Skip to content

fix(test): match Go output buffering and JSON streaming - #2724

Merged
xushiwei merged 4 commits into
xgo-dev:mainfrom
cpunion:codex/fix-test-output-go-compat-20261003
Oct 4, 2026
Merged

fix(test): match Go output buffering and JSON streaming#2724
xushiwei merged 4 commits into
xgo-dev:mainfrom
cpunion:codex/fix-test-output-go-compat-20261003

Conversation

@cpunion

@cpunion cpunion commented Oct 3, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator

Align llgo test output buffering and streaming with go test.

Scenario go test llgo test If a test loops forever
No package arguments Stream stdout/stderr immediately Same as Go Output printed before the loop remains visible
-v ./pkg (one package) Stream verbose output immediately Same as Go === RUN and earlier logs remain visible
-v -p=1 ./... Stream immediately; run packages sequentially Same as Go Earlier logs remain visible; later packages wait
-v ./... (multiple packages in parallel) Buffer per package; print in package order Same as Go The hung package's logs stay buffered; later packages' output may also wait
-json ./... Stream JSON events with package attribution Same as Go Earlier events and output from other running packages remain visible
Explicit package list, no -v / -json / -bench / -fuzz Buffer output; show only summaries on success, full output on failure Same as Go No progress output until the package exits or times out
-bench / -fuzz Stream immediately Same as Go Earlier output remains visible
-list Stream for one package or -p=1; otherwise buffer per package Same as Go Test bodies are not executed
-timeout Default: 10m; 0 disables the test timeout Same as Go Buffered logs become available when the timeout ends the process; -timeout=0 may leave them hidden indefinitely

Single-package output follows the same Go policy:

Command (same flags for go test) Output in both Go and LLGo If a test loops forever
llgo test (no package argument) Stream stdout/stderr immediately; t.Log needs -v Earlier stdout/stderr remains visible
llgo test ./pkg Buffer; show only a summary on success, full output on failure Logs remain buffered until exit or timeout
llgo test -v ./pkg Stream test progress and logs immediately === RUN and earlier logs remain visible
llgo test -json ./pkg Stream JSON events immediately Earlier events remain visible

fennoai[bot]
fennoai Bot reviewed Oct 3, 2026
View reviewed changes

@fennoai fennoai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Review: Go-compatible test output

Solid, well-tested rework. The new testOutputPolicy cleanly mirrors cmd/go's local-directory vs. package-list output modes, the test2json converter is a faithful upstream vendor with correct BSD attribution/LICENSE, and the package-ordered reporter is covered by targeted tests. I specifically investigated the main risk and found it not present:

  • No data race in testResultReporter. The stateful closure (results/ready/next) is only ever touched on the serialized scheduler goroutine. In test_dag.go, report(...) is invoked only from skipped/blocked/complete, all of which run on the single main loop; worker goroutines only run run, and runResults[rootIndex] is ordered through the completion channel (happens-before). In run.go runTestPrograms, report is called only from the collection loop. Writes to the shared stdout are serialized via lockedTestWriter.
  • Correctness verified: testJSONArgs normalization (stops at --, no caller-slice mutation, no duplicate -test.v), the build.go emulator-test refactor (compile-only path still no-ops via runInEmulator; executed emulator tests now join the batch), the \x16 FAIL-line prefix being consumed only inside a converter, and injection-safe exec.Command arg passing.

Notes (non-blocking)

  • One inline comment below: a wasted full-output buffer copy in JSON mode (the only concrete issue in the PR's own new code).
  • Memory characteristic (informational): in buffered/package-ordered mode, output of completed-but-not-yet-flushed packages stays resident until the head-of-line package finishes. This matches cmd/go in-order printing semantics, so it's by design — just worth awareness under skewed package durations with large output.
  • internal/test2json/test2json.go carries two inaccurate doc annotations copied verbatim from upstream Go: the NewConverter comment says framing is prefixed with ^V (\x22) (should be \x16) and references a ModeTime constant that is actually named Timestamp. Comment-only, no behavioral impact; left to you whether to diverge from upstream for a faithful-vendor sync.

No blocking issues.

Comment thread internal/build/test_output.go Outdated
@codecov

codecov Bot commented Oct 3, 2026 •
edited
Loading

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 97.30942% with 12 lines in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
internal/test2json/test2json.go 97.17% 7 Missing ⚠️
internal/build/run.go 88.46% 3 Missing ⚠️
internal/build/coverage_run.go 92.00% 2 Missing ⚠️

📢 Thoughts on this report? Let us know!

