Audio used to work until it changed to AZAL

[DarkSilentSC]

I had previously working audio with AppleALC layoutid=7 (well sort of, since I had to use one of the surround port to get audio)
But after purchasing Fenvi T-919, I had to reconnect my external audio card from one PCIe Slot from one slot to another.

After this is done, I made sure I go back and forth between gfxutil and/or Hackintool PCIe table list and update my OC DeviceProperties so that external-audio=1 and alctcsel=1 is assigned to the correct Pciroot directory.

However, Mac just seems to assign HDEF to my external card, and assigns AZAL to my onboard audio device, when I looked in IOReg, Even tho such device still shows up in PCIe list in Hackintool.

In Hackintool > System > Peripherals, Audio table shows only my USB connected C922 Webcam, and Navi 21 HDMI Audio [Radeon RX 6900XT]
Are there any additional edits am I supposed to do in my config.plist?

Reply below if I need to post my EFI or config.plist, and I shall. Thanks in advance!

 

[Edhawk]

Yes you need to post a copy of your EFI folder, so we can see what you are using and just as importantly what you are not using.

Which PCIe Audio Card are you using?
Which version of macOS are you using?

If you can also post a copy of your IOReg and the pciedevices.plist from Hackintool, that would be helpful.

 

[DarkSilentSC]

Here you go.

Yes you need to post a copy of your EFI folder, so we can see what you are using and just as importantly what you are not using.

Which PCIe Audio Card are you using?
Which version of macOS are you using?

If you can also post a copy of your IOReg and the pciedevices.plist from Hackintool, that would be helpful.

 

[DarkSilentSC]

PCIe audio card installed is Creative AE-5, which I’m trying to make AppleALC to ignore by adding external-audio=01. MacOS is 12.5.1

 

[Edhawk]

Your MacPro71IOReg is corrupt, according to two versions of IORegistryExplorer I used to try to open your IOReg.



Can you download the IoregistryExplorer v 2.1 dmg from this Github page and remake a copy of your IOReg.

GitHub - vulgo/IORegistryExplorer: Apple IORegistryExplorer 2.1

Apple IORegistryExplorer 2.1. Contribute to vulgo/IORegistryExplorer development by creating an account on GitHub.

github.com

 

[Edhawk]

You have 2 x WiFi/BT cards installed in your system, as shown in the screenshots from your pciedevices.plist

Broadcom WiFi/BT
below.

Intel WiFi/BT


Are both enabled in your system?

Your USBMap.kext doesn't contain or activate any 'Internal' header ports. These would be set with the connector type (255), if set correctly.

Motherboard layout highlighting M.2 WiFi/BT connector and USB2 header port on motherboard

With either of the 2 x WiFi/BT cards active you would need at least one Internal port active.

• The built-in port serving the M.2 connector containing the Intel card, or
• One of the USB2 header ports (there are 2 x USB2 ports per header), for the Broadcom Bluetooth device to work.

You have 1 x Intel WiFi and 3 x Intel Bluetooth kexts present in your /EFI/OC/Kexts folder but none are enabled. So I assume you have the Intel WiFi/BT card enabled and active in your system, but only for Windows?

Having both WiFi/BT cards active is not usually a good option, as the Bluetooth modules can clash when running macOS. You can use your Broadcom WiFi/BT card in Windows 10 or 11, with the correct Bootcamp drivers/installers. I have attached the drivers I use in Windows 10 and 11 for my Broadcom WiFi/BT cards.

Your Gigabyte X570 AORUS Extreme motherboard has the following USB ports.

Integrated in the CPU:​

1. 2 x USB 3.2 Gen 2/Gen 1 Type-A ports (red) on the back panel
2. 2 x USB 3.2 Gen 1 ports on the back panel

Integrated in the Chipset:​

1. 1 x USB Type-C™ port with USB 3.2 Gen 2 support, available through the internal USB header
2. 1 x USB Type-C™ port on the back panel, with USB 3.2 Gen 2 support
3. 3 x USB 3.2 Gen 2 Type-A port (red) on the back panel
4. 2 x USB 3.2 Gen 1 ports available through the internal USB header

Chipset+USB 3.2 Gen 1 Hub:​

1. 2 x USB 3.2 Gen 1 ports available through the internal USB header
2. 2 x USB 2.0/1.1 ports available through the internal USB header

Chipset+USB 2.0 Hub:​

1. 4 x USB 2.0/1.1 ports on the back panel

This means you motherboard can provide a maximum of 32 x USB ports. It has up to 4 x USB controllers, each of which can activate up to 15 ports each.

