3 ms·
Would the perf cost of such an abstraction have been acceptable on early 386 hardware? I lean towards no but that is just a feeling without data…
by bpye 5y ago
Would the perf cost of such an abstraction have been acceptable on early 386 hardware? I lean towards no but that is just a feeling without data…
- anyfoo 5y agoYeah, it sounds like KMag essentially wanted the BIOS to be a sort of microkernel, with other kernels being implemented under it. Beyond performance concerns, given the PC's history I'm also not sure whether e.g. Linux developers (which initially was, well, Linus) would have been happy with whatever microkernel and API had been given there.