Date   : Tue, 24 May 2005 23:51:16 +0100
From   : Philip Pemberton <philpem@...>
Subject: Re: 8-bit IDE i/f debugging

In message <050524205504@...>
          jgh@... (Jonathan Graham Harston) wrote:

> In going through my pile of hard drives I've found one that
> breaches IDE protocol, in that if you attempt to read past the end
> of the disk it hangs instead of giving a "sector not found" error.

I found a 120MB Fujitsu that did that. Made HDInit return "0K" for the disk
size - very funny :)
Still, the Conner CFS425A that's on there now (ex-RiscPC) seems nice and
stable. I'll fire up the "Lancaster 65C02 Assembler" tomorrow and see how
badly I can get it to thrash the hard drive.

> This may the the problem you experienced here. I'm going to add
> another option to HDInit to allow the user to explicity specify
> the drive size instead of investigating the drive for it.

Another idea (when the 16bit interface has been debugged) would be to send an
IDENTIFY command to the drive, then grab the size from that. A much better
idea than "investigating" the drive, IMO.

Later.
-- 
Phil.                              | Acorn Risc PC600 Mk3, SA202, 64MB, 6GB,
philpem@...                        | ViewFinder, 10BaseT Ethernet, 2-slice,
http://www.philpem.me.uk/          | 48xCD, ARCINv6c IDE, SCSI
* <- Tribble           # <- Tribble after Borg assimilation