@github-actions

github-actions Bot commented Oct 3, 2026 •
edited
Loading

Copy link
Copy Markdown

LLGo WebAssembly build benchmarks

ef30d3467dc3 | workflow run | long-term charts

WebAssembly output sizes
Example, profile and compiler Wasm module vs base Generated JS glue vs base
cprintf/j32-emscripten/LLGo 154193 B 0 B / +0.0% 88742 B 0 B / +0.0%
cprintf/j32-goos-js/LLGo 152295 B 0 B / +0.0% 73165 B 0 B / +0.0%
cprintf/j64-emscripten-memory64/LLGo 141248 B 0 B / +0.0% 92630 B 0 B / +0.0%
cprintf/w32-goos-wasip1/LLGo 152848 B 0 B / +0.0% 0 B 0 B / 0.0%
cprintf/w32-wasi/LLGo 153560 B 0 B / +0.0% 0 B 0 B / 0.0%
fmtprintf/j32-emscripten/LLGo 3213561 B 0 B / +0.0% 132442 B 0 B / +0.0%
fmtprintf/j32-goos-js/Go 2526852 B 0 B / +0.0% 0 B 0 B / 0.0%
fmtprintf/j32-goos-js/LLGo 3192660 B 0 B / +0.0% 101635 B 0 B / +0.0%
fmtprintf/j64-emscripten-memory64/LLGo 2950168 B 0 B / +0.0% 139285 B 0 B / +0.0%
fmtprintf/w32-goos-wasip1/Go 2500019 B 0 B / +0.0% 0 B 0 B / 0.0%
fmtprintf/w32-goos-wasip1/LLGo 2347479 B 0 B / +0.0% 0 B 0 B / 0.0%
fmtprintf/w32-wasi/LLGo 2344141 B 0 B / +0.0% 0 B 0 B / 0.0%
j32-emscripten/LLGo 153428 B 0 B / +0.0% 88742 B 0 B / +0.0%
j32-goos-js/Go 1895533 B 0 B / +0.0% 0 B 0 B / 0.0%
j32-goos-js/LLGo 151768 B 0 B / +0.0% 73165 B 0 B / +0.0%
j64-emscripten-memory64/LLGo 140582 B 0 B / +0.0% 92630 B 0 B / +0.0%
reflectcall/j32-emscripten/LLGo 1543898 B 0 B / +0.0% 105908 B 0 B / +0.0%
reflectcall/j32-goos-js/Go 2191221 B 0 B / +0.0% 0 B 0 B / 0.0%
reflectcall/j32-goos-js/LLGo 1546371 B 0 B / +0.0% 90331 B 0 B / +0.0%
reflectcall/j64-emscripten-memory64/LLGo 1428570 B 0 B / +0.0% 111641 B 0 B / +0.0%
reflectcall/w32-goos-wasip1/Go 2205707 B 0 B / +0.0% 0 B 0 B / 0.0%
reflectcall/w32-goos-wasip1/LLGo 1276821 B 0 B / +0.0% 0 B 0 B / 0.0%
reflectcall/w32-wasi/LLGo 1274429 B 0 B / +0.0% 0 B 0 B / 0.0%
w32-goos-wasip1/Go 1909947 B 0 B / +0.0% 0 B 0 B / 0.0%
w32-goos-wasip1/LLGo 152495 B 0 B / +0.0% 0 B 0 B / 0.0%
w32-wasi/LLGo 153207 B 0 B / +0.0% 0 B 0 B / 0.0%
LLGo WebAssembly build measurements
Example and profile Build vs base
j32-emscripten 6.291 s +91.18 ms / +1.5% (worse)
j32-goos-js 6.056 s +141.8 ms / +2.4% (worse)
j64-emscripten-memory64 5.272 s +31.9 ms / +0.6% (worse)
reflectcall/w32-wasi 23.195 s -546.1 ms / -2.3% (better)
w32-goos-wasip1 4.289 s -19.75 ms / -0.5% (better)
w32-wasi 4.053 s +58.16 ms / +1.5% (worse)

Compared with 783ec4fd3d57 measured in the same runner job.

