Re: Kernels > 1M

H. Peter Anvin (hpa@transmeta.com)
Mon, 09 Aug 1999 18:03:26 -0700

Werner Almesberger wrote:
>
> H. Peter Anvin wrote:
> > Well, the thing is that if you don't have access to the BIOS anyhow,
> > then you're really down to the stuff you can do out of initrd anyhow
>
> Which is pretty much everything you need to retrieve and boot a Linux
> kernel ... Obviously, boot manager functionality is lacking here, so
> for this, you need to have LILO or such.
>
> > Either way, what I'm trying to accomplish with lbcon is different --
>
> I think it's a great idea to have a better execution and development
> environment than what I use for LILO (16 bit assembler is a pain).

Well, that part works now, actually. Feel free to steal anything you
want... it's GPL.

> I'm just sceptical about the code replication issues. Also, it seems
> that the good old days of file system heterogenity are coming back
> again, so FS-specific code will be tricky (assuming you read file
> systems without using the LILO approach of mapping them before use).
>
> Well, I'm busy with other things right now, but I hope to find time
> to do some hacking on a "boot next kernel" component in about a month.
> Then we'll see if it's a good or a bad idea.
>

I don't, although I'm trying to make it as easy as possible to port
filesystems from the Linux kernel, by keeping a VFS layer which is
reasonably similar to the one in Linux. My goal is to initially support
ext2, iso9660 and FAT. NTFS would be nice for NT users, of course; not
sure how practical that is though. I've considered using libext2 for
the ext2 part of the filesystem; either that or using the Linux 2.0.37
ext2; I have been using 2.0.37 as my main source of stolen code.

-hpa

-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/