Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!uunet!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.sys.next
Subject: Re: DMA Channels: Where'd they go?
Keywords: dma
Message-ID: <14638@cbmvax.commodore.com>
Date: 25 Sep 90 17:33:43 GMT
References: <5593@mace.cc.purdue.edu> <50950@brunix.UUCP> <5606@mace.cc.purdue.edu>
Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 33

In article <5606@mace.cc.purdue.edu> asd@mace.cc.purdue.edu (Kareth) writes:
>In <50950@brunix.UUCP> rca@cs.brown.edu (Ronald C.F. Antony) writes:

>>In article <5593@mace.cc.purdue.edu> asd@mace.cc.purdue.edu (Kareth) writes:

>>>But didn't there used to be 12 DMA channels on the NeXT Cube???
>>>Now there are only 9!  Only 8 on the slab.  What happened to those 3 DMA
>>>channels?

>>But what happened to the 3 others I can't tell.

>One of the guys here suggest that he believed that the missing channels
>were due to the fact that the MMU, FPU, and cache which are all on the
>040 now.

Nope, at least not by any proper definition of "DMA channel".  The cache and
MMU situation is identical for the 68030 and 68040 with the NeXT; in both
cases, you have on-chip cache and MMU.  As for the FPU, the 68882 follows a
private, Motorola defined communication protocol with the 68030; no use for
any DMA channel here.  

My uninformed guess would be that NeXT considered each empty expansion bus slot
to be a separate DMA channel from their DMA controller's viewpoint.  Since the
"slab" doesn't have these slots, it would be missing 3 channels.  But that's
really just a guess...

>-k


-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
	Standing on the shoulders of giants leaves me cold	-REM