The rest of V4t. The composite, particle and shared-atlas texture paths now
hand out the device's GpuTextureSlot instead of a raw 64-bit
ARB_bindless_texture handle, and GroupKey, CachedBatch and ObjectRenderBatch
carry that slot. WbDrawDispatcher, EnvCellRenderer and ParticleRenderer retire
their interim GlBindlessHandleTable instances and share the device's one
table, exactly as V4t-1 did for terrain. Nothing about world submission
changes otherwise: these three renderers are still raw GL, still bind binding
9 themselves, and still draw the same geometry in the same order.
**What produces a slot now.** CompositeTextureArrayCache's GL backend interns
each array's handle when it makes it resident and retires the entry when it
makes it non-resident, so the pair is created and destroyed together and the
cache above it never learns a device exists — the fake backend its tests use
mints a stand-in slot. TextureCache.AcquireParticleTexture does the same for
the one-layer particle arrays it owns, including on its rollback path.
ObjectMeshManager registers each shared atlas's wrap/clamp handles at batch
upload; registration is idempotent by handle, so the many batches sharing an
atlas share its entry.
**Slot release is stricter than what it replaces, not looser.** The interim
tables never released anything — the class comment said so — and they grew
without bound. The device's table has a fixed 16,384-slot capacity, so an
unreleased entry is now a leak with an end. Every producer therefore retires
its entry: the composite backend at MakeNonResident, the particle backend at
MakeNonResident, and ObjectMeshManager when a retiring atlas's PHYSICAL
retirement completes — the point at which its handles are already non-resident
and its texture already deleted. That last one needs the handles snapshotted
at eviction, because ManagedGLTextureArray.Dispose zeroes its own copies as
its first act. Teardown deliberately does not release: the device is being torn
down alongside its callers, so there is nothing left to recycle a slot into,
and deferring work through a possibly-disposed retirement queue would turn a
clean shutdown into a throw.
**The default value became load-bearing, and that is the one real hazard here.**
BindlessTextureLocation could say "not resolved" with handle 0, because no
texture has handle 0. A slot index has no spare value — default(GpuTextureSlot)
is real slot 0 — so a positional record would have turned every
budget-rejected or still-uploading composite into a silent read of whichever
texture registered first. That is the magenta-placeholder failure shape one
layer down. The type is now a struct storing the slot one-based, so default IS
Unresolved, with a test pinning both halves: default is unresolved, and a
location naming slot 0 is resolved and distinguishable from it. Elsewhere the
sentinel is already exact — GpuTextureSlot.Unassigned is 0xFFFFFFFF, which is
common.glsl's ACDREAM_TEXTURE_NONE — so the classify path's "no texture yet"
test and the particle billboard's untextured branch are unchanged in meaning.
**GroupKey ordering is preserved because the key never ordered anything.**
Handle→slot is a bijection (the device interns one slot per resident handle),
so the same (entity, batch) pairs bucket together as before. The key reaches
equality, hashing and the scene-digest fingerprints — never a comparator:
opaque and translucent groups sort by cull mode then camera distance, the
delayed-alpha path by viewer distance then submission ordinal, and group
enumeration follows the persistent dictionary's insertion order, which a
changed hash does not disturb. The digests hash the slot index where they
hashed the handle; both sides of the render-shadow comparison compute them the
same way, so the value changing is invisible to it. Read
CompareOpaqueSubmissionOrder, CompareTransparentSubmissionOrder and
AlphaFingerprintComparer before doubting this — sort-order drift is a
pixel-visible regression class this project has hit, and it is why the check
was made before the retype rather than after.
**One visibility change, forced rather than chosen.** BindlessTextureLocation
was public and now holds an internal contract type, so it is internal;
ObjectRenderBatch.TextureSlot is internal on an otherwise public class for the
same reason. Nothing outside this assembly and its InternalsVisibleTo test
assemblies named either.
**SkyRenderer keeps its interim table**, and the report should say why: the
sky's textures are minted by SkyRenderer itself from TextureCache's raw GL
texture names, which this slice does not retype, so it would be the one
consumer registering handles it produced — a different shape from the world
stack. The offline gate also masks the sky band, so the one automated
instrument here cannot see a sky regression. V4f owns that renderer.
**Gates.** GL offline pixel gate vs cb2a70b8, measured twice: 31 and 22
differing pixels of 563,200 (5.50e-05, 3.91e-05). The first is above the
plan's documented 15-23 px band, so a control was measured rather than
assumed: two same-commit captures at this tree differ by 19 px, and — the
decisive number — a capture at V4t-1 and a capture at this commit differ by
9 px, fewer than the same-commit control. Maximum channel delta is 41-52 in
every pair including the controls, i.e. the differing pixels are drawn from
one flickering population, not from moved geometry. tools/run-repeat-connected-gate.ps1
-Runs 3: 3/3 RENDERED on both the desktop witness and the client capture. One
Vulkan composition-host run with VK_LAYER_KHRONOS_validation proven inserted
by the loader: zero errors, zero warnings, converged ownership ledger. App
tests 4,077 / 3 skips and the complete Release suite 9,140 / 5 — both the
4,075 and 9,138 baselines plus the two tests added here.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The decisive probe (3cf6bcc) caught it live in ONE session: a 43-part
staircase entity (src=0x020003F2, healthy MeshRefs tZ=[0.35..15.15])
drew with cache=hit:3 restZero=3 - THREE batches belonging to a 1-part
entity - then under a different hint the correct hit:119. Two
compounding bugs:
1. interiorIdBase = 0x40000000 | (landblockId & 0x00FFFF00) resolved to
0x40YYFF00 for landblock keys 0xXXYYFFFF - the landblock X byte
DISCARDED. Every landblock in a map Y-row shared one id space:
Holtburg town A9B3's 9th interior stab == the AAB3 tower's spiral
staircase, both 0x40B3FF09. Fixed to 0x40000000|(lbX<<16)|(lbY<<8)
(the scenery 0x80XXYY## scheme).
