Subject: Re: Removig generic optstr_get
To: Quentin Garnier <cube@cubidou.net>
From: Jachym Holecek <freza@dspfpga.com>
List: tech-kern
Date: 10/25/2006 17:26:06
# Daniel Carosone 2006-10-25:
> But more seriously, parsing a=b c=d strings handed to us by random
> platform bootloader or environment strings *into* plists in memory,
> for consumption by MI or almost-MI configuration code, might be a good
> idea.
... unless you need some of that information before you can internalize
the dictionary (internalize needs at least malloc(), which I think needs
uvm up and running, which needs to know the size and layout of physmem,
which may come from the bootloader ;-). Or did you mean that the (NetBSD)
bootloader should pass the info already internalized to early kernel
bootstrap?
(Just to point out there are cases where the scheme you're outlining
may not work well. Maybe it's not a problem for Multiboot code.)
-- Jachym