You are currently activating these 3 x USB controllers in your USBPorts.kext - XHC0, XHC2 and XHC3. What happened to XHC1?

Your USBPorts.kext actives 29 of the possible 32 ports. Are you purposefully not including the 3 x Internal ports?

The 4 x physical USB2 ports on the rear I/O plate seem to be spread through your kext.

• XHC0 - HS03 & HS06
• XHC2 - HS05 & HS06

Do these 4 x ports match with the 4 x USB2 physical ports on the rear I/O?

Rear I/O plate for X570 AORUS Extreme with 4 x USB2 ports highlighted.

There is nothing Audio related in your config.plist that would cause any issues.

There were a few unnecessary entries, which your AMD system would never use. I have removed them to make the config easier to read and less cluttered.

I have added the correct names for your Audio devices, based on the Sound Blaster card and Realtek codec used by your motherboard.

The only other change I have made is to enable Misc > Security > AllowSetDefault option, so you can use the CMD + Enter key to set your default boot drive in the OC boot screen.

When you provide a readable copy of your IOReg I will look to see what needs to be done to get the Audio devices named correctly.

 

[Edhawk]

Copy of the cleaned up and minor revised config.plist is attached, as mentioned in the post above.

 

[DarkSilentSC]

Oh wow sounds like I got massively wrong when I was working on USBMap then. I tried using corpnewt’s usbmap utility to identify which goes where. XHCI1 never appeared in hackintool USB view. I’ll review these over and make proper changes. You are correct in that I have disabled bluetooth and intel wireless kexts since now I have Fenvi card installed, but wanted to use intel wifi and BT on Windows. I heard Fenvi drivers have issues with Windows, but I could be wrong? I’ll post up the IOreg file as soon as I get back to my rig.

 

[Edhawk]

The Bluetooth and WiFi drivers I attached to post #6 should work with your Fenvi card in Windows 10 or 11. I use them with my Broadcom Apple cards in 6+ systems, and have not had any issues.

 

[DarkSilentSC]

Your MacPro71IOReg is corrupt, according to two versions of IORegistryExplorer I used to try to open your IOReg.

View attachment 8736

Can you download the IoregistryExplorer v 2.1 dmg from this Github page and remake a copy of your IOReg.

GitHub - vulgo/IORegistryExplorer: Apple IORegistryExplorer 2.1

Apple IORegistryExplorer 2.1. Contribute to vulgo/IORegistryExplorer development by creating an account on GitHub.

github.com

Re-attaching IOReg Export via provided installed app via dmg...

 

[DarkSilentSC]

The Bluetooth and WiFi drivers I attached to post #6 should work with your Fenvi card in Windows 10 or 11. I use them with my Broadcom Apple cards in 6+ systems, and have not had any issues.

I understand-- but I prefer using BT5 and Wifi6 in Windows, and want to use Airdrop and other useful feature in MacOS. I'll ponder and decide at a later time.

 

[DarkSilentSC]

Your USBMap.kext doesn't contain or activate any 'Internal' header ports. These would be set with the connector type (255), if set correctly.

View attachment 8741 Motherboard layout highlighting M.2 WiFi/BT connector and USB2 header port on motherboard

With either of the 2 x WiFi/BT cards active you would need at least one Internal port active.

• The built-in port serving the M.2 connector containing the Intel card, or
• One of the USB2 header ports (there are 2 x USB2 ports per header), for the Broadcom Bluetooth device to work.

You have 1 x Intel WiFi and 3 x Intel Bluetooth kexts present in your /EFI/OC/Kexts folder but none are enabled. So I assume you have the Intel WiFi/BT card enabled and active in your system, but only for Windows?

Having both WiFi/BT cards active is not usually a good option, as the Bluetooth modules can clash when running macOS. You can use your Broadcom WiFi/BT card in Windows 10 or 11, with the correct Bootcamp drivers/installers. I have attached the drivers I use in Windows 10 and 11 for my Broadcom WiFi/BT cards.

Your Gigabyte X570 AORUS Extreme motherboard has the following USB ports.

