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
{{ message }}
Repository navigation
IDE foundation: establish cold/warm/incremental latency and memory baselines #263
Parent: #250
Recommended tier: medium/local-agent
Can begin after: #251
Should inform: #262, #264, #265
Goal
Create a reproducible performance and latency baseline for the current Owen pipeline and the emerging Rust core before introducing LSP/incremental architecture or declaring a cutover performance win.
The primary IDE metric is edit-to-diagnostic latency, not a vague batch multiplier.
Workloads
Record representative workloads for:
minimal .own file;
medium .own module;
corpus batch;
single C# file through Roslyn extractor → OwnIR → core;
representative .csproj;
representative multi-project .sln;
selected large real-world solution;
five pinned OSS repositories where practical.
Pin source commits and exact commands.
Measurements
Measure separately where possible:
process startup;
frontend/Roslyn extraction;
OwnIR serialization and parse;
bridge/lowering;
analysis;
report/SARIF rendering;
end-to-end wall time;
peak RSS;
allocation/heap profile for selected workloads;
repeated warm run;
no-op recheck;
single-file edit recheck;
diagnostic publication latency proxy.
For the Rust path, do not benchmark incomplete placeholders as if they represent the final production route. Clearly mark unavailable stages.
Method discipline
Use pinned hardware/runtime/toolchain metadata.
Record warm-up policy.
Separate cold and warm runs.
Report distributions or repeated-run statistics, not one flattering number.
Keep batch and IDE-oriented measurements distinct.
Include raw machine-readable results.
Profile before recommending optimizations.
Outputs
Add:
a methodology note;
machine-readable baseline data;
scripts/commands that reproduce measurements;
a stage-by-stage timing table;
explicit latency budgets proposed for .own LSP and C# hybrid host;
a list of dominant costs confirmed by profiling.
Guardrails
No optimization PR in this issue.
No allocator/interner/rayon change based only on intuition.
No claiming Rust speedup from solver-only microbenchmarks while excluding extraction or rendering.
No hidden use of different inputs between engines.
Do not add unstable benchmarks as mandatory PR gates until variance is understood.
Acceptance
Cold, warm and no-op/edit-oriented baselines are documented.
End-to-end stages are separated sufficiently to locate bottlenecks.
Nothing here is stale conceptually: work resumes at host qualification and proceeds along the already-frozen protocol. No trial runs in the meantime; the suite's training-no-incidental-measurement control keeps step 7 shut and rejects any workflow line carrying --calibrate or --measure, so the park is enforced mechanically rather than by discipline.
Activity
feat(perf): #263-A — the measurement instrument, calibrated and firew…
fix(perf): a contended runner is a result, not a job failure
fix(perf): the harness digest named the checkout, not the instrument
fix(perf): a runner that will not hold still is not a broken instrument
fix(perf): implement D7 §6 literally — two commits, not two sections
fix(perf): the D7 anchor is provenance, and a preregistration is not …
fix(perf): close the boolean N/A escape hatch and check the whole cel…
fix(perf): D7's metrics are not the instrument's attribution taxonomy
docs(perf): ratify the D7 metric mapping, and preregister Round 6
docs(perf): propose the #263-A calibration reproducibility policy, wi…
docs(t0): apply the owner rulings — structure resolved, numbers still…
docs(t0): close the five freeze-blockers the full review found
style: the linter yields to the instrument, not the other way round
Merge pull request #351 from PhysShell/claude/p-022-analysis-wiring-m…
P-022 Stage 3: derive the #262 decision packet from evidence, not fro…
P-022 Stage 3: reconcile the status surfaces so they cannot disagree
P-022 Stage 3: COMPLETE — reconcile at the terminal candidate
PhysShell commentedon Sep 19, 2026
Status: PARKED (owner ruling, 2026-09-19), deliberately unmeasured.
Nothing here is stale conceptually: work resumes at host qualification and proceeds along the already-frozen protocol. No trial runs in the meantime; the suite's
training-no-incidental-measurementcontrol keeps step 7 shut and rejects any workflow line carrying--calibrateor--measure, so the park is enforced mechanically rather than by discipline.Generated by Claude Code
P-037 A2.2-0: freeze the relevance taxonomy, the named exclusions and…