Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!asuvax!ncar!zaphod.mps.ohio-state.edu!van-bc!
From: lphillips@lpami.wimsey.bc.ca (Larry Phillips)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Diskperf with A3000
Keywords: Strange behavior of Diskperf
Message-ID: <1735@lpami.wimsey.bc.ca>
Date: 14 Jun 90 11:26:35 GMT
Lines: 30
Return-Path: <lpami!lphillips>
To: van-bc!rnews

In <404@nyit.UUCP>, aca@nyit.UUCP (Al Arthur) writes:
>In article <12526@cbmvax.commodore.com> jesup@cbmvax (Randell Jesup) writes:
>>	Check to make sure the mask for the partition is 0xfffffffc.  If it's
>>0xfffffe, then it's set up wrong.
>
>I never did figure-out what the mask was really used for,  nor what it
>should be set to on my machine.  Can you explain it to me? (Thanks)

It 'masks' the data address such that any address having bits _only_ within the 1
bits in the mask will initiate a DMA transfer, while any bit in the address
corresponding to a 0 bit in the mask will cause a programmed IO transfer to to
be initiated. Put another way...

   if ((!Mask) & (DataAddress))
   	DoProgrammedIO
   else
   	DoDMA

One proper value for a 2000 with a 2630 would be 0xfffffe, which says that you
can DMA to any even address in the first 16 MBytes.

-larry

--
The raytracer of justice recurses slowly, but it renders exceedingly fine.
+-----------------------------------------------------------------------+ 
|   //   Larry Phillips                                                 |
| \X/    lphillips@lpami.wimsey.bc.ca -or- uunet!van-bc!lpami!lphillips |
|        COMPUSERVE: 76703,4322  -or-  76703.4322@compuserve.com        |
+-----------------------------------------------------------------------+