Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 18 Sep 2000 08:17:23 -0700 (PDT)
From:      Matthew Jacob <mjacob@feral.com>
To:        "Daniel C. Sobral" <dcs@newsguy.com>
Cc:        Peter Wemm <peter@netplex.com.au>, alpha@FreeBSD.ORG
Subject:   Re: Loader badly broken on the alpha
Message-ID:  <Pine.LNX.4.21.0009180814490.832-100000@zeppo.feral.com>
In-Reply-To: <39C6311E.19B7337B@newsguy.com>

next in thread | previous in thread | raw e-mail | index | archive | help

Whoa! Take it easy...

Fine insofar as it goes, and, err, umm, I wasn't asking you to pull the 3KB
out of the FICL code. I was trying to squeeze it elsewhere coz I *like* the
FICL code.

There are some fun && games for alpha boot space and how FreeBSD uses it which
makes this dance somewhat interesting.

But see later mail- we've solved the problem for now. A bit more cleverness
will extend things even further. I doubt that we'll need to require any
scrunch from FICL, which, as you correctly are hinting it, is the queen bee of
the boot code which everything else should bust butt to serve.

-matt


> Matthew Jacob wrote:
> > 
> > We need to pull out 3408 bytes- but it's not that simple- bcopy seems to take
> > 2544 bytes, which is ridiculously large for a boot block- so I replaced it
> > with a dopey one line C function but that didn't change _end for me, so I have
> > to look at the linker magic some I guess.
> 
> Well, here is my position on this. No, we need not pull out 3408 bytes.
> While the loader must not bloat, the whole point of having boot[12]
> _and_ loader is allowing us to add extra features and intelligence to
> loader without having to worry about how many bytes a function is
> taking.
> 
> boot[12] _must_ be capable of providing a decent environment for loader.
> That's it's main function, nowadays, no matter what bde wants or says.
> 
> For the short term, one area of investigation that might provide us a
> quick respite is compressing softwords.c on FICL. This file contains a
> huge string of Forth commands that is interpreted when FICL is
> initialized. Since we do have gz uncompression on libstand, it should
> probably be easy to change the build process as to make that string a
> compressed version of what it is right now (softwords.c is generated by
> softwords.awk -- there's also a softwords.pl, if whoever does it is more
> comfortable with perl), and then uncompress it before interpreting it
> during FICL initialization.
> 
> Alas, I'm not doing it, I'm not even looking into it. I have absolutely
> _no_ time to do it. Sorry.
> 
> 



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0009180814490.832-100000>