I'll tell you how I did it. I applied the changes of my merge request in the tag 22.03.2 and I downloaded the .config from the OpenWrt website to run the make:
# First apply changes from merge request
./scripts/feeds update -a
./scripts/feeds install -a
wget 'https://downloads.openwrt.org/releases/22.03.2/targets/ramips/mt7621/config.buildinfo' -O .config
make menuconfig # Here choose target Cudy v2
Once I have the image, when I flash it on the router I check the NOT keep settings option (this is the most important thing).
Once this is done when finished it is accessible via IP 192.168.1.1 only by wired network on any of the four LAN ports (wifi is disabled by default as it is not configured).
The config from openwrt website made the difference and worked like a charm. Thank you very much!
The image works as described. Only the 5G LED doesn't work but WiFi works. Hope that the code will be merged soon. My small Linux box took 4 hours to compile everything
Which dts are you referring to? Here we are discussing about v2 of the router. In the thread you link it seems that something about v1 is being discussed which is completely outside the scope of this discussion.
Not necessarily. Cudy might (intentionally or accidentally¹) have improved the SPI circuit in v2 so that it can deal with higher frequencies. At least I haven't heared of any SPI issues with 80 MHz on v2 yet.
That said, I'd opt for reducing it to 40 MHz as OpenWrt's default on v1 and v2 for now to avoid stability issues, and ask for volunteers to test it at 80 on v2 and report their results.
¹ I think accidentally would be more likely, given that they still use only 10 MHz in their own images.
Yes, but they use only 10 MHz, which is most likely much lower than the circuits in both versions are actually capable of. It looks like vendors tend to be overcautious here, so it does make sense that OpenWrt uses higher values, but 80 MHz seem to be too much at least for v1.
Reducing the value in the dts is trivial, but I don't see the need to do it beforehand in v2 because there are v1 routers that need to have that value reduced. So far no one has reported failures in v2 with the value at 80. In fact it is likely that having to reduce the value in v1 is due to the USB port, which v2 does not have.
In conclusion. Let's not get ahead of ourselves. If someone with v2 reports errors we will check to see if it is necessary to reduce the value. In the meantime, reducing it because there are bugs with v1 does not seem logical to me.
OpenWrt 22.03.3 was released today. Unfortunately it seems that WR1300-v2 didn't make it into the release. Does anybody know why it is not part of 22.03.3? I cannot find a precompiled image in the OpenWrt repository.
I would guess (and hope) that a patch that adds support for a new revision with only minor changes of an already supported board would also be accepted in the 22.03 branch, but somebody (ideally somebody with access to the hardware) would have to prepare and submit it.