Jump to content

Marvell (Aquantia) 10 Gb Ethernet support thread


d5aqoep
520 posts in this topic

Recommended Posts

10 minutes ago, Cyberdevs said:

@i0ntempest

Why are you using IntelMausi.kext? In the IOReg I see it uses en0 are using any PCIe adapters or just the AQC?

Post your efi without the private data (SN MLB etc.) please.

Because my motherboard has a built in Intel ethernet interface, and I want it along with the Broadcom Wi-Fi take en0 and en1 and AQC stays at en2. Will post EFI when I get back to my hack 

Link to comment
Share on other sites

19 minutes ago, Cyberdevs said:

@i0ntempest

Why are you using IntelMausi.kext? In the IOReg I see it uses en0 are using any PCIe adapters or just the AQC?

Post your efi without the private data (SN MLB etc.) please.

Here. Also removed itlwm kexts cuz the archive is too big.

EFI.zip

Link to comment
Share on other sites

13 minutes ago, Cyberdevs said:

Doesn't work unfortunately. I took a look inside, it's basically the changes I made except the XGBE -> PEGP change, but XGBE does not exist on my system. Changing it to PEGP (which is the device path I see on my system) also does not work... which is really weird, I'm using basically the same _DSM method property injection to disable my GPU's digital audio and that works fine.

Link to comment
Share on other sites

10 hours ago, Cyberdevs said:

@i0ntempest

I have this patch enabled in my config maybe that's why it doesn't work on your rig:

 

Screen Shot 2022-09-30 at 12.24.17 PM.png

Still doesn't work unfortunately. Maybe SSDT injection just sometimes does not work for PCIe AIC. For disabling GPU digital audio, OC device property injection does not work and only SSDT _DSM method works, which is exactly the opposite situation. Wonder what's the difference between these two methods?

Link to comment
Share on other sites

10 hours ago, Slice said:

But there are no miracles. If device property injection doesn't work then it is written wrong.

Well.. in what ways could it be wrong? The path is given by gfxutil, I really don't know what else of a device-id injection could go wrong (for OC property injection at least).

Link to comment
Share on other sites

On 10/1/2022 at 8:15 PM, i0ntempest said:

Well.. in what ways could it be wrong? The path is given by gfxutil, I really don't know what else of a device-id injection could go wrong (for OC property injection at least).

I can say something using crystal ball but I lost it.

Link to comment
Share on other sites

Anyways, I just installed a M.2 AQC113 card into my mini Hackintosh and the airplay problem is there as well. Looks like a common problem for these chips with ID 14c0. I guess worst case the next major OS upgrade should solve this.image.thumb.jpeg.f7d974c90739d054acc8b4a342051174.jpeg

Edited by i0ntempest
Link to comment
Share on other sites

  • 2 weeks later...
1 hour ago, DeadSinn said:

Hello everyone.

Has anyone managed ( or there is any way ) to get the AQtion AQC113CS work on Monterey?

Thanks!

See my adventure above. It works but isn’t perfect.

Link to comment
Share on other sites

  • 5 months later...
  • 2 months later...

Was posted in the wrong thread.

New Aquantia patch for macOS 14 beta 2 (23A5276g).

 

                <dict>
                    <key>Arch</key>
                    <string>Any</string>
                    <key>Base</key>
                    <string>__ZN27AppleEthernetAquantiaAqtion5startEP9IOService</string>
                    <key>Comment</key>
                    <string>Aquantia AQC107 patch 14.0 b2</string>
                    <key>Count</key>
                    <integer>0</integer>
                    <key>Enabled</key>
                    <true/>
                    <key>Find</key>
                    <data>D4RQAwAA</data>
                    <key>Identifier</key>
                    <string>com.apple.driver.AppleEthernetAquantiaAqtion</string>
                    <key>Limit</key>
                    <integer>0</integer>
                    <key>Mask</key>
                    <data></data>
                    <key>MaxKernel</key>
                    <string></string>
                    <key>MinKernel</key>
                    <string>23.0.0</string>
                    <key>Replace</key>
                    <data>Zg8fRAAA</data>
                    <key>ReplaceMask</key>
                    <data></data>
                    <key>Skip</key>
                    <integer>0</integer>
                </dict>

Maybe @PMheart can add it later to OC.

Edited by Shikumo
  • Like 3
  • Thanks 1
Link to comment
Share on other sites

16 hours ago, Shikumo said:

Was posted in the wrong thread.

New Aquantia patch for macOS 14 beta 2 (23A5276g).