Integrated in the CPU:​

1. 2 x USB 3.2 Gen 2/Gen 1 Type-A ports (red) on the back panel
2. 2 x USB 3.2 Gen 1 ports on the back panel

Integrated in the Chipset:​

1. 1 x USB Type-C™ port with USB 3.2 Gen 2 support, available through the internal USB header
2. 1 x USB Type-C™ port on the back panel, with USB 3.2 Gen 2 support
3. 3 x USB 3.2 Gen 2 Type-A port (red) on the back panel
4. 2 x USB 3.2 Gen 1 ports available through the internal USB header

Chipset+USB 3.2 Gen 1 Hub:​

1. 2 x USB 3.2 Gen 1 ports available through the internal USB header
2. 2 x USB 2.0/1.1 ports available through the internal USB header

Chipset+USB 2.0 Hub:​

1. 4 x USB 2.0/1.1 ports on the back panel

This means you motherboard can provide a maximum of 32 x USB ports. It has up to 4 x USB controllers, each of which can activate up to 15 ports each.

You are currently activating these 3 x USB controllers in your USBPorts.kext - XHC0, XHC2 and XHC3. What happened to XHC1?

Your USBPorts.kext actives 29 of the possible 32 ports. Are you purposefully not including the 3 x Internal ports?

The 4 x physical USB2 ports on the rear I/O plate seem to be spread through your kext.

• XHC0 - HS03 & HS06
• XHC2 - HS05 & HS06

Do these 4 x ports match with the 4 x USB2 physical ports on the rear I/O?

View attachment 8740 Rear I/O plate for X570 AORUS Extreme with 4 x USB2 ports highlighted.

There is nothing Audio related in your config.plist that would cause any issues.

There were a few unnecessary entries, which your AMD system would never use. I have removed them to make the config easier to read and less cluttered.

I have added the correct names for your Audio devices, based on the Sound Blaster card and Realtek codec used by your motherboard.

The only other change I have made is to enable Misc > Security > AllowSetDefault option, so you can use the CMD + Enter key to set your default boot drive in the OC boot screen.

When you provide a readable copy of your IOReg I will look to see what needs to be done to get the Audio devices named correctly.

First of all, let me insert the quote of my findings via corpnewt usb mapping utility (Thread from my R9 5900X on Gigabyte Aorus X570 Xtreme with RX 6900XT in Success Stories:

Updated to Monterey 12.2.1, OC 0.7.8 and now added Custom USBMap.kext for Aorus Xtreme X570 rev.0 motherboard.

I'll take this opportunity on how I used USBMap tool thanks to corpnewt.

Before talking about actual mapping, some realization:

1. Dorthania's guide on using FixShutdown SSDT with searching for _PRW in SB.PCI0.XHC is COMPLETELY useless in AMD systems, since this is more for Intel system. So proper USBMap kext was needed. (However it seems disabling Wake on LAN in BIOS resolved the problem of system starting itself as soon as I unplug any USB devices on the PC)
2. USBMap Kext apparently does not have any executables, and filling the line of executablePath in config.plist causes OC to no longer boot.

As far as Gigabyte Aorus X570 motherboard USB goes, there are 3 XHC Controllers involved:

1. The CPU (XHC0@0,3/60) - System in a chip where it also has its own USB controller
2. The X570 Chipset (XHC1@0,1) - Obviously
3. The 3.2 G1 Hub which is slave to X570 Chipset (XHC1@0,3/61)

The USB map table against the motherboard USB port layouts are as follows:
There is one USB A port (#12) that it just simply won't recognize.

No USB Type Loc Host Type/Port USBMapType
========================================================
1 2.0 A Back Chipset+2.0 Hub (ITE) XHC1@0,1/PRT4 255
2 2.0 A Back Chipset+2.0 Hub (ITE) XHC1@0,1/PRT4
3 2.0 A Back Chipset+2.0 Hub (ITE) XHC1@0,1/PRT4
4 2.0 A Back Chipset+2.0 Hub (ITE) XHC1@0,1/PRT4
5 3.2 G2 A Back CPU XHC0@0,3/60/PRT1 3
6 3.2 G2 A Back CPU XHC0@0,3/60/PRT2 3
7 3.2 G1 A-BIOS Back CPU XHC0@0,3/60/PRT3 3
8 3.2 G1 A Back CPU XHC0@0,3/60/PRT4
9 3.2 G2 A Back Chipset XHC1@0,1/PRT1 3
10 3.2 G2 A Back Chipset XHC1@0,1/PRT5 3
11 3.2 G2 C Back Chipset XHC1@0,1/PRT2 9
12 3.2 G2 A Back Chipset XHC1@0,1/PRT6? 3
13 3.2 G1 A FPanel Chipset+3.2 G1 XHC0@0,3/61/PRT1 3
14 3.2 G1 A FPanel Chipset+3.2 G1 XHC0@0,3/61/PRT2 3
15 3.2 G2 C FPanel Chipset+3.2 G1 XHC0@0,3/61/PRT4 9
16 3.2 G1 A-X FPanel Chipset
17 3.2 G1 A-X FPanel Chipset
18 2.0 A Internal Chipset+3.2 G1 XHC0@0,3/61/PRT3 255
19 2.0 A Internal Chipset+3.2 G1 XHC0@0,3/61/PRT3
20 BT HC A BT Chipset XHC1@0,1/PRT3 255

