Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!utcs!mnetor!seismo!nbires!hao!hplabs!pyramid!decwrl!amdcad!cae780!leadsv!eps2!msunix!jon From: jon@msunix.UUCP Newsgroups: net.arch,net.micro.68k Subject: CSA 68020 board - works now, with mods Message-ID: <240@msunix.UUCP> Date: Fri, 18-Jul-86 01:00:44 EDT Article-I.D.: msunix.240 Posted: Fri Jul 18 01:00:44 1986 Date-Received: Sat, 19-Jul-86 08:26:05 EDT Organization: Via Visuals Inc. Lines: 31 Keywords: tri-state strobes, 74F241, 68881 Xref: utcs net.arch:3650 net.micro.68k:1718 Hi everyone! When I last posted it looked like DMA was not working on the CSA 68020 piggyback board. Last Friday I added some DMA test routines to our diagnostic/boot PROMs to exercise the MC68450 DMA controller on the CPU board. Sure enought, they ran fine on the 010 but the 020 died and although the 68450 had the bus, nothing was happening. We noticed that Address Strobe, and Upper and Lower Data Strobe were not tri-stated by the CSA board, so this was a definite problem. On Monday we called CSA, and they said, "Oh yeah, we don't support DMA on that rev." What??!!! They provided a fix though, which was to add a 74F241 and use BGACK to tri-state Address Strobe, and Upper and Lower Data Strobe. We lost the schematic for the board, if we had it we could've figured it out ourselves. So we fired it up and it runs fine. Next step is to enable the cache, (which may slow it down, right?) and make sure the code I added to handle the 020's exception stack works so virtual memory functions. It would be nice if the 020's barrel shifter speeds up our application, which does lots of 11 and 5 bit shifts. Also, they provided a fix for the 881, I can't recall what the problem was, but I think it doesn't work without a cut and a jumper. I haven't tried it out yet, so I can't comment on it. "If we did it like everyone else, Jonathan Hue what would distinguish us from Via Visuals Inc. every other company in Silicon Valley?" sun!sunncal\ >!leadsv!msunix!jon "A profit?" amdcad!cae780/