[RFC] feat: Generate requires directives

[aleksanderkatan]

See https://www.w3.org/TR/WGSL/#language-extensions-sec

Requires directive must be located before any declarations in the WGSL code.
This breaks when concatenating multiple resolves, even with a namespace enabled.
A possible solution:

• add a flag (default=true) for generating 'requires' in resolve/root,
• ignore new requires in subsequent resolves with a namespace.

Or we can just don't generate the 'requires' directive at all, it is not necessary.
Or we could check the navigator (if available), and warn if features are unavailable on the resolving host instead of emitting 'requires'.

[github-actions]

pkg.pr.new

packages
Ready to be installed by your favorite package manager ⬇️

https://pkg.pr.new/software-mansion/TypeGPU/typegpu@e3086a1f3608a5e01d6f6d83a12c12bcbc7f6d6b

benchmark
view benchmark

commit
view commit

[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 [0.96, 1.97, 4.31, 6.30, 7.29, 12.72, 23.51, 25.41]
  line [0.92, 1.89, 3.95, 6.21, 7.64, 12.67, 23.68, 26.64]
  line [0.93, 1.79, 4.02, 5.81, 6.68, 11.33, 23.29, 25.36]

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 [0.29, 0.49, 0.69, 0.87, 1.11, 1.21, 1.42, 1.58]
  line [0.29, 0.49, 0.65, 0.83, 1.09, 1.16, 1.36, 1.49]
  line [0.34, 0.50, 0.65, 0.75, 1.03, 1.10, 1.29, 1.45]

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 [0.81, 2.04, 3.82, 6.76, 12.39, 25.75, 54.71, 114.64]
  line [0.83, 2.01, 3.36, 6.67, 11.92, 27.96, 54.71, 106.89]
  line [0.86, 2.04, 4.37, 6.51, 11.77, 24.83, 52.69, 106.18]

Loading

[github-actions]

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

🟢 Decreased	➖ Unchanged	🔴 Increased (max 1.36%)	❔ Unknown
0	84	265	0

import * as ... in PR vs import * as ... in target (did bundle size increase?):

Test	tsdown
tgpu_lazy.ts	63.56 kB ($${\color{red}+1.4\%}$$)
tgpu_const.ts	63.76 kB ($${\color{red}+1.4\%}$$)
tgpu_workgroupVar.ts	64.34 kB ($${\color{red}+1.3\%}$$)
tgpu_privateVar.ts	64.34 kB ($${\color{red}+1.3\%}$$)
tgpu_accessor.ts	68.72 kB ($${\color{red}+1.3\%}$$)
tgpu_mutableAccessor.ts	68.72 kB ($${\color{red}+1.3\%}$$)
tgpu_bindGroupLayout.ts	70.95 kB ($${\color{red}+1.2\%}$$)
tgpu_resolveWithContext.ts	174.10 kB ($${\color{red}+0.9\%}$$)
tgpu_resolve.ts	174.17 kB ($${\color{red}+0.9\%}$$)
tgpu_initFromDevice.ts	283.96 kB ($${\color{red}+0.5\%}$$)
tgpu_init.ts	284.50 kB ($${\color{red}+0.5\%}$$)
STATIC_tgpu.ts	296.66 kB ($${\color{red}+0.5\%}$$)

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