ci: parallelize the repeated test job; add concurrency cancellation

[ReneWerner87]

Motivation

Reduce CI wall-clock per commit without changing results, output or coverage.

Key finding: the Test workflow's long pole is the repeated job, and it is dominated by fixed test latency (133 time.Sleep calls: 14x 3s, 17x 1s, ...), not CPU. Removing -race alone saved only ~33s. The effective lever is parallelism.

Changes

test.yml

• Concurrency group that cancels superseded in-progress PR runs (never cancels main).
• Coverage computed only on the ubuntu/1.25.x job (the only one that uploads it); the other 5 matrix combinations skip -coverprofile -covermode=atomic.
• repeated job sharded into 3 parallel matrix shards of -count=5 each = still 15 shuffled runs total (3 distinct shuffle seeds), -race kept.

benchmark.yml

• Same PR-only concurrency cancellation. Otherwise unchanged (single job, default benchtime).

Measured impact (real run, not estimate)

Job	Before	After
Test repeated	~9m41s	~3m49s (slowest of 3 shards)
Test workflow overall	~10 min	~4 min

Functionality preserved

• Coverage: identical, still uploaded only from ubuntu/1.25.x.
• Race detection: unit matrix unchanged; repeated keeps -race.
• Flaky/order detection: 15 shuffled runs total, unchanged.

Why the benchmark job is NOT sharded

An attempt to shard benchmarks across 4 runners cut wall-clock to ~4min but failed the comparison: a no-code-change commit reported a uniform ~1.5-2.5x ns/op "regression" with identical B/op/allocs/op, i.e. pure CPU variance. The benchmark comparison relies on a cpu-model cache key so it only compares same-CPU results; spreading work across GitHub's heterogeneous runners breaks that guarantee. Benchmark therefore stays a single job (correct over fast).

🤖 Generated with Claude Code

[gemini-code-assist]

Note

Gemini is unable to generate a review for this pull request due to the file types involved not being currently supported.

[coderabbitai]

Important

Review skipped

Review was skipped due to path filters

⛔ Files ignored due to path filters (2)

• .github/workflows/benchmark.yml is excluded by !**/*.yml
• .github/workflows/test.yml is excluded by !**/*.yml

CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including **/dist/** will override the default block on the dist directory, by removing the pattern from both the lists.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro

Run ID: 12a54adf-0fa8-44f2-93b7-88e90c7c7b03

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch ci/speed-up-test-and-benchmark

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Copilot]

Pull request overview

This PR optimizes CI wall-clock time for Fiber’s test and benchmark workflows while keeping coverage upload behavior and existing validation intent intact.

Changes:

• Add PR-only workflow concurrency cancellation to avoid wasted CI on superseded commits.
• Pin gotestsum to v1.13.0 and reduce unnecessary coverage instrumentation to the single job that uploads coverage.
• Speed up benchmark runs by reducing go test benchtime.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
.github/workflows/test.yml	Cancels superseded PR runs, pins gotestsum, computes coverage only on the upload job, and removes -race from the repeated/flakiness job.
.github/workflows/benchmark.yml	Cancels superseded PR runs and reduces benchmark runtime via -benchtime=100ms.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.33%. Comparing base (86397b0) to head (a55fb1f).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4309   +/-   ##
=======================================
  Coverage   91.33%   91.33%           
=======================================
  Files         132      132           
  Lines       13089    13089           
=======================================
  Hits        11955    11955           
  Misses        716      716           
  Partials      418      418           

Flag	Coverage Δ
unittests	91.33% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.50.

