Soul: no defect; independent port of stegu 22434e04d7 matches (3D and 2D), goldens
confirmed independently, value jump 2.2e-6 over 20k crossings, gradient FD 9.7e-6,
every hand mutant killed. 2D follows upstream by ruling. Release follow-ups: rebuild
the Unity CultMath.dll, CHANGELOG entry, pin the two HLSL copies equal.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Closed after three Soul passes: the double iterate and progress stop hold over
~9000 cases against exact enumeration, zero allocation, deterministic; the stall
rule and fScale are now pinned (all six survivors killed). Ships in the next
CultMath release with Asura. Recorded, outside the documented range: 18/6000
gap misses at cond 1e5-3e5 with A scaled 1e-4..1e4 (solver or reference unknown).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Batches of small boxed problems whose passes end with the KKT test failing and
the cost unmoved (dependent free columns, releases the factor drops, weak-direction
noise-floor problems). They must report Converged, and land within the documented
gap of an exact active-set enumeration. Pins the stall comparison and the cost
scale in Objective, which no test reached before.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Seven values change: four snoise(float3) in HlslSemanticsTests and three snoise(float2)
in MathTests (the 2D (0,0) golden is 0 either way). Regenerated from the code after the
continuity probe passed. The tests no longer claim to match Unity.Mathematics, whose
snoise keeps the old constants.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Drops the unreproducible floors and detection-scale claims, and the seam-fluke
consistency filter (the spikes it hid were the snoise seam, now fixed and pinned by the
continuity probe). Tightens the tolerances to at least 5x the harness maximum (ridged
0.5 -> 0.05), corrects the point count in the bit-equality test, and caps the ridged
compounding loop so a mutant fails loudly instead of hanging.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Kernel radius^2 0.6 -> 0.5, scale 42 -> 105, permute (34x+1)x -> (34x+10)x, per
stegu/webgl-noise 22434e04d7 (fix 21d9fe23d7, scale ff3b5d34ea, permute a3e6d57095).
snoise_grad, fbm_grad, ridged_grad follow. C#, both HLSL copies, design.md and
THIRD-PARTY-NOTICES.md change together. The 2D snoise shares the permute helper, so
its values change too. Golden values are regenerated in the next commit.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
With b along the smallest left singular vector, |A^T b| is tiny, so the KKT
tolerance (1e-6 of it) sits below the gradient's rounding noise at cond(A) >=
1e6 and can never be met. The solver must stop on the progress rule, in at most
six passes, never at the cap. This is the path the progress rule exists for;
the earlier fixtures reached it only by accident.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Two tolerance behaviours had no observer: the scale taken at the box point
nearest the origin (a 6e-5 release on a 1000-scale problem, in either column
position), and the strict comparisons when the scale and the gradient are both
exactly zero (b = 0 at the origin takes no iteration).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Reconcile's withdrawal sent an untyped error with no subscription id and left the
peer holding the rows the subscription had delivered. It now sends removals for
those rows, then a variant_unsupported error whose details carry
field "subscriptionId" and the id (the existing details slots; no new wire key).
The subscribe and per-change paths use the same code. The error schema and the
Rust CultNetErrorCode accept variant_unsupported, so a Rust peer decodes it
instead of refusing an unknown code.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The full-step clamp repairs an out-of-box start on a free column, so the
warm-start clamp was invisible to the old test (Stryker: five survivors). The
new table runs one column with b beyond the box and a cap of 0, where the
solver returns the clamped start untouched, across finite, infinite and NaN
starts and finite and infinite bounds.
Workspace junk (including NaN doubles) is now checked on random boxed problems
that release bounds and use the tolerance, with bit-identical x and equal
iteration counts.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Measured with Stryker once Solve was actually mutated (21 survivors), most of
them were code that decided nothing:
- The finite-input loops over A and b duplicated the non-finite test on the
accumulated A^T b, which is exactly equivalent (floats cannot overflow a
double product, and NaN/infinity never cancel to a finite value). One check
on each A^T b entry remains.
- The tolerance scale is the gradient at the box point nearest the origin,
computed inline; the max with |A^T b| duplicated it whenever the origin is in
the box, and the scratch vector for it is gone.
- Bound violation is state * gradient, which is zero for a free column; the
three-way ternary duplicated the free-stationarity check.
- The warm-start clamp is Math.Clamp plus a NaN test.
