DO-NOT-MERGE: #657 revert for keyed-regression confirm

[azchohfi]

Measurement-only: reverts #657 (628fb7f) off current main to confirm its keyed-list reconcile/diff regression as an inverse signal. Do not merge; will be torn down.

[azchohfi]

/perf

[github-actions]

⚡ Reactor perf comparison

Workload: StressPerf.ReactorOptimized StocksGrid · --percent 50 --duration 10 · x64 Release · median of 12 paired runs (2 warmup dropped); Δ is the mean change with a 95% CI · PR head and main built and run interleaved on the same runner.

Regression vs main baseline

Metric	main (baseline)	This PR	Δ (95% CI)	Status
Renders/sec ↑	2.66	2.63	-1.5% 95% CI [-7.3, +4.2]	≈ within noise
Avg Reconcile (ms) ↓	121.7	122.5	+2.6% 95% CI [-1.9, +7.1]	≈ within noise
Avg Diff (ms) ↓	112.1	111.8	+2.4% 95% CI [-2.3, +7.2]	≈ within noise
Avg Memory (MB) ↓	283.8	283.8	-0.2% 95% CI [-1.2, +0.8]	≈ within noise

Low-mutation skip-floor (--percent 0)

At --percent 0 the workload mutates few cells per tick (always at least one), so reconcile/diff isolate the O(n) per-tick child skip-walk floor that higher mutation rates dilute — ChildReconciler re-walks every child each tick even when nothing moved. The closer --percent is to 0, the more this floor is the signal, so a structural-skip optimization shows up cleanly where the headline table above buries it. Δ is the mean paired change with a 95% CI.

Metric	main (baseline)	This PR	Δ (95% CI)	Status
Renders/sec ↑	16.59	16.48	-1.2% 95% CI [-8.5, +6.0]	≈ within noise
Avg Reconcile (ms) ↓	37.1	35.0	-1.8% 95% CI [-7.7, +4.1]	≈ within noise
Avg Diff (ms) ↓	35.0	33.0	-2.1% 95% CI [-8.2, +4.0]	≈ within noise
Avg Memory (MB) ↓	266.0	265.4	-0.2% 95% CI [-0.5, +0.1]	≈ within noise

Allocation (Reactor) — lower is better

Metric	main (baseline)	This PR	Δ (95% CI)	Status
Alloc bytes/render ↓	4848013	4884956	+1.4% 95% CI [+0.2, +2.6]	⚠️ regression
Gen0 GC / 1k renders ↓	192.31	200.00	+8.1% 95% CI [-3.9, +20.1]	≈ within noise

Keyed-list workload (StressPerf.KeyedList, --percent 50)

A separate macro workload: a ~500-row stably keyed list whose rows are reordered / inserted / removed each tick. Because every child carries a key, the child reconciler takes its keyed arm (ReconcileKeyed → ReconcileKeyedMiddle, the LIS-based minimal-move pass) instead of the positional re-walk the StocksGrid tables above measure — so this is the sensitive macro signal for keyed-diff work the positional cells can never reach. Same interleaved paired-Δ 95% CI as the headline table.

Metric	main (baseline)	This PR	Δ (95% CI)	Status
Renders/sec ↑	16.34	18.73	+16.1% 95% CI [+12.3, +20.0]	✅ improvement
Avg Reconcile (ms) ↓	20.6	17.6	-15.6% 95% CI [-17.9, -13.4]	✅ improvement
Avg Diff (ms) ↓	20.4	17.4	-15.6% 95% CI [-17.8, -13.3]	✅ improvement
Avg Memory (MB) ↓	164.2	167.9	+1.7% 95% CI [+1.0, +2.4]	⚠️ regression

Allocation (keyed-list) — lower is better

Metric	main (baseline)	This PR	Δ (95% CI)	Status
Alloc bytes/render ↓	216279	314216	+45.2% 95% CI [+44.4, +46.0]	⚠️ regression
Gen0 GC / 1k renders ↓	11.83	15.67	+36.1% 95% CI [+27.7, +44.5]	⚠️ regression

Reconciler micro-benchmarks (PerfBench.ControlModel)

Production --variant Reactor control-model path, ns-resolution and WinUI-undiluted (spec-047 M1–M13) — ↓ lower is better. Status tracks allocated bytes/op, the authoritative signal here; it is deterministic for structurally-fixed benches, while dispatcher / background-thread benches carry a small process-to-process offset, so a bench is flagged only when its 95% CI clears a ±3% minimum-effect band (real structural alloc changes are several percent to many-x). ns/op is shown for context but is not auto-flagged (its paired CI is rep-interleaved but the flag remains dormant pending a real-CI identical-binary band calibration). Δ is the mean paired change with a 95% CI.

