI testen und mich wieder melden.

Mit der Grafikkarte und auch mit der Tatsache, dass ich keine WLAN- oder Bluetooth-Karten eingebaut habe, hattest du vollkommen recht:

Grafikkarte: Radeon RX550/550 Serie (4 GB)
3D-Beschleuniger: AMD Radeon RX 550 (Polaris 21)
Soundkarte: ATI Radeon HDMI @ AMD Baffin/Lexa – High Definition Audio Controller
Soundkarte: Realtek ALC887 @ AMD K17 – High Definition Audio Controller
CPU-Typ: Octa-Core AMD Ryzen 7 2700X, 3700 MHz (37 x 100)
Hauptplatine: Gigabyte GA-AB350-Gaming
Motherboard-Chipsatz: AMD B350, AMD Taishan, AMD K17 IMC
 

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
I have resumed my attempt to install macOS Sequoia on my system. Currently, I am using an EFI from a user who has nearly identical hardware. His motherboard is the same as mine, but he uses a different Ryzen CPU, a different AMD Radeon graphics card, and Bluetooth devices that I don’t use. The BIOS settings are configured exactly the same, but no macOS version will successfully boot with this setup.

Here are the related images. I’m also attaching my modified EFI folder, hoping that someone can make a successful diagnosis based on the information I’ve provided and the contents of my EFI upload.
 

Attachments

  • IMG-20241114-WA0003.jpg
    IMG-20241114-WA0003.jpg
    696.3 KB · Views: 15
  • IMG-20241114-WA0005.jpg
    IMG-20241114-WA0005.jpg
    698.4 KB · Views: 6
  • IMG-20241114-WA0006.jpg
    IMG-20241114-WA0006.jpg
    711.3 KB · Views: 3
  • IMG-20241114-WA0007.jpg
    IMG-20241114-WA0007.jpg
    800 KB · Views: 3
  • IMG-20241114-WA0008.jpg
    IMG-20241114-WA0008.jpg
    426.5 KB · Views: 3
  • IMG-20241114-WA0012.jpg
    IMG-20241114-WA0012.jpg
    599.6 KB · Views: 4
  • IMG-20241114-WA0013.jpg
    IMG-20241114-WA0013.jpg
    379.1 KB · Views: 4
  • IMG-20241114-WA0015.jpg
    IMG-20241114-WA0015.jpg
    413.4 KB · Views: 8
  • IMG-20241114-WA0016.jpg
    IMG-20241114-WA0016.jpg
    432.5 KB · Views: 4
  • EFI.zip
    36.6 MB · Views: 2

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
Non-unique USB ports are causing your kernel panic with the EFI in the post above. This Ismay be down to the USBMap.kext in the EFI using the iMacPro1,1 model name, while your SMBIOS data is set for a MacPro7,1. The wrong model name will mean the kext won't work. I Having reviewed the USBMap.kext it is clear the creator of the kext didn't really know what he was doing. All the USB ports are set as USB3. This will be wrong and another reason why this kext won't work.

Having similar CPU's is not a problem, as long as the CPU core count in the first 4 x Kernel Patches in the config.plist are set to match your CPU. In this case it does, 8-cores.

The SSDT-HPET.aml table is missing the accompanying ACPI > Patches. So this table won't work in your or the other persons setup. It may be better to delete the SSDT-HPET table, or at least disable the table in the config.plist temporarily.

I have made a number of changes to the EFI in post #6, removing the defunct issues listed above and any unnecessary kexts, entries related to Bluetooth etc.

Try this EFI and see where it gets you.
 

Attachments

  • EFI-spakk-2.zip
    11.5 MB · Views: 8

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
@Edhawk

Thank you for your thorough research and detailed work! Your analysis of the USBMap.kext, particularly regarding the defined USB ports and the differences in SMBIOS configurations, was extremely helpful. I had completely overlooked this aspect, and your insights clarified many connections that I wouldn’t have noticed otherwise.

