Binary compatibility is bare minimum for any cpu project. Riscv as the new kid on town, and after basically promising a CPU revolution, needs to achieve far more than that.
ARM is currently moving towards a standard and open boot system, and making previously optional things mandatory.
> but you made it yours by tweaking it to your liking.
Serious question: why do you assume that?
My "tweaking" of a new Ubuntu machine is normally limited to changing the background. I am busy getting work done and don't really want or need to tweak things.
At a conference years ago, someone proposed the idea of using this as the _only_ security mechanism in really tiny embedded systems. The implementation was 100-200 bytes in total.
If L3 really is 4 cycles slower, how much does it _actually_ affect performance? Specially given that L3 can now be made much larger.