Benoit Daloze

Fixing this turns out to be rather complicated to not break tests, and writing a portable test for this seems pretty hard since `nohup` behave differently on different OS (some...

On CRuby it's: ``` $ ruby -e "puts RUBY_DESCRIPTION; break" |& cat -e:1: Invalid break -e: compile error (SyntaxError) ``` So probably something to handle in the parser or translator....

That's a slightly different case actually. It seems `break` is actually forbidden in `eval`, be it a break-from-call-with-block or a break-from-while: ``` $ ruby -e 'self.then { eval "break" }'...

Note: https://github.com/oracle/truffleruby/issues/1672#issuecomment-797446928 is the issue with `pry-byebug`.

> but was hoping than the memory utilization would drop instead It is expected TruffleRuby (and JRuby) use more memory than CRuby. After all they have a JIT and that...

The test suites of both `ethon` and `typhoeus` pass on TruffleRuby, and in fact TruffleRuby is in CI for both of those gems. And as Chris replied, the test suite...

I've tried to reproduce and I could but not reliably. I ran on JVM with some debug logging (using the new `--engine.TraceStackTraceInterval`): ``` ruby --experimental-options --engine.TraceStackTraceInterval=500 --log.level=INFO -S wpscan -v...

Tracked internally as GR-30480, it could potentially be a Truffle NFI issue.

@aardvark179 is working on implementing `rb_io_wait`. The current conflict is unfortunate but expected: we import headers from CRuby, changed as little as possible to keep it maintainable. `have_func` correctly detects...

Maybe a simple workaround here is to do some renames/prefix for the polyfill? I think that would be the fastest to solve this specific issue. (We're also working on the...