Thomas,
The link doesn't work. Do you have another link to that paper?
Phil
On Thu, Jun 5, 2008 at 1:39 AM, Thomas Danecker <[EMAIL PROTECTED]> wrote:
> Hi Phil,
>
> I also want to mention that the GC only compacts the heap whenever it
> thinks
> that the heap is already too fragmented. Most of the time he only collects
> dead objects without moving around anything. Also collecting may be done by
> only walking through gen0. Whenever a reference in gen1, gen2 or any other
> root is changed to point to an object in gen0, it gets especially marked.
> That's another performance boost. Maybe google can turn up more details.
>
> There was a guest lecture about Heap Reference Analysis on my Uni in April.
> The lecturer more precisely defined when an object may be considered as
> dead
> (i.e. not when there are no references anymore but as soon as the runtime
> knows that the reference won't be used anymore).
> It makes an interesting read:
> http://www.cse.iitb.ac.in/~uday/soft-copies/etaps-08-dfa.pdf
>
> Thomas
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> SharpOS-Developers mailing list
> SharpOS-Developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/sharpos-developers
>
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
SharpOS-Developers mailing list
SharpOS-Developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sharpos-developers
