Repository navigation
perf: release failed high-density route searches - #2894
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Benchmark This PRRun benchmarks by commenting on this PR: Comment Everything after Use Any PR whose title contains |
|
/benchmark --same-machine |
|
/benchmark --same-machine --dataset 18 |
|
/benchmark --same-machine --dataset 24 |
Same Machine Benchmark ResultsBoth revisions ran sequentially in one Blacksmith job on Dataset:
Outcome changes: 0 improved, 0 regressed. DRC issues are totaled across solved samples. Timing percentiles include all samples, with failed and timed-out samples counted at their configured timeout; negative timing deltas are faster. Historical failures without timeout metadata make timing percentiles unavailable. Base pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 85/85 samples (0 partial, including failed or timed-out samples).
PR pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 85/85 samples (0 partial, including failed or timed-out samples).
Workflow: View run |
Same Machine Benchmark ResultsBoth revisions ran sequentially in one Blacksmith job on Dataset:
Outcome changes: 0 improved, 0 regressed. DRC issues are totaled across solved samples. Timing percentiles include all samples, with failed and timed-out samples counted at their configured timeout; negative timing deltas are faster. Historical failures without timeout metadata make timing percentiles unavailable. Base pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 16/16 samples (2 partial, including failed or timed-out samples).
PR pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 16/16 samples (2 partial, including failed or timed-out samples).
Workflow: View run |
Same Machine Benchmark ResultsBoth revisions ran sequentially in one Blacksmith job on Dataset:
Outcome changes: 0 improved, 0 regressed. DRC issues are totaled across solved samples. Timing percentiles include all samples, with failed and timed-out samples counted at their configured timeout; negative timing deltas are faster. Historical failures without timeout metadata make timing percentiles unavailable. Base pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 26/26 samples (11 partial, including failed or timed-out samples).
PR pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 26/26 samples (11 partial, including failed or timed-out samples).
Workflow: View run |
|
/benchmark --same-machine --dataset 18 |
Same Machine Benchmark ResultsThe paired benchmark ended with cancelled before both reports were produced. Workflow: View run |
|
/benchmark --same-machine --dataset 18 |
Same Machine Benchmark ResultsBoth revisions ran sequentially in one Blacksmith job on Dataset:
Outcome changes: 0 improved, 0 regressed. DRC issues are totaled across solved samples. Timing percentiles include all samples, with failed and timed-out samples counted at their configured timeout; negative timing deltas are faster. Historical failures without timeout metadata make timing percentiles unavailable. Base pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 16/16 samples (2 partial, including failed or timed-out samples).
PR pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 16/16 samples (2 partial, including failed or timed-out samples).
Workflow: View run |
|
/benchmark --same-machine --dataset 24 |
Same Machine Benchmark ResultsBoth revisions ran sequentially in one Blacksmith job on Dataset:
Outcome changes: 0 improved, 1 regressed. DRC issues are totaled across solved samples. Timing percentiles include all samples, with failed and timed-out samples counted at their configured timeout; negative timing deltas are faster. Historical failures without timeout metadata make timing percentiles unavailable. Base pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 26/26 samples (10 partial, including failed or timed-out samples).
PR pipeline stage timingsAutoroutingPipelineSolver9_PreloadedTraceGraph Recorded timings: 26/26 samples (11 partial, including failed or timed-out samples).
Changed outcomes (1)
Workflow: View run |
|
Thank you for your contribution! 🎉 PR Rating: ⭐⭐⭐ Track your contributions and see the leaderboard at: tscircuit Contribution Tracker |
Summary
Failed high-density route searches were kept in memory even when detailed search-history capture was disabled.
This releases each failed search after preserving its error. When
captureSearchDebugis enabled, the search is still retained for the debugger. Core's production preview graphics are unchanged.Benchmark
Seven fresh-process A/B runs on the real Dataset 18 sample 002 large node: