home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!usenet.coe.montana.edu!giac1.oscs.montana.edu!osymh
- From: osymh@giac1.oscs.montana.edu (Michael L. Hitch)
- Newsgroups: comp.os.minix
- Subject: Re: AmigaMinix
- Keywords: n
- Message-ID: <1992Aug28.170741.349@coe.montana.edu>
- Date: 28 Aug 92 17:07:41 GMT
- Article-I.D.: coe.1992Aug28.170741.349
- References: <1992Aug28.021419.7329@newshub.ariel.yorku.ca>
- Sender: usenet@coe.montana.edu (USENET News System)
- Organization: Montana State University, Bozeman MT
- Lines: 25
-
- In article <1992Aug28.021419.7329@newshub.ariel.yorku.ca> cs902125@ariel.yorku.ca (SHUK C YIP) writes:
- >Is Amiga version of Minix compatible with 68040 CPU?
- >Does it support MMU?
-
- Unfortunately no. The Prentice Hall version only supports the 60000 CPU, and
- the 1.5.10.1 and 1.5.10.2 updates will only support the 020 and 030 CPU. The
- problem is that the 040 has different instructions for the cache control and
- the MMU control, which are not compatible with the 030 instructions. If
- AmigaDOS recognized the 040 CPU (AmigaDOS 2.04 does), the minix loader
- program attempts to turn off MMU translation using the 030 instruction which
- is invalid on the 040. The MOVEC CCR instruction used on the 030 to control
- the cache will execute on the 040, but the 040 has new instructions that
- actually control the cache. The cache invalidate routines in mpx.s need to
- be changed to recognize that the CPU is an 040 and execute the proper
- instructions.
-
- I think I was able to run Amiga Minix 1.5.10.1/1.5.10.2 on my PP&S 040 card
- when I was still running AmigaDOS 1.3 and had the 040 copyback cache off
- (which was required under 1.3 anyway). AmigaDOS 1.3 recognized the CPU as
- a 020 and the minix loader didn't attempt to turn off MMU translation.
- --
- Michael L. Hitch BITNET: OSYMH@MTSUNIX1.BITNET
- Computer Services Consultant INTERNET: osymh@msu.oscs.montana.edu
- Office of Systems and Computing Services
- Montana State University Bozeman, MT USA
-