X299 Big Sur Support

[izo1]

Is the usb audio device powered solely by usb? Just suspecting a usb power issue, sounds like when an input device doesn’t get enough power and you get random glitchy input. I’m just kind of thinking out loud but that could still be software related hence introduction at 11.1 etc not sure if there’s any way to test that but I’d be tempted to put it on a front panel port so it’s on a dedicated m/b header and see if it makes any difference or failing that, see if there’s any difference with powered usb hub - might help point in the right direction if there’s any difference in behaviour.

Another good point ^

Might be worth passing through a USB hub that has a power brick?
Something like this that's available in your country.

 

[pieropontra]

Is the usb audio device powered solely by usb? Just suspecting a usb power issue, sounds like when an input device doesn’t get enough power and you get random glitchy input. I’m just kind of thinking out loud but that could still be software related hence introduction at 11.1 etc not sure if there’s any way to test that but I’d be tempted to put it on a front panel port so it’s on a dedicated m/b header and see if it makes any difference or failing that, see if there’s any difference with powered usb hub - might help point in the right direction if there’s any difference in behaviour.

Another good point ^

Might be worth passing through a USB hub that has a power brick?
Something like this that's available in your country.

Unfortunately the only 2.0 mapped ports are the ones on the back of the mobo, I don't want to remap everything on a new kext... I'm exhausted... believe me... And I don't wanna spend money, actually I listed this machine on eBay tryin to get rid of it and wait for M1X Mac mini using my Chromebook in the meantime. Don't really need recording gear right now, It's a slow decline away from hacks ahahah.

Anyway attached the USB info screen that does confirm (apparently?!?!) the card is receiving the 500mA it needs.

The Radeon VII has some coil whine but it doesn't seem to be synced with kernel errors

But, again, it worked with zero issues until 11.1

 

[izo1]

Unfortunately the only 2.0 mapped ports are the ones on the back of the mobo, I don't want to remap everything on a new kext... I'm exhausted... believe me... And I don't wanna spend money, actually I listed this machine on eBay tryin to get rid of it and wait for M1X Mac mini using my Chromebook in the meantime. Don't really need recording gear right now, It's a slow decline away from hacks ahahah.

Anyway attached the USB info screen that does confirm (apparently?!?!) the card is receiving the 500mA it needs.

The Radeon VII has some coil whine but it doesn't seem to be synced with kernel errors

You can get a pretty penny for the Radeon VII since miners love it right now haha

 

[pieropontra]

You can get a pretty penny for the Radeon VII since miners love it right now haha

I thought about this, but I will than pay a boatload of cash for an inferior card!

 

[hughesyadaddy]

Unfortunately the only 2.0 mapped ports are the ones on the back of the mobo, I don't want to remap everything on a new kext... I'm exhausted... believe me... And I don't wanna spend money, actually I listed this machine on eBay tryin to get rid of it and wait for M1X Mac mini using my Chromebook in the meantime. Don't really need recording gear right now, It's a slow decline away from hacks ahahah.

Anyway attached the USB info screen that does confirm (apparently?!?!) the card is receiving the 500mA it needs.

The Radeon VII has some coil whine but it doesn't seem to be synced with kernel errors

But, again, it worked with zero issues until 11.1

Thank GOD someone else is talking about this. I've about had it with my Hackintosh and the USB audio issues since Big Sur 11.1. I am getting the same exact error that you are getting. I've tried 3 USB cards and somehow got the Sonnet Allegro Pro USB 3.2 Type A PCIe Card to work without the crashes in Audio; however, for some reason the other cards do! I've played with literally everything you can imagine. Throwing a shitload of different kexts at it. Remapping my USB using different bundleIDs, changing nearly every setting in the BIOS just to test, changing mac definitions from iMacPro1,1 to MacPro7,1, filling in all of the suggested DSDT modifications you can imagine, disabling the Vega VII audio with AppleHDADisabler.kext, disabling all but one of my three monitors, changing USB ports, patching com.apple.iokit.IOHIDFamily, removing the bluetooth Broadcom from the USB ports, ensuring the broadcom port is 255, disabling wifi, ect. The only clue I have is that the Sonnet Allegro card isn't using a AppleUSBHubPort for the USB devices themselves as you see below compared to the XHCI intel ports. I suspect another Kext is overriding this? Potentially the Broadcom chip because it is sharing bandwidth with my XHCI devices? But even disabling it still happened so who knows. I would literally pay someone to help me at this point because the Allegro Pro of course doesn't work with my windows drive and now I literally have to switch a bunch of USB devices around each time I want to boot into another fricken OS.

