In theory, it would simplify concurrency primitives in the kernel space, especially on the main thread, which is something I’ve notice is not an area that a lot of kernels explore.