shrieken213 Posted October 20, 2018 Share Posted October 20, 2018 I'm working on these port injections, and I'm confounded by what's happening with these port assignments. Before the injection, SS05 and SS06 corresponded to the front USB 3.0 ports. The two rear ASMedia 3.0 intermittently worked depending on which ports were occupied. Another PCIe add-on card (4 USB 3.0 ports) worked perfectly fine, picked up by GenericUSBXHCI just like the ASMedia chip. The DSDT was missing a lot of USB ports in the Clover raw export, so I filled those in appropriately. Port assignments between the ACPI device and the physical port didn't change. What did happen is that of the 6 devices I added (2 3.0/2.0 pairs and 2 2.0 ports), the 3.0 ports were dummy devices that didn't do anything. The 2.0 devices however were the empty 2.0 header on the mobo, which when I plugged a device into the header worked fine (i.e. not a dummy but a clearly omitted device). A consequence of this was that due to the dummy entries being recognized the system hit the 15-port limit and needed the delimit kext patch. Now with the DSDT fully filled in I made the USBInjectAll.kext SSDT for all 20 ports, dummy and all. What happened was extremely interesting. The two rear 3.0 ports from the chipset switched ACPI assignments from SS05 and SS06 to SS04 and SS05. SS06 was no longer recognized and does not show up on IOReg. The ASMedia chip is recognized under RP05, but does not function at all, unlike the add-on card which works without a hitch. And yes the delimit patch is still in place just in case. I don't use all the ports so it's not that big of a problem, but that the injection is messing with device assignment is really weird. Does anyone else have this kind of issue? Link to comment Share on other sites More sharing options...
Recommended Posts