Per verdict s1: the judge keeps the fixture tolerance as correctness and adds a recorded sha256 of each family's
result per route (a change in the computation fails until re-recorded); a test refuses mediump/lowp/precision
other than highp and min16float/min10float/min16int/half in the committed shaders; phacelle's tolerance text states
a device clause (2^-18) that the judge parses and enforces. The committed bounds file lacks the new column until
the next WRITE_BOUNDS=1 job, so this commit's device run fails and it does not pass DeviceResultsHoldFixture.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Only the trophic families after the inserted segment families change; every family before them is identical.
Builds; the parent commit does not pass CommittedGlslEqualsLowering.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Shader overloads stay native (ruling shader-overloads-stay-native): cultmath_segment_closest and
cultmath_distance_to_segment have float2 and float3 forms again, with mirror asserts and fixture families.
The rejected "avoid overloads" comment and design text are gone. The border step's outside-cell behaviour
(distance 0 with that plane's normal) is documented in C# and the HLSL header and pinned by a test; the mirror
and fixture gain distinct sites closer than 1e-2. GLSL and fixture regenerate in the next commit; this one does
not pass CommittedGlslEqualsLowering.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tools/device/gen.py spells one HLSL and one GLSL ES 3.10 compute shader per work
item, vkrun.c dispatches it on the first Vulkan device, run-lavapipe.sh runs the
whole conforming job. None of them compares. Also fixes the metric test's family
choice. Not yet run: these may not build or run.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
GlslMirrorTests becomes partial; GlslMirrorTests.Device.cs holds WriteDeviceWork
(the work list, with reference families for the four sampled-form families) and
DeviceResultsHoldFixture (exact, check, sampled-form and recorded-bound judging),
plus simulated-device tests for each rule. Not yet built or run: this commit may
not build.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The earlier version read past a one-element array and passed under the mutant by hitting a non-finite site.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
power_border_step/power_border give the exact distance to the owner's power cell boundary: the least radical-plane
distance over every other site, built from the offset and (w1-w2)(w1+w2) so it does not cancel. site_cells_border is
documented as the two-winner bisector distance. Cell spans refuse non-finite positions, queries and spans over 2^24.
segment_closest is public in C#; the HLSL segment functions are 3D only (no new overload sets). Tests added.
Does not build the GLSL checks: CultMath.glsl and glsl-parity.json are regenerated in the next commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The wrapper built one empty argument for an empty parameter list, so cultmath_site_cells_empty() failed glslang.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Regenerated with CULTMATH_WRITE_GLSL=1 on Yggdrasil; the fixture adds seven families and deletes no line.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
CommittedGlslEqualsLowering and GoldenFixtureMatchesCSharp fail at this commit until the next one regenerates the GLSL and the fixture.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
CommittedGlslEqualsLowering fails at this commit until the GLSL is regenerated in the next one.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Verdicts cut-stage-shaders-device.s1/s2 hold. main's interim changes are Rust packages only (cultnet-rs, cultcache-rs, cultmesh-rs), disjoint from the verified C#/shader tree.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Generated by CULTMATH_WRITE_GLSL=1 on Yggdrasil; glsl-parity.json is unchanged (the af and
span families keep their bytes). Builds.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
span_slack and af_snoise each spelled 2^-14 (value) and 2^-12 (gradient) inline, in C# and
HLSL. Both now read the constants beside SNOISE_HESSIAN (math.Affine.cs, CultMath.Affine.hlsl,
Unity copies), so a device-judged widening moves every reader. Bits unchanged. The test
oracles in AffineTests and GlslMirrorTests read the constants; a new test pins af_snoise's
allowance to span_slack's terms. span_slack's comment no longer claims it is the only
allowance. Does not build green yet: CultMath.glsl is not regenerated in this commit.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
cultnet-rs 0.2.0: every received MessagePack byte decodes once through
cultcache-rs's door; the dynamic document lane is parked at
parked/cultnet-rs-dynamic-document-lane; one decoder per message family;
duplicate keys refused; refusals name schema and fields, never values;
cultmesh-rs 0.2.0. Cut cultnet-rs-door r2, held by Soul s1-s3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
EveryMirrorFunctionMatchesCSharpMath compared both new twins bit for bit against the C# seam;
only its pinned count of mirrored functions moves from 125 to 127. Builds.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Generated by CULTMATH_WRITE_GLSL=1 on Yggdrasil (dotnet/sdk:10.0) at 5ccbb094; the fixture
diff is additions only, every earlier family byte-identical. Builds.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Four coordinate-band snoise_grad / snoise_hessian families with Verify AssertSpanAllowance,
SpanAllowanceCoversReadsAlongTheSpan, NoiseBoundTests.GradDouble and its pin, design.md
paragraph. The fixture and CultMath.glsl are regenerated in the next commit; this one compiles.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The inline rounding and the three allowance literals of span_stage and cultmath_span_stage
move into internal span_rounding / span_slack and their HLSL twins; operations and order are
unchanged. CultMath.glsl is not regenerated in this commit, so
GlslMirrorTests.CommittedGlslEqualsLowering fails here; it compiles.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Raw v1 is decoded by the enum's derive alone (SnapshotResponseRawV1Rows,
the v1 arm and its helpers deleted); the five raw families carry
skip_deserializing so the hand reader is their only decoder. The four hand
readers refuse a duplicated key; refusals carry a schema-shaped
schemaVersion context and never a received value. README describes only
the raw lane. A same-length str-payload credential forgery is pinned.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
describe_decode_error keeps serde's own value-free texts for missing and
duplicate fields (static names from the type), maps unknown-variant and
invalid-type to kind descriptions, and leaves everything else opaque.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