@github-actions

github-actions Bot commented Oct 3, 2026

Copy link
Copy Markdown

LLGo baseline benchmarks

ef30d3467dc3 | workflow run | long-term charts

Program measurements

Platform Workload File size vs base Text size vs base Build vs base Run vs base
Linux cprintf 8568 B 0 B / +0.0% 387 B 0 B / +0.0% 919.939 ms -16.02 ms / -1.7% (better) 1.279 ms +69.2 us / +5.7% (worse)
Linux cprintf-lto 8408 B 0 B / +0.0% 368 B 0 B / +0.0% 929.834 ms -503 us / -0.1% (better) 1.246 ms +29.85 us / +2.5% (worse)
Linux fmtprintf 4880952 B +8 B / +0.0001639% (worse) 501619 B 0 B / +0.0% 7.357 s +93.13 ms / +1.3% (worse) 3.008 ms +18.08 us / +0.6% (worse)
Linux fmtprintf-lto 3600808 B 0 B / +0.0% 438538 B 0 B / +0.0% 16.167 s -15.58 ms / -0.1% (better) 3.090 ms +93.45 us / +3.1% (worse)
Linux println 674952 B 0 B / +0.0% 16855 B 0 B / +0.0% 996.860 ms +59.74 ms / +6.4% (worse) 1.622 ms +52.78 us / +3.4% (worse)
Linux println-lto 190024 B 0 B / +0.0% 14273 B 0 B / +0.0% 1.235 s -41.04 ms / -3.2% (better) 1.622 ms +35.75 us / +2.3% (worse)
macOS cprintf 50736 B 0 B / +0.0% 4409 B 0 B / +0.0% 959.833 ms +42.56 ms / +4.6% (worse) 5.683 ms +3.237 ms / +132.4% (worse)
macOS cprintf-lto 50496 B 0 B / +0.0% 161 B 0 B / +0.0% 903.644 ms -240.9 ms / -21.0% (better) 2.347 ms -830.4 us / -26.1% (better)
macOS fmtprintf 1773456 B 0 B / +0.0% 877780 B 0 B / +0.0% 4.616 s -186.3 ms / -3.9% (better) 8.466 ms -490.5 us / -5.5% (better)
macOS fmtprintf-lto 1360928 B 0 B / +0.0% 762428 B 0 B / +0.0% 10.379 s +423 ms / +4.2% (worse) 5.828 ms +427 us / +7.9% (worse)
macOS println 99344 B 0 B / +0.0% 24216 B 0 B / +0.0% 935.817 ms -157.1 ms / -14.4% (better) 4.460 ms +488.7 us / +12.3% (worse)
macOS println-lto 83664 B 0 B / +0.0% 21457 B 0 B / +0.0% 1.128 s -20.73 ms / -1.8% (better) 5.630 ms +1.726 ms / +44.2% (worse)
Windows MinGW cprintf 651264 B 0 B / +0.0% 4550 B 0 B / +0.0% 1.676 s +40.6 ms / +2.5% (worse) 4.063 ms +118.4 us / +3.0% (worse)
Windows MinGW cprintf-lto 43520 B 0 B / +0.0% 4486 B 0 B / +0.0% 1.718 s +21.86 ms / +1.3% (worse) 3.600 ms -510.2 us / -12.4% (better)
Windows MinGW fmtprintf 5432832 B 0 B / +0.0% 600310 B 0 B / +0.0% 7.214 s +84.75 ms / +1.2% (worse) 9.310 ms +682.1 us / +7.9% (worse)
Windows MinGW fmtprintf-lto 4127232 B 0 B / +0.0% 546582 B 0 B / +0.0% 14.797 s +415.7 ms / +2.9% (worse) 8.523 ms +105.7 us / +1.3% (worse)
Windows MinGW println 706560 B 0 B / +0.0% 25190 B 0 B / +0.0% 1.671 s +26.22 ms / +1.6% (worse) 7.732 ms +663.6 us / +9.4% (worse)
Windows MinGW println-lto 208896 B 0 B / +0.0% 22054 B 0 B / +0.0% 1.955 s +38.35 ms / +2.0% (worse) 7.317 ms +15.6 us / +0.2% (worse)
Windows MinGW 386 cprintf 601600 B 0 B / +0.0% 5326 B 0 B / +0.0% 1.226 s -33.81 ms / -2.7% (better) 3.206 ms -253.2 us / -7.3% (better)
Windows MinGW 386 cprintf-lto 103424 B 0 B / +0.0% 5094 B 0 B / +0.0% 1.247 s -112 ms / -8.2% (better) 3.647 ms +502 us / +16.0% (worse)
Windows MinGW 386 fmtprintf 4745216 B 0 B / +0.0% 472478 B 0 B / +0.0% 6.185 s +440.6 ms / +7.7% (worse) 6.537 ms -955.2 us / -12.7% (better)
Windows MinGW 386 fmtprintf-lto 4148736 B 0 B / +0.0% 451114 B 0 B / +0.0% 11.102 s -617.4 ms / -5.3% (better) 6.349 ms -211.4 us / -3.2% (better)
Windows MinGW 386 println 653312 B 0 B / +0.0% 21490 B 0 B / +0.0% 1.442 s +254.1 ms / +21.4% (worse) 6.241 ms +412 us / +7.1% (worse)
Windows MinGW 386 println-lto 258560 B 0 B / +0.0% 19306 B 0 B / +0.0% 1.500 s -297.9 ms / -16.6% (better) 5.362 ms -738 us / -12.1% (better)
