You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
[.NET 10 vs main] WASM-Mono-AOT performance regression tracking
This issue tracks the actionable WASM-Mono-AOT results from the reviewed release/10.0 vs main windowed comparison. Draft generated 2026-08-17. GitHub issue/PR states refreshed from live metadata on 2026-08-17. User decisions and AI-attributed notes applied on 2026-08-17.
Use the checkboxes in place: leave an item unchecked while investigation, filing, ownership, or verification remains; check it when that work is finished. Feel free to edit or append status directly on the same checklist line as work progresses. There is intentionally no separate Completed section.
Within each section, items are ordered by worst reviewed percentage (largest first). Regression count is used only to break ties.
dotnet/runtime#126361 — needs-investigation; closed-not-planned; 13 WASM-Mono-AOT reviewed regressions; worst +50.7%; regression date 2026-03-27. Review note: Marked acceptable as the impacts on CoreCLR were small. Still needs a Wasm-Mono-AOT look over. Representative chart.
Active tracked regressions
No items selected by the current inclusion rules.
Autofiling issues needing transfer or resolution
Show 22 items
Uncorrelated autofiling issues that are unmarked or marked needs-investigation.
dotnet/perf-autofiling-issues#62589 — needs-investigation; open; 219 WASM-Mono-AOT reviewed regressions; worst +109.1%; regression date 2025-09-09. Review note: Some still are regressed from this regression: such as System.Globalization.Tests.StringSearch.IsSuffix_SecondHalf(Options: (pl-PL, None, False)) Representative chart.
dotnet/perf-autofiling-issues#64552 — needs-investigation; open; 19 WASM-Mono-AOT reviewed regressions; worst +109.1%; regression date 2025-10-27. Review note: Initial look, but no follow-up or closeout. AI Note: PR #121096 is a potential culprit and warrants targeted follow-up. (evidence) Representative chart.
dotnet/perf-autofiling-issues#76345 — needs-investigation; open; 1 WASM-Mono-AOT reviewed regressions; worst +67.9%; regression date 2026-07-13. Review note: AI Note: A real two-step WASM-AOT regression is documented; bisection remains deferred. (evidence) Representative chart.
dotnet/perf-autofiling-issues#75373 — needs-investigation; open; 47 WASM-Mono-AOT reviewed regressions; worst +63.6%; regression date 2026-06-25. Review note:Upgrade Emscripten to 5.0.6 & LLVM to 23 runtime#129299 was linked as the likely cause, but no specific ok/acceptable has been given. Some regressions have unregressed. AI Note: Sibling issue #75780 was attributed to the same change; toolchain-upgrade disposition remains unresolved. (evidence) Representative chart.
dotnet/perf-autofiling-issues#70611 — needs-investigation; open; 2 WASM-Mono-AOT reviewed regressions; worst +48.3%; regression date 2026-03-19. Review note: No updates since initial look basically (march) AI Note: A July 28 re-examination reconfirmed the Regex and MemorySpan regressions. (evidence) Representative chart.
dotnet/perf-autofiling-issues#65460 — needs-investigation; closed-completed; 2 WASM-Mono-AOT reviewed regressions; worst +40.8%; regression date 2025-11-21. Review note: These still seem like real regression (both) Representative chart.
dotnet/perf-autofiling-issues#73326 — needs-investigation; open; 2 WASM-Mono-AOT reviewed regressions; worst +27.2%; regression date 2026-05-15. Review note: System.Collections.Sort<IntClass>.List(Size: 512) is regressed from this but the test on the issue is improved. Representative chart.
dotnet/perf-autofiling-issues#73166 — needs-investigation; open; 1 WASM-Mono-AOT reviewed regressions; worst +21.3%; regression date 2026-05-13. Review note: This still seems to be regressed. (Though date is slightly off). Representative chart.
dotnet/perf-autofiling-issues#73724 — needs-investigation; open; 1 WASM-Mono-AOT reviewed regressions; worst +17.6%; regression date 2026-05-21. Review note: Seems like it improved and then regressed again, but we only had 1 good run after the improvements. Representative chart.
dotnet/perf-autofiling-issues#61031 — needs-investigation; open; 1 WASM-Mono-AOT reviewed regressions; worst +14.1%; regression date 2025-08-16. Review note: Linked to a PR but still real. Maybe before the branching of release 10 as the net10 baseline is post this it seems. Representative chart.
