[Guide] 10.11+ USB changes and solutions

[RehabMan]

The only other cmp I could find against 0x0Fh.

That would make this the second patch:

Find: 4183fd0f 753a
Replace: 4183fd0f 9090
Name: com.apple.driver.usb.AppleUSBXHCI
MatchOS: 10.14.x

It doesn't look very promising though.
View attachment 361782

Test for equality not likely part of the problem.

There are other kext binaries that may have additional checks (ex. AppleUSBXHCIPCI).

 

[headkaze]

Test for equality not likely part of the problem.

There are other kext binaries that may have additional checks (ex. AppleUSBXHCIPCI).

Yeah it's already confirmed as not working. I had a look in AppleUSBXHCIPCI and the only obvious patch I could find would be:

Find: 83f90f0f 87bb0000 00
Replace: 83f90f90 90909090 90
Name: com.apple.driver.usb.AppleUSBXHCIPCI
MatchOS: 10.14.x

If that doesn't work someone needs to have a better look but even if we do find a working patch the fact we need more than one does make it a very undesirable issue going forward.

 

[headkaze]

To discover all ports, you can simply do them in groups... for example, do all HSxx in one pass, all SSxx ports in second pass, being careful to keep at least the HSxx port where your mouse/keyboard are attached.

As I understand it your solution is to add boot flags to USBInjectAll that will allow the user to enable certain groups of USB ports at a time. ie. HPxx, PRxx, HSxx, SSxx etc. and after each time you configure them you change the boot arg and configure the next group. It just sounds too complicated to me. I still think a port limit patch would be more desirable even if it means two of them.

 

[RehabMan]

Yeah it's already confirmed as not working. I had a look in AppleUSBXHCIPCI and the only obvious patch I could find would be:

Find: 83f90f0f 87bb0000 00
Replace: 83f90f90 90909090 90
Name: com.apple.driver.usb.AppleUSBXHCIPCI
MatchOS: 10.14.x

If that doesn't work someone needs to have a better look but even if we do find a working patch the fact we need more than one does make it a very undesirable issue going forward.
View attachment 361810

Not working.

 

[RehabMan]

As I understand it your solution is to add boot flags to USBInjectAll that will allow the user to enable certain groups of USB ports at a time. ie. HPxx, PRxx, HSxx, SSxx etc. and after each time you configure them you change the boot arg and configure the next group. It just sounds too complicated to me. I still think a port limit patch would be more desirable even if it means two of them.

Yes.
Hackintosh is hard work, not for the lazy. People should get used to it.

 

[headkaze]

Yes.
Hackintosh is hard work, not for the lazy. People should get used to it.

That's true. Don't get me wrong I still think your idea is worth implementing. It may even be the only practical way going forward.

 

[RehabMan]

That's true. Don't get me wrong I still think your idea is worth implementing. It may even be the only practical way going forward.

Already implemented (uia_include kernel flag in USBInjectAll.kext 0.7.0)

Guide updated:
https://www.tonymacx86.com/threads/guide-creating-a-custom-ssdt-for-usbinjectall-kext.211311/

Which means... the "port limit patch" is no longer part of my world.

 

[stiligFox]

Quick question; the 15 port limit only applies to ports on the machine, correct? I.E. I could have 15 ports on the machine and a 8 port hub connected to one of those ports (22 in all) and still be fine?

 

[RehabMan]

Quick question; the 15 port limit only applies to ports on the machine, correct? I.E. I could have 15 ports on the machine and a 8 port hub connected to one of those ports (22 in all) and still be fine?

Correct.
It is actually not a "ports on the machine" but rather "ports on each controller".
(port limit is per controller).

And the only controller you run into this problem is the xHCI controller as it is the only place where it is possible to go over 15 (some xHCI can have up to 26 ports, but Apple accommodates only 15).

 

[stiligFox]

Correct.
It is actually not a "ports on the machine" but rather "ports on each controller".
(port limit is per controller).

And the only controller you run into this problem is the xHCI controller as it is the only place where it is possible to go over 15 (some xHCI can have up to 26 ports, but Apple accommodates only 15).

Oh sweet, that makes much more sense now!

One last question, does the “Intel FB-patcher” tool help to do the discovery and SSDT creation part of this guide, or is it better to still follow this guide solely? I can’t find anything that says they do the same thing although they look like the same process...