Making process spawning faster would allow for workloads to spawn "so many" processes/threads significantly more practically, and therefore more simply than use

Making process spawning faster would allow for workloads to spawn "so many" processes/threads significantly more practically, and therefore more simply than userspace async-style runtimes. (Note that threads on Linux are just processes which share memory and some other properties.) Merely halving it may not be all that's needed to totally replace those runtimes but it's a good step in the right direction.