Benchmark suite	Current: 5fad032	Previous: 86397b0	Ratio
Benchmark_NewError (github.com/gofiber/fiber/v3)	85.77 ns/op 24 B/op 1 allocs/op	33.9 ns/op 24 B/op 1 allocs/op	2.53
Benchmark_NewError (github.com/gofiber/fiber/v3) - ns/op	85.77 ns/op	33.9 ns/op	2.53
Benchmark_Ctx_AutoFormat_JSON (github.com/gofiber/fiber/v3)	428.9 ns/op 16 B/op 1 allocs/op	285.3 ns/op 16 B/op 1 allocs/op	1.50
Benchmark_Ctx_AutoFormat_JSON (github.com/gofiber/fiber/v3) - ns/op	428.9 ns/op	285.3 ns/op	1.50
Benchmark_Ctx_Write (github.com/gofiber/fiber/v3)	19.14 ns/op 70 B/op 0 allocs/op	12.74 ns/op 76 B/op 0 allocs/op	1.50
Benchmark_Ctx_Write (github.com/gofiber/fiber/v3) - ns/op	19.14 ns/op	12.74 ns/op	1.50
Benchmark_App_RebuildTree (github.com/gofiber/fiber/v3)	25290 ns/op 15960 B/op 48 allocs/op	16340 ns/op 15960 B/op 48 allocs/op	1.55
Benchmark_App_RebuildTree (github.com/gofiber/fiber/v3) - ns/op	25290 ns/op	16340 ns/op	1.55
Benchmark_Router_WithCompression (github.com/gofiber/fiber/v3)	320.6 ns/op 8 B/op 1 allocs/op	209.5 ns/op 8 B/op 1 allocs/op	1.53
Benchmark_Router_WithCompression (github.com/gofiber/fiber/v3) - ns/op	320.6 ns/op	209.5 ns/op	1.53
Benchmark_PaginateMiddlewareWithCustomConfig (github.com/gofiber/fiber/v3/middleware/paginate)	21657 ns/op 11492 B/op 43 allocs/op	14378 ns/op 11474 B/op 43 allocs/op	1.51
Benchmark_PaginateMiddlewareWithCustomConfig (github.com/gofiber/fiber/v3/middleware/paginate) - ns/op	21657 ns/op	14378 ns/op	1.51
Benchmark_Rewrite/Next_always_true (github.com/gofiber/fiber/v3/middleware/rewrite)	556.1 ns/op 24 B/op 2 allocs/op	360.7 ns/op 24 B/op 2 allocs/op	1.54
Benchmark_Rewrite/Next_always_true (github.com/gofiber/fiber/v3/middleware/rewrite) - ns/op	556.1 ns/op	360.7 ns/op	1.54
Benchmark_Bind_URI (github.com/gofiber/fiber/v3)	3593 ns/op 768 B/op 32 allocs/op	2349 ns/op 768 B/op 32 allocs/op	1.53
Benchmark_Bind_URI (github.com/gofiber/fiber/v3) - ns/op	3593 ns/op	2349 ns/op	1.53
Benchmark_Ctx_Cookie (github.com/gofiber/fiber/v3)	205.9 ns/op 0 B/op 0 allocs/op	135.7 ns/op 0 B/op 0 allocs/op	1.52
Benchmark_Ctx_Cookie (github.com/gofiber/fiber/v3) - ns/op	205.9 ns/op	135.7 ns/op	1.52
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckSimple (github.com/gofiber/fiber/v3)	68.96 ns/op 8 B/op 1 allocs/op	44.71 ns/op 8 B/op 1 allocs/op	1.54
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckSimple (github.com/gofiber/fiber/v3) - ns/op	68.96 ns/op	44.71 ns/op	1.54
Benchmark_Ctx_IsProxyTrusted/WithProxyCheck (github.com/gofiber/fiber/v3)	76.16 ns/op 8 B/op 1 allocs/op	50.32 ns/op 8 B/op 1 allocs/op	1.51
Benchmark_Ctx_IsProxyTrusted/WithProxyCheck (github.com/gofiber/fiber/v3) - ns/op	76.16 ns/op	50.32 ns/op	1.51
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckAllowPrivate (github.com/gofiber/fiber/v3)	81.49 ns/op 8 B/op 1 allocs/op	52.33 ns/op 8 B/op 1 allocs/op	1.56
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckAllowPrivate (github.com/gofiber/fiber/v3) - ns/op	81.49 ns/op	52.33 ns/op	1.56
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckAllowAll (github.com/gofiber/fiber/v3)	103.8 ns/op 8 B/op 1 allocs/op	66.56 ns/op 8 B/op 1 allocs/op	1.56
Benchmark_Ctx_IsProxyTrusted/WithProxyCheckAllowAll (github.com/gofiber/fiber/v3) - ns/op	103.8 ns/op	66.56 ns/op	1.56
Benchmark_Router_Handler_Unescape (github.com/gofiber/fiber/v3)	181.6 ns/op 0 B/op 0 allocs/op	116 ns/op 0 B/op 0 allocs/op	1.57
Benchmark_Router_Handler_Unescape (github.com/gofiber/fiber/v3) - ns/op	181.6 ns/op	116 ns/op	1.57
Benchmark_Compress_Levels/Brotli_LevelBestCompression (github.com/gofiber/fiber/v3/middleware/compress) - B/op	17 B/op	11 B/op	1.55
Benchmark_Communication_Flow (github.com/gofiber/fiber/v3)	81.98 ns/op 0 B/op 0 allocs/op	49.1 ns/op 0 B/op 0 allocs/op	1.67
Benchmark_Communication_Flow (github.com/gofiber/fiber/v3) - ns/op	81.98 ns/op	49.1 ns/op	1.67

This comment was automatically generated by workflow using github-action-benchmark.