Windows MinGW ARM64 cprintf 661504 B 0 B / +0.0% 4408 B 0 B / +0.0% 1.863 s -57.8 ms / -3.0% (better) 5.944 ms -995.7 us / -14.3% (better)
Windows MinGW ARM64 cprintf-lto 43520 B 0 B / +0.0% 4340 B 0 B / +0.0% 1.880 s -48.74 ms / -2.5% (better) 6.725 ms -89 us / -1.3% (better)
Windows MinGW ARM64 fmtprintf 5343744 B 0 B / +0.0% 510876 B 0 B / +0.0% 7.251 s +389.3 ms / +5.7% (worse) 13.618 ms +1.683 ms / +14.1% (worse)
Windows MinGW ARM64 fmtprintf-lto 4302336 B 0 B / +0.0% 477264 B 0 B / +0.0% 13.614 s -129.9 ms / -0.9% (better) 11.963 ms +315.5 us / +2.7% (worse)
Windows MinGW ARM64 println 714240 B 0 B / +0.0% 23884 B 0 B / +0.0% 1.849 s -47.76 ms / -2.5% (better) 10.139 ms -239.9 us / -2.3% (better)
Windows MinGW ARM64 println-lto 215552 B 0 B / +0.0% 21232 B 0 B / +0.0% 2.088 s -57.13 ms / -2.7% (better) 10.415 ms -1.282 ms / -11.0% (better)
Windows MSVC cprintf 893952 B 0 B / +0.0% 65798 B 0 B / +0.0% 1.557 s +30.46 ms / +2.0% (worse) 3.952 ms +7.6 us / +0.2% (worse)
Windows MSVC cprintf-lto 289792 B 0 B / +0.0% 65734 B 0 B / +0.0% 1.563 s +11.06 ms / +0.7% (worse) 4.091 ms +321.9 us / +8.5% (worse)
Windows MSVC fmtprintf 5732864 B 0 B / +0.0% 695862 B 0 B / +0.0% 6.900 s -30.48 ms / -0.4% (better) 10.424 ms -386.8 us / -3.6% (better)
Windows MSVC fmtprintf-lto 4449792 B 0 B / +0.0% 646134 B 0 B / +0.0% 13.815 s +71.57 ms / +0.5% (worse) 9.486 ms -919.9 us / -8.8% (better)
Windows MSVC println 1017344 B 0 B / +0.0% 120854 B 0 B / +0.0% 1.546 s +14.1 ms / +0.9% (worse) 7.638 ms -740.5 us / -8.8% (better)
Windows MSVC println-lto 528384 B 0 B / +0.0% 118390 B 0 B / +0.0% 1.817 s -16.74 ms / -0.9% (better) 9.001 ms +659.5 us / +7.9% (worse)
Windows MSVC 386 cprintf 513536 B 0 B / +0.0% 3931 B 0 B / +0.0% 1.433 s -140.3 ms / -8.9% (better) 3.605 ms -1.315 ms / -26.7% (better)
Windows MSVC 386 cprintf-lto 44032 B 0 B / +0.0% 3853 B 0 B / +0.0% 1.273 s -97.26 ms / -7.1% (better) 3.444 ms -274.6 us / -7.4% (better)
Windows MSVC 386 fmtprintf 4479488 B 0 B / +0.0% 455868 B 0 B / +0.0% 6.494 s +498.1 ms / +8.3% (worse) 8.504 ms +167.6 us / +2.0% (worse)
Windows MSVC 386 fmtprintf-lto 3899392 B 0 B / +0.0% 426651 B 0 B / +0.0% 10.836 s -55.4 ms / -0.5% (better) 8.363 ms -2.851 ms / -25.4% (better)
Windows MSVC 386 println 567296 B 0 B / +0.0% 20340 B 0 B / +0.0% 1.479 s +149.8 ms / +11.3% (worse) 6.536 ms -1.093 ms / -14.3% (better)
Windows MSVC 386 println-lto 199168 B 0 B / +0.0% 18501 B 0 B / +0.0% 1.768 s +164.4 ms / +10.3% (worse) 6.264 ms -1.497 ms / -19.3% (better)
Windows MSVC ARM64 cprintf 662528 B 0 B / +0.0% 4192 B 0 B / +0.0% 1.589 s +38.84 ms / +2.5% (worse) 6.771 ms +685.4 us / +11.3% (worse)
Windows MSVC ARM64 cprintf-lto 47616 B 0 B / +0.0% 4084 B 0 B / +0.0% 1.614 s +52.91 ms / +3.4% (worse) 6.697 ms +252.4 us / +3.9% (worse)
Windows MSVC ARM64 fmtprintf 5339648 B 0 B / +0.0% 510808 B 0 B / +0.0% 6.592 s -20.45 ms / -0.3% (better) 14.600 ms +1.579 ms / +12.1% (worse)
Windows MSVC ARM64 fmtprintf-lto 4309504 B 0 B / +0.0% 477924 B 0 B / +0.0% 12.931 s +370.6 ms / +3.0% (worse) 14.272 ms +755.1 us / +5.6% (worse)
Windows MSVC ARM64 println 715264 B 0 B / +0.0% 23908 B 0 B / +0.0% 1.586 s +28.87 ms / +1.9% (worse) 12.342 ms +456.4 us / +3.8% (worse)
Windows MSVC ARM64 println-lto 220672 B 0 B / +0.0% 21380 B 0 B / +0.0% 1.801 s -7.707 ms / -0.4% (better) 11.883 ms +453.6 us / +4.0% (worse)
Core language and compiler benchmarks
