Menu
▾
▴
RE: [Jikesrvm-researchers] Implementation of incremental/current collector in JikesRVM
|
From: Genady B. <me...@ge...> - 2005-10-10 23:46:07
|
I'm not sure what other environments you consider, so I can't really compare. Jikes was built with stop-the-world collector in mind and in the code there are many implicit assumptions that GC code won't be executed in parallel with user code. I'm not sure whether I've found all of them yet. If you are new to Jikes, you should first try to write some simple stop the world collector to get started, since diving directly into on-the-fly issues brings lots of extra complexity. I don't have a tutorial on that subject, but if you have specific questions (like how to implement an On-the-Fly Phase), you can ask me either directly or through this list. Regards, Genady _____ From: Delvin Defoe [mailto:del...@gm...] Sent: Monday, October 10, 2005 22:54 To: jik...@li... Subject: Re: [Jikesrvm-researchers] Implementation of incremental/current collector in JikesRVM Thanks much. I am considering building an on-the-fly collector as well but I was not sure whether Jikes RVM is an appropriate environment since examples are not available. Any info in this regard is appreciated. Delvin On 10/8/05, Genady Beryozkin <me...@ge...> wrote: I'm trying to build an on-the-fly collector. I've almost completed the work with the old MMTk design, but some weird bug caused random crashes on 4 processor simulation. Now I'm working on a new version for the latest RVM ( 2.4.x). I can post the modifications I found to be necessary to make anything that is not stop-the-world to work. Genady The Technion, IIT -----Original Message----- From: Robin Garner [mailto: rob...@an... <mailto:rob...@an...> ] Sent: Saturday, October 08, 2005 03:39 To: jik...@li... Cc: jik...@li... Subject: Re: [Jikesrvm-researchers] Implementation of incremental/current collector in JikesRVM Hi Delvin, > Is there an example implementation of an incremental/concurrent > collector in Jikes RVM? The last time I looked the collectors were > based on the stop-the-world model. Daniel and I are both intending to build this type of collector - this was one of our primary motivations in the recent refactoring of MMTk. I currently have a read-barrier based incremental copying collector written but not quite debugged, and I've put it aside to work on other things. I'd be happy to share the patch. cheers Robin ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Jikesrvm-researchers mailing list Jik...@li... <mailto:Jik...@li...> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl <http://solutions.newsforge.com/ibmarch.tmpl> _______________________________________________ Jikesrvm-researchers mailing list Jik...@li... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers -- *********************************************************** Delvin Defoe Doctoral Candidate Department of Computer Science Washington University in Saint Louis 314.935.4502 http://www.cs.wustl.edu/~dcd2/ *********************************************************** |