Benoit Daloze

I changed the default to BucketsHashStore in https://github.com/oracle/truffleruby/pull/3456 until thread-safe Hash lands.

Thank you for the report. I wonder if the problem could be related to the fact hexapdf creates so many Fibers: https://github.com/Shopify/yjit-bench/pull/47#issuecomment-1478313399 If the Fibers are not executed to completion...

> we have to check, I don't know if that happens for the hexapdf benchmark That's not the case, there is always 2 Fibers and 2 Threads when running this...

I can reproduce it on 23.0.0 both in Native and JVM mode (on Oracle GraalVM). Note that for benchmarking one should use JVM mode: https://github.com/oracle/truffleruby/blob/master/doc/user/benchmarking.md With `--engine.TraceCompilation` I noticed we...

On master there are many compilations of `Truffle::Splitter.add_substring` but it does stabilize.

Thank you for the report, we'll try to take a look. I might wait after https://github.com/oracle/truffleruby/issues/3118 as that might possibly solve it.

Thank you for the great issue report. So these allocations are from C extensions and for native pointers, as used by the gems. A Cleaner thread should take care of...

Thank you for the report. I wonder if this is architecture-specific or so, we'll check.

Does it work if you increase it like the message suggests? It should be something like `--vm.XX:NonProfiledCodeHeapSize=512m`, in TRUFFLERUBYOPT or RUBYOPT or as a command line argument. I can't double-check...

Could you share the siege command line and which database & gem is used? Even better would be a repository I can just use to reproduce.