Commit graph

10 commits

Author SHA1 Message Date
Erik
39967e78bd perf #418: parallelize landblock builds across a striped worker pool
Login publishes the 25x25 window at a flat 32 blocks/s (~27 s in the
tunnel). The reveal-timing probe A/B (695a27b4) showed the consumer
budget env ceilings change nothing, which was read as producer-limited:
one "acdream.streaming.worker" thread, ~31 ms/block. This replaces the
single worker with min(ProcessorCount-2, 8) workers, floor 1.

Design: striped/affinity dispatch. Each worker owns one unbounded lane
channel plus its own high/low priority queues; jobs route to
lane = ((id >> 16) * 2654435761) % N (the low word of a landblock id is
constant, so the id is mixed before reduction). Striping was chosen
over a shared queue + in-flight conflict tracker because it preserves
the per-landblock contract structurally rather than by bookkeeping:
every job for one id lives on one lane, so per-id enqueue order IS
execution and completion-arrival order, and the same-landblock
supersede rules (PromoteToNear removes queued LoadFar/Unload) keep
seeing every queued job for that id. Contract, point by point:

- Per-landblock ordering: same id -> same lane -> serial FIFO.
- ClearLoads: broadcast to every lane inside the same _inboxGate lock
  that serializes enqueues, so any load enqueued before
  ClearPendingLoads() returns sits ahead of its lane's ClearLoads copy
  in that lane's FIFO and is dropped at read time, exactly like the
  single-thread path. Already-dequeued builds still complete (now up
  to one per worker instead of one total); StreamingController's
  SweepCollapsed already unloads those uniformly.
- Priority: per-lane high/low split unchanged. Cross-lane, priority is
  not globally ordered (a lane cannot run another lane's job), which
  the contract permits; near-tier jobs hash-spread across lanes and
  are preferred within each.
- Outbox: SingleWriter flipped to false; nothing assumed single-writer
  (PublishResult already used TryWrite + an Interlocked backlog, and
  the consumer's peek->read head-stability holds because only the
  single reader ever moves the head). Cross-landblock arrival order
  was verified arbitrary-tolerant before relying on it:
  StreamingController.AdmitCompletions classifies each result
  independently into per-priority FIFOs (generation staleness +
  per-landblock retirement blocking); per-landblock arrival order is
  preserved by striping.
- Crash surface: per-worker. The first real crash publishes
  WorkerCrashed (prefixed "worker N:" in pools > 1), sets
  _workerFailure, completes every lane, and cancels the pool (a crash
  still ends all processing, as before); siblings that merely observe
  the closed lanes (ChannelClosedException) exit quietly instead of
  reporting spurious crashes; the outbox completes only when the LAST
  worker exits so no in-flight completions are dropped.
- Disposal: joins every worker under the same _disposeGate; Start
  stays idempotent and dispose-serialized.

Thread-safety audit of the production build closures
(SessionPlayerComposition), per shared object:

- DatCollection (every read in LandblockBuildFactory.BuildLocked:
  LandblockLoader.Load, SceneryGenerator.Generate, SetupMesh.Flatten,
  CellMesh.Build, GfxObjBounds.Get, GfxObjDegradeResolver): NOT
  thread-safe; already serialized under the shared _datLock, which
  BuildLocked holds for the whole read transaction. Unchanged; the
  probe run measured hold 0-13 ms / wait <= 12 ms during the login
  window, so the lock is not the new bottleneck and the build was NOT
  serialized beyond it.
- PakPreparedAssetSource / PakReader (BuildPreparedCollisionClosure,
  outside the lock): immutable TOC array + read-only
  MemoryMappedViewAccessor random-access reads + ConcurrentDictionary
  verdict caches - safe for N concurrent readers (Slice I3 design;
  the headless SharedPreparedCollisionCache wrapper is fully
  lock-protected).
- LandblockMesh.Build (outside the lock): pure math over the dat
  record + the composition-time height table + the immutable
  TerrainBlendingContext record; the shared SurfaceCache is a
  ConcurrentDictionary and BuildSurface is deterministic, so its
  lookup-or-build race is last-write-wins-benign (the code already
  documented exactly this).
- PhysicsDiagnostics probe statics: read-only bools + thread-safe
  Console writes.

