mikeb mikeb has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/30749 )
Change subject: sb/amd/agesa/hudson/Kconfig: Disable XHCI by default, enable only with firmware ......................................................................
Patch Set 4:
- and also add the correct xHCI firmware (e.g. available at https://github.com/informer2016/shared_devfiles/blob/master/0.10.0_1022_7814... ) which has been extracted using Felix Held's fch_xhci_rom_dumper - https://github.com/felixheld/fch_xhci_rom_dumper - from G505S while it has been running the proprietary UEFI.
Huh, fch_xhci_rom_dumper is completely unrelated and not needed here, since it dumps the rom part of the firmware and not the runtime-loaded part. The right XHCI blob for bolton is in the blobs repo.
Sorry I misunderstood him, usb 3.0 worked for Ivan at g505s with this bolton/xhci.bin from the blobs repo, and this another blob (obtained with your fch_xhci_rom_dumper) wasn't tested by him at all! So I just tested it and (maybe as expected because you said it is wrong blob) my laptop didn't boot at all - black screen :P Then I tried building coreboot with bolton/xhci.bin and now it boots, however I see the following messages at Linux kernel log:
xhci_hcd 0000:00:10.0: xHCI Host Controller xhci_hcd 0000:00:10.0: new USB bus registered, assigned bus number 8 xhci_hcd 0000:00:10.0: can't setup: -110 xhci_hcd 0000:00:10.0: USB bus 8 deregistered xhci_hcd 0000:00:10.0: init 0000:00:10.0 fail, -110 xhci_hcd: probe of 0000:00:10.0 failed with error -110
So those "blue" USB ports are working but maybe at 2.0 mode only, don't know why I can't reproduce the successful results. Tested with Linux kernel 4.15.0 if it matters