Platform Benchmark ns/op vs base
Linux BenchmarkLookupPCRandom 14.560 ns/op +0.16 ns/op / +1.1% (worse)
Linux BenchmarkMergeCompilerFlags 199.500 ns/op -9.1 ns/op / -4.4% (better)
Linux BenchmarkMergeLinkerFlags 125.900 ns/op +0.7 ns/op / +0.6% (worse)
Linux BenchmarkChannelBuffered 55.480 ns/op +0.27 ns/op / +0.5% (worse)
Linux BenchmarkChannelHandoff 12968 ns/op +203 ns/op / +1.6% (worse)
Linux BenchmarkDefer 48.230 ns/op +1.03 ns/op / +2.2% (worse)
Linux BenchmarkDirectCall 1.209 ns/op +0.043 ns/op / +3.7% (worse)
Linux BenchmarkGlobalRead 1.172 ns/op +0.008 ns/op / +0.7% (worse)
Linux BenchmarkGlobalWrite 7.765 ns/op +0.007 ns/op / +0.1% (worse)
Linux BenchmarkGoroutine 23790 ns/op -790 ns/op / -3.2% (better)
Linux BenchmarkInterfaceCall 6.376 ns/op +0.544 ns/op / +9.3% (worse)
Linux BenchmarkRuntimeGetG 3.028 ns/op +0.008 ns/op / +0.3% (worse)
macOS BenchmarkLookupPCRandom 11.370 ns/op -0.96 ns/op / -7.8% (better)
macOS BenchmarkMergeCompilerFlags 84.810 ns/op -21.19 ns/op / -20.0% (better)
macOS BenchmarkMergeLinkerFlags 54.160 ns/op -13.01 ns/op / -19.4% (better)
macOS BenchmarkChannelBuffered 24.650 ns/op -0.94 ns/op / -3.7% (better)
macOS BenchmarkChannelHandoff 6800 ns/op +675 ns/op / +11.0% (worse)
macOS BenchmarkDefer 30.840 ns/op -13 ns/op / -29.7% (better)
macOS BenchmarkDirectCall 0.960 ns/op -0.0681 ns/op / -6.6% (better)
macOS BenchmarkGlobalRead 1.076 ns/op +0.008 ns/op / +0.7% (worse)
macOS BenchmarkGlobalWrite 1.129 ns/op +0.069 ns/op / +6.5% (worse)
macOS BenchmarkGoroutine 36492 ns/op +637 ns/op / +1.8% (worse)
macOS BenchmarkInterfaceCall 3.459 ns/op -0.435 ns/op / -11.2% (better)
macOS BenchmarkRuntimeGetG 2.068 ns/op 0 ns/op / +0.0%
Windows MinGW BenchmarkLookupPCRandom 11.670 ns/op +0.08 ns/op / +0.7% (worse)
Windows MinGW BenchmarkMergeCompilerFlags 567.600 ns/op -10 ns/op / -1.7% (better)
Windows MinGW BenchmarkMergeLinkerFlags 518.300 ns/op +26.3 ns/op / +5.3% (worse)
Windows MinGW BenchmarkChannelBuffered 44.890 ns/op +0.72 ns/op / +1.6% (worse)
Windows MinGW BenchmarkChannelHandoff 1288 ns/op -210 ns/op / -14.0% (better)
Windows MinGW BenchmarkDefer 66.380 ns/op +6.6 ns/op / +11.0% (worse)
Windows MinGW BenchmarkDirectCall 1.152 ns/op +0.063 ns/op / +5.8% (worse)
Windows MinGW BenchmarkGlobalRead 1.056 ns/op -0.012 ns/op / -1.1% (better)
Windows MinGW BenchmarkGlobalWrite 8.597 ns/op +0.061 ns/op / +0.7% (worse)
Windows MinGW BenchmarkGoroutine 78625 ns/op -6636 ns/op / -7.8% (better)
Windows MinGW BenchmarkInterfaceCall 6.077 ns/op +0.044 ns/op / +0.7% (worse)
Windows MinGW BenchmarkRuntimeGetG 2.003 ns/op +0.042 ns/op / +2.1% (worse)
Windows MinGW 386 BenchmarkLookupPCRandom 15.710 ns/op -0.65 ns/op / -4.0% (better)
Windows MinGW 386 BenchmarkMergeCompilerFlags 994.300 ns/op +141.2 ns/op / +16.6% (worse)
Windows MinGW 386 BenchmarkMergeLinkerFlags 901.100 ns/op -88.7 ns/op / -9.0% (better)
Windows MinGW 386 BenchmarkChannelBuffered 34.780 ns/op -1.25 ns/op / -3.5% (better)
Windows MinGW 386 BenchmarkChannelHandoff 573.800 ns/op -13.9 ns/op / -2.4% (better)
Windows MinGW 386 BenchmarkDefer 26.050 ns/op -1.73 ns/op / -6.2% (better)
Windows MinGW 386 BenchmarkDirectCall 0.887 ns/op -0.0046 ns/op / -0.5% (better)
Windows MinGW 386 BenchmarkGlobalRead 0.888 ns/op -0.007 ns/op / -0.8% (better)
Windows MinGW 386 BenchmarkGlobalWrite 7.983 ns/op -0.023 ns/op / -0.3% (better)
Windows MinGW 386 BenchmarkGoroutine 51329 ns/op +2148 ns/op / +4.4% (worse)
