Claes Redestad

Yes, `String` remains effectively UTF-16-encoded, with an implementation detail to encode using ISO-8859-1 when possible. Representing `String`s using UTF-8 internally has been brought up a few times. Getting there would...

@lemire interesting, thanks. I don't have any experience integrating third party code into the OpenJDK, though, but if you want to collaborate on contributing something I'd be happy to facilitate....

A classical JVM intrinsic could work great, yeah, but porting to HotSpot intrinsics will likely take significant effort. I don't have the expertise to do it, and those I know...

> @cl4es, what set of platforms do we usually consider for evaluating performance changes like A mix of x86 and aarch64 systems, some old, some new. We could run the...

Seems like I should make another attempt at fixing this. Which version of the JDK are you targetting?

Ok, I'll see what I can do.

Got it. It'd be nice if you could get some data on how much this actually matters performance-wise in your case, e.g., for startup. I think fixing it for JDK...

Error events are not sampled (java.lang.Error is instrumented to trigger the events on creation). I'd guess the overhead is small, too, and mostly happen during startup. It's an unnecessary distraction...

This looks very much like what I had in mind: implement a C1 clone intrinsic by reusing existing arraycopy code as much as possible while not going through too much...

> Looks great to me. Sorry for the pings, but we may need @rgiulietti to verify the math correctness and @cl4es to comment on whether having these 2 separate code...