UIFallbackParity renders one scene through both binding models and
compares the frames. It starts itself twice, once as is and once with
CRAFTER_GRAPHICS_NO_DESCRIPTOR_HEAP, so the fallback cannot drift from
the heap path on a machine that never takes it. The scene reaches what
HelloUI does not: the four standalone kernels next to the fused one, a
sampled image with two samplers indexed per item, an image slot and a
sampler slot that are freed and registered again, a resize, an image
created with ImageVulkan::Create's default consumer stage, and a
generic Dispatch whose push payload is not a whole number of words.
On a driver without the heap both runs take the fallback, and the test
then only checks that the scene renders. Validation errors raised
while rendering fail a run, which only bites when
CRAFTER_GRAPHICS_VALIDATION is set. They are counted before the frame
capture because Window::SaveFrame raises one itself (#157).
UIFusedShader now also compiles and validates the descriptor-set build
of the fused kernel and pins its push-constant layout: both builds are
fed the same UIFusedHeader bytes.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Add a fifth UI compute path — DispatchFused — alongside the four
per-element Dispatch* calls. It composites quads → circles → images →
text (canonical back-to-front order) in ONE dispatch that loads the
destination image once and stores once, eliminating the per-pass
load+store and the inter-pass VkMemoryBarriers a run of consecutive
Dispatch* calls would pay. The win materialises at >= 2 non-empty
categories; an absent category is a zero-trip, push-constant-uniform
loop (~free).
The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's
exact cooperative shared-memory tile-cull + per-pixel accumulate, so a
fused category is pixel-identical to its standalone pass. Categories run
sequentially per thread and share one set of shared-memory scratch, so
the VGPR/shared high-water mark stays at the per-category level, not the
sum.
Additive and non-breaking: the per-element Dispatch* calls and the
frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own
128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four
per-category clip vec4s). To interleave a custom ui.Dispatch() between
standard passes, bracket it with two DispatchFused calls — the dispatch
boundary is the explicit, app-declared flush point.
WebGPU (Vulkan-second): DispatchFused falls back to the per-element
Dispatch* calls in canonical order — one texture/sampler per dispatch
there can't feed the bindless image phase — so the result matches; only
the load/store fusion is Vulkan-only.
HelloUI now composites its background quads, mouse circle, and label
text in a single DispatchFused. Verified rendering identical on a live
Vulkan/Wayland session.
Resolves#47
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
