Path: utzoo!attcan!uunet!samsung!uakari.primate.wisc.edu!sdd.hp.com!hplabs!hpfcso!hpfelg!koren
From: koren@hpfelg.HP.COM (Steve Koren)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Hurricane 2800 doesn't like 16 bit fast ram?
Message-ID: <16530004@hpfelg.HP.COM>
Date: 11 Oct 90 14:15:33 GMT
References: <16530001@hpfelg.HP.COM>
Organization: HP Elec. Design Div. -FtCollins
Lines: 33


> time the 68000 bus, and by extention the Zorro II bus, were defined.  So, in
> order to support data caching of 16 bit RAM, your 68030 board needs extra
> logic.  The Hurricane board didn't add this extra logic, the A2630 and
> GV 68030 boards did.  I don't hold this against the Imtronics, the current

This may be true, but there is apparently something else going on also.

I did a few memory tests last night on the 16 bit memory as follows:

  Test 1: 16 bit memory enabled, 68030,    I-cache,    D-cache:  FAILED
  Test 1: 16 bit memory enabled, 68030,    I-cache, no D-cache:  FAILED
  Test 1: 16 bit memory enabled, 68030, no I-cache,    D-cache:  FAILED
  Test 1: 16 bit memory enabled, 68030, no I-cache, no D-cache:  FAILED

  Test 1: 16 bit memory enabled, 68000, no I-cache, no D-cache:  SUCCEDED

All configurations failed with the '030 card; however, booting with the
68000 allows the mem to work.  (This would tend to indicate that the
16 bit mem card itself is seated correctly, has good chips, etc).

The mem test program I have fails on the first test of that memory.
This test writes "0"s to each byte, then reads them back to verify that
they are really "0"s.  They weren't.  They were things like "0x01",
"0xfd", etc.  It was not totally random, as only about 10 or so bytes
seemed to show up.

At first I suspected that possibly the 16 bit mem and the 32 bit
mem were being mapped to the same spot.  However, the Hurricane docs
say that their mem is mapped to a higher spot in the mem map, thus
allowing for something like 4 meg of other fast ram.

  - steve "clueless..." koren