- Workspace slices come off a cursor (Take) instead of hand-summed offsets.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Stryker left the refinement step (removed, negated, or subtracted) unobserved,
and it is unobservable for a reason: refinement with a same-precision residual
does not improve forward error, and Cholesky is backward stable, so the
residual the KKT test reads is already at rounding level. Measured on n=24
problems at cond(A) 1e3..1e6, the solver with and without the step gives the
same iteration counts and the same x error against a Gram-Schmidt QR
reference (both limited by cond(A^T A) * eps).
The refinement pass, its dp buffer and the scratch vector are deleted;
WorkspaceLength returns to 4n^2 + 15n. The doc states where the normal
equations stop being optimal (cost gap below 1e-6 |b|^2 through cond(A) 3e5,
about 2e-5 at 1e6, independent of the scale of A) and that a ridge row beyond
that is the caller's policy.
status is now initialised: the unassigned local made Stryker discard every
mutant inside Solve as a compile error, so 255 mutants were never measured.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- n=24 fully free problems at cond 1e2/1e3/1e4, and A x 1e6 with unscaled
bounds, converge in at most two iterations (a regression to cap-spinning
fails on the count, not on a clock).
- b = 0 from a far start reaches the optimum instead of the start.
- Warm-start clamp table, free duplicate column gets no step, parallel columns
at absolute scales 1e3 and 1e-3, near-singular free sets confirm once and
stop, near-collinear optimum at A x 1e6, and zero allocation on the invalid,
converged and cap paths (first and second call).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The working iterate now lives in double inside the workspace (WorkspaceLength
4n^2 + 17n); A, b and the API x stay float and x is converted in at the start
and out at the end. The gradient, the free-stationarity check and the KKT test
all run in the double domain, which removes the float noise floor that made the
free-stationarity re-solve spin to the cap on optimal points (cond >= ~20).
A fixed tolerance is no longer the only stopping rule: after a stationarity
miss or a release, a pass that lowers the double cost by no more than rounding
reports Converged.
The tolerance scale is the larger of the gradient at the origin and at the box
point nearest the origin, so it never depends on the warm start; b = 0 with a
far start is no longer judged converged by its own steepness.
Two clauses were equivalent or unobservable and are deleted: the d <= 0 pivot
test (implied by d <= tol * diag) and the dependent[] exclusion from the
free-stationarity check (the progress rule now terminates it).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The four hard-coded sets become flags, with @file response-file splicing.
A consumer keeps its own .args file instead of committing a patch against
the example. fixture.args carries CultLib own sets; the README documents
the flags. Committed fixture bytes are unchanged.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
C#, TypeScript, Python, Rust and CultMesh refuse an unknown
cultcache.store.* header and a fifth record slot, naming the version or the
key and schemaId. The silent slot-skipping and C#'s ignored header are
deleted, and CultMesh's second record decoder is gone. Shared vectors are in
tests/vectors/document-variants-c0/. v1 stores stay byte-identical. At base,
the single-file commit's read-modify-write rewrote a v2 store as v1, which
lost data; it now refuses.
Soul passed it. Cross-runtime interop was run and passes. Three C#
Caching tests fail on Linux at both base and tip: they are Windows
share-mode tests (a FileShare.Read handle blocking delete/replace), which
cannot pass on Linux.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Rust runtime gains the reference's content chunk messages on the raw
path, CultMeshCdnArtifactManifest with wire-ordered metadata,
normalize_hash, pack_content, validate_manifest,
answer_content_chunk_request (every bad request is answered, byte
identical to the reference server) and fetch_content (one owner of
receiving-side verification; refuses zero-size floods and unnormalised
hashes before any request). Vectors run both ways between C# and Rust.
The operator ruled Q-BP1 A: chunks travel on the session already held.
This is the explicit Rust content path until an authenticated connector
exists.
