Re: Compile HLL to microcode on VLIW - possible?

WStreett@shell.monmouth.com (Wilbur Streett) writes:


>and based on what I used to hear my father talk about, the Burroughs 6700..


Either your father was mistaken, or you have misremembered.


The Burroughs microcoded machines (the "D-machines") were the
B1700/B1800/... series. That's SEVENTEEN hundred. The memory was
supposedly bit addressable; programming language implementations could
choose the word length they preferred. The "User Programming
Language" implementation chose 24 bits.


The Burroughs "Large Systems" machines where the
B6700/B6800/... series. That's SIXTY-SEVEN hundred. The memory was
48 data bits + 3 tag bits + 1 or more parity/ecc bits depending on
model; programming language implementations had to take big numbers
and like them. The instruction set was hardwired. At my parents'
home I still have a listing of an old B6700 MCP and Algol compiler,
and I can assure you that there was no provision for any kind of
dynamic microcode.


In both systems, the operating system was called MCP, but they were
different programs.


Perhaps someone along the line may have been misled by the Maintenance
Diagnostic Processor, which could be programmed in an Algol dialect,
or the Data Communications Processor, which could be programmed in a
different Algol dialect called NDL. The DCP certainly wasn't
microcoded. I don't know about the MDP, but in any case user programs
didn't run on the DCP or the MDP.


I *believe* that the Burroughs cheque reader/sorters also had
D-machines inside them, programmed in COBOL. Aritmetic operations
decoded descriptors at run time. It made for compact code, and the
software was able to keep up with the mechanical side.
--
Richard A. O'Keefe; http://www.cs.rmit.edu.au/~ok; RMIT Comp.Sci.
[The B1700 was the machine I characterized in a previous note as running
as though affixed to a gastropod. -John]


--

Return to the comp.compilers page.
Search the comp.compilers archives again.