Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!samsung!munnari.oz.au!cluster!metro!natmlab.dap.csiro.au!ditsyda!evans
From: evans@ditsyda.oz (Bruce Evans)
Newsgroups: comp.os.minix
Subject: Re: Non standard at problem
Message-ID: <2457@ditsyda.oz>
Date: 20 Feb 90 05:29:00 GMT
References: <1190@cherokee.tusun2.knet.utulsa.edu>
Reply-To: evans@ditsyda.oz (Bruce Evans)
Organization: CSIRO DIT Sydney, Australia
Lines: 21

In article <1190@cherokee.tusun2.knet.utulsa.edu> stevens@cherokee.UUCP (Michael Stevens) writes:
*I am trying to install minix on a non-standard AT 
*(Diversified Technology cat 901).  This machine has an xt hard drive
*controller built onto the motherboard, MS dos stuff works just fine,
*and the thing is mapped in at 320H (where everyone expects it to be).
*Since this controller is an xt controller, I am using xt_wini.c for 

An XT drive in an AT is like not to work without changing the kernel. In
protect.c and main.c, the kernel assumes that ATs use AT_WINI_VECTOR for
disk interrupts, independent of whether xt_wini.c is used. This may be
correct depending on how standard the "non-standard" really is.

*The problem is this:  When minix trys to reset the hard drive, it sends 
*a zero to the status byte of the controller and expects to receive
*2|4|8 when it checks status,  Instead it receives a zero.  Does anyone

Something like this was fixed in 1.5. Try a w_reset() if the win_init()
just before zeroing nr_drives fails. w_reset() calls win_init() so if
w_reset() succeeds you don't have to branch back to retry the win_init().
-- 
Bruce Evans		evans@ditsyda.oz.au