From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?Kolbj=C3=B8rn_Barmen?= Subject: Re: Linux/m68k on the Apollo FPGA softcore? Date: Fri, 5 Feb 2016 15:49:53 +0100 (CET) Message-ID: References: <56B36771.9010006@physik.fu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from epost.ufisa.uninett.no ([158.38.152.65]:43501 "EHLO epost.ufisa.uninett.no" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750958AbcBEOtz (ORCPT ); Fri, 5 Feb 2016 09:49:55 -0500 In-Reply-To: <56B36771.9010006@physik.fu-berlin.de> Sender: linux-m68k-owner@vger.kernel.org List-Id: linux-m68k@vger.kernel.org To: John Paul Adrian Glaubitz Cc: linux-m68k@vger.kernel.org On Thu, 4 Feb 2016, John Paul Adrian Glaubitz wrote: > On 02/02/2016 06:16 PM, Kolbj=C3=B8rn Barmen wrote: > > From what I understand, the core implements a new FPU, not > > compatible with existing FPUs, and there has > > been talks about a new MMU implementation too.=20 >=20 > Well, that would be too bad given the fact that there is already a hu= ge > number of applications and kernels which work with the old FPU/MMU > combination. Sure, but the idea is to move forward from what I understand. > I'd buy such an accelerator in a heartbeat, but it should be binary > compatible with existing 68020+ code. Apparently it will be usermode compatible with all existing 68k, but no= t supervisor mode compatible, not MMU compatibale, not FPU compatible. An= d it is already blazingly fast compared to most 68060 systems, without an= y optimization done. I guess it will be an AmigaOS only CPU core :) -- kolla