*BSD News Article 26430


Return to BSD News archive

Xref: sserve comp.os.386bsd.development:1718 comp.sys.ibm.pc.hardware.chips:1909 comp.periphs:4945
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!news.uoknor.edu!ns1.nodak.edu!netnews.nwnet.net!news.clark.edu!spool.mu.edu!agate!howland.reston.ans.net!vixen.cso.uiuc.edu!sdd.hp.com!col.hp.com!csn!hellgate.utah.edu!cc.usu.edu!ivie
From: ivie@cc.usu.edu
Newsgroups: comp.os.386bsd.development,comp.sys.ibm.pc.hardware.chips,comp.periphs
Subject: Re: The DMA problem again!
Message-ID: <1994Jan24.103126.8590@cc.usu.edu>
Date: 24 Jan 94 10:31:26 MDT
References: <jmonroyCJxHBH.2x0@netcom.com> <1994Jan22.120513.8484@cc.usu.edu> <jmonroyCK4tFo.3Jx@netcom.com>
Organization: Utah State University
Lines: 36

In article <jmonroyCK4tFo.3Jx@netcom.com>, jmonroy@netcom.com (Jesus Monroy Jr) writes:
> ivie@cc.usu.edu wrote:
> : In other words, DMA channel 0 is not involved in refresh.
> :
> 	From the facts you have previously stated, 
> 	no logic can be derived to state DMA channel 0 is
> 	not involved in the refresh.

Sorry. Point being that if DMA controller channel 0 were used for refresh,
either the DMA request or the grant would go somewhere involved with refresh;
since it only goes to the connectors and the DMA page registers, it can't
be used to do refresh.

> 	This makes two of us... I don't think I've ever
> 	stated that the FDC can hang the "refresh".
> 	I am saying the inverse (or is it the converse, anyhow).
> 	I beleive the DMA RAM refresh signal is interfering
> 	with the FDC transfer.

Again, sorry. Lost track of what the argument was.

From sheet 21 of the schematics, it looks like you're right, Jesus. A
DMA request is captured and synchronized to the DMA clock by the flops on
top of the sheet. However, the request then goes to the two flops in the
middle of the sheet; these flops decide who to give the grant to. When
the refresh request is made, the top flop becomes set. This does two
things: it grants the DMA request to the refresh controller and it clears
the grant to the DMA controller.

There's nothing in there to stop the grant from being ripped away from the
DMA controller if a refresh request comes along in the middle of a 
DMA transaction.
-- 
----------------+------------------------------------------------------
Roger Ivie      | Don't think of it as a 'new' computer, think of it as
ivie@cc.usu.edu |     'obsolete-ready'