Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 6 Nov 2000 10:14:06 +0000 (GMT)
From:      Doug Rabson <dfr@nlsystems.com>
To:        Peter Jeremy <peter.jeremy@alcatel.com.au>
Cc:        Mike Smith <msmith@FreeBSD.ORG>, alpha@FreeBSD.ORG
Subject:   Re: cvs commit: src/sys/boot/alpha/common main.c src/sys/boot/al
Message-ID:  <Pine.BSF.4.21.0011061012270.256-100000@salmon.nlsystems.com>
In-Reply-To: <00Nov6.091801est.115213@border.alcanet.com.au>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 6 Nov 2000, Peter Jeremy wrote:

> On Fri, Oct 27, 2000 at 10:28:38AM +0100, Doug Rabson wrote:
> >On Thu, 26 Oct 2000, Mike Smith wrote:
> >> Forgive the silly question, but is there any reason that boot1 doesn't 
> >> read the load address from the loader object?
> >
> >On alpha, /boot/loader is just a binary (the headers have been stripped
> >off). When I wrote it, I didn't think I would be able to squeeze an ELF
> >loader into the space available. Currently there are 280 bytes spare -
> >probably not enough.
> 
> Any reason why you couldn't have a primitive header: eg magic number
> and start address in /boot/loader.  The code to read this and
> appropriately relocate /boot/loader shouldn't be too big.  I agree that
> it's undesirable to wind up with yet another incompatible object
> format, but having an unbootable machine because /boot/loader expects
> a different load address to boot1 is even less desirable.

That would also involve changing the ABI between boot1 and loader. The
problem with that is that loader is updated on installworld but boot1
isn't so its easy to create a mismatch. The load address of loader is
essentially fixed.

-- 
Doug Rabson				Mail:  dfr@nlsystems.com
					Phone: +44 20 8348 6160




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.BSF.4.21.0011061012270.256-100000>