Skip to content

[perf-improver] Monthly Activity 2026-10 #11691

Closed

Description

@github-actions

Activity for October 2026

Suggested Actions for Maintainer

No suggested actions at this time.

Performance Opportunities Backlog

  • PrivateObject.Helpers.cs BuildGenericMethodCacheForType (net-framework-only) - rebuilds cache per PrivateObject instance construction. Medium risk, low priority.
  • TestExecutionManager.ParallelExecution.cs per-test array wrapping for parallel scheduling chunks - inherent design, medium risk/effort.
  • AggregatedConfiguration indexer scan - no per-key memoization, but consumers already cache on first access. Not a real hot path.
  • ServerTestHost.RequestExecution.cs Select+ToArray - per-request, not per-test.
  • RetryArtifactProcessor.cs GroupBy/Count double-enumeration - measured with real data: GroupBy ~1.5us/op vs manual dict ~0.5-1.1us/op, called O(1) times per run. Confirmed not worth fixing.
  • RetryOrchestrator.ArtifactRecovery.cs Any() per-manifest-line - low volume.

All items above have been re-verified multiple times across runs and remain low priority / not worth fixing. The codebase has been in a stable, well-optimized state for runtime hot paths since the Assert.That fix (PR #11561, merged 2026-09-27) - 29 consecutive scan cycles with no new fixable hot-path finding as of this run.

Discovered Commands

./build.sh                              # restore + build (Debug)
./build.sh -test                        # build + unit tests
./build.sh -pack                        # produce NuGet packages
./build.sh -pack -test -integrationTest # full suite (slow)
# SDK: .NET 12 preview (bootstrapped via ./build.sh into .dotnet/)
# Note: MSTestAdapter.UnitTests and other net48-targeting tests can't run in Linux env (no .NET Framework SDK)

Benchmark/perf infra: test/Performance/MSTest.Performance.Runner (end-to-end process timing), test/Performance/MSTest.Performance.Benchmarks (BenchmarkDotNet allocation microbenchmarks), .github/workflows/perf-timing-nightly.yml (nightly run with rolling-baseline regression detection for both end-to-end timings via compare_perf_timings.py and allocation microbenchmarks via compare_benchmark_allocations.py, added PR #11743, merged 2026-10-05).

Run History

2026-10-20 - Run

2026-10-19 - Run

2026-10-18 13:33 UTC - Run

2026-10-17 13:32 UTC - Run

2026-10-16 13:36 UTC - Run

Earlier September-October 2026 (condensed)

🤖 Automated content by GitHub Copilot. Generated by the Perf Improver workflow. · copilot · auto · 107.8 AIC · ⌖ 13.4 AIC · ⊞ 19.6K · [◷]( · ◷)
Comment /perf-assist to run again

Add this agentic workflow to your repo

To install this agentic workflow, run

gh aw add githubnext/agentics/workflows/perf-improver.md@main
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area/performanceRuntime / build performance / efficiency.type/automationCreated or maintained by an agentic workflow.

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions