On Wed, 2 Aug 2000, Gilbert Baron wrote:
> >> > I told you. Nobody ever asked before. It is the BIOS that handles this
> >> > anyhow is it not. Version 7/1 should not se anything that 7.0 did not.
> >> > This message is useless, it does not ell why it is refusing to
> >accept it.
> >
> >I think you will find that with Linux, the BIOS only handles the drive
> >until the relevant parts of the kernel have loaded. After that, Linux
> >handles the job itself.
> >
>
> This would be stupid in my opinion. There are too many drives. It is like no
> drivers would be written and it will all be in the Kernel..
> It is The BIOS that makes an API that is constant and that is the way I am
> sure it is done.
No, Ozz is correct. Linux does not rely on the BIOS for disk access - the
BIOS is only invoked during initial boot, just long enough to allow the
system to read the master boot record.
Windows relies on the BIOS for I/O -- Linux does not.
Ironically, this lets you use large drives in Linux with *very* old BIOS
and motherboards - Windows cannot make the same claim, as it relies
entirely on the BIOS. Yes, there are many drives -- but the BIOS still has
to be able to talk to them, right? How big is the BIOS? 512k? Not many
drivers would fit in there, I'd say.
That's why we have standardization.
If you still think this is a stupid way of doing it, please write a new
operating system and let me know when you are finished. I will be sure to
try it.
-Stephen-