Also additionally, I've been studying the logs and basically isolated it to this. When I plug in the speaker or any audio device into the Sonnet card - HALS_Object_GetPropertyData_DAI32 gets the property BUT when I plug it in any other USB port it shows this error.

This is before the crash. When the audio cuts out and eventually crashes, I am getting an error that looks like the computer is searching for audio devices but has trouble with the ID and then looses the connection.

Any help would be so greatly appreciated! I'm am apparently too dumb to get this one.

<3 Alex

 

[djlild7hina]

Few questions @pieropontra and @hughesyadaddy:
1. Any specific audio interface you guys are using that has crackling?
2. Does this also happen on a real mac on Big Sur 11.1+?
3. Any examples that causes it? I've tried my Scarlett 2i4 and haven't heard any issues though I've only tried HEVC videos and Apple Music so far.

Tried a front USB 3 port and a back USB 2 port. I remember some people had issues with audio popping with the Radeon VII but not sure if that's related or not. I also have a RX 580 that's different from you guys

 

[pieropontra]

@hughesyadaddy

Dumb? You basically spent the last months tryin to solve this and surely you went far deeper than I did! Surely you're not dumb. I think Apple changed something in the config of the MacPro7,1 in 11.1 'cause it's the only one with Vega20 GPU (Radeon VII and Vega II Pro use the same architecture) and this broke coreaudiod. I've also read on a Swedish forum that a genuine Mac Pro owner has the same issue and he has the Vega II. We should try I different GPU but right now it's complicated.

@djlild7hina (tagging your name is harder than reaching Mars mate, ahahah )

1. I got a Scarlett 2i2 Gen 3 (similar to yours with less outputs I think).
2. read above | Radeon VII is known to be not officially supported by Apple (even if they adopted the deviceID) and even Sonnet discourages its use in their TB boxes.
3. play this

it destroys the audio since it's an high bitrate 4K VP9 video.

 

[rustEswan]

AppleUSBHubPort for the USB devices themselves as you see below compared to the XHCI intel ports. I suspect another Kext is overriding this? Potentially the Broadcom chip because it is sharing bandwidth with my XHCI devices? But even disabling it still happened so who knows. I would literally pay someone to help me at this point because the Allegro Pro of course doesn't work with my windows drive and now I literally have to switch a bunch of USB devices around each time I want to boot into another fricken OS.
View attachment 508742View attachment 508741
Also additionally, I've been studying the logs and basically isolated it to this. When I plug in the speaker or any audio device into the Sonnet card - HALS_Object_GetPropertyData_DAI32 gets the property BUT when I plug it in any other USB port it shows this error.

The only glaringly obvious thing here is that you're plugged into an internal hub.

This is why I was asking @pieropontra about which port it is plugged into - a lot of boards have a group of 2,4 or more ports connected to an internal hub - behind that bank of 4 ports there could actually be just a single USB port, as, in the case above HS05@14500000 - its just one port with multiple devices connected, I don't know what they are because the grey lines dissappear off the bottom of the screenshot.

Again, I don't know, but I strongly suspect power related, perhaps due to power regulation being applied to one port but having multiple devices attached. I know it only changed 11.1 but perhaps something changed under the hood for power delivery/regulation. Pretty sure real Macs all have a single ACPI port per physical port.

...and just a side note on the Allegro Pro, looks like that is supported in Windows? Happened to see that when looking at its product page to see its power features:

Allegro Pro Type A USB 3.2 Gen 2 - Sonnet

These are individually addressed as well - like we see in your first screenshot.

First screenshot
Controller address(Allegro Pro): 03000000
USB2.0 XCHI ACPI Port 03300000
Your Device: 03300000

Your device is on the same ACPI address as the port, which presumably regulates power delivery for the attached devices (just one).

Second screenshot
Controller address(XCHI internal X299): 14000000
USB2.0 XHCI ACPI Port HS05: 14500000
sub-address of internal hub port: 14510000
Your Device: 14510000

Now your device has a different ACPI address to the internal HS05 port which again presumably power delivery is regulated to that port and multiple devices may be sharing available power.

Thats why I wanted to see a full USBMap - It would at least tell us what's common here.

I've used two different gigabyte X299's both with internal USB hubs for all Type-A ports. I've seen it on some ASUS boards too. In fact this MSI X299 Pro I'm using now is the first one I've seen where every USB-A is on its own ACPI port number - like a real Mac. More annoying for overall port limit of 15, but I think avoids potential issues like this.

Honestly, I don't know if this is your problem, I'm not an expert, but that would certainly be my first point of investigation - Its why I suggested the Front Panel USB ports because most of the maps I've made have shown those internal USB3.0 headers to have dedicated ACPI addresses and not be attached to internal hubs.

 

[pieropontra]

The only glaringly obvious thing here is that you're plugged into an internal hub.

This is why I was asking @pieropontra about which port it is plugged into - a lot of boards have a group of 2,4 or more ports connected to an internal hub - behind that bank of 4 ports there could actually be just a single USB port, as, in the case above HS05@14500000 - its just one port with multiple devices connected, I don't know what they are because the grey lines dissappear off the bottom of the screenshot.

Again, I don't know, but I strongly suspect power related, perhaps due to power regulation being applied to one port but having multiple devices attached. I know it only changed 11.1 but perhaps something changed under the hood for power delivery/regulation. Pretty sure real Macs all have a single ACPI port per physical port.

...and just a side note on the Allegro Pro, looks like that is supported in Windows? Happened to see that when looking at its product page to see its power features:

Allegro Pro Type A USB 3.2 Gen 2 - Sonnet

This are individually addressed as well - like we see in your first screenshot.

First screenshot
Controller address(Allegro Pro): 03000000
USB2.0 XCHI ACPI Port 03300000
Your Device: 03300000

Your device is on the same ACPI address as the port, which presumably regulates power delivery for the attached devices (just one).

Second screenshot
Controller address(XCHI internal X299): 14000000
USB2.0 XHCI ACPI Port HS05: 14500000
sub-address of internal hub port: 14510000
Your Device: 14510000

Now your device has a different ACPI address to the internal HS05 port which again presumably power delivery is regulated to that.

Thats why I wanted to see a full USBMap - It would at least tell us what's common here.

I've used two different gigabyte X299's both with internal USB hubs for all Type-A ports. I've seen it on some ASUS boards too. In fact this MSI X299 Pro I'm using now is the first one I've seen where every USB-A is on its own ACPI port number - like a real Mac. More annoying for overall port limit of 15, but I think avoids potential issues like this.

Honestly, I don't know if this is your problem, I'm not expert, but that would certainly be my first point of investigation - Its why I suggested the Front Panel USB ports because most of the maps I've made have shown those internal USB3.0 headers to have dedicated ACPI addresses and not be attached to internal hubs.

As soon as I have a free hour, maybe tonight, I'll try remapping my USBs and use the audio card on the ASMedia controller instead of the Intel one currently handling my rear USB 2.0 and let you know.

 

[rustEswan]

As soon as I have a free hour, maybe tonight, I'll try remapping my USBs and use the audio card on the ASMedia controller instead of the Intel one currently handling my rear USB 2.0 and let you know.

Just make sure when you map the ASMedia that it isn't also using internal hubs - my Aorus board did.

See previous posts for examples - Aorus USB layout (using internal hubs) and MSI X299 USB layout (using individually addressed ports)