Indeed, the 68000 and 68010 pushed the same amount of data onto the stack,
but the 68010 produced more different types of stack-frames than the 68000.
If you wanted to speed up the ATARI-ST you could replace the 68000 with an
68010, but you needed to patch TOS. Otherwise the O.S. crashed.
Nice was the fact that the 68000 and the 68010 are *pin-compatible*.
I know, because my StarShip first ran on a 6802, then a 68000 and then on
a 68010. I had to re-write a small part of the embedded OS that handles the
stack-frame processing. Tight-loops (2 instruction) are cached.
[My StarShip runs on a 68020 at 30 MHz. now]
- Henk.
> -----Original Message-----
> From: Ethan Dicks [mailto:erd_6502_at_yahoo.com]
> Sent: dinsdag 15 januari 2002 21:20
> To: classiccmp_at_classiccmp.org
> Subject: Re: 68010 (was Re: Mac IIci)
>
>
>
> --- Pete Turnbull <pete_at_dunnington.u-net.com> wrote:
> > On Jan 14, 14:56, Bruce Robertson wrote:
> >
> > > Yes, I seem to remember that with the 68000, there was an
> interaction
> > > with the Bus Fault signal... something to do with what
> state got saved
> > > on the stack; I don't remember the exact details
>
> After a bus fault, there was not enough information on the stack to
> properly do an instruction restart from where the bus-fault occured.
>
> > Oops, I forgot about that. You're right; the 68010 saves
> slightly more
> > state on the stack than the early 68000. I have a feeling
> that was fixed
> > in later 68000; some traps save more state than others.
>
> Not as far as I know... the quantity of bytes pushed on the
> stack should
> be constant for a given member of the 68K family. A lot of
> older software
> for the Amiga that did things with the stack (debugging
> tools, mostly)
> assumed certain things relating to the stack - it became confused on
> the '010 and up because the number of bytes did change for certain
> traps (like bus error). I have the details at home, not with
> me, in my
> Motorola books, or I'd post them here. Eventually, people learned to
> ask the OS what was going on, rather than paw through the stack
> indescriminantly, kinda like when people got burned on the
> first Fatter
> Agnus Amigas - 1Mb of CHIP and 0Mb of FAST RAM - broke all kinds of
> software that asked for a buffer of FAST RAM instead of
> "fastest available
> RAM".
>
> > You can't get 68010s any more, unless you can find old
> stock somewhere
> > :-(
> > You can still get 68000s and 68020s.
>
> That's not surprising. Even when they were current, we had a
> hard time
> getting 68010 chips for our products. We paid $45 each for them at a
> time when the 68000P8 was about $3 (eventually, I found them at a
> surplus/overstock electronics dealer for $10).
>
> At the moment, I have dozens of 68000L8s and one tube of 68010P10s. I
> hope I never have to look for any more 68010s.
>
> -ethan
>
>
> __________________________________________________
> Do You Yahoo!?
> Send FREE video emails in Yahoo! Mail!
> http://promo.yahoo.com/videomail/
>
Received on Wed Jan 16 2002 - 06:23:02 GMT