Contribute
Register

Problem with Asus XG-C100F Aquantia Chipset

Hey Grasshopper, hope it works :)
Hey, sadly no. With your aml file the IOReg Name changed to ethernet, but the card is still not showing up under network, kext isnt loading


sudo kextstat | grep AppleEthernetAquantiaAqtion

Executing: /usr/bin/kmutil showloaded
No variant specified, falling back to release

did 2 nvmram resets in opencore

Changing to AMD (hence the "exotic" ACPI paths) and trying Monterey may be too much change in one go…
"ETH0" or "ETH1" is not the issue: Any legal ACPI name would do, and _SB.PCI0.GPP2.PT02.PT24.ETH0 is not the same as ETH0 in another path. (You may try "ERSR" or any other mangling of "erasor2010" to four characters if you want ;) )
Maybe a full copy of your IOReg could help understand what's going on?
yes, maybe that was to much, but i had to make a fresh install so i thaught why not :)

here is the complete ioreg
 

Attachments

  • Max’s Mac Pro.ioreg
    4.4 MB · Views: 68
Thank you for the file bit it didnt work, after the first reboot with NVMRAM reset nothing changed, than i changed eth0 to eth1 in the aml file because my onboard ethernet already was eth0, after that it changed to 06:00.0 1D6A 07B1 1043 874A like written in post #5 but the card still dont show up

@milksjeek here is the IOReg Name

/PCI0@0/GPP2@1,3/PT02@0,2/PT24@4/ETH1@0

sudo kextstat | grep AppleEthernetAquantiaAqtion gives me this output
Executing: /usr/bin/kmutil showloaded
No variant specified, falling back to release
80 0 0xffffff8002bcc000 0x29000 0x29000 com.apple.driver.AppleEthernetAquantiaAqtion (1.0.64) 494E324D-9FBC-32B6-BB8D-F1B547BBC450 <79 51 50 16 8 7 6 3 1>

what i forgot to mention, with the motherboard change i switched to an AMD Based System and upgraded to MontereyView attachment 533363
Your string is wrong, disable/remove quantia.aml and disable kernel patches, give me the hackintool string -> PCIe again.
should be: "vendor 0x1D6A", "device 0x00B1", Device Name AQC 100..." need the IOReg Name
 
Last edited:
Your string is wrong, disable/remove quantia.aml and disable kernel patches, give me the hackintool string -> PCIe again.
should be: "vendor 0x1D6A", "device 0x00B1", Device Name AQC 100..." need the IOReg Name
Hi, here the string

vendor: 0x1D6A
device: 0x00B1
Sub Vendor: 0x1043
Sub Device 0x874A
Vendor Name: Aquantia Corp.
Device Name: AQC100 10G Ethernet MAC Controller
IOReg Name: /PCI0@0/GPP2@1,3/PT02@0,2/PT24@4/ethernet@0
Bildschirmfoto 2021-11-08 um 17.04.04.png


sudo kextstat | grep AppleEthernetAquantiaAqtion
Executing: /usr/bin/kmutil showloaded
No variant specified, falling back to release
 

Attachments

  • Max’s Mac Pro.ioreg
    4.3 MB · Views: 76
Thank you for the file bit it didnt work, after the first reboot with NVMRAM reset nothing changed, than i changed eth0 to eth1 in the aml file because my onboard ethernet already was eth0, after that it changed to 06:00.0 1D6A 07B1 1043 874A like written in post #5 but the card still dont show up

@milksjeek here is the IOReg Name

/PCI0@0/GPP2@1,3/PT02@0,2/PT24@4/ETH1@0

sudo kextstat | grep AppleEthernetAquantiaAqtion gives me this output
Executing: /usr/bin/kmutil showloaded
No variant specified, falling back to release
80 0 0xffffff8002bcc000 0x29000 0x29000 com.apple.driver.AppleEthernetAquantiaAqtion (1.0.64) 494E324D-9FBC-32B6-BB8D-F1B547BBC450 <79 51 50 16 8 7 6 3 1>

what i forgot to mention, with the motherboard change i switched to an AMD Based System and upgraded to MontereyView attachment 533363
Thank you for the file bit it didnt work, after the first reboot with NVMRAM reset nothing changed, than i changed eth0 to eth1 in the aml file because my onboard ethernet already was eth0, after that it changed to 06:00.0 1D6A 07B1 1043 874A like written in post #5 but the card still dont show up