dotnet/perf-autofiling-issues#73163 — needs-investigation; open; 3 WASM-Mono-AOT reviewed regressions; worst +13.6%; regression date 2026-05-13. Review note: System.Collections.IterateForEach<String>.HashSet(Size: 512) seems to still be affected. (Among other regressions at the time from the issue. AI Note: The issue was closed and immediately reopened, indicating the regression remains active. (evidence) Representative chart.
dotnet/perf-autofiling-issues#73170 — linked; closed-completed; 4 WASM-Mono-AOT reviewed regressions; worst +11.1%; regression date 2026-05-14. Review note: All 4 of these are still real and regressed from this date. AI Note (user-approved): Closed as a triage redirect and linked to perf-autofiling-issues#73325; this is not proof the regression was fixed. (evidence) Representative chart.
Missed regression groups needing filing
Show 3 items
Dated groups containing at least one needs-investigation or truly-missed regression. The checkbox represents the group as one candidate public tracking issue.
Regression group 2026-03-26 — 7 actionable of 7 related regressions; worst +20.0%; queues Ubuntu-x64-Viper. Review notes: Specific tiem is March 26th, 15:05 on the graph
Benchstone.BenchF.Whetsto.Test — Ubuntu-x64-Viper; +13.9%; 696,732,421 → 793,903,250; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graph
…and 1 more related regressions in this date group.
Regression group 2026-06-25 — 5 actionable of 7 related regressions; worst +20.0%; queues Ubuntu-x64-Viper. Review notes: Specific tiem is March 26th, 15:05 on the graph
…and 1 more related regressions in this date group.
Regression group 2025-12-13 — 1 actionable of 1 related regressions; worst +5.1%; queues Ubuntu-x64-Viper. Review notes: Only current regression time is AFAIK uninvestigated.
[.NET 10 vs main] WASM-Mono-AOT performance regression tracking
This issue tracks the actionable
WASM-Mono-AOTresults from the reviewed release/10.0 vs main windowed comparison. Draft generated 2026-08-17. GitHub issue/PR states refreshed from live metadata on 2026-08-17. User decisions and AI-attributed notes applied on 2026-08-17.Use the checkboxes in place: leave an item unchecked while investigation, filing, ownership, or verification remains; check it when that work is finished. Feel free to edit or append status directly on the same checklist line as work progresses. There is intentionally no separate Completed section.
Within each section, items are ordered by worst reviewed percentage (largest first). Regression count is used only to break ties.
Inclusion rules
Included:
needs-investigation, confirmed runtime regressions, unmarked runtime/autofiling/missed items, and missed regressions markedtruly-missed.Excluded:
fixed,acceptable,not-applicable,noise,improved,baseline-outlier,covered, andlinked.Actionable summary
Needs investigation
Show 2 items
Items requiring a second look before deciding whether to file, transfer, accept, or close them.
dotnet/runtime#129860 — needs-investigation;
open; 72WASM-Mono-AOTreviewed regressions; worst +85.3%; regression date 2026-06-24. Review note: Need to double check it stays, but it seems fixed. (as stated in [Perf] Linux/arm64: 9 Regressions on 6/24/2026 7:31:32 AM +00:00 runtime#129860 (comment)) Representative chart.dotnet/runtime#126361 — needs-investigation;
closed-not-planned; 13WASM-Mono-AOTreviewed regressions; worst +50.7%; regression date 2026-03-27. Review note: Marked acceptable as the impacts on CoreCLR were small. Still needs a Wasm-Mono-AOT look over. Representative chart.Active tracked regressions
No items selected by the current inclusion rules.
