11:18 <eballetbo[m]> mmind00: Do you know what's the purpose of have this extcon here?

11:24 <eballetbo[m]> mmind00: ok, i'm stupid, fwiw I introduced this change

11:24 <eballetbo[m]> mmind00: I guess that's completely wrong, that's not used at all

12:11 <mmind00> eballetbo[m]: not sure if I'm missing context, but which extcon do you mean?

12:12 <eballetbo[m]> sorry, I didn't paste the line :-O

12:12 <eballetbo[m]> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi#n635

12:13 <eballetbo[m]> that one. It's wrong in mainline, is copied from the chromeos kernel, that was my mistake

12:14 <eballetbo[m]> well, not wrong, just that doesn't have any effect

12:14 <mmind00> eballetbo[m]: I remember us introducing the default for host-mode if the extcon is not present ... but I think that is the cros-ec showing the cable state?

12:15 <mmind00> eballetbo[m]: I remember at least the dp-state working through that ... although I'm not sure it does right now

12:16 <mmind00> eballetbo[m]: of course, the whole thing should instead go through the type-c framework of the kernel ... with the cros-ec-typec-thingy also being part of that

12:16 <mmind00> eballetbo[m]: because right now, all the devices using fusb302 controllers cannot role-switch at all

12:21 <eballetbo[m]> the fallback to host-mode was in the rockchip-typec not the dwc3 controller

12:25 <eballetbo[m]> what I'm try to make otg working on the kevin, for that the dwc3 controller needs the cable state from the extcon, but this needs to be defined in the usbdrd_dwc3_0 node, not usbdrd3_0. ChromeOS kernel uses a different glue-layer that needs an extcon to switch the role but this is not the case in mainline.

12:29 <mmind00> eballetbo[m]: ah ok

12:30 <mmind00> eballetbo[m]: so both typec and dwc3 need to be property setup for otg mode?

12:31 <eballetbo[m]> both need know the cable state, and also the usb-phy