Your adjustments to the config.plist and your recommendation to disable or remove the SSDT-HPET.aml table were also very valuable. Cleaning up my EFI folder and removing unnecessary entries is now something I can approach more confidently thanks to your support.

Thank you once again for your time and detailed assistance! I tried your adjusted EFI configuration, but unfortunately, the boot process is still failing, and the system doesn’t start up successfully, see the attached images of the boot process.
 

Attachments

  • IMG-20241114-WA0022.jpg
    IMG-20241114-WA0022.jpg
    218.6 KB · Views: 7
  • IMG-20241114-WA0021.jpg
    IMG-20241114-WA0021.jpg
    287.9 KB · Views: 6
  • IMG-20241114-WA0020.jpg
    IMG-20241114-WA0020.jpg
    356.5 KB · Views: 6
  • IMG-20241114-WA0019.jpg
    IMG-20241114-WA0019.jpg
    355.5 KB · Views: 5
  • IMG-20241114-WA0018.jpg
    IMG-20241114-WA0018.jpg
    361 KB · Views: 5

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
Ok, I will have another look at the EFI folder when I am next at my desk.
 

Law_ri

Donator
Donator
AMD OS X Member
Joined
Dec 3, 2021
Messages
40
As I'm told this gpu doesn't need WhateverGreen, just device-id FF670000 and it works in my case.
Also why SSDT-PLUG, isn't that for Intel.
Is there any special reason why you didn't use prebuilt SSDT-EC-USBX-DESKTOP.aml?
 
Last edited:

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
@Law_ri

I’ve tried everything and came across a user who mentioned successfully booting their system with almost identical hardware to what I’ve described above.
Even in my numerous previous tests, where I only used SSDT-EC-USBX-DESKTOP.aml, as you correctly pointed out, I ended up with the same boot result.
Before that, I had created and tested SSDTs specifically tailored for my system, but those attempts also led to the same results.

Additionally, during my research, I found reports from users suggesting completely omitting WhateverGreen.kext and instead simply configuring the device ID in the config.plist. Unfortunately, even with this approach, the boot result remains unchanged.

After so many tests, I am completely at a loss and don’t know what else to try.

XML Code:
Code:
                <dict>

                    <key>Arch</key>

                    <string>Any</string>

                    <key>BundlePath</key>

                    <string>WhateverGreen.kext</string>

                    <key>Comment</key>

                    <string>V1.6.8 | WhateverGreen.kext</string>

                    <key>Enabled</key>

                    <false/>

                    <key>ExecutablePath</key>

                    <string>Contents/MacOS/WhateverGreen</string>

                    <key>MaxKernel</key>

                    <string></string>

                    <key>MinKernel</key>

                    <string></string>

                    <key>PlistPath</key>

                    <string>Contents/Info.plist</string>

                </dict>


XML Code for AMD Radeon RX550:

Code:
        <key>DeviceProperties</key>

        <dict>

            <key>Add</key>

            <dict>

                <key>PciRoot(0x0)/Pci(0x1,0x3)/Pci(0x0,0x2)/Pci(0x8,0x0)</key>

                <dict>

                    <key>AAPL,slot-name</key>

                    <string>Internal@0,1,3/0,2/8,0</string>

                    <key>built-in</key>

                    <data>AQ==</data>

                    <key>device-id</key>

                    <string>FF670000</string>

                    <key>device_type</key>

                    <string>Graphics controller</string>

                    <key>model</key>

                    <string>AMD Radeon RX 550</string>

                </dict>

            </dict>



Video: https://www.transfernow.net/dl/20241115fO3Z77NC/YUzNrD67
 

Attachments

  • IMG-20241115-WA0001.jpg
    IMG-20241115-WA0001.jpg
    644.5 KB · Views: 2
  • IMG-20241115-WA0002.jpg
    IMG-20241115-WA0002.jpg
    625 KB · Views: 2
  • IMG-20241115-WA0003.jpg
    IMG-20241115-WA0003.jpg
    624.6 KB · Views: 2
  • IMG-20241115-WA0004.jpg
    IMG-20241115-WA0004.jpg
    629.5 KB · Views: 2
  • IMG-20241115-WA0005.jpg
    IMG-20241115-WA0005.jpg
    536.4 KB · Views: 2
  • IMG-20241115-WA0006.jpg
    IMG-20241115-WA0006.jpg
    475.2 KB · Views: 2
  • IMG-20241115-WA0007.jpg
    IMG-20241115-WA0007.jpg
    567.3 KB · Views: 7
Last edited:

