a8ed369c71
fontTextureSlot/fontSamplerSlot are push constants, hence provably dynamically uniform. Wrapping them in nonuniformEXT forced the per-invocation divergent-descriptor path and blocked hoisting the uniform descriptor load out of the per-pixel glyph loop. Removing the decoration is zero correctness risk — it is purely a read-index hint. Not applied to ui-images.comp.glsl, whose slots come from an SSBO load the compiler cannot prove uniform. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
68 lines
2.4 KiB
GLSL
68 lines
2.4 KiB
GLSL
#version 460
|
||
#extension GL_GOOGLE_include_directive : enable
|
||
#include "ui-shared.glsl"
|
||
|
||
// One workgroup per 8×8 screen tile. Iterates every glyph in order; each
|
||
// pixel keeps a local accumulator so order in the buffer == draw order.
|
||
layout(push_constant) uniform PC {
|
||
UIDispatchHeader hdr;
|
||
uint fontTextureSlot;
|
||
uint fontSamplerSlot;
|
||
uint _p0;
|
||
uint _p1;
|
||
} pc;
|
||
|
||
layout(local_size_x = 8, local_size_y = 8, local_size_z = 1) in;
|
||
|
||
// SDF tuning — must match Crafter::FontAtlas::kOnEdgeValue / kPixelDistScale.
|
||
const float ON_EDGE = 128.0 / 255.0;
|
||
const float DIST_SCALE = 32.0;
|
||
|
||
void main() {
|
||
ivec2 screenPx;
|
||
if (!uiResolveScreenPixel(pc.hdr, screenPx)) return;
|
||
|
||
vec4 dst = imageLoad(uiImages[pc.hdr.outImage], screenPx);
|
||
vec2 sp = vec2(screenPx) + 0.5;
|
||
|
||
for (uint i = 0u; i < pc.hdr.itemCount; ++i) {
|
||
GlyphItem it = LoadGlpyhtem(pc.hdr.itemBuffer, i);
|
||
|
||
vec2 lo = it.rect.xy;
|
||
vec2 hi = it.rect.xy + it.rect.zw;
|
||
if (sp.x < lo.x || sp.y < lo.y) continue;
|
||
if (sp.x >= hi.x || sp.y >= hi.y) continue;
|
||
|
||
vec2 t = (sp - it.rect.xy) / it.rect.zw;
|
||
vec2 uv = mix(it.uv.xy, it.uv.zw, t);
|
||
|
||
// Font slots are push constants — provably dynamically uniform, so no
|
||
// nonuniformEXT: lets the compiler hoist the descriptor load out of the
|
||
// per-pixel glyph loop and avoid the divergent-descriptor path.
|
||
float sdf = texture(
|
||
sampler2D(uiTextures[pc.fontTextureSlot],
|
||
uiSamplers[pc.fontSamplerSlot]),
|
||
uv
|
||
).r;
|
||
|
||
// Distance in atlas-pixels (negative inside the glyph).
|
||
float dAtlas = (ON_EDGE - sdf) * DIST_SCALE;
|
||
|
||
// Atlas-px per screen-px along this glyph's transform — keeps AA crisp
|
||
// at any rendering size. uvSpan * atlasSize / screenSpan.
|
||
vec2 uvSpan = it.uv.zw - it.uv.xy;
|
||
// FontAtlas::kAtlasSize = 1024.
|
||
vec2 atlasPerScreen = (uvSpan * 1024.0) / it.rect.zw;
|
||
float scalePx = max(atlasPerScreen.x, atlasPerScreen.y);
|
||
// 1-screen-px AA band, expressed in atlas-pixel units of dAtlas.
|
||
float band = max(scalePx, 0.0001);
|
||
|
||
float a = clamp(0.5 - dAtlas / band, 0.0, 1.0);
|
||
if (a <= 0.0) continue;
|
||
|
||
vec4 src = vec4(it.color.rgb, it.color.a * a);
|
||
dst = uiBlendOver(dst, src);
|
||
}
|
||
|
||
imageStore(uiImages[pc.hdr.outImage], screenPx, dst);
|
||
}
|