Re: Best Ref-counting algorithms?

Hans Aberg <haberg_20080406@math.su.se>
Fri, 17 Jul 2009 13:05:31 +0200

          From comp.compilers

Related articles
[15 earlier articles]
Re: Best Ref-counting algorithms? cr88192@hotmail.com (BGB / cr88192) (2009-07-15)
Re: Best Ref-counting algorithms? cr88192@hotmail.com (BGB / cr88192) (2009-07-15)
Re: Best Ref-counting algorithms? gene.ressler@gmail.com (Gene) (2009-07-15)
Re: Best Ref-counting algorithms? torbenm@pc-003.diku.dk (2009-07-16)
Re: Best Ref-counting algorithms? bartc@freeuk.com (BartC) (2009-07-16)
Re: Best Ref-counting algorithms? gneuner2@comcast.net (George Neuner) (2009-07-16)
Re: Best Ref-counting algorithms? haberg_20080406@math.su.se (Hans Aberg) (2009-07-17)
Re: Best Ref-counting algorithms? haberg_20080406@math.su.se (Hans Aberg) (2009-07-17)
Re: Best Ref-counting algorithms? cppljevans@gmail.com (Larry) (2009-07-17)
Re: Best Ref-counting algorithms? lerno@dragonascendant.com (=?ISO-8859-1?Q?Christoffer_Lern=F6?=) (2009-07-17)
Re: Best Ref-counting algorithms? gah@ugcs.caltech.edu (glen herrmannsfeldt) (2009-07-17)
Re: Best Ref-counting algorithms? gneuner2@comcast.net (George Neuner) (2009-07-17)
Re: Best Ref-counting algorithms? gneuner2@comcast.net (George Neuner) (2009-07-17)
[12 later articles]
| List of all articles for this month |

From: Hans Aberg <haberg_20080406@math.su.se>
Newsgroups: comp.compilers
Date: Fri, 17 Jul 2009 13:05:31 +0200
Organization: Aioe.org NNTP Server
References: 09-07-018 09-07-023 09-07-027 09-07-037 09-07-042
Keywords: GC
Posted-Date: 17 Jul 2009 14:44:33 EDT

Christoffer LernC6 wrote:
>> But what language are you planning use for the implementation? If you
>> choose C++, then it is hard to find the root set. Suppose you create
>> object for use in your language; then global objects and those in th
>> stack should be registered somehow for the tracing, but those on the
>> heap should not, as the are live when they can be traced from the other.
>
> I was thinking of using C for implementation, but does that matter? As
> long as I implement the language in a non GC:ed language I will have
> to take care of writing the GC myself (or for C, use the Bohm
> conservative GC).


If you want to make a tracing GC in C++, then you will have to keep
track of where elements are located by hand.


For example, if you have class Object which keep track of a polymorphic
pointer, and a
      class S {
          Object x, y;
          ...
      };


Then in
      T f(...) {
          Object w = new S(...);
          ...
      }
the w belongs to the root set, as it is allocation on the function
stack, whereas the S::x, S::y are on the heap, as they are allocated via
the "new S".


So at least for dynamic languages, this is problem.


      Hans


Post a followup to this message

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