Tobias Meggendorfer

> Incompatible Just to add: Yes, agree, I meant noting that they are logically incompatible, i.e. if you provide `ipv6_suffix` then setting `provider_ip` is unneeded / does not make sense....

Since I brewed this - jeah probably fine. The issue is that the output format of jdeps is not fixed. Maybe one can do ` ?` instead of adding just...

Did you profile these changes? (With a proper JMH setup) I'm doubtful that these changes have any noticeable performance improvements (the JVM is incredibly good at figuring out these things)....

> Based on the benchmark results, caching the array object and size of the list reduces iteration time by over 70% compared to not caching. ... on this benchmark that...

PS: I don't want to sound overly negative! I've just seen way too often that people try to outsmart a compiler, which more often than not actually is counterproductive.

> It's interesting. I ran the very same benchmark out of curiosity on Hotspot 17.0.2 both on Mac M1 and on x86 and see the same improvements as @mouse0w0 when...

@szarnekow @mouse0w0 Could you try this benchmark variant: ```java @State(Scope.Benchmark) @Warmup(time = 2, iterations = 2) @Measurement(time = 2, iterations = 4) @Fork(1) @BenchmarkMode(Mode.AverageTime) @OutputTimeUnit(TimeUnit.NANOSECONDS) public class IterationBenchmark { @Param({"100000"})...

Ok this is very surprising. Let me try if I can reproduce with `HotSpot` and, if so, if it is hotspot being slow or openjdk being fast. EDIT: Nope. `JDK...

> I think the reason for this result is that the benchmark variant only consumes the array object itself, rather than consuming the data within the array object. But the...

> But my benchmark aligns with the function of forEach, obtaining and consuming one element at a time. We need to clarify what the JVM is doing. Absolutely. This was...