Autofiling issues needing transfer or resolution
Show 22 items
Uncorrelated autofiling issues that are unmarked or marked
needs-investigation.dotnet/perf-autofiling-issues#73330 — needs-investigation;
open; 18WASM-Mono-AOTreviewed regressions; worst +111.1%; regression date 2026-05-19. Review note: Probably due to BDN update [Perf] Linux/arm64: 86 Regressions on 5/19/2026 5:33:16 PM +00:00 perf-autofiling-issues#73304, still seeing regressions so marking needs-investigation. Representative chart.dotnet/perf-autofiling-issues#62589 — needs-investigation;
open; 219WASM-Mono-AOTreviewed regressions; worst +109.1%; regression date 2025-09-09. Review note: Some still are regressed from this regression: such as System.Globalization.Tests.StringSearch.IsSuffix_SecondHalf(Options: (pl-PL, None, False)) Representative chart.dotnet/perf-autofiling-issues#64552 — needs-investigation;
open; 19WASM-Mono-AOTreviewed regressions; worst +109.1%; regression date 2025-10-27. Review note: Initial look, but no follow-up or closeout. AI Note: PR #121096 is a potential culprit and warrants targeted follow-up. (evidence) Representative chart.dotnet/perf-autofiling-issues#75372 — needs-investigation;
open; 12WASM-Mono-AOTreviewed regressions; worst +109.1%; regression date 2026-06-25. Representative chart.dotnet/perf-autofiling-issues#76345 — needs-investigation;
open; 1WASM-Mono-AOTreviewed regressions; worst +67.9%; regression date 2026-07-13. Review note: AI Note: A real two-step WASM-AOT regression is documented; bisection remains deferred. (evidence) Representative chart.dotnet/perf-autofiling-issues#75373 — needs-investigation;
open; 47WASM-Mono-AOTreviewed regressions; worst +63.6%; regression date 2026-06-25. Review note: Upgrade Emscripten to 5.0.6 & LLVM to 23 runtime#129299 was linked as the likely cause, but no specific ok/acceptable has been given. Some regressions have unregressed. AI Note: Sibling issue #75780 was attributed to the same change; toolchain-upgrade disposition remains unresolved. (evidence) Representative chart.dotnet/perf-autofiling-issues#70611 — needs-investigation;
open; 2WASM-Mono-AOTreviewed regressions; worst +48.3%; regression date 2026-03-19. Review note: No updates since initial look basically (march) AI Note: A July 28 re-examination reconfirmed the Regex and MemorySpan regressions. (evidence) Representative chart.dotnet/perf-autofiling-issues#65460 — needs-investigation;
closed-completed; 2WASM-Mono-AOTreviewed regressions; worst +40.8%; regression date 2025-11-21. Review note: These still seem like real regression (both) Representative chart.dotnet/perf-autofiling-issues#76183 — needs-investigation;
open; 6WASM-Mono-AOTreviewed regressions; worst +37.4%; regression date 2026-07-10. Review note: Large gap and still real. Representative chart.dotnet/perf-autofiling-issues#75371 — needs-investigation;
open; 2WASM-Mono-AOTreviewed regressions; worst +37.0%; regression date 2026-06-25. Review note: Real and still valid, no obvious runtime issue yet. Representative chart.dotnet/perf-autofiling-issues#64174 — needs-investigation;
open; 11WASM-Mono-AOTreviewed regressions; worst +31.5%; regression date 2025-10-17. Review note: Due to partial revert [browser][mono][perf] partial revert 119307 119800 runtime#120742, seems some reverted to worse than before (System.Globalization.Tests.StringSearch). Representative chart.dotnet/perf-autofiling-issues#76544 — noise;
open; 8WASM-Mono-AOTreviewed regressions; worst +31.5%; regression date 2026-07-13. Review note: User update: Closed as Noise. Representative chart.dotnet/perf-autofiling-issues#73326 — needs-investigation;
open; 2WASM-Mono-AOTreviewed regressions; worst +27.2%; regression date 2026-05-15. Review note: System.Collections.Sort<IntClass>.List(Size: 512) is regressed from this but the test on the issue is improved. Representative chart.dotnet/perf-autofiling-issues#73166 — needs-investigation;
open; 1WASM-Mono-AOTreviewed regressions; worst +21.3%; regression date 2026-05-13. Review note: This still seems to be regressed. (Though date is slightly off). Representative chart.dotnet/perf-autofiling-issues#73724 — needs-investigation;
open; 1WASM-Mono-AOTreviewed regressions; worst +17.6%; regression date 2026-05-21. Review note: Seems like it improved and then regressed again, but we only had 1 good run after the improvements. Representative chart.dotnet/perf-autofiling-issues#71720 — needs-investigation;
open; 4WASM-Mono-AOTreviewed regressions; worst +16.8%; regression date 2026-04-14. Representative chart.dotnet/perf-autofiling-issues#62762 — needs-investigation;
open; 1WASM-Mono-AOTreviewed regressions; worst +15.5%; regression date 2025-09-18. Review note: Seems to be real and uninvestigated. Representative chart.dotnet/perf-autofiling-issues#73325 — needs-investigation;
open; 4WASM-Mono-AOTreviewed regressions; worst +15.4%; regression date 2026-05-14. Representative chart.dotnet/perf-autofiling-issues#61031 — needs-investigation;
open; 1WASM-Mono-AOTreviewed regressions; worst +14.1%; regression date 2025-08-16. Review note: Linked to a PR but still real. Maybe before the branching of release 10 as the net10 baseline is post this it seems. Representative chart.dotnet/perf-autofiling-issues#73163 — needs-investigation;
open; 3WASM-Mono-AOTreviewed regressions; worst +13.6%; regression date 2026-05-13. Review note: System.Collections.IterateForEach<String>.HashSet(Size: 512) seems to still be affected. (Among other regressions at the time from the issue. AI Note: The issue was closed and immediately reopened, indicating the regression remains active. (evidence) Representative chart.dotnet/perf-autofiling-issues#61032 — needs-investigation;
open; 3WASM-Mono-AOTreviewed regressions; worst +12.6%; regression date 2025-08-17. Representative chart.dotnet/perf-autofiling-issues#73170 — linked;
closed-completed; 4WASM-Mono-AOTreviewed regressions; worst +11.1%; regression date 2026-05-14. Review note: All 4 of these are still real and regressed from this date. AI Note (user-approved): Closed as a triage redirect and linked to perf-autofiling-issues#73325; this is not proof the regression was fixed. (evidence) Representative chart.Missed regression groups needing filing
Show 3 items
Dated groups containing at least one
needs-investigationortruly-missedregression. The checkbox represents the group as one candidate public tracking issue.Regression group 2026-03-26 — 7 actionable of 7 related regressions; worst +20.0%; queues
Ubuntu-x64-Viper. Review notes: Specific tiem is March 26th, 15:05 on the graphSystem.Memory.Slice<Byte>.MemoryStartLengthSpan—Ubuntu-x64-Viper; +20.0%; 5.71 → 6.85; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Memory.Slice<Byte>.ReadOnlyMemoryStartLengthSpan—Ubuntu-x64-Viper; +18.0%; 5.69 → 6.72; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphBenchstone.BenchF.Whetsto.Test—Ubuntu-x64-Viper; +13.9%; 696,732,421 → 793,903,250; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Collections.TryAddDefaultSize<String>.ConcurrentDictionary(Count: 512)—Ubuntu-x64-Viper; +6.3%; 56,799 → 60,360; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Collections.AddGivenSize<String>.ConcurrentDictionary(Size: 512)—Ubuntu-x64-Viper; +6.0%; 51,929 → 55,058; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Collections.TryAddGiventSize<String>.ConcurrentDictionary(Count: 512)—Ubuntu-x64-Viper; +5.9%; 52,211 → 55,295; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphRegression group 2026-06-25 — 5 actionable of 7 related regressions; worst +20.0%; queues
Ubuntu-x64-Viper. Review notes: Specific tiem is March 26th, 15:05 on the graphSystem.Memory.Slice<Byte>.MemoryStartLengthSpan—Ubuntu-x64-Viper; +20.0%; 5.71 → 6.85; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Memory.Slice<Byte>.ReadOnlyMemoryStartLengthSpan—Ubuntu-x64-Viper; +18.0%; 5.69 → 6.72; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Memory.Slice<Byte>.MemoryStartSpan—Ubuntu-x64-Viper; +11.3%; 5.95 → 6.62; mark coveredSystem.MathBenchmarks.Single.Pow—Ubuntu-x64-Viper; +9.0%; 33,065 → 36,033; mark coveredSystem.Collections.AddGivenSize<String>.ConcurrentDictionary(Size: 512)—Ubuntu-x64-Viper; +6.0%; 51,929 → 55,058; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphSystem.Collections.TryAddGiventSize<String>.ConcurrentDictionary(Count: 512)—Ubuntu-x64-Viper; +5.9%; 52,211 → 55,295; mark needs-investigation; note: Specific tiem is March 26th, 15:05 on the graphRegression group 2025-12-13 — 1 actionable of 1 related regressions; worst +5.1%; queues
Ubuntu-x64-Viper. Review notes: Only current regression time is AFAIK uninvestigated.PerfLabTests.CastingPerf2.CastingPerf.FooObjIsNull—Ubuntu-x64-Viper; +5.1%; 135,678 → 142,639; mark truly-missed; note: Only current regression time is AFAIK uninvestigated.