We added Perfetto tracing support to ZJIT so we could visualize and > query slow events. Take a look at the pretty colors and see how you can add > this to your > system too.

The RubyLA blog offers insights, tutorials, and community updates for Ruby developers in Los Angeles and beyond. Developers can explore Ruby language features, web development frameworks, and best practices for building Ruby applications. Additionally, the blog covers Ruby meetups, events, and job opportunities in the Los Angeles area, fostering collaboration and networking within the local Ruby community.

RUBYLAND

The ZJIT team (Ruby's new JIT compiler) integrated Perfetto tracing to visualize and analyze JIT side-exits — moments when compiled code falls back to the interpreter. The post explains why side-exits matter for performance, shows how --zjit-stats provides aggregate counters but lacks spatial/temporal context, and demonstrates how Perfetto's UI and SQL interface reveal hotspots in Rails app code. Implementation details cover writing traces in Chromium JSON format (which ballooned to 8GB) versus the more compact Fuchsia Trace Format (FXT) with string interning (~100MB), plus sampling via --zjit-trace-exits-sample-rate to reduce output further. The Rust-based trace writer is linked on GitHub.

Using Perfetto in ZJIT