Windows MinGW 386 BenchmarkInterfaceCall 4.441 ns/op -0.029 ns/op / -0.6% (better)
Windows MinGW 386 BenchmarkRuntimeGetG 0.934 ns/op -0.0015 ns/op / -0.2% (better)
Windows MinGW ARM64 BenchmarkLookupPCRandom 12.140 ns/op +0.08 ns/op / +0.7% (worse)
Windows MinGW ARM64 BenchmarkMergeCompilerFlags 580.600 ns/op +12.1 ns/op / +2.1% (worse)
Windows MinGW ARM64 BenchmarkMergeLinkerFlags 545.600 ns/op +13.4 ns/op / +2.5% (worse)
Windows MinGW ARM64 BenchmarkChannelBuffered 38.950 ns/op -0.54 ns/op / -1.4% (better)
Windows MinGW ARM64 BenchmarkChannelHandoff 1767 ns/op +329 ns/op / +22.9% (worse)
Windows MinGW ARM64 BenchmarkDefer 55.780 ns/op +3.74 ns/op / +7.2% (worse)
Windows MinGW ARM64 BenchmarkDirectCall 0.663 ns/op -0.0001 ns/op / -0.01507% (better)
Windows MinGW ARM64 BenchmarkGlobalRead 0.664 ns/op +0.0004 ns/op / +0.1% (worse)
Windows MinGW ARM64 BenchmarkGlobalWrite 0.739 ns/op +0.0022 ns/op / +0.3% (worse)
Windows MinGW ARM64 BenchmarkGoroutine 59913 ns/op +511 ns/op / +0.9% (worse)
Windows MinGW ARM64 BenchmarkInterfaceCall 4.144 ns/op 0 ns/op / +0.0%
Windows MinGW ARM64 BenchmarkRuntimeGetG 1.784 ns/op +0.014 ns/op / +0.8% (worse)
Windows MSVC BenchmarkLookupPCRandom 11.640 ns/op -0.09 ns/op / -0.8% (better)
Windows MSVC BenchmarkMergeCompilerFlags 553.900 ns/op -22.4 ns/op / -3.9% (better)
Windows MSVC BenchmarkMergeLinkerFlags 495.300 ns/op -14.1 ns/op / -2.8% (better)
Windows MSVC BenchmarkChannelBuffered 44.370 ns/op -5.01 ns/op / -10.1% (better)
Windows MSVC BenchmarkChannelHandoff 1792 ns/op +955.2 ns/op / +114.1% (worse)
Windows MSVC BenchmarkDefer 56.460 ns/op -6.75 ns/op / -10.7% (better)
Windows MSVC BenchmarkDirectCall 1.240 ns/op -0.008 ns/op / -0.6% (better)
Windows MSVC BenchmarkGlobalRead 1.067 ns/op -0.071 ns/op / -6.2% (better)
Windows MSVC BenchmarkGlobalWrite 8.711 ns/op +0.018 ns/op / +0.2% (worse)
Windows MSVC BenchmarkGoroutine 78140 ns/op +1373 ns/op / +1.8% (worse)
Windows MSVC BenchmarkInterfaceCall 6.682 ns/op -0.676 ns/op / -9.2% (better)
Windows MSVC BenchmarkRuntimeGetG 1.982 ns/op 0 ns/op / +0.0%
Windows MSVC 386 BenchmarkLookupPCRandom 15.600 ns/op +0.23 ns/op / +1.5% (worse)
Windows MSVC 386 BenchmarkMergeCompilerFlags 1068 ns/op +311.4 ns/op / +41.2% (worse)
Windows MSVC 386 BenchmarkMergeLinkerFlags 1017 ns/op +376.9 ns/op / +58.9% (worse)
Windows MSVC 386 BenchmarkChannelBuffered 35.030 ns/op +0.19 ns/op / +0.5% (worse)
Windows MSVC 386 BenchmarkChannelHandoff 576.900 ns/op +68.1 ns/op / +13.4% (worse)
Windows MSVC 386 BenchmarkDefer 27.520 ns/op +0.92 ns/op / +3.5% (worse)
Windows MSVC 386 BenchmarkDirectCall 0.901 ns/op -0.0091 ns/op / -1.0% (better)
Windows MSVC 386 BenchmarkGlobalRead 0.892 ns/op +0.0009 ns/op / +0.1% (worse)
Windows MSVC 386 BenchmarkGlobalWrite 8.004 ns/op -0.026 ns/op / -0.3% (better)
Windows MSVC 386 BenchmarkGoroutine 52740 ns/op -2718 ns/op / -4.9% (better)
Windows MSVC 386 BenchmarkInterfaceCall 4.648 ns/op +0.132 ns/op / +2.9% (worse)
Windows MSVC 386 BenchmarkRuntimeGetG 0.942 ns/op +0.0031 ns/op / +0.3% (worse)
Windows MSVC ARM64 BenchmarkLookupPCRandom 12.090 ns/op 0 ns/op / +0.0%
Windows MSVC ARM64 BenchmarkMergeCompilerFlags 576.600 ns/op -2.1 ns/op / -0.4% (better)
Windows MSVC ARM64 BenchmarkMergeLinkerFlags 531 ns/op -9.5 ns/op / -1.8% (better)
Windows MSVC ARM64 BenchmarkChannelBuffered 38.590 ns/op +1.04 ns/op / +2.8% (worse)
Windows MSVC ARM64 BenchmarkChannelHandoff 1801 ns/op -190 ns/op / -9.5% (better)