 

                <dict>
                    <key>Arch</key>
                    <string>Any</string>
                    <key>Base</key>
                    <string>__ZN27AppleEthernetAquantiaAqtion5startEP9IOService</string>
                    <key>Comment</key>
                    <string>Aquantia AQC107 patch 14.0 b2</string>
                    <key>Count</key>
                    <integer>0</integer>
                    <key>Enabled</key>
                    <true/>
                    <key>Find</key>
                    <data>D4RQAwAA</data>
                    <key>Identifier</key>
                    <string>com.apple.driver.AppleEthernetAquantiaAqtion</string>
                    <key>Limit</key>
                    <integer>0</integer>
                    <key>Mask</key>
                    <data></data>
                    <key>MaxKernel</key>
                    <string></string>
                    <key>MinKernel</key>
                    <string>23.0.0</string>
                    <key>Replace</key>
                    <data>Zg8fRAAA</data>
                    <key>ReplaceMask</key>
                    <data></data>
                    <key>Skip</key>
                    <integer>0</integer>
                </dict>

Maybe @PMheart can add it later to OC.

EDIT: Wrong content. Removed.

Edited by PMheart
Wrong content
Link to comment
Share on other sites

Just checked again to be sure. On OC 0.9.3 with ForceAquantiaEthernet enabled and the patch disabled, no network connection. After enabling the new patch, network is working as expected. So something is different.

 

The same behavior was reported in this post.

 

PS: Here is the code in question. It looks like where before there was a `cmp` now is a `mov`. Meaning the checking of the second return value was put into a subroutine.

0xffffff80024feced 2145D4                 and        dword [rbp+var_2C], eax    ; CODE XREF=__ZN27AppleEthernetAquantiaAqtion5startEP9IOService+942
0xffffff80024fecf0 4C89F7                 mov        rdi, r14
0xffffff80024fecf3 E8F8A05DFE             call       0xffffff8000ad8df0
0xffffff80024fecf8 4989C4                 mov        r12, rax
0xffffff80024fecfb 498BBE28010000         mov        rdi, qword [r14+0x128]
0xffffff80024fed02 E8E9A05DFE             call       0xffffff8000ad8df0
0xffffff80024fed07 448B45CC               mov        r8d, dword [rbp+var_34]
0xffffff80024fed0b 448B4DD4               mov        r9d, dword [rbp+var_2C]
0xffffff80024fed0f 488D3DFF1F0200         lea        rdi, qword [a0x08llxSProvid] ; "[0x%08llx] %s() provider ID 0x%llx, supported %d, update %d\\n"
0xffffff80024fed16 488D156A1F0200         lea        rdx, qword [aStart]        ; "start"
0xffffff80024fed1d 4C89E6                 mov        rsi, r12
0xffffff80024fed20 4889C1                 mov        rcx, rax
0xffffff80024fed23 31C0                   xor        eax, eax
0xffffff80024fed25 E8A6E95CFE             call       0xffffff8000acd6d0

; patch area start
0xffffff80024fed2a 837DCC00               cmp        dword [rbp+var_34], 0x0
0xffffff80024fed2e 0F8450030000           je         loc_ffffff80024ff084

0xffffff80024fed34 448B65D4               mov        r12d, dword [rbp+var_2C]
; patch area end

0xffffff80024fed38 4C89F7                 mov        rdi, r14
0xffffff80024fed3b E8B0A05DFE             call       0xffffff8000ad8df0
0xffffff80024fed40 4585E4                 test       r12d, r12d
0xffffff80024fed43 0F8451010000           je         loc_ffffff80024fee9a

 

Edited by Shikumo
  • Like 1
Link to comment
Share on other sites

2 hours ago, PMheart said:

@Shikumo I am sorry for my disturbance. If possible, could you please confirm if with the latest version of OC 0.9.4 ForceAquantiaEthernet works again? You may get the binaries from https://github.com/acidanthera/OpenCorePkg/actions/runs/5370102701. Thanks in advance!

No problem and thank you for updating OC. I tested the latest 0.9.4 build successfully. My Aquantia is now working with only ForceAquantiaEthernet enabled.

  • Like 1
Link to comment
Share on other sites

31 minutes ago, miliuco said:

@surenmunoo

Latest 0.9.4 build? After "OcAppleKernelLib: FixForceAquantiaEthernet quirk on macOS 14 beta 2, thx @SHikumo" and "OcAppleKernelLib: FixForceAquantiaEthernet patch by @Mieze"?

I am on latest nightly build, if i disable the patch i can boot Sonoma, the quirk does not work for the Asus Z790 ProArt. 

  • Sad 2
Link to comment
Share on other sites

10 minutes ago, cankiulascmnfye said:

@surenmunoo If you are certain, the quirk is the cause for Sonoma not booting, I would create a boot log with the debug version of OpenCore and create a bug report

Silly question, is the boot log the debug systext file it creates when it dumps the ssdt's and dsdt's. 

Link to comment
Share on other sites

×
×
  • Create New...