[Home] [Groups] - Message: [Prev in Group] [Next in Group]

nu.kanga.list.mud-dev

8535: [MUD-Dev] Re: Bruce Sterling on Virtual Community goals

[Full Header] [Plain Text]
From: Chris Gray <cg@ami-cg.GraySage.Edmonton.AB.CA>
Newsgroups: nu.kanga.list.mud-dev
Date: Thu, 22 Oct 1998 22:05:26 -0600
Organization: Kanga.Nu
[Jon A. Lambert:]

 >It also claims the operand stack is FIFO.  I thought this was odd.

Definitely. If its FIFO it isn't a stack. By definition.

 >Each method gets it's own stack frame.  Perhaps this is peculiar
 >to Microsoft's or Symantec's VM.  The book is not really clear on
 >whose implementation is being referenced.   And I don't know if 
 >it's authoritative, it may well be misleading.

Could well be. The bit about the stack frame is correct, but when I
looked just now, I saw no mention of a frame pointer to point at it.
In the instruction format, locals are addressed via an offset within
the frame, so most implementations will likely use a frame pointer.

"Optop" sounds like the stack pointer, which they do define. Not
clear what "Vars" refers to. All non-local variables are inside classes.
I'd guess its the constant table.

--
Chris Gray     cg@ami-cg.GraySage.Edmonton.AB.CA