Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!emory!ogicse!milton!uw-beaver!cornell!johnhlee
From: johnhlee@flute.cs.cornell.edu (John H. Lee)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Hardwareless hack?
Message-ID: <49037@cornell.UUCP>
Date: 28 Nov 90 23:25:24 GMT
References: <1990Nov23.012503.1150@vax5.cit.cornell.edu>
Sender: nobody@cornell.UUCP
Reply-To: johnhlee@cs.cornell.edu (John H. Lee)
Distribution: comp
Organization: Cornell Univ. CS Dept, Ithaca NY
Lines: 44
Summary:
Followup-To:

In article <1990Nov23.012503.1150@vax5.cit.cornell.edu> inuy@vax5.cit.cornell.edu writes:
[...]
>looking I noaticed that track 0 line (TK0*, pin 15) and the disk write protect
>line (WPRO*, pin 14) are both open collector.  I would be willing to bet that
>one could just unplug there external drive and plug in a db23 with pins 14 and
>15 jumpered and that would make it very hard for a viri to write to track zero.
>What do you think?  Can it really be this easy?
>
>Matthew

Close, but there are problems.  First off, simply jumpering those two together
will:
	1)  cause TK0* to be active whenever a write-protected floppy is
	    inserted (so the drive appears to be always at track 0), and
	2)  cause *all* drives to be non-track-0 writeable.

You can fix these two problems by using a open-collector OR-gate to neg-logic
AND the TK0* and SEL* signals together to drive WPRO*.

But a more serious problem has to do with the fact that the signal is output-
only from the drive and only a data-bit to the controller.  You're not really
write-protecting the drive, but merely telling the controller that the disk
is protected whenever the head is at track 0.

For the Amiga, I believe the floppy controller does not enforce write-
protection.  Software and the drive itself does.  Also, the trackdisk device
may not check the write-protect signal except when a disk is inserted (not
unreasonable since it's difficult for the user to (de)write-protect a disk
loaded in the drive.)  In this case, a write-enabled disk will: (a) always
appear write-protected (head at track 0 when disk is checked), or (b) will
never appear write-protected (head not at track 0.)  In the latter case,
the trackdisk device will write on track 0 anyways despite the circuit.

Then how to do what you want?  One possible method is to disable the write
gate signal to the drive whenever TK0* and SEL* are active.  Possibly using
one quad OC-NAND chip.  But there are timing considerations (like making
sure the signal is off at the right time w/respect to the drive select.)

To sum it all up:  I wish it were that simple.

-------------------------------------------------------------------------------
The DiskDoctor threatens the crew!  Next time on AmigaDos: The Next Generation.
	John Lee		Internet: johnhlee@cs.cornell.edu
The above opinions of those of the user, and not of this machine.