WAN performance on nanopi r2s

I've recently upgraded to this device to improve my WAN performance. But with the latest official build (21.02.3), i can only seem to be able hit around ~750-780Mbit. Completly stock image with some port fowards. I've enabled software offloading and tried packet steering with no luck.

Are there any tweaks i should do to boost that performance? According to this, this device should have no issues saturating 1Gbit WAN.

Run htop on the console, while pushing it, see what happens to the CPU load.

This is my load when downloading a single big file. To me, there doesn't seem to be any obvious bottlenecks. For some reason, i can't seem to install htop from official repos.

So did some more testing. Looks like i'm indeed cpu bottlenecking. I was able to get htop working, which made it evident, that core #0 is getting trashed to almost 100% and core #2 is also around 80%.

So what is there to do?

I've done some further testing and my conclusion is this; official openwrt images seem to be clocking the cpu too low to hit those WAN speeds. Forked version doesn't have those issues (you can clock it to ~1,6GHz).
Additionally, eth1 disconnecting issue also seems to be gone or mitigated very well with above build.