From: MagerValp (MagerValp_at_cling.gu.se)
Date: 2004-02-10 12:56:11
André, you also need to resubscribe with the new address.
-----begin-----
Date: Tue, 10 Feb 2004 10:22:37 +0100
From: Andre Fachat <fachat@web.de>
To: cbm-hackers@cling.gu.se
Subject: Re: Clock Stretching...
If I remember correctly, the RDY input is acknowledged immediately
for read cycles, but not at all for writes. This was indented to
halt the CPU for reads from slow memory.
So to make sure to have the CPU stopped, as it is needed in the
C64, three wait cycles have to be inserted as the CPU can write
up to three bytes in sequence - during which RDY is not recognized
and thus the CPU is not stopped. Only the first read after the
writes is then stretched and the CPU stopped.
Andre
On Mon, Feb 09, 2004 at 09:33:57PM +0200, Marko Mäkelä wrote:
> > Wasn't the problem with RDY that it needs up to 3 cycles to be
> > acknowledged, i.e. that write cycles can't be interrupted? Clock
> > stretching, in term, would work "just in time".
>
> True, I forgot about that.
>
> Marko
>
> Message was sent through the cbm-hackers mailing list
-----end-----
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.