v4.0.0

🎉 What's Changed

This major release aims to reduce variance in the measurements, by updating the V8 flags and forcing other optimizations.

🚀 Migration Path

This is a breaking change and it will affect your current performance baseline. But updating will improve result stability!

We recommend you create a standalone PR to update the @codspeed/* dependencies, acknowledge the potential regressions, and merge it to update the baseline on your default branch.

📚 Details

• ci: bump actions/checkout to v4 by @fargito in #39
• feat!: update V8 flags and force optimization by @adriencaccia in #41

Full Changelog: v3.1.1...v4.0.0

v3.1.1

What's Changed

• fix(vitest-plugin): ensure typings are found when using exports by @adriencaccia in #35

Full Changelog: v3.1.0...v3.1.1

v3.1.0

🎉 What's Changed

The inimum required version of vitest is now the 1.2.2.
@codspeed/vitest-plugin now supports the before* and after* hooks from vitest 🪝
This brings a benchmark authoring experience similar to writing unit and integration tests.
Take a look at the documentation on Database instrumentation to see it in action!

Since this feature is not available yet in the actual vitest's bench API, enabling the codspeedPlugin will now always activate the CodSpeed benchmark runner instead of the default one, even in local environments.

🚀 Migration Path

We recommend you create a standalone PR to update the @codspeed/* dependencies, acknowledge the potential regressions, and merge it to update the baseline on your default branch.

📚 Details

• feat/vitest hooks by @adriencaccia in #32

Full Changelog: v3.0.0...v3.1.0

v3.0.0

🎉 What's Changed

This major release brings support for the MongoDB instrumentation beta, bringing MongoDB queries insights directly in the CodSpeed report.
Documentation is coming soon...

🚀 Migration Path

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🎉!

We recommend you create a standalone PR to update the @codspeed/* dependencies, acknowledge the potential regressions, and merge it to update the baseline on your default branch.

📚 Details

• chore(ci): update to action v2 by @adriencaccia in #29
• chore/update vitest requirements by @adriencaccia in #30
• feat/integrate mongodb instrument by @adriencaccia in #22

Full Changelog: v2.3.1...v3.0.0

v2.3.1

What's changed

• fix(vitest-plugin): ensure filename is defined by @adriencaccia in 900f5c4

Full Changelog: v2.3.0...v2.3.1

v2.3.0

What's Changed

• feat/vitest plugin by @adriencaccia in #28

Full Changelog: v2.2.0...v2.3.0

v2.2.0

🎉 What's Changed

Trace generation now works out of the box with both plugins 🚀
Find the flame graphs directly in the reports on the codspeed app.

@codspeed/tinybench-plugin

• call internal setupCore and teardownCore at the right places by @adriencaccia in #23

@codspeed/benchmark.js-plugin

• call internal setupCore and teardownCore at the right places by @adriencaccia in #23
• support benchmark.js's setup and teardown options by @adriencaccia in #21

Internals

• chore(ci): split workflows by @adriencaccia in #24
• use bench.table() in all tinybench examples by @adriencaccia in #19

Full Changelog: v2.1.0...v2.2.0

v2.1.0

What's Changed

• feat(tinybench-plugin): support afterAll, afterEach, beforeAll and beforeEach by @adriencaccia in #20

Full Changelog: v2.0.0...v2.1.0

v2.0.0

🎉 What's Changed

• Flame graph generation is now supported with Node 16+, allowing more granularity in the performance reports. More details can be found in the documentation.
• Each benchmark is now warmed up a single time before each run to reduce the flakiness and the side effects related to the order in which the benchmarks are run.
• The uri of the benchmarks now contains the exact path of the file where they are added, either with bench.add(...) (with @codspeed/tinybench-plugin) or suite.add(...) (with @codspeed/benchmark.js-plugin). Before, they were all defined under the file where the bench or suite variable was instantiated. This feature enables better code splitting of your different benchmarks, read more on this in our documentation about bigger projects. This means that your existing benchmarks might be dropped in favor of new ones with correct URIs. Just acknowledge the dropping in the CodSpeed app, and you are good to go!

🚀 Migration Path

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🎉!

We recommend you create a standalone PR to update the @codspeed/* dependencies, acknowledge the potential regressions, and merge it to update the baseline on your default branch.

📚 Details

Features

• feat: expose the v8 flags from @codspeed/core by @art049 in PR #13
• feat/callgraph generation by @art049 in PR #12
• feat/support node 18 by @adriencaccia in PR #11
• feat: enable stack trampoline with v8 flags by @art049 in PR #14
• perf: warmup V8 symbols generation of the performance map by @adriencaccia in PR #15
• feat: custom perf map generation by @art049 in PR #16

Fixes

• fix: fix file path when running in ESM by @art049 in PR #10

Internals

• ci: disable benchmarks concurrency by @art049 in PR #17

---

Full Changelog: v1.1.0...v2.0.0

v1.1.0

What's Changed

@codspeed/core

• fix: explicitly zero callgrind stats at the beginning of each bench by @art049 in #8
• feat: make forced v8 optimization optional by @art049 in #9

Internals

• fix: remove version dependency from snapshot tests by @art049 in #7

Full Changelog: v1.0.2...v1.1.0