00:00 <sven> it looks like the controller isn’t completely reset (the old io queue still seems to exist)

00:01 <sven> but that stuff still shouldn’t be happening. It appears as if it never gets the irq for the first admin command. It’s strange it tries to identify before setting up the new io queue though

00:02 <sven> anyway, hope your holidays are enjoyable as well :)

12:52 <bluerise> sven: hiho. kettenis told me I can ask you about which part of dwc3 to reset? on the Mini I have here the 'empty' type-c port works fine, but the one I used for m1n1 as gadget can't be turned back into host mode

12:59 <jannau> bluerise: afaik it has to be reset completely. see https://lore.kernel.org/all/20211108170946.49689-2-sven@svenpeter.dev/ for reference

13:00 <bluerise> ouch :)

13:00 <bluerise> thanks!

13:00 <sven> that’s not even enough :(

13:00 <sven> you’ll also have to do a hard reset in the PHY

13:01 <sven> so essentially tear everything down, flip reset bit somewhere in the phy and then bring everything up again

13:01 <bluerise> which phy? the dwc3 usb2/usb3 stuff or some external chip?

13:01 <sven> the thing apple calls “ATC PHY”

13:01 <bluerise> ok

13:05 <sven> 0x382A8401C bit 0 and 4 I think

13:05 <sven> If you flip those everything in dwc3 will reset to its defaults and the port will work again

18:05 <marcan> jannau: looks like I have to update my AP (which I haven't since the UniFi mess) and create a WPA3 network.. :-)

18:05 <marcan> jannau: j314c issue was a regression from the T2 stuff; pushed take4 to fix that

18:19 <jannau> already seen and works now. same wpa3 problem

18:30 <jannau> macos can act as a wpa3 personal access point