Francesco Nigro

I can see why @cescoffier thanks for sharing, that helped: It is likely due to the presence of cleaner's API accessible, see https://github.com/netty/netty/blob/4.1/common/src/main/java/io/netty/util/internal/PlatformDependent.java#L202 which transitively depends by Unsafe presence, see...

@cescoffier > Well, and it will likely be different between JVM and native. which is kind of surprising (to me - because I didn't profiled native image apps enough -...

@cescoffier @zakkak I'm a bit concerned about > It's not what we saw with Julien yesterday, where we got prefer direct true in some cases and false in other cases...

PTAL @cescoffier I've added a new property `vertx.reuseNettyAllocators`, which is `false` by default, to preserve the existing behaviour. When merged in Quarkus we can set this property to `true` saving...

Given that it depends by a static final property, if I set it to try it, is going to impact all the tests after it, randomly (because it depends how...

@vietj Any other concern for this @cescoffier @vietj ?

Agree on your points @wfouche as well: as @lampajr said we are in the process of improving the load generator scheduler to better handle submillis intervals AND as you said,...

I am considering to add a jmh module to Hyperfoil, for another enhancement, and it could be a nice playground to evaluate different serialization mechanisms

yes, actually I see this could be "partially" applied here -> https://github.com/FasterXML/jackson-core/blob/2.19/src/main/java/com/fasterxml/jackson/core/json/UTF8JsonGenerator.java#L1758-L1765 What you got here is exactly the same as https://lemire.me/blog/2024/10/14/table-lookups-are-efficient/ which I have improved in https://github.com/lemire/Code-used-on-Daniel-Lemire-s-blog/blob/6236934cffb0de1d3f17dca43d242b735a4a2125/2024/10/14/src/main/java/me/lemire/MyBenchmark.java#L258-L269 i.e. -...

yep it looks like it could be part of such, but please :pray: suggest to that user to use JMH and not handrolled bench... :cry: