Fwd: big.LITTLE status for rk3399/rockpro64?

[myfreeweb]

On July 28, 2020 3:43:19 PM UTC, Vincent Milum Jr <freebsd-arm at darkain.com> wrote:
>The issue is most certainly from the big.LITTLE setup. Talking with others
>via Twitter, we've determined the issue is the CAM system doesn't play
>nicely in this setup. Anything that touches CAM on a big.LITTLE setup has a
>chance (but not guarantee) of breaking the system.

Interesting..

>The downside is that the two big cores
>are not fully clocking up to their max speed, because FreeBSD doesn't have
>support yet to clock cores or groups of cores independently.

Back when I was actively experimenting with RK3399, I could use the sysctl to set maximum frequency allowed by dtb for the big cores (2200). It would *display* weirdly since heterogeneous systems are not properly supported, I think the readout of the sysctl was the max of the little cores, but all cores *were* at max clock, since the performance was there.