Re: Questions about Bytecode

Eric <englere_geo@yahoo.com> writes:


> The UCSD p-machine is also somewhat easy to understand, and sources
> have recently surfaced for that, along with detailed documentation.
> This is a non-OOP model and it extrememly efficient with a memory
> constrained environment. It could run programs coded in many
> languages, not just Pascal. I remember seeing COBOL and Fortran
> support for it. But it pre-dated the popularity of C, so I don't think
> C was a supported language.


Now, I may have my history wrong (and somewhere in the HOPL this
should be documented). However, if I understand correctly the p-code
of the UCSD p-machine was part of a family of VMs. I presume that the
root was the pcode invented by Wirth. For C, there was the pcode
derivative called ucode invented by Fred Chow at Stanford, as part of
the MIPS project, which became the basis of the Mips Inc. compiler
suite, which included C and Fortran compilers. At DEC, they did a C++
front end to ucode. The main distinction of ucode was that it had the
concept of registers which was good for compiling and generating code
for general register machines (e.g. vax, mips, 68k, alpha). However,
you could remove the register extension and interpret the code on a
pure stack machine with no loss of correctness, which was one
advantage of the pcode heritage.


Hope this helps,
-Chris


*****************************************************************************
Chris Clark Internet : compres@world.std.com
Compiler Resources, Inc. Web Site : http://world.std.com/~compres
23 Bailey Rd voice : (508) 435-5016
Berlin, MA 01503 USA fax : (978) 838-0263 (24 hours)

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