Soul passed it after two fix batches. cultnet-rs: content 24, cultnet 53,
rudp 7, lib 110. cargo-mutants on content.rs misses 0.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A C# caller that leaves ChunkHash, MessageId or RecordKey null writes msgpack nil; the
reference reads it back as null, treats it as blank and answers found:false. Rust now
decodes nil as blank in content chunk requests, so those requests are answered with the
reference bytes instead of dropped. Three nil vectors from the reference server are
judged by decoding them in Rust. Record two content-plane limits in the parity scope.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The function took state and lease digest from the caller, so anyone holding the
authority could sign active without reportHealth or assertWriteLease. It is now
private to the signer. Test: the subpath and root do not expose it. Docs:
waitForWriteLease throws out of publish if health is already non-warming;
MAX_RECENT_WARMING_PROOFS=64 is a recorded limit.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
`cultnet-ts/idunn-runtime` exposes the signer and loader alone. Web apps import
it instead of the package root, which also carries the Odin publisher and
node:dgram. The root export is unchanged. Test loads both under a stubbed
dgram: the subpath loads, the root fails.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The loaded authority is deeply frozen (bind, health contract, Expected,
Activation, identities), so a caller cannot steer what the signer signs; the
typed Buffers that could not be frozen are dropped (Expected canonical bytes,
Activation public key kept as hex only for the load-time comparison). A second
createIdunnRuntimeSigner on the same authority throws, so two signers cannot
emit sequence 1 under one runtimeInstanceId.
Tests: stateful set pinned against Rust (warming, active, degraded), lease
accepted for a route-answered warming, refused for another warming and another
target, one signer per authority, frozen authority.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The signer holds the last reported health, starting at warming, and gains
reportHealth(state). sign(detail), answerRouteObservation(request) and the Odin
publisher's publish(detail) no longer take a state, so the route answer and the
Odin publish cannot disagree and a fresh signer cannot answer active. Every
warming presence the signer signs becomes a valid write-lease target; the
publisher's recordPublishedWarming is deleted. Tests now assert every committed
stateless vector, dead ones included.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Output of the Rust generator (3c9ce38) on Yggdrasil. Every set is re-enrolled, so
all bytes change; adds service-stateful with lease.cc, the two wrong leases and
the warming/active/degraded vectors.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds `service-stateful` (streampixels-service shape, state lineage set,
write_lease_required) with a Rust-issued lease.cc, warming/active/degraded
vectors bound to it, and two wrong leases (other warming digest, other target).
Each vector still passes authenticate + correlate before it is written.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The peek left snapshot slots unread and rmp-serde refused the length
mismatch, so every store fell through to the legacy decode.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A hostile manifest of empty chunks no longer becomes a request flood: fetch_content
refuses a zero-size chunk beside other chunks, and settles the content hash and every
chunk hash (normalised, non-empty) before the first request.
Every bad chunk request is now answered found:false in the reference handler's
validation order, never refused at decode. validate_message no longer refuses content
chunk requests or responses, so a blank message id or hash, a prefix-only hash and a
negative size reach answer_content_chunk_request. The answer follows the reference's
double normalisation (handler, then CreateRecordKey), which fixes the prefix-only
hash spelling to ArgumentException on 'chunkHash'. The cs vectors gain nine
reference-server answers (blank and whitespace id, blank, whitespace and prefix-only
hash, negative size, record-key case, blank record key, double prefix) and a
metadata-order manifest; Rust answers all of them byte-identically.
Manifest metadata is Vec<(String, String)> in wire order and refuses a duplicate key,
replacing the BTreeMap that re-sorted keys.
Tests pin the surviving mutants: unusable chunk sizes, the payload-length and
claimed-size halves of the chunk check, stable sort at equal offsets, the normalised
hash and canonical key actually sent, and the kind/mime defaults never trimming.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
idunn-runtime-authority.ts now loads and signs only: no Odin endpoint, no Odin
dependency check, no RUDP. createIdunnRuntimeSigner signs presence and answers
Idunn's route challenge with answerRouteObservation(request, state), a
synchronous local signature in the state the app reports. The RUDP publisher
moves to idunn-odin-presence-publisher.ts for services that declare
odin.verse-rendezvous, and keeps the Expected-endpoint check.
Deleted: the TS capability-minimum refusal (the Rust correlation records the
typed disagreement), latestPresenceDocument, the forced "active", and the
publisher's redundant health-contract and bind checks. Route validation now
accepts http, tcp and rudp as Rust does; the machine-id falls back to
/var/lib/dbus/machine-id. Tests run against the Rust-written fixtures.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Expected, Activation, identities and signed presence vectors from the cultnet-rs
example; the README says how to regenerate them.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
PersistedRecord refuses a fifth slot naming its key and schema id, and an
unknown cultcache.store.* header no longer falls through to the legacy
envelope decode that hid the cause.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Delete the raw_record[:4] slice and the legacy-envelope fallthrough for
unknown cultcache.store.* headers.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The store reader and the inspector share one header and record-slot rule;
an unknown cultcache.store.* header no longer falls through to the legacy
envelope decoder.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
