test: Test case for tgpu.unroll emulating dynamic dispatch

[iwoplaza]

No description provided.

[github-actions]

pkg.pr.new

packages
Ready to be installed by your favorite package manager ⬇️

https://pkg.pr.new/software-mansion/TypeGPU/tgpu-gen@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/tinyest-for-wgsl@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/typegpu@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/cli@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/color@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/gl@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/noise@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/radiance-cascades@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/react@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/sdf@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/three@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

https://pkg.pr.new/software-mansion/TypeGPU/unplugin-typegpu@f886e403fa7c862a94d7fd6c8e33f4fe8cce0228

benchmark
view benchmark

commit
view commit

[Copilot]

Pull request overview

Adds a new tgpu.unroll test that demonstrates a pattern for emulating “dynamic dispatch” on the GPU by unrolling over a compile-time list of keys and selecting an implementation via an index.

Changes:

• Added a new tgpu.unroll test case that builds a “dispatch table” (Tonemap) and selects an implementation based on fnIdx.
• Introduced a snapshot asserting the generated WGSL contains the unrolled selection logic and direct calls to the chosen functions.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Bundle size comparison (import * as ... in PR vs import * as ... in target):

🟢 Decreased	➖ Unchanged	🔴 Increased	❔ Unknown
0	321	0	0

import { ... } in PR vs import * as ... in PR (is the library tree-Shakeable?):

Test	tsdown
tgpu_init.ts	256.12 kB ($${\color{green}-3.2\%}$$)
tgpu_initFromDevice.ts	255.61 kB ($${\color{green}-3.4\%}$$)
tgpu_resolve.ts	161.53 kB ($${\color{green}-38.9\%}$$)
tgpu_resolveWithContext.ts	161.47 kB ($${\color{green}-39.0\%}$$)
tgpu_bindGroupLayout.ts	69.02 kB ($${\color{green}-73.9\%}$$)
tgpu_mutableAccessor.ts	66.00 kB ($${\color{green}-75.1\%}$$)
tgpu_accessor.ts	65.98 kB ($${\color{green}-75.1\%}$$)
tgpu_privateVar.ts	65.32 kB ($${\color{green}-75.3\%}$$)
tgpu_workgroupVar.ts	65.32 kB ($${\color{green}-75.3\%}$$)
tgpu_const.ts	64.54 kB ($${\color{green}-75.6\%}$$)
tgpu_fn.ts	37.99 kB ($${\color{green}-85.6\%}$$)
tgpu_fragmentFn.ts	37.92 kB ($${\color{green}-85.7\%}$$)
tgpu_vertexFn.ts	37.74 kB ($${\color{green}-85.7\%}$$)
tgpu_computeFn.ts	37.44 kB ($${\color{green}-85.8\%}$$)
tgpu_vertexLayout.ts	26.63 kB ($${\color{green}-89.9\%}$$)
tgpu_comptime.ts	14.43 kB ($${\color{green}-94.5\%}$$)
tgpu_unroll.ts	2.60 kB ($${\color{green}-99.0\%}$$)
tgpu_slot.ts	1.57 kB ($${\color{green}-99.4\%}$$)
tgpu_lazy.ts	1.22 kB ($${\color{green}-99.5\%}$$)

If you wish to run a comparison for other, slower bundlers, run the 'Tree-shake test' from the GitHub Actions menu.

[github-actions]

Resolution Time Benchmark

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Random Branching (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [1.01, 2.01, 4.53, 6.57, 7.55, 11.32, 22.26, 25.89]
  line [0.99, 1.95, 4.38, 7.43, 7.84, 11.32, 22.46, 23.44]
  line [0.97, 1.99, 4.31, 6.64, 7.56, 10.50, 22.25, 25.70]

Loading

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Linear Recursion (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.31, 0.52, 0.70, 0.88, 1.18, 1.27, 1.56, 1.67]
  line [0.33, 0.52, 0.69, 0.86, 1.14, 1.24, 1.46, 1.65]
  line [0.32, 0.59, 0.79, 0.89, 1.23, 1.30, 1.46, 1.55]

Loading

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Full Tree (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.91, 2.10, 4.29, 6.50, 12.57, 26.93, 57.36, 116.51]
  line [0.98, 2.09, 4.23, 6.74, 13.05, 26.52, 56.59, 119.71]
  line [1.02, 2.18, 4.23, 6.52, 12.91, 27.16, 56.55, 117.16]

Loading

[pullfrog]

✅ No new issues found.

Reviewed changes — added a single test demonstrating dynamic dispatch emulation via tgpu.unroll over compile-time function variants.

• Adds the allows for emulating dynamic dispatch test in packages/typegpu/tests/unroll.test.ts — verifies the generated WGSL for unrolled conditionals over Tonemap keys, including an empty skip for the None variant and per-function index checks for Power and Tanh.

  ^{｜ Fix it ➔ ｜ View workflow run ｜ Using Kimi K2 (free via Pullfrog for OSS) ｜ 𝕏}