Performance is basically the same (in microbenchmarks), they went as far as preserving the use of red black trees for an apples to apples comparison, but it’s going to improve security as binder runs inside every process.
That means binder is going to join Asahi graphics, the Android Bluetooth stack, and puzzlefs in the serious drivers written in Rust club.
What else would it stand for?
It’s instructions per clock/cycle in a hardware context, because you can’t use clock speeds to compare performance between processors.
Instructions per cycle
Was what I first read anyways