Good news, bad news

[John Rose]

On May 19, 2011, at 10:06 PM, Charles Oliver Nutter wrote:

> I've confimed that the i386 build performs just as poorly. Have not
> investigate at inlining or assembly level yet...and it's late, so I'm
> going to get some sleep.

It's likely that method handle inlining is failing.

Try -XX:+UnlockDiag* -XX:-UseRicochetFrames for differential testing.

If you can find a microbenchmark which shows differential slowdown, we can debug it.

-- John