[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Why still using Apple II?

Roger Johnstone wrote:

In <MPG.1d1211ba7f493dee9896a2@news.individual.net> Paul Schlyter  wrote:


In article <_dWdnUq3Cs1RQTrfRVn-2w@comcast.com>, alexf@mirrow.com says...



4GB address space


That came later.  The first versions of the 68K CPU had only a 24-
bit
I'm talking about LOGICAL memory space, not the number of physical address pins.
Did the 68K have some paging mechanism which allowed access to 4 GBytes address space through its 24 address pins? If not, that LOGICAL memory space becomes quite useless, since no physical RAM can ever be attached to most of that memory space.
The logical address space was very important for forwards compatibility. It meant that when newer CPUs came out with the full 32-bit address space, the old programs writen for the original 68000 could access more than 16MB. 

Of course Apple still managed to stuff it up! Some of the earliest 68020/
030 Macs had ROM code which used the unused top 8 bits in the addresses. Well, they were unused until System 7 came out with 32-bit addressing support. Macs without '32-bit clean' ROMs could only support 24-bit addressing and so were limited to just 8MB of RAM. 

There is a universal tendency to see sizable "headroom" as virtually
infinite, so some of it is perceived as "excess" to be preempted
for other purposes (like tag bits).

The IMB 360 architecture provided for 32-bit logical addressing
from the start, but no machine was envisioned that would support
more than 24-bit physical addresses (16MB was _huge_ and unaffordable
at the time).

As advances in silicon technology put >16MB machines within reach,
IBM had to scramble to re-write lots of software that had used
bits in the upper byte of addresses.  Even their procedure calling
convention used the upper bit for a flag, creating a 2GB limit that
also had to be dealt with!

-michael

8-voice music synthesizer using NadaNet networking!
Home page:  http://members.aol.com/MJMahon/