RFR (M) 8213753: SymbolTable is double walked during class unloading and clean up table timing in do_unloading
Aleksey Shipilev
shade at redhat.com
Thu Jan 31 15:03:20 UTC 2019
On 1/30/19 8:41 PM, coleen.phillimore at oracle.com wrote:
> I added the timing for triggering cleanup. It's likely not interesting. The main cleanup time is
> code cache cleaning still.
>
> This is my fastdebug timing:
> [4.992s][debug][gc,phases] GC(2) ClassLoaderData 0.016ms
> [4.992s][debug][gc,phases] GC(2) Trigger cleanups 0.010ms
> [5.039s][debug][gc,phases] GC(2) Class Unloading 47.290ms
>
> open webrev at http://cr.openjdk.java.net/~coleenp/2019/8213753.02/webrev
Looks good.
-Aleksey
More information about the hotspot-dev
mailing list