Re: Programming the 65816

["Michael J. Mahon" <mjmahon@aol.com>]

Michael J. Mahon wrote:
> sfahey wrote:
>
> > I was thinking about buying a copy of Eyes and Lichty's "Programming 
> > the 65816
> > (Including the 6502, 65C02 and 65802)" on eBay, when I noticed it's 
> > available
> > for free on the Western Design Center site as a PDF.
> >
> > I didn't know if this was common knowledge or not, so for anyone 
> > interested:
> >
> > http://www.westerndesigncenter.com/wdc/datasheets/Programmanual.pdf
> >
> > Sean Fahey
> > www.a2central.com
>
>
> Thanks, Sean--very nice!
>
> Now I have a "portable" copy when the hard copy is at home.  ;-)

I couldn't resist--I opened the "book" and started reading.  ;-)

I was immediately struck by the relatively high density of minor
(reader correctable) typos.  Are all these in the paper book, or
might they be a result of OCRing the book?

I also--finally--understand the reason for Load Indexed sometimes
taking a cycle less than the maximum (when the indexing operation
does not result in a page crossing), while Store Indexed always
takes the maximum time.

For a Load Indexed, the 6502 does a "speculative load" before the
page crossing is determined, and, if no crossing occurs, it bypasses
the second "guaranteed correct" page access.  In the case of a load,
this causes no problem (unless the lower page access causes some
side-effect, as in I/O space ;-).  However, in the case of a store,
a "speculative store" is not permissible, since it could corrupt
data not actually addressed.  Therefore, no "cycle-skipping" is
permitted for stores.

Thanks again, Sean, for providing a nice "aha"!  ;-)

-michael

NadaNet and AppleCrate II: parallel computing for Apple II computers!
Home page: http://home.comcast.net/~mjmahon

"The wastebasket is our most important design
tool--and it's seriously underused."