Catcrafts/Crafter.Graphics
2
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

perf(ui): hoist loop-invariant per-item math out of the per-pixel loop (#125)

Browse source 
The cooperative-load section already streams each item into shared memory
once per workgroup, but the per-pixel inner loop then recomputed values
that are constant for the whole item/glyph. The compiler can't hoist them
itself — they read shared memory at a varying index.

Precompute them once per item at load time into new shared slots:

  - ui-images / ui-text / ui-fused (images+text phases): invRectSize =
    1.0/rect.zw, turning the per-pixel `(sp-rect.xy)/rect.zw` vec2 divide
    into a multiply.
  - ui-text / ui-fused (text phase): the SDF AA `band` (a vec2 divide +
    two maxes depending only on the glyph's uv span and rect size).

In ui-fused the new slots (s_inv, s_band) are reused across phases like
the existing s_v* scratch, so the LDS bump is per-category, not summed.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
catbot 2026-06-18 13:22:07 +00:00
commit ac9180076c
3 changed files with 45 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

27
shaders/ui-fused.comp.glsl
View file

@ -70,6 +70,13 @@ shared vec4 s_v1[UI_CHUNK];
shared vec4 s_v2[UI_CHUNK];
shared vec4 s_v3[UI_CHUNK];
shared uvec4 s_v4[UI_CHUNK];
// Per-item constants precomputed once at cooperative-load time, REUSED across
// phases like the s_v* scratch (the per-pixel inner loops read them instead of
// recomputing per pixel × item):
// images: s_inv = 1.0/rect.zw
// text: s_inv = 1.0/rect.zw, s_band = SDF AA scale (a divide + two maxes)
shared vec2 s_inv[UI_CHUNK];
shared float s_band[UI_CHUNK];
shared uint s_keep[UI_CHUNK];
shared uint s_order[UI_CHUNK];
shared uint s_count;
@ -232,6 +239,7 @@ void main() {
s_v1[lid] = uiImageHeap[heap].items[idx].uv;
s_v2[lid] = uiImageHeap[heap].items[idx].tint;
s_v4[lid] = uiImageHeap[heap].items[idx].slots;
s_inv[lid] = 1.0 / s_v0[lid].zw;
keep = uiAabbOverlapsTile(s_v0[lid].xy, s_v0[lid].xy + s_v0[lid].zw,
tileMin, tileMax);
}
@ -249,7 +257,7 @@ void main() {
if (sp.x < lo.x || sp.y < lo.y) continue;
if (sp.x >= hi.x || sp.y >= hi.y) continue;
vec2 t = (sp - s_v0[c].xy) / s_v0[c].zw;
vec2 t = (sp - s_v0[c].xy) * s_inv[c];
vec2 uv = mix(s_v1[c].xy, s_v1[c].zw, t);
uint texSlot = s_v4[c].x;
@ -282,6 +290,13 @@ void main() {
s_v0[lid] = uiGlyphHeap[heap].items[idx].rect;
s_v1[lid] = uiGlyphHeap[heap].items[idx].uv;
s_v2[lid] = uiGlyphHeap[heap].items[idx].color;
s_inv[lid] = 1.0 / s_v0[lid].zw;
// SDF AA band — atlas-px per screen-px, constant per glyph
// (uvSpan * kAtlasSize(1024) / screenSpan), max'd to 1px floor.
vec2 uvSpan = s_v1[lid].zw - s_v1[lid].xy;
vec2 atlasPerScreen = (uvSpan * 1024.0) * s_inv[lid];
float scalePx = max(atlasPerScreen.x, atlasPerScreen.y);
s_band[lid] = max(scalePx, 0.0001);
keep = uiAabbOverlapsTile(s_v0[lid].xy, s_v0[lid].xy + s_v0[lid].zw,
tileMin, tileMax);
}
@ -299,7 +314,7 @@ void main() {
if (sp.x < lo.x || sp.y < lo.y) continue;
if (sp.x >= hi.x || sp.y >= hi.y) continue;
vec2 t = (sp - s_v0[c].xy) / s_v0[c].zw;
vec2 t = (sp - s_v0[c].xy) * s_inv[c];
vec2 uv = mix(s_v1[c].xy, s_v1[c].zw, t);
// Font slots are push constants — provably dynamically uniform,
@ -312,12 +327,8 @@ void main() {
float dAtlas = (ON_EDGE - sdf) * DIST_SCALE;
vec2 uvSpan = s_v1[c].zw - s_v1[c].xy;
vec2 atlasPerScreen = (uvSpan * 1024.0) / s_v0[c].zw;
float scalePx = max(atlasPerScreen.x, atlasPerScreen.y);
float band = max(scalePx, 0.0001);
float a = clamp(0.5 - dAtlas / band, 0.0, 1.0);
// band (the AA scale) was precomputed once per glyph at load.
float a = clamp(0.5 - dAtlas / s_band[c], 0.0, 1.0);
if (a <= 0.0) continue;
vec4 col = s_v2[c];