Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 1 Jun 1998 09:11:41 +0100 (BST)
From:      Doug Rabson <dfr@nlsystems.com>
To:        Jason Thorpe <thorpej@nas.nasa.gov>
Cc:        Curt Sampson <cjs@portal.ca>, Mike Smith <mike@smith.net.au>, hackers@FreeBSD.ORG
Subject:   Re: AlphaBIOS documentation 
Message-ID:  <Pine.BSF.3.95q.980601091011.330O-100000@herring.nlsystems.com>
In-Reply-To: <199806010743.AAA20008@lestat.nas.nasa.gov>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 1 Jun 1998, Jason Thorpe wrote:

> On Mon, 1 Jun 1998 00:24:23 -0700 (PDT) 
>  Curt Sampson <cjs@portal.ca> wrote:
> 
>  > There's plenty of device driver source code in Linux and NetBSD, too. :-)
>  > 
>  > My point is, every time a new type of Alpha comes along, you have
>  > to write new PALcode and compile it in to the boot loader. This is
>  > not really all that different than having to write and compile in
>  > a new device driver every time a new device comes along.
> 
> Not really.  Once you have working PALcode for a given platform, you never
> have to touch it again.  This is not true with drivers, which may need
> modification for the next rev of a device.
> 
> Also, the VAST majority of PALcode is common from one platform to another.
> Usually, the only differences are in the interrupt routing code, and these
> differences are usually very minor (esp. between same implementation revs,
> e.g. two 21064-based systems or two 21164-based systems).  The other main
> difference is in the code that builds the CPU logout area; this code usually
> copies platform-specific core logic registers.  However, this will be the
> same between two platforms which use the same core logic chipset (e.g.
> the AlphaStation 500 and the AlphaPC 164, both of which use an ALCOR2).
> 
> However, it is worth noting that the MILO PALcode doesn't actually work
> very well, and doesn't implement an interface completely compatible with
> the OSF/1 PALcode shipped with the SRM (logout area is different, for
> example, and some exception vectors are different).
> 
> In other words, I wouldn't exactly recommend using the MILO PALcode as a base.

Point taken.  I am going to try to buy a copy of the EBSDK.  Maybe that
will have more up-to-date code in it.  I still think that DEC should put a
copy of the OSF palcode into AlphaBIOS. 

--
Doug Rabson				Mail:  dfr@nlsystems.com
Nonlinear Systems Ltd.			Phone: +44 181 951 1891
					Fax:   +44 181 381 1039


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" 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.3.95q.980601091011.330O-100000>