On Thu, May 26, 2011 at 9:02 PM, John Rose <[email protected]> wrote:
> Fully optimized invalidations will use safepoints.  The point of using an 
> array is to allow any number of switchpoints to ride on one safepoint.  Can 
> you batch all the invalidations in a given hierarchy of N classes?  Or is 
> there an ordering problem?

It might be possible to walk the full hierarchy, constructing new
switchpoints as I go, and then invalidate the old ones en masse. I'll
prototype it that way and see how it goes.

The ordering is not a problem, as long as the currently-in-use
switchpoint gets invalidated at some point.

- Charlie
_______________________________________________
mlvm-dev mailing list
[email protected]
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev

Reply via email to