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
I'm told this gpu doesn't need WhateverGreen
That is not as unusual as you would think.

Probably related to the fact you are spoofing a device-id for a compatible GPU, which is natively supported by the AMD drivers in macOS.

Is there any special reason why you didn't use prebuilt SSDT-EC-USBX-DESKTOP.aml?
Because the Custom SSDT's created with his system DSDT.aml table and Corpnewt's SSDTTime script are a better fit to his system than the generic/catchall SSDT's available from Dortania.

Also why SSDT-PLUG, isn't that for Intel.
Yes this SSDT is used for Intel systems, but it can also be used for AMD systems. The one in Spakk's EFI was generated from his system DSDT.aml and Corpnewt's SSDTTime script. He can choose to remove the SSDT and config.plist entry if he wants. But should make sure the errors in the Verbose text are CPU related before doing so.

From what I read it appears to be related to graphic issues, not CPU issue.
 

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
@spakk can you confirm which is your graphics card, from the table in the link below.

 

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
This is the AMD Radeon RX 550 (Polaris 21) graphics card with the following PCI device IDs: 1002-67FF / 1682-9550 (Rev FF).
The graphics card should be compatible with macOS Sequoia according to the hardware specifications.

Edit:
This is the same graphics card; however, my version has 4 GB of RAM, whereas the one described in the link only has 2 GB: https://www.techpowerup.com/gpu-specs/radeon-rx-550x-640sp.c3203
 

Attachments

  • Screenshot 2024-11-15 213214.png
    Screenshot 2024-11-15 213214.png
    72.3 KB · Views: 10
Last edited:

Law_ri

Donator
Donator
AMD OS X Member
Joined
Dec 3, 2021
Messages
40
Thanks to both for explanation, I was just being curious.
Man you have original Buffin core, your card should work with no problem, that's odd.

Long shot but can you change last 2 Kernel patches, deactivate number 23, and activate number 24. I was told that it might help with gpu performance.
This is how they look at my config.
 

Attachments

  • Screenshot 2024-11-16 at 00.55.07.png
    Screenshot 2024-11-16 at 00.55.07.png
    97.4 KB · Views: 7
Last edited:

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
Thanks to both for explanation, I was just being curious.
Man you have original Buffin core, your card should work with no problem, that's odd.

Long shot but can you change last 2 Kernel patches, deactivate number 23, and activate number 24. I was told that it might help with gpu performance.
This is how they look at my config.
No, this does not affect whether a graphics card ultimately works or not. As far as I remember, Sinetek and Bronya encountered graphical artifacts during the AMD kernel project about ten years ago, particularly among users with AMD Radeon graphics cards. I personally had no issues at the time with my Nvidia GTX 550 TI. These artifacts occurred in connection with patching the Mavericks AMD kernel. To address these issues, they added the relevant code snippet to the mtrr.c file.

I am not familiar with binary patches, but I assume that Algrey and Shaneee adapted this AMD patch accordingly. Based on what I’ve read, there are two different approaches described by users:

Algrey’s Original Patch:
  • Works for all GPUs.
  • Does not affect audio.
  • Does not improve GPU performance.
Shaneee’s Patch:
  • Can boost GPU performance on AMD GPUs.
  • Prevents Nvidia GPUs from booting.
  • Causes audio issues when using HDMI and DP audio.
One user wrote:
"The standard-compatible patch is Algrey’s original patch. It works for all GPUs, does not affect audio, and does not improve performance. The alternative is Shaneee’s patch, which can boost GPU performance on AMD GPUs but prevents Nvidia GPUs from booting. Additionally, it causes audio issues when using HDMI and DP audio. It’s up to you which patch you prefer. Please do not try to use both simultaneously."