Bench	main ns/op	Δ ns (95% CI)	main B/op	Δ alloc (95% CI)	Status
M1 Mount_Leaf_NoCallback	149999.2	+0.6% 95% CI [-2.3, +3.6]	1140.9	0.0% 95% CI [0.0, 0.0]	≈ within noise
M2 Mount_Leaf_OneCallback	109267.4	-1.0% 95% CI [-5.7, +3.7]	3383.3	0.0% 95% CI [0.0, 0.0]	≈ within noise
M3 Mount_Leaf_ThreeCallbacks	225433.3	-1.8% 95% CI [-5.8, +2.1]	8395.4	+1.6% 95% CI [+0.2, +2.9]	≈ within noise
M4 Dispatch_Switch_Cold	112542.6	-3.6% 95% CI [-8.3, +1.0]	1767.8	0.0% 95% CI [0.0, 0.0]	≈ within noise
M5 Dispatch_Switch_Warm	111827.4	+1.1% 95% CI [-7.6, +9.7]	1805.9	-1.4% 95% CI [-3.6, +0.8]	≈ within noise
M6 Dispatch_ExternalType	91199.2	+0.6% 95% CI [-0.5, +1.6]	1028.6	-2.4% 95% CI [-6.4, +1.5]	≈ within noise
M7 Update_NoChange	55403.2	+0.2% 95% CI [-0.5, +0.8]	370.1	+8.4% 95% CI [-3.1, +19.8]	≈ within noise
M8 Update_OneLeafChanged	42066.4	+0.8% 95% CI [-2.0, +3.7]	536.0	0.0% 95% CI [0.0, 0.0]	≈ within noise
M9 Update_AllChanged	2884322.0	+0.1% 95% CI [-1.2, +1.4]	184278.1	0.0% 95% CI [0.0, 0.0]	≈ within noise
M10 EventHandlerState_Alloc	86233.9	-0.1% 95% CI [-2.6, +2.4]	3095.2	0.0% 95% CI [0.0, +0.1]	≈ within noise
M11 ModifierEHS_Frequency	45952.8	+1.3% 95% CI [-0.5, +3.2]	638.9	0.0% 95% CI [0.0, 0.0]	≈ within noise
M12 Pool_Rent_HotPath	117647.9	+1.6% 95% CI [+0.1, +3.1]	1099.9	0.0% 95% CI [0.0, 0.0]	≈ within noise
M13 Setters_Suppression_Scope	107.1	+28.9% 95% CI [+5.0, +52.8]	26.7	0.0% 95% CI [0.0, 0.0]	≈ within noise
M14 Dsl_Rebuild_Cascade	1580037.0	+0.7% 95% CI [-1.7, +3.1]	2231828.9	0.0% 95% CI [0.0, 0.0]	≈ within noise
C207 ChangeHandler_DpRead_Coalesce	1262.9	+6.4% 95% CI [-9.9, +22.8]	0.6	0.0% 95% CI [0.0, 0.0]	≈ within noise
OAlloc Optional_Element_Alloc	214.5	+4.4% 95% CI [-2.6, +11.5]	528.0	0.0% 95% CI [0.0, 0.0]	≈ within noise
OUpdate Optional_Reconciler_Update	12558.9	-0.8% 95% CI [-3.1, +1.5]	2772.3	0.0% 95% CI [0.0, 0.0]	≈ within noise

Cross-framework reference (same StocksGrid workload)

Metric	vanilla WinUI3¹	Rust windows-reactor²	Reactor (this PR)
Renders/sec ↑	3.06	4.65	2.63
Avg Reconcile (ms) ↓	n/a	19.7	122.5
Avg Diff (ms) ↓	n/a	18.3	111.8
Avg Memory (MB) ↓	263.3	197.8	283.8

_{↑ higher is better · ↓ lower is better. Within noise = the 95% confidence interval of the paired Δ includes 0 (no change resolvable at this sample size); ✅ improvement / ⚠️ regression require the CI to exclude 0.
Allocation metrics (alloc bytes/render, Gen0 GC) are the sensitive signal for allocation-reduction work, where the mean-ms / memory figures are largely flat. They read n/a for a harness built from a revision that predates them (rebase the PR onto main to populate them).
Reconciler micro-benchmarks run PerfBench.ControlModel --variant Reactor (M1–M13) as a headless loop bracketed by per-thread alloc + GC counters — ns-resolution and free of WinUI render / working-set dilution, so they resolve Core/Reconciler allocation deltas the macro StocksGrid workload cannot. main and PR each link their own src/Reactor build and are rep-interleaved (a fresh alternated process per rep); Δ is the paired 95% CI over per-rep means. The Status column tracks allocated bytes/op (deterministic for identical code); ns/op is informational — its paired CI is now unbiased but the flag stays dormant pending a real-CI identical-binary band calibration.
¹ vanilla WinUI3 = StressPerf.Direct (imperative; no virtual-DOM, so it has no reconcile/diff phase — those cells read n/a). Measured live on this runner.
² Rust = test_reactor_perf from microsoft/windows-rs — a port of this harness (same StocksGrid, same --percent/--duration CLI). Built from source and measured live on this runner.
Absolute numbers are runner-dependent — trust the Δ vs main, not the absolute values. Memory (working set) is the noisiest metric.
Runner: CPU: AMD EPYC 7763 64-Core Processor · 4 logical cores · 16 GB RAM · runner: GitHub Actions 1043025925.
Generated by .github/workflows/perf-compare.yml · PR 5274c5a vs main 0002f19 · 2026-06-27T16:46:48Z · run log.}