GRUB vs LILO and how STUPID the X86 architecture really is...

Michael Wu edf_iceburner at SPAMKILLSCATS_yahoo.com
Wed Oct 16 00:58:00 PDT 2002


It's possible to boot linux from linux using the two kernel monte, but that 
doesn't seem to be maintained. Essentially, it shuts the current kernel 
down, switches to real mode, and boots the next one.

Though I still prefer grub, because LILO is IMHO, too crappy. If there was 
a decent minimal bootloader, I'd use it, but I don't want stuff between 
dding the kernel to the drive (which I've successfully used to boot, while 
screwing up my partition table) or a minimal OS, grub. Someone needs to 
come up with a safer way of starting linux directly without using dd, and 
with some commandline options. (okay, probably impossible. I'll stick to 
grub then)

Justin Hibbits wrote:

> On Mon, Oct 14, 2002 at 10:01:18PM +0100, Ian Molton wrote:
>> On Mon, 14 Oct 2002 16:54:06 -0400
>> Zack Winkles <sativa93 at bellsouth.net> wrote:
>> 
>> > 
>> > Actually that's wrong. The reason operating systems like Windows 2000
>> > and XP and such can't boot linux is because they have already enabled
>> > protected mode on the processor, but the 16-bit boot code in the linux
>> > 
>> > kernel requires that it be started in real mode. This could be
>> > attributed to yet another stupidity in the x86 processor line, but it
>> > prevents booting of some operating systems from others, hence
>> > disproving that statement.
>> 
>> Nothing to stop you re-entering real mode. (not that I recall though).
>> 
>> You'd probably need to be a driver to do it, but it must be possible,
>> because LINUX can do it.
>> 
> 
> How can Linux do it?  All I know how it can do it is to enter vm86.  Once
> you enter 'real' real mode, you lose everything that linux gives you, and
> hence
> you're on your own then.  IIRC, linux only does it when it shuts down.
> 
> But, as always, I could be wrong :)
> 
> 

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list