iMac Pro X99 - Live the Future now with macOS 10.14 Mojave [Successful Build/Extended Guide]

[kgp]

All right thanks, that I'm doing now

But note that in fact this not any solution..

We need to make the method _ini call work at some point!

 

[nmano]

I update QINI
Still DSB0 not work with out c cable

 

[kgp]

I update QINI
Still DSB0 not work with out c cable

But are you now with SSDT-9.aml or SSDT-9-minimal hotplug injector, where essentially most of the conflicting SSDT-9.aml stuff has been removed?

SSDT-9.aml does not work at all! No sense to remove THB_C cable or to check sleep/wake funtionality at present.. it does not even work with THB-C cable connected.. entire TB Local Node and a lot of TB properties missing! hotplug should be avaiable as soon the TB node is installed, all TB properties are correctly implemented and all TB kexts are fully loaded.. i don‘t think it will be necessary to remove the THB_C at all..

Chipset of Alpine Ridge is the same as used for iMac Pro.. I also do not understand why you changed now to TBEX 3..

In my opinion, you are just wasting time.. and method _ini cannot be renamed.. it must be _ini in order to do what it should do.. it is not a merhod definition but rather a method execution!!! Also regarding the other two positions when implementing OSDW..

The simple renaming by quini will never work! There is no method quini to be executed for defining hardware properties!

 

[nmano]

But are you now with SSDT-9.aml or SSDT-9-minimal hotplug injector, where essentially most of the conflicting SSDT-9.aml stuff has been removed?

SSDT-9.aml does not work at all! No sense to remove THB_C cable or to check sleep/wake funtionality at present.. it does not even work with THB-C cable connected.. entire TB Local Node and a lot of TB properties missing! hotplug should be avaiable as soon the TB node is installed, all TB properties are correctly implemented and all TB kexts are fully loaded..

In my opinion, you are just wasting time.. and method _ini cannot be renamed.. it must be _ini in order to do what it should do.. it is not a merhod definition but rather a method execution!!! Also regarding the other two positions when implementing OSDW..

The simple renaming by quini will never work! There is no method quini to be executed for defining hardware properties!

If you want _ini
You have to rename DSDT
Clover config acpi patch
_INI to QINI
Find
5f494e49
Replace
51494e49
After this you can add _INI

I remove all device IDs

 

[kgp]

If you want _ini
You have to rename DSDT
Clover config api patch
_INI to QINI
Find
5f494e49
Replace
51494e49
After this you can add _INI

I guess you don‘t understand my friend..

_ini is a predefined method to define and change hardware properties! All we do is executing _ini with a set of predefined parameters, which apparently does not work...

You cannot use for this a method different from method _ini..

True.. the code at least loads when removing the _ini call.. but it absolutely does nothing at this place in this case!!

The same states for your OSDW and GPWR approach, which does not work and will always return zero.. also in this case the code skips everything that requires OSDW to be 1 when loaded and there are no errors, as the code does absolutly nothing at these instances..

Check! You will even miss else implemented TB properties like PCIHotplugCapable, PCI-Thunderbold, etc... because it skips all parts of the code which require OSDW to be one!!!!

If (OSDW ())
                {

There are no errors because the code simply does nothing in this case!

Certainly, there are also instances where

If (LNot (OSDW ()))
                {

Finally the screenshot that you attached above just shows another call of method _ini, but is not the definition of method _ini!

By the way, with respect to OSDW, method _ini is called at a second place!

These are not method _ini definitions but method _ini executions!:

i.e. hardware/software definitions.. you cannot replace these calls by calls of another undefined method!

Do you understand???

Try to Google if you do not believe..

 

[SmariMo]

Enable above 4G decoding in mainboard BIOS..

Also use most recent clover distribution with most recent AptioMemoryfix-64.efi.

I tried this, but it didn‘t work
Any other ideas?

Thanks in advance

 

[Loloflatsix]

@Kgp
@nmano

Work in progress : I compliled the SSDT-9-Lolo-dump.aml.zip for my Gigabyte X299-UD4

And I was able to load this SSDT at booting but with ACPI error messages.

The result in my IOreg is half-tone :

As you can see :
- RP21, UPSP and DBS0 is about OK
- NHI0 is not finished due to missing entries in IOthunderbolt Controller (which I expected ..)
- DBS2/ XHC2 : here is the real problem : while hot plug is working for my screen connected with USB_C I lost entries I did have before with basic SSDT : I can't see if a device is connected on USB_C and more wrong if I connect an USB plug on the USB 3,1 I does not work...

Here is attached my SSDT9

To be continued

 

[kgp]

View attachment 346248 View attachment 346249 View attachment 346250 View attachment 346251 @Kgp
@nmano

Work in progress : I compliled the SSDT-9-Lolo-dump.aml.zip for my Gigabyte X299-UD4

And I was able to load this SSDT at booting but with ACPI error messages.

The result in my IOreg is is half-tone :

As you can see :
- RP21, UPSPand DBS0 is about OK
- NHI0 is not finished due to missing entries in IOthunderbolt Controller (which I expected ..)
- DBS2/ XHC2 : here is the real problem : while hot plug is working for my screen connected with USB_C I lost entries I did have before with basic SSDT : I can see if a device is connected on USB_C and more wrong if I connect an USB plug on the USB 3,1 I does not work...

As expected.. method _ini implementation is missing.. and a proper implementation of OSDW as well..

SSDT9.aml loads but fails with most implementations.

Honestly no idea what all your tests are for.

Anyway.. don‘t forget that we are aming here at a full TB implementation, while the hot plug workaround we achieved previously works anyway..

We don‘t want another TB workaround, we aim at full TB implementation.

Hope I can count on you in this endevour..

It won‘t be easy though...

Good night..

KGP

 

[Loloflatsix]

As expected.. method _ini implementation is missing.. and a proper implementation of OSDW as well..

SSDT9.aml loads but fails with most implementations.

Honestly no idea what all your tests are for.

Anyway.. don‘t forget that we are aming here at a full TB implementation, while the hot plug workaround we achieved previously works anyway..

We don‘t want another TB workaround, we aim at full TB implementation.

Hope I can count on you in this endevour..

It won‘t be easy though...

Good night..

KGP

I totally agree with you: we want a full implementation of TB and even if the solution with the hot plug workaround we used previously works anyway , I really think we can achieve the full implementation in continuing to work.
Good night too.

 

[Loloflatsix]

Work in progress:
- I recovered my previous missing entries on SSP1 and SSP2
- Hot-plug works on SSP1 screen disconnected and reconnected on USB_C
- Hot-plug works on SSP2 for example with an USB on USB 3.1



So persists:
- missing entries in IOthunderbolt Controller
- method _ini problem even if I rename all _ini entries by qini