[Guide] HackrNVMeFamily co-existence with IONVMeFamily using class-code spoof

[Lsucorey]

Attach EFI/Clover in use.

I got it working. Just booted from 850 EVO into Squid OS X RAID 10 Array (the mirror adds stability to the setup, and my write speed is still plenty fast for me). I'm somewhat disappointed my 16x PCIe socket seems to max out just a hair over 4 GB/sec though .. or it could be the max bandwidth of the Squid M.2 Carrier Board.

BACK ON TOPIC: I still do not understand why I needed the IONVMeFamily patches if I had your HackrNVMeFamily spoof installed properly; that was the whole point of your spoof, right? To avoid having to patch IONVMeFamily and streamline future updates? Do you think it is probable or possible that it is due to my SSDT-NVMe-Pcc.aml file being coded improperly?

 

[RehabMan]

I got it working. Just booted from 850 EVO into Squid OS X RAID 10 Array (the mirror adds stability to the setup, and my write speed is still plenty fast for me). I'm somewhat disappointed my 16x PCIe socket seems to max out just a hair over 4 GB/sec though .. or it could be the max bandwidth of the Squid M.2 Carrier Board.

BACK ON TOPIC: I still do not understand why I needed the IONVMeFamily patches if I had your HackrNVMeFamily spoof installed properly; that was the whole point of your spoof, right? To avoid having to patch IONVMeFamily and streamline future updates? Do you think it is probable or possible that it is due to my SSDT-NVMe-Pcc.aml file being coded improperly?

No need for IONVMeFamily patches if you have spoof implemented correctly.
Your previous ioreg shows class-code/ACPI identity not working. Hence request for EFI/Clover.

 

[Lsucorey]

No need for IONVMeFamily patches if you have spoof implemented correctly.
Your previous ioreg shows class-code/ACPI identity not working. Hence request for EFI/Clover.

Okay.

 

[RehabMan]

This EFI/Clover does not correspond to the ioreg you provided previously.

This EFI/Clover has two *.aml files in ACPI/patched:
SSDT-i7-6700K.aml
SSDT-NVMe-Pcc.aml

But your Clover bootlog (extracted from ioreg) shows only one loading:

17:029  0:000  Start: Processing Patched AML(s): Unsorted
17:029  0:000  Inserting SSDT-NVMe-Pcc.aml from EFI\CLOVER\ACPI\patched ... Success
17:033  0:003  End: Processing Patched AML(s)

Need files that are consistent with respect to each other...

 

[Lsucorey]

This EFI/Clover does not correspond to the ioreg you provided previously.

This EFI/Clover has two *.aml files in ACPI/patched:
SSDT-i7-6700K.aml
SSDT-NVMe-Pcc.aml

But your Clover bootlog (extracted from ioreg) shows only one loading:

17:029  0:000  Start: Processing Patched AML(s): Unsorted
17:029  0:000  Inserting SSDT-NVMe-Pcc.aml from EFI\CLOVER\ACPI\patched ... Success
17:033  0:003  End: Processing Patched AML(s)

Need files that are consistent with respect to each other...

Done. (See prior post by me)

 

[RehabMan]

I got it working. Just booted from 850 EVO into Squid OS X RAID 10 Array (the mirror adds stability to the setup, and my write speed is still plenty fast for me). I'm somewhat disappointed my 16x PCIe socket seems to max out just a hair over 4 GB/sec though .. or it could be the max bandwidth of the Squid M.2 Carrier Board.

BACK ON TOPIC: I still do not understand why I needed the IONVMeFamily patches if I had your HackrNVMeFamily spoof installed properly; that was the whole point of your spoof, right? To avoid having to patch IONVMeFamily and streamline future updates? Do you think it is probable or possible that it is due to my SSDT-NVMe-Pcc.aml file being coded improperly?

This clover & ioreg will be edited out.

Clover is doing strange things to your ACPI files.

Remove (or set false) config.plist/ACPI/Patches:
AddHDMI_8000000 (probably is causing strange HDAU showing up in PEGP)
AddPNLF_1000000 (nonsense for desktop)
FIX_ACST_4000000 (nonsense for desktop)
FIX_ADP1_800000 (nonsense for desktop)
FixRegions_10000000 (does not work correctly all the time, and you don't need it anyway)
DeleteUnused_400000 (probably not needed)

Reply with new EFI/Clover and ioreg.

 

[Lsucorey]

Clover is doing strange things to your ACPI files.

Remove (or set false) config.plist/ACPI/Patches:
AddHDMI_8000000 (probably is causing strange HDAU showing up in PEGP)
AddPNLF_1000000 (nonsense for desktop)
FIX_ACST_4000000 (nonsense for desktop)
FIX_ADP1_800000 (nonsense for desktop)
FixRegions_10000000 (does not work correctly all the time, and you don't need it anyway)
DeleteUnused_400000 (probably not needed)

Reply with new EFI/Clover and ioreg.

Will do. I will need a couple of hours to work on a non related but mandatory project, and will get back on it afterwards (I'll have a couple of hours until sleep comes calling). Thanks again for your help.

 

[RehabMan]

Will do. I will need a couple of hours to work on a non related but mandatory project, and will get back on it afterwards (I'll have a couple of hours until sleep comes calling). Thanks again for your help.

Note also that you need to change each "class-code" to "#class-code" as previously discussed.
We will not turn on real "class-code" injection until we are sure the injections are working correctly.

 

[Lsucorey]

Note also that you need to change each "class-code" to "#class-code" as previously discussed.
We will not turn on real "class-code" injection until we are sure the injections are working correctly.

Done. It boots now, but it broke my Nvidia Web Driver (even manually selected Nvidia Web Driver in Clover at boot up). Tried switching it from OS X driver to Nvidia via the Nvidia preference panel, but it defaulted back to OS X display driver after restart.

Note that I am currently using config0 (took out Pike's patches) and changed class-code to #class-code as requested.

Only natural that you go away as soon as everything broke . I was unable to restore my Nvidia Web Drivers unless I booted using the UniBeast USB.. & then I noticed that I had forgotten to log out of my initial 960..

 

[RehabMan]

Done. It boots now, but it broke my Nvidia Web Driver (even manually selected Nvidia Web Driver in Clover at boot up). Tried switching it from OS X driver to Nvidia via the Nvidia preference panel, but it defaulted back to OS X display driver after restart.

Note that I am currently using config0 (took out Pike's patches) and changed class-code to #class-code as requested.

Only natural that you go away as soon as everything broke . I was unable to restore my Nvidia Web Drivers unless I booted using the UniBeast USB.. & then I noticed that I had forgotten to log out of my initial 960..

Nvidia not related to NVMe. config.plist and config0.plist have more differences than just NVMe patches.

Note: Good to see my original proposed code works as expected.