Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 9 Sep 2001 11:12:38 -0700
From:      Marcel Moolenaar <marcel@xcllnt.net>
To:        Doug Rabson <dfr@nlsystems.com>
Cc:        ia64@freebsd.org
Subject:   Re: Ski port
Message-ID:  <20010909111238.A575@dhcp01.pn.xcllnt.net>
In-Reply-To: <20010909184220.K406-100000@salmon.nlsystems.com>
References:  <20010909101700.A531@athlon.pn.xcllnt.net> <20010909184220.K406-100000@salmon.nlsystems.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Sep 09, 2001 at 06:44:20PM +0100, Doug Rabson wrote:
> On Sun, 9 Sep 2001, Marcel Moolenaar wrote:
> 
> > On Sun, Sep 09, 2001 at 05:44:02PM +0100, Doug Rabson wrote:
> > > On Sun, 9 Sep 2001, Marcel Moolenaar wrote:
> > >
> > > > We're currently not really interested in userland code, so we're
> > > > going to run ski in "raw" mode anyway. I should be able to hack
> > > > something together that allows us to work on a real EFI bootloader.
> > > > With a combination of fake and real ACPI entries we should be able
> > > > to boot a kernel as if it was running on real hardware, right?
> > > >
> > >
> > > That would probably be pretty useful. It sounds like a *lot* of work
> > > though. EFI is pretty extensive.
> >
> > Yes, it would be a lot of work. And I can imagine that the ski owners
> > are not going to keep the feature if it wasn't done right. I think
> > it's good to have. I just want to make sure I spend my time well and
> > not waste it on a feature that will never complete and/or won't be
> > used.
> 
> Before you start, I suggest you pick up a copy of the EFI sample
> implementation source and have a look at that. Actually, having that
> source around is pretty useful for clarifying bits of the EFI spec.

I got it. The license allows its use for EFI emulation, so maybe we're
lucky here. I'll get in touch with Mike (ski owner) and David (Linux
porter) and see what they think.

> >
> > Another thought struck me. As soon as we have our own toolchain, we
> > probably will use DBX as the debugging format and not dwarf. I'd
> > better make sure that ski supports DBX if we want it to be as useful
> > then as it is now.
> 
> I have no idea what format is considered best these days. I thought that
> dwarf2 was the direction things were moving in though.

Yes, I think it is. If we're going to stick with dwarf, now is probably
the best time to decide (with now I mean without having any ia64
releases and backward compatibility problems; and me not having coded
DBX support in ski of course :-).

I think we should stick with dwarf on ia64. It gives us the most freedom
to experiment with different compilers. I don't want to get stuck with
gcc while SGI has a compiler that works for us and produces much better
code.

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel@xcllnt.net

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ia64" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010909111238.A575>