Sander Vanheule

Instead of merging the OpenWrt master branch, you should rebase onto it. That will give conflicts when you try to push to your github mirror, but you just have to...

> note GitHub add code function is really bad and it doesn't see line feeds!? See Github's [Mardown docs](https://docs.github.com/en/get-started/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax#quoting-code) on how to provide blocks of code. (You need to use...

> Yes it has the same PoE controllers as the -10 where the lua-rs232 actually came from. > > How much hardware info is really meaningful to have in the...

The "Test Formalities" have failed because you have "`flygarn12 `" as the commit author. Please update this (`git commit --amend --reset-author`).

@flygarn12 I'm going to reduce the noise in my inbox a bit and unsubscribe here. Please ping me when your patches are ready for review. Thanks in any case for...

Aha, so those LEDs are an indication of the current mode of the port LEDs. Thanks for clarifying. I still think the `_poe.dtsi` should be included with the first device...

What you need to compare against, is a `MIPS_GENERIC` based platform. The `MIPS_GENERIC` setup code will use the DT info to fetch the clock, and to initialise `mips_hpt_frequency` in a...

Right, `CLK_OF_DECLARE()` is the right thing to do, and is not a platform driver (just like on MIPS ingenic). Perhaps I need more coffee... Still, we should be working towards...

Merged the platform init patch ahead of the other changes, since that's useful to have in any case. 396e190f0be7 realtek: more generic platform initialization

On RTL8381M (GS110TPP v1) my results are a bit strange (gzip a 10MB file of random data): ``` @ 500 MHz root@OpenWrt:/sys/devices/system/cpu/cpufreq/policy0# time cat /tmp/null.bin | gzip > /dev/null real...