After discovering that I had boneheadedly multiplied the os::sleep()
values by a 1000, it fixed the stall problems I had with the Java2D
The production/non-debug builds were crashing last night, but I reduced
the optimization level to -O0 and progressively got it to run more
At -O0 the HotSpot production version runs reliably for the Java2D demo.
I'm turning off specific optimization options incrementally as I increase
the global level of optimization to find what's blowing things up.
With the no optimizations at -O0, I get within 50 percent of the
native Linux JDK in speed running Java2D. This is really encouraging.
To Unsubscribe: send mail to majo...@FreeBSD.org
with "unsubscribe freebsd-java" in the body of the message