On Fri, Oct 11, 2013 at 9:46 AM, Sandro Magi <[email protected]> wrote:
> Not sure when you last checked LtU, but I just posted a paper on a > competitive refcounting GC based on Immix: > > http://lambda-the-ultimate.**org/node/4825<http://lambda-the-ultimate.org/node/4825> > > They conclude that the heap organization and resulting cache misses > account for at least 10% throughput degradation of refcounting GCs compared > to tracing. Yup. That post is what got me looking at RC-immix. Is there a paper somewhere that digs into the state of the art for the reference counting portion of things? How much (if at all) are people doing in the way of conservative liveness analysis?
_______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