@milksjeek here is the IOReg Name

/PCI0@0/GPP2@1,3/PT02@0,2/PT24@4/ETH1@0

sudo kextstat | grep AppleEthernetAquantiaAqtion gives me this output
Executing: /usr/bin/kmutil showloaded
No variant specified, falling back to release
80 0 0xffffff8002bcc000 0x29000 0x29000 com.apple.driver.AppleEthernetAquantiaAqtion (1.0.64) 494E324D-9FBC-32B6-BB8D-F1B547BBC450 <79 51 50 16 8 7 6 3 1>

what i forgot to mention, with the motherboard change i switched to an AMD Based System and upgraded to MontereyView attachment 533363
Hi, the quanta.aml worked, Vendor Device 0x1D6A 0x7B1 ACQ107 and therefore is the quantia kext loaded it self

kextstat | grep AppleEthernetAquantiaAqtion

80 0 0xffffff8002bcc000 0x29000 0x29000 com.apple.driver.AppleEthernetAquantiaAqtion (1.0.64) 494E324D-9FBC-32B6-BB8D-F1B547BBC450 <79 51 50 16 8 7 6 3 1> (kernel loaded)


As a test swap your video card and network card and make a new .aml file, I saw your board layout slots PCIe 2.0 and 3.0 only as test. or a change to iMac Pro(2017) don't know if that's possible with AMD
 
Hi, the quanta.aml worked, Vendor Device 0x1D6A 0x7B1 ACQ107 and therefore is the quantia kext loaded it self

kextstat | grep AppleEthernetAquantiaAqtion

80 0 0xffffff8002bcc000 0x29000 0x29000 com.apple.driver.AppleEthernetAquantiaAqtion (1.0.64) 494E324D-9FBC-32B6-BB8D-F1B547BBC450 <79 51 50 16 8 7 6 3 1> (kernel loaded)


As a test swap your video card and network card and make a new .aml file, I saw your board layout slots PCIe 2.0 and 3.0 only as test. or a change to iMac Pro(2017) don't know if that's possible with AMD
Hi, i switched back to the aml file from etorix and changed to iMacPro, kext is loaded and vendor and device ids changed to 0x1D6A and 0x07B1 but the card is not showing up like bevor.

Switching PCIE Slots is not possible, not enough space in the case
 
For the ASUS XG-C100F, NO SSDT is needed - just a kernel-patch and the aquantia kernel-quirk with one entry in the device-properties of the card and it works....ok properly configured AppleVTD is a prerequisite.

As always, 'AAPL,slot name' has only a cosmetic effect...the other entry does the trick.
2.png

1.png



3.png


4.png
 

Attachments

  • ASUS-XG-C100F.plist
    1.3 KB · Views: 83
Last edited:
For the ASUS XG-C100F, NO SSDT is needed - just a kernel-patch and the aquantia kernel-quirk with one entry in the device-properties of the card and it works....ok properly configured AppleVTD is a prerequisite.

As always, 'AAPL,slot name' has only a cosmetic effect...the other entry does the trick.View attachment 555672

View attachment 555568



View attachment 555571

View attachment 555572

Thanks for your info!! I got further then before…
Unfortunately, after I copied everything from your plist (with my PCIe settings) my card is still not working.
Driver is loaded, AppleVTD is configured, but the card is showing Cable Unplugged!
If I didn’t try and confirm that the cable works with Asus card (in Windows and ESXi and Truenas) as well as the other card that it is connected to, I would blame the cable… Butt!

Also, the other machine (Truenas server) shows Interface LINK STATE UP!

Hope you have an idea… really need this to work!
Thanks again!
 
Last edited:
Thanks for your info!! I got further then before…
Unfortunately, after I copied everything from your plist (with my PCIe settings) my card is still not working.
Driver is loaded, AppleVTD is configured, but the card is showing Cable Unplugged!
If I didn’t try and confirm that the cable works with Asus card (in Windows and ESXi and Truenas) as well as the other card that it is connected to, I would blame the cable… Butt!

Also, the other machine (Truenas server) shows Interface LINK STATE UP!

Hope you have an idea… really need this to work!
Thanks again!
Info for someone…

After the situation in my last message all that was needed was a reboot with cable plugged in. The it showed that the cable was connected! I thought that the problem was solved.

Unfortunately I’m experiencing hard locks after connecting or disconnecting the cable.
 
This works.

I was wondering, since the kext advertises suport for pci1d6a,80b1 (AQC100S) why won't the kext load only by providing correct compatible string and device-id for it? AQC100 and AQC100S have same capabilities according to linux driver.

Also devices like https://www.qnap.com/en/product/qna-t310g1s will work out of the box (not sure if it has the S version)

I also brought it up using compatible string of pci1d6a,80b1 and patch 81F9B180 0000 -> 81F9B100 0000
Does this mean configRead16 will bypass device-id property? That's why we need binary patch?
 
Last edited:
For the ASUS XG-C100F, NO SSDT is needed - just a kernel-patch and the aquantia kernel-quirk with one entry in the device-properties of the card and it works....ok properly configured AppleVTD is a prerequisite.

As always, 'AAPL,slot name' has only a cosmetic effect...the other entry does the trick.View attachment 555672

View attachment 555568



View attachment 555571

View attachment 555572
any luck with macOS Sonoma? It worked correctly with Ventura but not with Sonoma
 
Back
Top