Another user claimed that switching from Algrey’s patch to Shaneee’s patch significantly improved their graphics card performance.

I’d like to leave these statements as they are without further comment.

Nonetheless, I tested both patches, disabling the other each time. Unfortunately, I always ended up with the same result.

Edit:

It turns out the issue was actually caused by my AMD Radeon RX 550. Just for fun, I swapped my NVIDIA GTX 1060 for the Radeon RX 550, and voilà – I successfully booted into macOS Sequoia!
🤣🤣🤣

I’m now in the process of completing the macOS Sequoia installation. Thank you all for your patience and support!👌

1731767026820.jpeg
 
Last edited:

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
@spakk I recall using and testing a number of your patched kernels over at InsanelyMac all those years ago, on AMD Athlon and FX systems.

As you say your GPU should work OOB in all recent versions of macOS., so I am not sure why you are having these issues.

It might be worth installing an older version of macOS, Ventura for example and then seeing about updating your system to a newer macOS release. As Sequoia has thrown up a few issues.
 

spakk

New member
AMD OS X Member
Joined
Aug 13, 2020
Messages
20
@spakk I recall using and testing a number of your patched kernels over at InsanelyMac all those years ago, on AMD Athlon and FX systems.

As you say your GPU should work OOB in all recent versions of macOS., so I am not sure why you are having these issues.

It might be worth installing an older version of macOS, Ventura for example and then seeing about updating your system to a newer macOS release. As Sequoia has thrown up a few issues.
Ah yes, those were the good old days with the brightest minds that shaped the Hackintosh scene of the past.

I had previously tried macOS Catalina 10.15.5 as well, but it was also impossible to boot into the desktop. Then I started experimenting, and I began to doubt – the rest of the story you already know.

Now the installation is acting up. About 12 minutes before completion, while SharedSupport.dmg is being extracted, an error occurs: "An error occurred while loading the update." The installation then abruptly stops. I need to investigate the cause.
 

Edhawk

Guru
Guru
Joined
May 2, 2020
Messages
2,798
Yes they were good days, yourself and @Shaneee were both very helpful with your guides and patched kernels. Bronya and the others too, of course.


I've recently had a similar issue when using a Sequoia 15.1 USB installer. Installing Ventura and then updating to Sequoia via the Apple OTA option worked for me.

Try downloading Ventura, it will work with my EFI. Sonoma as well. But I've found Ventura gives less issues during the installation. Also if you have a Broadcom WiFi/BT card it will work in Ventura as the BCM drivers are still contained within the OS.


Catalina requires a change to the UEFI > APFS > MinDate and MinVersion entries in the config.plist, as it uses a different version of Apple's APFS. As do Mojave and High Sierra, both have different data for these two entries. Without these MinDate and MinVersion options being changed the OS won't boot. Big Sur and newer don't require any changes to these entries, as the default '0' entry works for all newer versions of macOS.

UEFI-APFS-MinDate+MinVersion.png MinDate and MinVersion entries for the three older OS's.
 

Law_ri

Donator
Donator
AMD OS X Member
Joined
Dec 3, 2021
Messages
40
...

It turns out the issue was actually caused by my AMD Radeon RX 550. Just for fun, I swapped my NVIDIA GTX 1060 for the Radeon RX 550, and voilà – I successfully booted into macOS Sequoia!
🤣🤣🤣
...
Just for curiosity, now your 1060 is just VESA drivers, no support but works from CPU?

I would love to use my 1080ti and not to have 2 gpus inside case.
 
Back
Top Bottom
  AdBlock Detected
Sure, ad-blocking software does a great job at blocking ads, but it also blocks some useful and important features of our website. For the best possible site experience please take a moment to disable your AdBlocker.