Windows MSVC ARM64 BenchmarkDefer 62.310 ns/op +0.74 ns/op / +1.2% (worse)
Windows MSVC ARM64 BenchmarkDirectCall 0.663 ns/op +0.0001 ns/op / +0.01507% (worse)
Windows MSVC ARM64 BenchmarkGlobalRead 0.663 ns/op -0.0004 ns/op / -0.1% (better)
Windows MSVC ARM64 BenchmarkGlobalWrite 3.801 ns/op +0.005 ns/op / +0.1% (worse)
Windows MSVC ARM64 BenchmarkGoroutine 58569 ns/op +751 ns/op / +1.3% (worse)
Windows MSVC ARM64 BenchmarkInterfaceCall 4.150 ns/op +0.007 ns/op / +0.2% (worse)
Windows MSVC ARM64 BenchmarkRuntimeGetG 1.769 ns/op -0.001 ns/op / -0.1% (better)
Timer runtime benchmarks
Platform Operation and runtime ns/op vs base
Linux AfterFuncZeroDelivery/Go 902.200 ns/op -15.7 ns/op / -1.7% (better)
Linux AfterFuncZeroDelivery/LLGo 35762 ns/op -1136 ns/op / -3.1% (better)
Linux CreateStop/Go 289.700 ns/op -0.9 ns/op / -0.3% (better)
Linux CreateStop/LLGo 1863 ns/op -34 ns/op / -1.8% (better)
Linux RearmStopped/Go 115.700 ns/op -0.2 ns/op / -0.2% (better)
Linux RearmStopped/LLGo 1195 ns/op +29 ns/op / +2.5% (worse)
Linux ResetActive/Go 68.580 ns/op -0.1 ns/op / -0.1% (better)
Linux ResetActive/LLGo 869.200 ns/op +0.5 ns/op / +0.1% (worse)
Linux ResetHeap1024/Go 67.070 ns/op -0.05 ns/op / -0.1% (better)
Linux ResetHeap1024/LLGo 173.900 ns/op -2.5 ns/op / -1.4% (better)
macOS AfterFuncZeroDelivery/Go 364.700 ns/op -74.8 ns/op / -17.0% (better)
macOS AfterFuncZeroDelivery/LLGo 64809 ns/op -9296 ns/op / -12.5% (better)
macOS CreateStop/Go 115.100 ns/op -47 ns/op / -29.0% (better)
macOS CreateStop/LLGo 424.800 ns/op -54 ns/op / -11.3% (better)
macOS RearmStopped/Go 50.890 ns/op -4.79 ns/op / -8.6% (better)
macOS RearmStopped/LLGo 310.500 ns/op +18.2 ns/op / +6.2% (worse)
macOS ResetActive/Go 37.450 ns/op -15.37 ns/op / -29.1% (better)
macOS ResetActive/LLGo 146.500 ns/op -3 ns/op / -2.0% (better)
macOS ResetHeap1024/Go 37.560 ns/op -4.42 ns/op / -10.5% (better)
macOS ResetHeap1024/LLGo 74.470 ns/op -15.25 ns/op / -17.0% (better)
Windows MinGW AfterFuncZeroDelivery/Go 697.400 ns/op +9.5 ns/op / +1.4% (worse)
Windows MinGW AfterFuncZeroDelivery/LLGo 144310 ns/op +2097 ns/op / +1.5% (worse)
Windows MinGW CreateStop/Go 194.600 ns/op +5 ns/op / +2.6% (worse)
Windows MinGW CreateStop/LLGo 643.200 ns/op -12.8 ns/op / -2.0% (better)
Windows MinGW RearmStopped/Go 70.610 ns/op +0.46 ns/op / +0.7% (worse)
Windows MinGW RearmStopped/LLGo 304.400 ns/op -27.1 ns/op / -8.2% (better)
Windows MinGW ResetActive/Go 31.300 ns/op +0.19 ns/op / +0.6% (worse)
Windows MinGW ResetActive/LLGo 196.400 ns/op -13.2 ns/op / -6.3% (better)
Windows MinGW ResetHeap1024/Go 31.640 ns/op +0.2 ns/op / +0.6% (worse)
Windows MinGW ResetHeap1024/LLGo 136.500 ns/op +3.3 ns/op / +2.5% (worse)
Windows MinGW 386 AfterFuncZeroDelivery/Go 585.500 ns/op -46 ns/op / -7.3% (better)
Windows MinGW 386 AfterFuncZeroDelivery/LLGo 100218 ns/op -6934 ns/op / -6.5% (better)
Windows MinGW 386 CreateStop/Go 153.800 ns/op -13.2 ns/op / -7.9% (better)
Windows MinGW 386 CreateStop/LLGo 436.200 ns/op +24.3 ns/op / +5.9% (worse)
Windows MinGW 386 RearmStopped/Go 58.530 ns/op -5.3 ns/op / -8.3% (better)
Windows MinGW 386 RearmStopped/LLGo 249.100 ns/op -25.2 ns/op / -9.2% (better)
Windows MinGW 386 ResetActive/Go 27.970 ns/op -2.53 ns/op / -8.3% (better)
Windows MinGW 386 ResetActive/LLGo 541.900 ns/op -220.2 ns/op / -28.9% (better)
Windows MinGW 386 ResetHeap1024/Go 28.020 ns/op -2.45 ns/op / -8.0% (better)
Windows MinGW 386 ResetHeap1024/LLGo 114.900 ns/op -11.7 ns/op / -9.2% (better)
