Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rutgers!cmcl2!sbcs!sbstaff2!allen
From: allen@sbstaff2.cs.sunysb.edu ( Allen Leung)
Newsgroups: comp.lang.functional
Subject: Re: GC triggering and stack limit checking by MMU hardware
Keywords: GC, stack, heap, MMU
Message-ID: <1990Jul20.010305.27654@sbcs.sunysb.edu>
Date: 20 Jul 90 01:03:05 GMT
References: <1990Jul19.151524.22544@diku.dk>
Sender: usenet@sbcs.sunysb.edu (Usenet poster)
Organization: State University of New York at Stony Brook
Lines: 31

In article <1990Jul19.151524.22544@diku.dk> torbenm@diku.dk (Torben [gidius Mogensen) writes:
>
>Most implementations (that I know) of programming languages with
>stacks or heaps, use some kind of explicit check to see when a stack
>or heap is full. This is used to trigger events such as error
>messages, garbage collection or allocation of more memory for a
>fragmented stack. Usually, good implementations avoid testing on every
>use of the stack / heap, but only once for every basic block or
>procedure.
>
>By using the MMU facilities present in most modern computers, these
>tests can be entirely done in hardware, and fragmented stacks can be
>done transparently. This is the idea:
>   [delete]
>My question is: has this been done? And if not, why not?

    If I remember correctly, earlier version of Standard ML compiler
from New Jersey uses a very similar idea of using virtual address fault
to trigger garbage collection( if I read you correctly, this is what you are
describing.)    This scheme was eventually abandoned
in favor of the explicit test due to problems with restarting an instruction 
on the 680x0 after the fault.  Also, this method does not port easily to 
OSs other than Unix( or even to different versions of Unix).

    I suspect using virtual memory to fault to gc is in general
a rather old trick.

>	Torben Mogensen (torbenm@diku.dk)

Cheers,
Allen Leung (allen@sbcs.sunysn.edu)