MEASURED OUTCOME (gate 4): the timing acceptance did NOT pass, and per
the task contract that is reported, not tuned around. With 8 workers
on this 16-core machine all 625 builds complete in ~203 ms
(ACDREAM_PROBE_TELEPORT BUILD lines t=3475390..3475593) - the producer
is off the critical path - but loaded= still advances at exactly
+32/1000 ms and SUMMARY totalMs measured 27395 and 27503 across two
runs (baseline 26728). The 32/s pacer is in the consumer
admission/publication path and is not governed by the
StreamingWorkBudgetOptions env ceilings. #418 stays IN-PROGRESS on the
consumer side; see docs/ISSUES.md for the evidence chain.

Tests: per-landblock ordering under 4-worker contention, cross-lane
ClearLoads drop, per-lane near-before-far preference, pool-of-1 serial
equivalence, disposal joining every worker, lane-spread guard, and
worker-count validation (LandblockStreamerPoolTests). Two existing
tests asserted a GLOBAL cross-landblock execution order - a serial
implementation detail, not the contract - and now pin workerCount: 1
with justification comments (LoadNear_OvertakesQueuedFarLoads,
TwoQueuedLoads_RetainTheirDistinctOriginAndGeneration).

Gates: Release build 0 errors; App suite 5575 passed / 3 skipped
(5568 + 7 new); Runtime suite 1756/0.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 19:34:20 +02:00
Erik
c6bc2bf7eb test(streaming): fix #351 — the Debug-only FarLoad strip-test failure was the deliberate near-payload Debug.Assert, not a flake
The test feeds a far-tier factory returning Near payload on purpose to
prove the strip safety net. LandblockStreamer.HandleJob is documented
"fail loud in Debug builds and strip in Release": the Debug.Assert fires
on exactly that input, the VSTest host translates it into a thrown
DebugAssertException, and the worker catch folds it into a Failed
completion — deterministic on every Debug run since the test and the
tripwire landed in the same commit (090b0354). Release compiles the
assert out ([Conditional("DEBUG")]), so the strip runs and the test
passed there, which is why the Release gate never saw it.

The test now pins BOTH halves of the config-divergent contract via
"#if DEBUG": Debug expects the loud Failed carrying the assert text,
Release keeps the strip assertions. Production code unchanged.
LandblockBuildOriginTests 11/11 in Debug AND Release.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-09 20:00:42 +02:00
Erik
ce41efb9e5 refactor(runtime): cut graphical host over to canonical root
Make every App composition phase borrow one GameRuntime, retire the duplicate view/event adapters, and dispose the root only after its graphical borrowers release. This preserves synchronous UI commands while giving shutdown one exact ownership ledger.

Co-authored-by: OpenAI Codex <codex@openai.com>
2026-07-26 19:06:09 +02:00
Erik
31e6e192b3 refactor(app): extract focused window lifetime
Move the exact retryable shutdown manifest, typed root snapshot, terminal reporting, and native-window-last release out of GameWindow. Keep session and GPU convergence as hard barriers while reporting persistent physical callback cleanup without stranding dependent owners.

Co-authored-by: Codex <codex@openai.com>
2026-07-22 19:43:52 +02:00
Erik
88f32dc4e2 refactor(app): compose live presentation startup 2026-07-22 17:55:15 +02:00
Erik
4a205a3e56 test(streaming): close lifecycle gate after cutover 2026-07-21 23:41:15 +02:00
Erik
c79d0a49da refactor(streaming): complete landblock presentation cutover
Remove the legacy GameWindow apply path and make the concrete render, physics, and static publishers the only production owner graph. Serialize full-window retirement with shared-origin teleport and session boundaries so old coordinate-frame resources cannot survive into a new world or login.
2026-07-21 22:47:30 +02:00
Erik
1029f8372e refactor(streaming): extract landblock render publisher 2026-07-21 20:47:01 +02:00
Erik
5af101b2f1 refactor(streaming): extract landblock build factory 2026-07-21 20:30:32 +02:00
Erik
090b0354ff refactor(streaming): capture landblock build origin 2026-07-21 20:12:56 +02:00