Multifail with multiboot
Sigh. How many things must a poor soul learn in order to boot some damn operating system code on a computer?
I originally planned to only support x86_64. My current plan is to support anything but x86 (32 bit). That's the painful part, where you have a machine with 8gb of memory but only 4gb adressable, have fun, go wild. It makes sense that I keep arm32 and arm64 in mind, and whatever new and cool stuff that the future holds.
But for now, multiboot1 and x86_64. No need for grand plans of future multi-awesomeness, I need to get something up and running. The original EFI plan failed, I just couldn't handle the feeling of tearing down a lot of crap I didn't want there in the first place. Seems like a much better rite of passage to do proper multiboot and have my OS talk assembly with the CPU. I used to think that talking assembly seemed dirty and old school. But I've learned that assembly is basically the API of a CPU, so it makes a whole lot of sense to use assembly when making an operating system, whose job it is to talk to the CPU.
I'm also glad I blogged about my thoughs and plans. If I hadn't, I've probably given up a long time ago. But I want to show the world that I'm not completely useless, so until I have at least attempted to do one of the interesting things I have in my grand plan, I'll probably keep on going, albeit with a slow pace. Andante os development. Some might even say grave.
Sidenote: Learned about ARM's boot process. No BIOS. A commonly used boot loader is Das U-Boot. I c wat u did thar.
Questions or comments?