Content-Type: text/plain;
  charset="utf-8"
From: Akon <akon@gmx.net>
To: linux-kernel@vger.kernel.org
Subject: Double-Harvard Architectures
Date: Fri, 15 Aug 2003 02:26:36 +0200
User-Agent: KMail/1.4.3
References: <000601c362bf$7eadc9a0$62a14943@joe> <200308142009.53875.admin@kentonet.net>
In-Reply-To: <200308142009.53875.admin@kentonet.net>
MIME-Version: 1.0
Content-Transfer-Encoding: 8BIT
Message-Id: <200308150226.36787.akon@gmx.net>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1132
Lines: 25

Hiall,

does someone have experience in porting some Kernel to Double-Harvard-Arch?
I don't think Lx was ever ported to such a kind of µP (please correct me!), 
all I found on the web were several ports to embedded, but still vNeumann-, 
or Single-Harvard µPs.

DH means, that the µprocessor (typically a DSP) has a seperated program 
memory, a seperate (X)Data memory and a seperate (Y)Data mem, so it can 
fetch two data adresses simultanely in one cycle via two physically 
independent mem ports. For DSPs, that's a common behaviour!

So, obviously one (me) will have to integrate two flavours of malloc() 
into the Kernel (vmallocX() and vmallocY()). Of course, i could leave this 
issue to a specialized (uC)glibc, but i think, it should be the job of the 
kernel to keep the oversight on memory issues...;)

Any ideas how to manage that trouble as "frictionless" as can?,
And¡

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/