2. The Tier-1 classification cache's #53 tuple key (EntityId,
LandblockHint) was fed the PLAYER's landblock at bucket-draw time
(RetailPViewRenderer.DrawEntityBucket fabricates its tuple with
ctx.PlayerLandblockId), so colliding ids from different landblocks
shared a key: whichever entity classified first under a hint won,
and the loser wore its batches all session (static fast path never
re-classifies). Also: bucket-hinted entries were never swept by
InvalidateLandblock(owner) - stale entries survived owner unload.
Fixed: ResolveCacheLandblockHint derives the hint from the entity's
owning cell (ParentCellId landblock, canonical 0xXXYYFFFF), falling
back to the tuple id for ownerless paths (outdoor stabs/scenery,
where the tuple IS the owner).
Explains the session-shaped repro exactly: town-login + run to the
tower hydrates/classifies town interiors first -> the tower staircase
cache-hits the town twin's batches (stairs missing/partial + a wrong
object near the floor - the "water barrel"); login-inside classifies
the tower first -> usually clean. meshMissing=0 / entSeen==entDrawn
both ways (everything draws, wrong batches). Likely also feeds #113's
distance-dependent phantom staircase (the town twin wearing the
tower's staircase batches).
3 new cache tests pin the collision contract + hint derivation.
Suites: App green / Core 1430+2skip / UI 420 / Net 294.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
User confirmed via A/B test (ACDREAM_DISABLE_TIER1_CACHE=1) that the
visual bug — buildings rendering up in the air outside Holtburg — is in
the cache wiring, not elsewhere. The matrix math (restPose * entityWorld
== model) was provably correct, so the bug had to be cache key collision.
Stabs were namespaced in commit 71d0edc, but scenery (0x80LLBB00 +
localIndex) and interior (0x40LLBB00 + localCounter) still have the
same 256-overflow risk. Dense LBs outside Holtburg (forest, urban) push
localIndex past 255, wrapping into the lbY byte and creating cross-LB
collisions.
Fix: change the cache key from uint entityId to (uint, uint) tuple of
(EntityId, LandblockHint). The cache is now correct-by-construction
regardless of any hydration path's Id-generation strategy. Defensive
against future regressions in any ID namespace.
InvalidateEntity becomes a sweep (was O(1)), but it's called rarely
(only on live-entity despawn). InvalidateLandblock was already a sweep.
Updated 14 existing cache tests + 1 dispatcher integration test to thread
landblockHint through TryGet / DebugCrossCheck calls.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds EntityClassificationCache.DebugCrossCheck(entityId, liveBatches) that
asserts cached state matches a live re-classification. Wires a simpler
predicate assert into WbDrawDispatcher's cache-hit branch (asserts
isAnimated == false on cache hit). Tests #13a and #13b cover the
batch-count mismatch and clean-match cases via a custom TraceListener
that captures Debug.Assert calls.
Zero cost in Release. In DEBUG, the assert fires immediately if a future
regression mutates static-entity state outside the audit's known write
sites — the same failure mode that bit the prior Tier 1 attempt.
Phase 4 complete. Cache + invalidation + safety net all in place.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
GameWindow.RemoveLiveEntityByServerGuid now invalidates the entity's
cache entry next to the existing _animatedEntities.Remove(). Fires for
DeleteObject (0xF747) and the dedup leg of ObjDescEvent (0xF625).
Adds test #15 (despawn-respawn under reused id repopulates fresh) per
spec section 7.5 — pins the audit's ObjDescEvent-as-despawn-respawn contract.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Idempotent removal of a cached entry by entity id. Tests #4 and #5 from
spec section 7.1 lock in the contract.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Implements Populate (insert-or-overwrite) and adds 5 tests covering the
populate->TryGet round-trip including the Setup pre-flatten shape. Per
spec test plan section 7.1 tests #2, #3, #9, #10, #14.
Tests use xUnit Assert.* (not FluentAssertions) to match the Task 2
implementer's choice and the existing 149 sibling assertions in the Wb
test directory.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds CachedBatch, EntityCacheEntry, and EntityClassificationCache with
just TryGet (returns false on empty). The skeleton compiles and the first
test (TryGet_EmptyCache_ReturnsFalse) passes. Subsequent tasks add
Populate, InvalidateEntity, InvalidateLandblock, and the dispatcher
integration. Per spec design Section 6.1.
Note: CachedBatch / EntityCacheEntry / EntityClassificationCache are
internal (not public as the plan snippet showed). Their members
transitively reference the internal GroupKey type, so promoting them to
public produces CS0051 inconsistent-accessibility errors. The cache is
dispatcher-internal coordination state anyway, and the AcDream.App
csproj already exposes internals to AcDream.Core.Tests via
InternalsVisibleTo, so the test sees everything it needs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
