Wohoo! Let us know when you have something we can test drive. I have some JFR recordings that show that the catch combinator indeed takes time in otherwise well behaved benchmarks (even though the catch is never entered)
/M On 12 Nov 2013, at 17:14, John Rose <john.r.r...@oracle.com> wrote: > On Nov 12, 2013, at 1:00 AM, Attila Szegedi <szege...@gmail.com> wrote: > >> Just saw this - thank you for the fix! > > I pushed Vladimir Ivanov's fix. Now we are thinking about unboxed paths for > try/catch, since you guys use them for deopt. > > — John > _______________________________________________ > mlvm-dev mailing list > mlvm-dev@openjdk.java.net > http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
_______________________________________________ mlvm-dev mailing list mlvm-dev@openjdk.java.net http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
