feat: std.intdiv

[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/typegpu@f012224736fa3ac54991a3beab337edd02c207ea

benchmark
view benchmark

commit
view commit

[github-actions]

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

🟢 Decreased	➖ Unchanged	🔴 Increased (max 0.54%)	❔ Unknown
0	209	115	1

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

Click to reveal the results table (62 entries).

Test	tsdown
std_dot.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_length.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_distance.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_dot4U8Packed.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_fract.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_frexp.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_mix.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_modf.ts	69.67 kB ($${\color{red}+0.5\%}$$)
std_abs.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_acos.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_acosh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_asin.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_asinh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_atan.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_atan2.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_atanh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_ceil.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_clamp.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_cos.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_cosh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_countLeadingZeros.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_countOneBits.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_countTrailingZeros.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_cross.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_degrees.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_determinant.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_dot4I8Packed.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_exp.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_exp2.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_extractBits.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_faceForward.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_firstLeadingBit.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_firstTrailingBit.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_floor.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_fma.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_insertBits.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_inverseSqrt.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_ldexp.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_log.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_log2.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_max.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_min.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_normalize.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_pow.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_quantizeToF16.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_radians.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_reflect.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_refract.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_reverseBits.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_round.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_saturate.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_sign.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_sin.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_sinh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_smoothstep.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_sqrt.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_step.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_tan.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_tanh.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_transpose.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_trunc.ts	69.68 kB ($${\color{red}+0.5\%}$$)
std_intdiv.ts	69.68 kB

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

Test	tsdown
tgpu_init.ts	274.88 kB ($${\color{green}-3.4\%}$$)
tgpu_initFromDevice.ts	274.34 kB ($${\color{green}-3.5\%}$$)
tgpu_resolve.ts	175.17 kB ($${\color{green}-38.4\%}$$)
tgpu_resolveWithContext.ts	175.11 kB ($${\color{green}-38.4\%}$$)
tgpu_bindGroupLayout.ts	74.49 kB ($${\color{green}-73.8\%}$$)
tgpu_mutableAccessor.ts	69.21 kB ($${\color{green}-75.7\%}$$)
tgpu_accessor.ts	69.21 kB ($${\color{green}-75.7\%}$$)
tgpu_privateVar.ts	67.90 kB ($${\color{green}-76.1\%}$$)
tgpu_workgroupVar.ts	67.90 kB ($${\color{green}-76.1\%}$$)
tgpu_const.ts	67.32 kB ($${\color{green}-76.3\%}$$)
tgpu_lazy.ts	67.12 kB ($${\color{green}-76.4\%}$$)
tgpu_fragmentFn.ts	39.65 kB ($${\color{green}-86.1\%}$$)
tgpu_fn.ts	39.60 kB ($${\color{green}-86.1\%}$$)
tgpu_vertexFn.ts	39.47 kB ($${\color{green}-86.1\%}$$)
tgpu_computeFn.ts	39.17 kB ($${\color{green}-86.2\%}$$)
tgpu_vertexLayout.ts	28.30 kB ($${\color{green}-90.1\%}$$)
tgpu_comptime.ts	15.91 kB ($${\color{green}-94.4\%}$$)
tgpu_unroll.ts	1.75 kB ($${\color{green}-99.4\%}$$)
tgpu_slot.ts	1.70 kB ($${\color{green}-99.4\%}$$)

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 [1.06, 2.13, 4.60, 7.31, 8.50, 13.46, 27.05, 26.89]
  line [1.08, 2.10, 4.44, 7.65, 8.20, 13.66, 25.72, 25.10]
  line [1.00, 2.07, 4.94, 7.20, 8.62, 11.40, 26.11, 26.17]

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 [0.32, 0.60, 0.78, 0.87, 1.21, 1.25, 1.54, 1.68]
  line [0.30, 0.53, 0.69, 0.84, 1.16, 1.26, 1.43, 1.62]
  line [0.29, 0.56, 0.73, 0.86, 1.21, 1.24, 1.51, 1.68]

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 [0.96, 2.49, 4.71, 8.10, 13.81, 29.24, 62.74, 125.34]
  line [0.92, 2.35, 4.76, 8.70, 14.04, 28.51, 60.29, 118.85]
  line [0.87, 2.27, 4.06, 7.42, 12.93, 27.04, 58.62, 120.32]

Loading

[Copilot]

Pull request overview

Adds a new std.intdiv helper to provide integer-division semantics (with proper type unification) in TypeGPU’s standard library, along with a dedicated test suite to validate CPU behavior, TGSL codegen, coercions, and unsupported-argument errors.

