Re: 8088 and 610 saga continues...

From: Michał Pleban <lists_at_michau.name>
Date: Mon, 24 Jan 2011 20:10:55 +0100
Message-ID: <4D3DCEBF.3010805@michau.name>
Hello!

W dniu 2011-01-23 16:49, Hoffmann-Vetter, Martin pisze:

> Is this the only address that was changed? Can you test it?

Now, it gets very interesting! I downloaded three different memory dumps
after MS-DOS hung, and here's what I found:

EVERY byte with address xx21 and xxA1 gets corrupted with random values.
At 0021 it is usually zero, although once I saw a $C1 value. But with
other bytes, it's totally random and unrelated to the original value.

How is it possible that every xx21 and xxA1 byte gets corrupted this
way? It seems likely to me that something is playing very rudely with
DRAM control lines and overwriting a location in all memory columns at
once. I am not knowledgeable with DRAM chips so don't know whether this
is at all possible?

Or maybe it is a different problem. Maybe these locations are not
overwritten by anything, but they are not refreshed? Something causes
these two rows to not be refreshed? Does it make any sense?

Regards,
Michau.



       Message was sent through the cbm-hackers mailing list
Received on 2011-01-24 20:00:10

Archive generated by hypermail 2.2.0.