Edit – Finally figured it out and with ERP disabled in BIOS the system no longer wakes up on its own!. EFI file update will follow soon.

This mapping, at least in Big Sur, worked beautifully. No sudden startup of the PC when it was shut down, sleep/wake, etc. I have been mingling around with USBMap.kext, where the EFI I have sent ... is probably no longer correct and may have to return back to my backup. Let me take a further look on it.

 

[DarkSilentSC]

Okay I remapped from ground up and disabled unused ports and ITE device and Intel bluetooth USB port… shutdown seems better, but the PC boots up when I put a USB in and pull it out while it is shut down.

 

[Edhawk]

That being the case I would check that you have the correct connector type for each port. As that doesn’t happen when things are set correctly.

Post a copy of the USB map, so we can see if anything looks wrong.

 

[DarkSilentSC]

Any thoughts with AZAL0 on my realtek? IOReg post is up in #10. Here's USBMap. Thanks for your help!

 

[Edhawk]

ACPI device names are 4 characters or less, with the missing characters replaced by an ’_’. So AZAL0 would not comply with the expected ACPI code.

What was it named before it changed to AZAL?
What has caused the device name to change to AZAL?
Did you add or replace any kexts in your OC Kexts folder/config.plist?

There was a fairly common ACPI Rename patch in use, before AppleALC.kext was released, which was used to rename the AZAL audio device to HDEF. It became obsolete with the release of AppleALC.kext. Are you sure that AppleALC is loading or being injected in your system?

 

[DarkSilentSC]

ACPI device names are 4 characters or less, with the missing characters replaced by an ’_’. So AZAL0 would not comply with the expected ACPI code.

What was it named before it changed to AZAL?
What has caused the device name to change to AZAL?
Did you add or replace any kexts in your OC Kexts folder/config.plist?

There was a fairly common ACPI Rename patch in use, before AppleALC.kext was released, which was used to rename the AZAL audio device to HDEF. It became obsolete with the release of AppleALC.kext. Are you sure that AppleALC is loading or being injected in your system?

I believe it was named to HDEF with AppleHDA drivers loaded. Now, no AppleHDA seem to latch to either HDEF or AZAL.
I did run kextstat | grep -E "AppleHDA|AppleALC|Lilu" and all 3 do show up.

 

[DarkSilentSC]

IOReg screenshot of AZAL (Realtek) and HDEF (Creative)

 

[DarkSilentSC]

Okay--- I ended up swapping the slot installed between Fenvi card and Creative Sound card to work.
So what this tells me is that if the PCIe sound card is installed right below GPU, this whacks up the whole AppleALC/AppleHDA driver install.
Veeeery interesting.

IOReg screenshots below: Realtek's AZAL is now assigned HDEF, and Creative PCIe card now becomes BYS2.


It's so strange as if AppleALC no longer picks up external-audio parameter if the device is installed behind GPU.

 

[Edhawk]

I don't think you are comparing the same device, from looking at the two IOReg screenshots you provided in post #18 I think you are comparing the Creative Labs audio (HDEF) and AMD audio (AZAL) devices.

When you look at the HDEF device

the device address is \SB.PCI0.GPP9.HDEF

Then look at the AZAL device
the device address is \SB.PCI0.GP13.AZAL

If this were a single device in a different slot then I would expect the Values attributed to the device to be a pretty close match, which they are not. As can be seen in the two screenshots below.

Coloured rectangles have been used to show comparable values, none match.

HDEF values - Creative Labs device

AZAL values - AMD device