Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!dsuvax!ghelmer
From: ghelmer@dsuvax.uucp (Guy Helmer)
Newsgroups: comp.os.minix
Subject: Re: More PH 1.5 problems
Message-ID: <1991Jan22.155420.19842@dsuvax.uucp>
Date: 22 Jan 91 15:54:20 GMT
References: <42396@nigel.ee.udel.edu>
Organization: Dakota State University
Lines: 30

In <42396@nigel.ee.udel.edu> D98E%DDATHD21.BITNET@pucc.princeton.edu (Peter Lautenschlaeger) writes:
>>
>> 4.  The readall program will lock the system waiting for the
>>     hard disc.  ( I was able to find a combination of partition
>>     sizes that it would work with.)
>>
>  Guy Helmer and I were faced with the same problem, if you are
>  eventually using a WD-1006 (RLL?,so do I) 1:1 controler, please
>  let me know. The only thing I can say so far is after I updated
>  to beta-1.6.11 the problem vanished.

I tried just applying the single fix to my 1.5 at_wini.c from MINIX 1.6 that
wasn't cosmetic, and my disk just locked up sooner.  Does anyone
have any technical material on the WD-1007/WD-1006 controllers
that lists exactly which bits to wait on before sending
commands to the controller, and what timing restrictions there
may be with these controllers?  I strongly suspect that the
wini driver is sending commands to the controller before it
is ready, and then waiting for a response that never comes.

BTW, the routine controller_ready() in at_wini.c is the place
where the CPU blocks until certain bits become set/unset.
MINIX 1.6 modifies the wait loop slightly, but that single change
by itself did not help my problem.

-- 
More information that you couldn't have existed another day without, from:
Guy Helmer                       helmer@sdnet.bitnet, uunet!dsuvax!ghelmer
work: DSU Computing Services                                (605) 256-5315
play: MidIX System Support Services - Crufty Hacks 'R' Us   (605) 256-2788