comp.lang.ada
 help / color / mirror / Atom feed
From: Martin Krischik <krischik@users.sourceforge.net>
Subject: Re: Memeory management
Date: Thu, 22 Sep 2005 19:11:11 +0200
Date: 2005-09-22T19:11:11+02:00	[thread overview]
Message-ID: <59346899.OiYf4XpetV@linux1.krischik.com> (raw)
In-Reply-To: F%zYe.5725$6e1.146@newssvr14.news.prodigy.com

adaworks@sbcglobal.net wrote:

> 
> <tmoran@acm.org> wrote in message
> news:naidnWP4ToNgJoHeRVn-2g@comcast.com...
>>
>>   If you allocate explicitly from the heap using access types (pointers)
>> and "new" then the Ada specification does not say whether the
>> implementation must do automatic garbage collection or not.  Most don't,
>>
> Because Ada does not prohibit automatic garbage collection, it would seem
> like a good idea to have it as a compile-time option in a well-designed
> compiler.   Yes, it is a lot of work, but for most programs, not
> safety-critical,
> and running under a commercial OS, it would make sense.

Well I had a go on this allready:

http://adacl.sourceforge.net/pmwiki.php/Main/GarbageCollector

Using storrage pools is actually better then compiler options - sure more
typing work - but what happens if GC and non-GC packages are linked with
each other?

The only thing what bothers me is finalisation - with lots of GNAT source
browsing and intermediate code listing I got it implemented - but I it
would be better if GC storrage pools where incorporated right into the
compiler .

Martin

-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com



  reply	other threads:[~2005-09-22 17:11 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-09-05 21:02 Memeory management Zheng Wang
2005-09-05 21:43 ` tmoran
2005-09-22 15:23   ` adaworks
2005-09-22 17:11     ` Martin Krischik [this message]
2005-09-05 21:45 ` jimmaureenrogers
2005-09-06 16:45 ` Martin Krischik
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox