Deallocating memory pages
jon.masamitsu at oracle.com
Thu Feb 21 20:04:41 UTC 2013
On 02/21/13 06:20, Clemens Eisserer wrote:
>> I'd say let's put this on hold for now. I don't personally have
>> experience/expertise in building/testing JDK on non-Linux platforms (or
>> access to machines, either), unfortunately. But I still think it's
>> compelling enough to have it contributed somehow.
> This is really unfourtunate, from my POV especially for client-side workloads.
> On the desktop, footprint is often more important than
> peak-performance, especially when running many java based desktop
> (although I have to admit CMS doesn't make a lot of sence there).
> What I find so intriguing about this patch is that it basically
> reduces one of the drawbacks of GC based systems: Unused memory is not
> given back to the memory manager of the OS. I've recently read a paper
This again being most useful when applications are idle?
> about the idea of a GC interface provided by the kernel to avoid this
> issue, but this patch essentially achieves the same just by using
> existing interfaces.
> Regards, Clemens
More information about the hotspot-gc-dev