Changes:

• Introduces std.intdiv as a dualImpl in packages/typegpu/src/std/numeric.ts.
• Re-exports intdiv from packages/typegpu/src/std/index.ts for public consumption.
• Adds packages/typegpu/tests/std/numeric/intdiv.test.ts to cover i32/u32, mixed integer types, float coercion, and vector-argument rejection.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 4 comments.

File	Description
packages/typegpu/src/std/numeric.ts	Adds std.intdiv implementation (CPU + WGSL codegen) and signature validation.
packages/typegpu/src/std/index.ts	Exposes intdiv via the public std entrypoint.
packages/typegpu/tests/std/numeric/intdiv.test.ts	Adds tests for correctness, codegen snapshots, type capture, and error reporting.

Suppressed comments (1)

packages/typegpu/tests/std/numeric/intdiv.test.ts:106

• Second inline error snapshot also hardcodes fn:div; update it to fn:intdiv to stay consistent with the function name shown by the resolver after renaming.

    - fn:div: Unsupported data types: abstractInt, vec3i. Supported types are: u32, i32.]

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[pullfrog]

ℹ️ Minor suggestions only - a couple of polish items, nothing blocking.

Reviewed changes

This PR adds std.intdiv, a dual-CPU/WGSL integer division operator restricted to scalar u32/i32 (vector/composite args reject with SignatureNotSupportedError), wired into the std barrel with a CPU reference implementation and a / codegen path.

• packages/typegpu/src/std/numeric.ts — adds cpuIntdiv and the intdiv dualImpl (signature unify to [u32, i32], codegenImpl emitting /, ignoreImplicitCastWarning, sideEffects: false).
• packages/typegpu/src/std/index.ts — exports intdiv from the std barrel.
• packages/typegpu/tests/std/numeric/intdiv.test.ts — 5 tests covering i32, u32, mixed u32/i32, float coercion, and vector rejection via inline snapshots.

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

[pullfrog]

✅ No new issues found.

Reviewed changes

This incremental review covers the single new commit (0d4c02d69) since the prior Pullfrog review (9baee402). It resolves both previously raised items and adds a small declaration-style touch-up:

• cpuIntdiv reuses divInteger — the CPU path now delegates to the shared divInteger helper in data/numberOps.ts, inheriting its rhs === 0 → lhs guard so CPU and WGSL division-by-zero semantics stay consistent. Also inverted to an early-throw on non-number args.
• name: 'div' → name: 'intdiv' — resolution/diagnostic errors for intdiv now report as fn:intdiv instead of colliding with std.div. Test snapshots updated accordingly.
• data/numberOps.ts — divInteger converted from arrow to function declaration and annotated with the WGSL div-by-zero comment.

All 5 intdiv tests pass. No blocking or noteworthy issues in the delta.

  ^{｜ View workflow run ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

[pullfrog]

ℹ️ No critical issues — one design observation worth a second look; everything tested and passing.

Reviewed changes

This incremental review covers the new commit f0122247 landed since the prior review (0d4c02d69).

• Removed ignoreImplicitCastWarning: true from the intdiv dualImpl in packages/typegpu/src/std/numeric.ts:1325 — mixed u32/i32 and float-coercion calls now surface [implicit-conversion] warnings.
• Added console-warning assertions to the mixed-int and float-coercion tests in packages/typegpu/tests/std/numeric/intdiv.test.ts, asserting the two implicit-conversion warnings (101u→i32, 5u→i32) and the float coercion warning.

All 5 intdiv tests pass. Prior feedback (reuse divInteger, name: 'intdiv') remains addressed.

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

[pullfrog]

Removing ignoreImplicitCastWarning: true means every mixed-u32/i32 and float-coercion call to std.intdiv now emits [implicit-conversion] warnings. Since this operator's signature deliberately unifies u32/i32 and coerces floats, warning on those expected inputs makes each documented usage noisy. Worth confirming the warning noise on intended inputs is desired, vs. keeping the flag (the coercion is inherent to intdiv, not an accidental implicit cast).

Suggested change

	export const intdiv = dualImpl({
	name: 'intdiv',
	signature: (lhs: BaseData, rhs: BaseData) => {
	const unified = unify([lhs, rhs], [u32, i32]);
	if (!unified) {
	throw new SignatureNotSupportedError([lhs, rhs], [u32, i32]);
	}
	return { argTypes: unified, returnType: unified[0] };
	},
	normalImpl: cpuIntdiv,
	codegenImpl: (ctx, [lhs, rhs]) => ctx.gen.emitBinaryOp(lhs, '/', rhs),
	codegenImpl: (ctx, [lhs, rhs]) => ctx.gen.emitBinaryOp(lhs, '/', rhs),
	ignoreImplicitCastWarning: true,
	sideEffects: false,