feat(render): V6e — move the world mesh's texture lookup to where Vulkan can express it

Campaign V slice V6e, first of three. mesh_modern is the shader every world
static, every piece of scenery and every EnvCell surface draws through, and it
was one of the four production pairs the SPIR-V toolchain still refused.

The blocker was a varying. Since V2 the vertex stage looked a batch's table slot
up in the binding=9 handle table and forwarded the resulting 64-bit
GL_ARB_bindless_texture handle to the fragment stage as a `flat uvec2`. That
works on GL because a bindless handle is just a number a shader may carry
anywhere. It cannot work on Vulkan at all: the equivalent object is a descriptor
in set 2, and a descriptor is not a value a stage can hand to another stage. So
what travels between the stages is now the SLOT — a `flat uint` — and the
fragment stage does the lookup at the point of sampling.

That relocation needs one shared idea, because the two backends disagree about
what the lookup IS. `ACDREAM_SAMPLE_ARRAY(slot, uvw)` asks the dialect-neutral
question — "sample table slot N" — and expands to
`texture(sampler2DArray(gTextureTable[slot]), uvw)` under GL and to
`texture(uTextures[nonuniformEXT(slot)], uvw)` under Vulkan. It is deliberately
a SAMPLING macro rather than a sampler-returning one: `nonuniformEXT` belongs on
the indexing expression itself, and binding the result to a local
`sampler2DArray` first is exactly where an implementation is free to drop it.
That is the same shape V6d already used for the retained UI's 2-D reads, and it
now covers the array reads the world path needs.

`ACDREAM_TEXTURE_NONE` lands alongside it, unused here and used by the next
commit. GL can ask "does this slot hold a texture" of the payload, because an
unregistered slot holds the null handle; Vulkan cannot, because set 2 is opaque
and reading an unwritten element of a partially-bound array is undefined rather
than zero. The sentinel moves that answer into the index, where both dialects
test it identically.

On GL nothing about the sampled result changes — the same slot resolves to the
same handle to the same texel. The SSBO read simply happens one stage later,
and `flat` keeps it one scalar load per primitive rather than per fragment.

Also: RenderBootstrap has been loading mesh_modern without common.glsl since V2,
which cannot have linked — `ACDREAM_UBO_SET` sits inside a layout qualifier
there. The UI Studio path is the only caller. One argument, same pair, same way
WorldRenderComposition has always loaded it.

Gates: Release build clean; App tests 4,057 passed / 3 skipped (baseline);
offline pixel gate against 95f8c25f differing fraction 3.37e-05 (~19 px of
563,200), inside the documented 15–23 px same-commit noise band and ~30x under
the 0.001 threshold. mesh_modern is the shader that gate covers most heavily,
so this is the strongest automated evidence any V6e commit gets.

Manifest: 4/9 pairs compile (debug_line, mesh_modern, ui_text, vk_probe).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Erik 2026-07-28 09:21:22 +02:00
parent 95f8c25f31
commit 935f4dc3d9
8 changed files with 80 additions and 28 deletions

View file

@ -12,9 +12,8 @@ struct InstanceData {
// Campaign V slice V2 (2026-07-27): textureHandle (uvec2, a 64-bit
// GL_ARB_bindless_texture handle) became textureIndex (uint) plus an explicit
// pad word. textureIndex is a slot into the binding=9 handle table
// (ACDREAM_TEXTURE_HANDLE, common.glsl) that main() below looks up once per
// vertex to reconstruct the exact same uvec2 handle main() used to receive
// directly — one indirection, identical value. The pad word keeps
// (common.glsl) which main() below forwards to the fragment stage, where slice
// V6e moved the lookup so the same source compiles for Vulkan. The pad word keeps
// textureLayer/flags at their original std430 offsets (8/12), so the struct
// is still 16 bytes and every existing CPU writer's layout is unchanged
// (GpuBindingModel.GpuBatchDataStrideBytes).
@ -289,7 +288,14 @@ out vec3 vNormal;
out vec2 vTexCoord;
out vec3 vWorldPos;
out vec3 vLit; // A7: per-vertex Gouraud lighting (ambient + capped lights)
out flat uvec2 vTextureHandle;
// Campaign V slice V6e: was `flat uvec2 vTextureHandle` — a raw 64-bit
// GL_ARB_bindless_texture handle handed across the stage boundary. A varying
// cannot carry a Vulkan descriptor, so what travels is the table SLOT and the
// fragment stage does the lookup (see mesh_modern.frag). Under GL the value
// sampled is bit-for-bit the one the vertex stage used to forward; the SSBO
// read simply happens one stage later, and `flat` keeps it one scalar load per
// primitive rather than per fragment.
out flat uint vTextureIndex;
out flat uint vTextureLayer;
out flat float vOpacityMultiplier; // #188
out flat vec2 vSelectionLighting;
@ -323,9 +329,9 @@ void main() {
vTexCoord = aTexCoord;
BatchData b = Batches[uDrawIDOffset + gl_DrawIDARB];
// Campaign V slice V2: reconstruct the SAME uvec2 handle the shader used
// to receive directly from BatchData, now via one binding=9 table lookup.
// vTextureHandle's type and every downstream frag-shader use are unchanged.
vTextureHandle = ACDREAM_TEXTURE_HANDLE(b.textureIndex);
// Campaign V slice V6e: forward the table SLOT untouched. V2 looked the
// handle up here and passed the handle; the lookup now lives at the sample
// site in mesh_modern.frag, which is the only form Vulkan can express.
vTextureIndex = b.textureIndex;
vTextureLayer = b.textureLayer;
}