Jonas Jelonek

realtek target support for 6.18 is being worked on: #21181

seems like those can be removed too: - hack-6.18/722-net-phy-aquantia-enable-AQR112-and-AQR412.patch - hack-6.18/723-net-phy-aquantia-fix-system-side-protocol-mi.patch - hack-6.18/725-net-phy-aquantia-add-PHY_IDs-for-AQR112-variants.patch the last one didn't compile for me and it seems like upstream now has support for these...

No deal-breaking objections from my side. Thanks for the work! A quick Google search shows that at least RTL9602C and RTL9603 seem to exist, being also SoCs used in PON...

Well then nevermind, I wasn't aware of that. Their naming policy drives me crazy. If it's really that bad, then just keep rtl9607c. I'm just curious, could you share some...

Thanks for the list. Since you use a real device, I would suggest to have that named specific to your router then, not `engboard`. EDIT: and while at it, maybe...

There should be more opinions from others, maybe I'm just wrong but ... In the current state this suggests that you actually have such an engineering board and ran OpenWrt...

Thanks for keeping up the work and steady progress on this. I join plappermauls opinion on the required minimum baseline of having at least (some) networking running on the device...

I did the rewrite of the RTL93XX I2C driver but haven't looked into how RTL9607C works regarding I2C. Nice that you designed the driver like the one for RTL93XX. But...

Thanks for that. I'm fine with the separate driver for now. Might be a future discussion whether it should stay separate or be included in a big, one-fits-all I2C driver.

> Unbelievable how less work is left in contrast to 6.6 and 6.12. Those credits go to you :) > Could you cherry-pick Haukes revert-patch for the time being? Verified...