Article View: comp.sw.components
Article #182Re: Real-time Garbage Collection
From: Stephen Crawley
Date: Wed, 27 Sep 1989 13:21
Date: Wed, 27 Sep 1989 13:21
28 lines
1405 bytes
1405 bytes
From wtwolfe@hubcap.clemson.edu:
> From johnson@p.cs.uiuc.edu:
>> In fact, I don't believe that managed and unmanaged storage paradigm
>> should coexist in one program. You should either use components that
>> use automatic garbage collection or components that do not. While it
>> is possible to link C code to Lisp or Smalltalk, it is always tricky
>> and error prone.
>
> Since Ada components can coexist with no difficulty, this is
> not a true statement in general.
Suppose we garbage collect an object that contains a pointer to a
non-GC'able object X. The garbage collector cannot reclaim X and
it cannot tell the ADT for X that the pointer has been blown away.
In general the object X is likely to be lost to a storage leak.
The only possible escape is if the garbage collection mechanism includes
some scheme for applying a "destructor" to an object that is about to
be reclaimed by the GC. I have not come across such a scheme described
in the ADA RM.
-- Steve
Technical Details
Message-ID:
<921@scaup.cl.cam.ac.uk>Path:
arc1-ca.pugleaf.net!utzoo!attcan!uunet!mcsun!ukc!cam-cl!sccReferences:
<130200013@p.cs.uiuc.edu> <6599@hubcap.clemson.edu>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!mcsun!ukc!cam-cl!scc From: scc@cl.cam.ac.uk (Stephen Crawley) Newsgroups: comp.sw.components Subject: Re: Real-time Garbage Collection Message-ID: <921@scaup.cl.cam.ac.uk> Date: 27 Sep 89 13:21:03 GMT References: <130200013@p.cs.uiuc.edu> <6599@hubcap.clemson.edu> Sender: news@cl.cam.ac.uk Organization: U of Cambridge Comp Lab, UK Lines: 28