Windows MinGW ARM64 AfterFuncZeroDelivery/Go 664.500 ns/op -1.3 ns/op / -0.2% (better)
Windows MinGW ARM64 AfterFuncZeroDelivery/LLGo 144680 ns/op +8763 ns/op / +6.4% (worse)
Windows MinGW ARM64 CreateStop/Go 195.900 ns/op -1.9 ns/op / -1.0% (better)
Windows MinGW ARM64 CreateStop/LLGo 406.100 ns/op +5.6 ns/op / +1.4% (worse)
Windows MinGW ARM64 RearmStopped/Go 70.610 ns/op -0.05 ns/op / -0.1% (better)
Windows MinGW ARM64 RearmStopped/LLGo 259.900 ns/op +4.6 ns/op / +1.8% (worse)
Windows MinGW ARM64 ResetActive/Go 31.050 ns/op +0.02 ns/op / +0.1% (worse)
Windows MinGW ARM64 ResetActive/LLGo 130 ns/op -9.1 ns/op / -6.5% (better)
Windows MinGW ARM64 ResetHeap1024/Go 31.060 ns/op -0.06 ns/op / -0.2% (better)
Windows MinGW ARM64 ResetHeap1024/LLGo 124.400 ns/op +0.1 ns/op / +0.1% (worse)
Windows MSVC AfterFuncZeroDelivery/Go 708.300 ns/op +0.1 ns/op / +0.01412% (worse)
Windows MSVC AfterFuncZeroDelivery/LLGo 147874 ns/op +1577 ns/op / +1.1% (worse)
Windows MSVC CreateStop/Go 192.300 ns/op -7.1 ns/op / -3.6% (better)
Windows MSVC CreateStop/LLGo 854 ns/op -97.8 ns/op / -10.3% (better)
Windows MSVC RearmStopped/Go 71.640 ns/op +0.07 ns/op / +0.1% (worse)
Windows MSVC RearmStopped/LLGo 387.400 ns/op +36.7 ns/op / +10.5% (worse)
Windows MSVC ResetActive/Go 31.760 ns/op +0.03 ns/op / +0.1% (worse)
Windows MSVC ResetActive/LLGo 175.800 ns/op -18.2 ns/op / -9.4% (better)
Windows MSVC ResetHeap1024/Go 32.020 ns/op -0.01 ns/op / -0.03122% (better)
Windows MSVC ResetHeap1024/LLGo 132.200 ns/op +1.3 ns/op / +1.0% (worse)
Windows MSVC 386 AfterFuncZeroDelivery/Go 619.900 ns/op +21.3 ns/op / +3.6% (worse)
Windows MSVC 386 AfterFuncZeroDelivery/LLGo 104415 ns/op -649 ns/op / -0.6% (better)
Windows MSVC 386 CreateStop/Go 159.700 ns/op +4.6 ns/op / +3.0% (worse)
Windows MSVC 386 CreateStop/LLGo 398.300 ns/op +84 ns/op / +26.7% (worse)
Windows MSVC 386 RearmStopped/Go 60.790 ns/op +1.89 ns/op / +3.2% (worse)
Windows MSVC 386 RearmStopped/LLGo 223.300 ns/op -8.9 ns/op / -3.8% (better)
Windows MSVC 386 ResetActive/Go 29.230 ns/op +1.05 ns/op / +3.7% (worse)
Windows MSVC 386 ResetActive/LLGo 737.200 ns/op -26.4 ns/op / -3.5% (better)
Windows MSVC 386 ResetHeap1024/Go 29.490 ns/op +1.42 ns/op / +5.1% (worse)
Windows MSVC 386 ResetHeap1024/LLGo 115.100 ns/op +3.7 ns/op / +3.3% (worse)
Windows MSVC ARM64 AfterFuncZeroDelivery/Go 667.300 ns/op -10.8 ns/op / -1.6% (better)
Windows MSVC ARM64 AfterFuncZeroDelivery/LLGo 165773 ns/op +1125 ns/op / +0.7% (worse)
Windows MSVC ARM64 CreateStop/Go 206.400 ns/op +5.6 ns/op / +2.8% (worse)
Windows MSVC ARM64 CreateStop/LLGo 461.700 ns/op +32.3 ns/op / +7.5% (worse)
Windows MSVC ARM64 RearmStopped/Go 70.560 ns/op -0.06 ns/op / -0.1% (better)
Windows MSVC ARM64 RearmStopped/LLGo 273.200 ns/op -4.6 ns/op / -1.7% (better)
Windows MSVC ARM64 ResetActive/Go 31.070 ns/op +0.17 ns/op / +0.6% (worse)
Windows MSVC ARM64 ResetActive/LLGo 153.900 ns/op +8.1 ns/op / +5.6% (worse)
Windows MSVC ARM64 ResetHeap1024/Go 31.070 ns/op +0.17 ns/op / +0.6% (worse)
Windows MSVC ARM64 ResetHeap1024/LLGo 135.700 ns/op -1.6 ns/op / -1.2% (better)

Compared with 783ec4fd3d57 measured in the same runner job.

@xushiwei
xushiwei merged commit 5f1f138 into xgo-dev:main Oct 4, 2026
98 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@fennoai fennoai[bot] fennoai[bot] left review comments

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@cpunion @xushiwei