Subject: Re: 64-bit daddr_t problems with libsa
To: Luke Mewburn <email@example.com>
From: Rafal Boni <firstname.lastname@example.org>
Date: 01/30/2003 19:32:47
In message <20030130025815.GV747@mewburn.net>, you write:
-> Of course; you're correct for the platforms with a 2 stage boot.
-> (first stage boot block loads second stage /boot which parses
-> arguments from the BIOS/PROM or whatever which loads /netbsd).
-> This is a lot of NetBSD platforms, including alpha & pmax (not sure
-> about sgimips), so it won't be a problem there.
-> As an aside, are there any platforms which load the kernel directly
-> from the file system from the first stage boot block? (IMHO, the two
-> stage method is much better)
sgimips does, but since its first-stage booter goes in the volume header,
it can be as big as we want it ("standard" volhdr is ~ 2MB, I'm not sure
if it's scaled up for bigger disks but I'd assume "no"); in fact, I know
of someone out there who was working on adding all sorts of features to
the sgimips bootloader to make it look more like the x86 bootblocks 8-)
Rafal Boni email@example.com
We are all worms. But I do believe I am a glowworm. -- Winston Churchill