Re: [rvm-research] Regarding GC
Hari Krishnan <[email protected]>
| Newsgroups | gmane.comp.java.jikes.rvm.devel |
|---|---|
| Message-ID | <CAMJ8-Rof0HjQZDV7D-eQwpD_Ku58YZFRzpKyQDstUC6XpvifJA@mail.gmail.com> |
Hi Mike, Thank you for the reply. I just don't want to assume anything. So just want to know that before I make progress in my work. On Wed, Aug 27, 2014 at 1:34 PM, Michael Hind <[email protected]> wrote: > The single statement in your snippet is eventually expanded by the > compiler into a sequence of indirect loads (think assembly level > programming). A GCmap is (logically) created for each such > (assembly-like) instruction that contains the current live stack roots. > So, before the snippet it will contain what head (a local variable) points > to and then as the sequence is executed (one indirection at a time) and > head is updated, the stack root set is updated. > > So the answer is > - yes, there are points in the sequence where a stack root is pointing > to each node in the list 1-6, but > - no, there is not a single point in the sequence where the stack roots > point to all these nodes > > Mike > _____________________________________________________________ > Michael Hind, Senior Manager, Programming Technologies Department > IBM T.J. Watson Research Center > http://www.research.ibm.com/people/h/hind > 914 945-2712 > My internal blog: https://w3-connections.ibm.com/blogs/hindsight > > > > From: Hari Krishnan <[email protected]> > To: "General discussion of Jikes RVM design, implementation, > issues, and plans" <[email protected]> > Date: 08/27/2014 12:21 PM > Subject: [rvm-research] Regarding GC > ------------------------------ > > > > Hi, > > I have a question. I just want to make if a scenario is possible to > happen. The scenario goes like this. > > I have a cycle of linked list. All nodes have id from 1 to 50 and they are > arranged in order. There is a head pointer to a node with id 1. I am moving > this head pointer to node with id 6. Does all the intermediate nodes have > stack root pointing to them at the interim time during traversal? > for example > > fun(head){ > head = head->next->next->next->next->next; > ......... > > > } > In this above snippet, do all of the intermediate nodes 2 through 5 has > stack root ("head") pointing to them briefly? or VM figure out and > calculate the address and move the root from node 1 to node 6 directly? > > Cheers > Hari. > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/_______________________________________________ > Jikesrvm-researchers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > _______________________________________________ > Jikesrvm-researchers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > ------------------------------------------------------------------------------ Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/ _______________________________________________ Jikesrvm-researchers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers