[reczkok]

(FYI this is not the feature discussed in #2816 proposal, this is the actual WebGPU feature)
d.buffer(n) resolves to the WGSL buffer<n> type, and std.bufferView, std.bufferArrayView and std.bufferLength read and write its bytes as any host-shareable type at runtime offsets

[github-actions]

pkg.pr.new

packages
Ready to be installed by your favorite package manager ⬇️

https://pkg.pr.new/software-mansion/TypeGPU/tgpu-gen@318d02a90cefb3f00f113aef9a157e4f1a9b235c

https://pkg.pr.new/software-mansion/TypeGPU/typegpu@318d02a90cefb3f00f113aef9a157e4f1a9b235c

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/cli@318d02a90cefb3f00f113aef9a157e4f1a9b235c

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/gl@318d02a90cefb3f00f113aef9a157e4f1a9b235c

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 [0.77, 1.55, 3.52, 5.21, 5.96, 9.35, 20.03, 22.35]
  line [0.72, 1.53, 3.35, 5.06, 5.85, 9.26, 18.22, 21.47]
  line [0.73, 1.48, 3.26, 5.30, 6.18, 10.70, 20.90, 21.50]

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 [0.27, 0.43, 0.57, 0.68, 0.91, 0.97, 1.18, 1.25]
  line [0.28, 0.45, 0.58, 0.68, 0.93, 1.00, 1.17, 1.20]
  line [0.27, 0.40, 0.55, 0.65, 0.83, 0.88, 1.10, 1.19]

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 [0.73, 1.80, 3.03, 5.19, 9.82, 21.31, 46.07, 91.60]
  line [0.74, 1.78, 2.84, 5.48, 11.21, 20.88, 43.45, 89.11]
  line [0.70, 1.70, 2.70, 5.69, 10.58, 20.83, 45.66, 91.49]

Loading

[github-actions]

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

🟢 Decreased	➖ Unchanged	🔴 Increased (max 13.02%)	❔ Unknown
0	36	319	4

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

Click to reveal the results table (300 entries).

Test	tsdown
STATIC_std.ts	114.82 kB ($${\color{red}+13.0\%}$$)
d_struct.ts	4.06 kB ($${\color{red}+4.6\%}$$)
STATIC_allImports.ts	320.23 kB ($${\color{red}+1.5\%}$$)
STATIC_d.ts	72.35 kB ($${\color{red}+1.3\%}$$)
d_arrayOf.ts	33.95 kB ($${\color{red}+1.2\%}$$)
d_builtin.ts	36.48 kB ($${\color{red}+1.2\%}$$)
d_deepEqual.ts	23.01 kB ($${\color{red}+1.1\%}$$)
d_schemasEqualNominally.ts	23.01 kB ($${\color{red}+1.1\%}$$)
d_schemasEqualStructurally.ts	23.01 kB ($${\color{red}+1.1\%}$$)
d_bool.ts	20.25 kB ($${\color{red}+1.1\%}$$)
d_f16.ts	20.25 kB ($${\color{red}+1.1\%}$$)
d_f32.ts	20.25 kB ($${\color{red}+1.1\%}$$)
d_i32.ts	20.25 kB ($${\color{red}+1.1\%}$$)
d_u32.ts	20.25 kB ($${\color{red}+1.1\%}$$)
d_ref.ts	20.26 kB ($${\color{red}+1.1\%}$$)
d_u16.ts	20.27 kB ($${\color{red}+1.1\%}$$)
d_abstractInt.ts	20.33 kB ($${\color{red}+1.1\%}$$)
d_abstractFloat.ts	20.33 kB ($${\color{red}+1.1\%}$$)
tgpu_resolveWithContext.ts	162.09 kB ($${\color{red}+1.1\%}$$)
tgpu_resolve.ts	162.16 kB ($${\color{red}+1.1\%}$$)
d_textureDepth2d.ts	20.69 kB ($${\color{red}+1.0\%}$$)
d_textureDepthCube.ts	20.70 kB ($${\color{red}+1.0\%}$$)
d_texture1d.ts	20.71 kB ($${\color{red}+1.0\%}$$)
d_texture2d.ts	20.71 kB ($${\color{red}+1.0\%}$$)
d_texture3d.ts	20.71 kB ($${\color{red}+1.0\%}$$)
d_textureCube.ts	20.71 kB ($${\color{red}+1.0\%}$$)
d_textureDepth2dArray.ts	20.71 kB ($${\color{red}+1.0\%}$$)
d_textureDepthCubeArray.ts	20.72 kB ($${\color{red}+1.0\%}$$)
d_textureDepthMultisampled2d.ts	20.72 kB ($${\color{red}+1.0\%}$$)
d_texture2dArray.ts	20.72 kB ($${\color{red}+1.0\%}$$)
d_textureCubeArray.ts	20.73 kB ($${\color{red}+1.0\%}$$)
d_textureMultisampled2d.ts	20.73 kB ($${\color{red}+1.0\%}$$)
std_isKnownAtComptime.ts	20.87 kB ($${\color{red}+1.0\%}$$)
std_subgroupElect.ts	21.44 kB ($${\color{red}+1.0\%}$$)
std_subgroupAll.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupAny.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_quadSwapX.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_quadSwapY.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupOr.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupAdd.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupAnd.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupMax.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupMin.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupMul.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_subgroupXor.ts	21.45 kB ($${\color{red}+1.0\%}$$)
std_quadSwapDiagonal.ts	21.46 kB ($${\color{red}+1.0\%}$$)
std_subgroupExclusiveAdd.ts	21.47 kB ($${\color{red}+1.0\%}$$)
std_subgroupExclusiveMul.ts	21.47 kB ($${\color{red}+1.0\%}$$)
std_subgroupInclusiveAdd.ts	21.47 kB ($${\color{red}+1.0\%}$$)
std_subgroupInclusiveMul.ts	21.47 kB ($${\color{red}+1.0\%}$$)
std_subgroupBroadcastFirst.ts	21.47 kB ($${\color{red}+1.0\%}$$)
std_storageBarrier.ts	21.49 kB ($${\color{red}+1.0\%}$$)
std_textureBarrier.ts	21.49 kB ($${\color{red}+1.0\%}$$)
std_workgroupBarrier.ts	21.50 kB ($${\color{red}+1.0\%}$$)
std_discard.ts	21.50 kB ($${\color{red}+1.0\%}$$)
std_textureNumLayers.ts	21.52 kB ($${\color{red}+1.0\%}$$)
std_textureNumLevels.ts	21.52 kB ($${\color{red}+1.0\%}$$)
std_textureNumSamples.ts	21.53 kB ($${\color{red}+1.0\%}$$)
std_typeOf.ts	22.25 kB ($${\color{red}+1.0\%}$$)
std_textureSampleCompare.ts	21.55 kB ($${\color{red}+1.0\%}$$)
std_textureSampleCompareLevel.ts	21.57 kB ($${\color{red}+1.0\%}$$)
std_textureStore.ts	21.57 kB ($${\color{red}+1.0\%}$$)
std_workgroupUniformLoad.ts	21.59 kB ($${\color{red}+1.0\%}$$)
tgpu_comptime.ts	21.60 kB ($${\color{red}+1.0\%}$$)
std_atomicLoad.ts	21.63 kB ($${\color{red}+1.0\%}$$)
std_atomicOr.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicAdd.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicAnd.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicMax.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicMin.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicSub.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicXor.ts	21.67 kB ($${\color{red}+1.0\%}$$)
std_atomicExchange.ts	21.68 kB ($${\color{red}+1.0\%}$$)
std_atomicStore.ts	21.71 kB ($${\color{red}+1.0\%}$$)
std_quadBroadcast.ts	21.72 kB ($${\color{red}+1.0\%}$$)
std_arrayLength.ts	21.72 kB ($${\color{red}+1.0\%}$$)
std_subgroupShuffleUp.ts	21.72 kB ($${\color{red}+1.0\%}$$)
std_subgroupShuffle.ts	21.72 kB ($${\color{red}+1.0\%}$$)
std_subgroupShuffleXor.ts	21.72 kB ($${\color{red}+1.0\%}$$)
std_subgroupShuffleDown.ts	21.73 kB ($${\color{red}+1.0\%}$$)
std_subgroupBroadcast.ts	21.73 kB ($${\color{red}+1.0\%}$$)
d_sizeOf.ts	32.97 kB ($${\color{red}+1.0\%}$$)
std_copy.ts	21.75 kB ($${\color{red}+1.0\%}$$)
d_isContiguous.ts	32.98 kB ($${\color{red}+1.0\%}$$)
d_getLongestContiguousPrefix.ts	32.99 kB ($${\color{red}+1.0\%}$$)
d_isBuiltin.ts	33.07 kB ($${\color{red}+1.0\%}$$)
std_extensionEnabled.ts	21.82 kB ($${\color{red}+1.0\%}$$)
std_range.ts	22.09 kB ($${\color{red}+1.0\%}$$)
std_isBeingTranspiled.ts	22.13 kB ($${\color{red}+1.0\%}$$)
std_getShaderStage.ts	22.19 kB ($${\color{red}+1.0\%}$$)
std_getTargetShaderLanguage.ts	22.20 kB ($${\color{red}+1.0\%}$$)
d_disarrayOf.ts	22.31 kB ($${\color{red}+1.0\%}$$)
d_size.ts	34.74 kB ($${\color{red}+1.0\%}$$)
d_align.ts	34.74 kB ($${\color{red}+1.0\%}$$)
d_location.ts	34.74 kB ($${\color{red}+1.0\%}$$)
d_interpolate.ts	34.74 kB ($${\color{red}+1.0\%}$$)
d_invariant.ts	35.11 kB ($${\color{red}+0.9\%}$$)
tgpu_vertexLayout.ts	35.00 kB ($${\color{red}+0.9\%}$$)
common_writeSoA.ts	35.59 kB ($${\color{red}+0.9\%}$$)
d_memoryLayoutOf.ts	36.56 kB ($${\color{red}+0.9\%}$$)
std_atomicCompareExchangeWeak.ts	24.27 kB ($${\color{red}+0.9\%}$$)
common_fullScreenTriangle.ts	49.15 kB ($${\color{red}+0.9\%}$$)
tgpu_initFromDevice.ts	257.29 kB ($${\color{red}+0.8\%}$$)
tgpu_init.ts	257.83 kB ($${\color{red}+0.8\%}$$)
tgpu_fn.ts	47.40 kB ($${\color{red}+0.8\%}$$)
STATIC_tgpu.ts	269.06 kB ($${\color{red}+0.8\%}$$)
std_subgroupBallot.ts	27.92 kB ($${\color{red}+0.8\%}$$)
std_textureSampleBaseClampToEdge.ts	28.00 kB ($${\color{red}+0.8\%}$$)
std_textureSampleGrad.ts	28.01 kB ($${\color{red}+0.8\%}$$)
std_textureSampleBias.ts	28.07 kB ($${\color{red}+0.8\%}$$)
d_uint8.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float16.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float16x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float16x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float32.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float32x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float32x3.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_float32x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint16.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint16x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint16x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint32.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint32x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint32x3.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint32x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint8.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint8x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_sint8x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm16.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm16x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm16x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm8.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm8x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_snorm8x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint16.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint16x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint16x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint32.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint32x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint32x3.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint32x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint8x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_uint8x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm10_10_10_2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm16.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm16x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm16x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm8.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm8x2.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm8x4.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_unorm8x4_bgra.ts	28.21 kB ($${\color{red}+0.8\%}$$)
std_textureSample.ts	28.09 kB ($${\color{red}+0.8\%}$$)
std_textureSampleLevel.ts	28.10 kB ($${\color{red}+0.8\%}$$)
std_textureGatherCompare.ts	28.13 kB ($${\color{red}+0.8\%}$$)
std_textureDimensions.ts	28.16 kB ($${\color{red}+0.8\%}$$)
d_formatToWGSLType.ts	28.21 kB ($${\color{red}+0.8\%}$$)
d_packedFormats.ts	28.23 kB ($${\color{red}+0.8\%}$$)
d_isPackedData.ts	28.28 kB ($${\color{red}+0.8\%}$$)
std_textureGather.ts	28.33 kB ($${\color{red}+0.8\%}$$)
STATIC_common.ts	56.70 kB ($${\color{red}+0.7\%}$$)
d_alignmentOf.ts	30.08 kB ($${\color{red}+0.7\%}$$)
tgpu_computeFn.ts	46.50 kB ($${\color{red}+0.7\%}$$)
std_textureLoad.ts	29.95 kB ($${\color{red}+0.7\%}$$)
tgpu_vertexFn.ts	47.05 kB ($${\color{red}+0.7\%}$$)
tgpu_fragmentFn.ts	47.24 kB ($${\color{red}+0.7\%}$$)
std_bitcast.ts	30.73 kB ($${\color{red}+0.7\%}$$)
tgpu_accessor.ts	68.39 kB ($${\color{red}+0.7\%}$$)
tgpu_mutableAccessor.ts	68.40 kB ($${\color{red}+0.7\%}$$)
std_pack2x16float.ts	31.74 kB ($${\color{red}+0.7\%}$$)
std_unpack2x16unorm.ts	31.74 kB ($${\color{red}+0.7\%}$$)
std_unpack2x16float.ts	31.75 kB ($${\color{red}+0.7\%}$$)
std_pack4xI8.ts	31.76 kB ($${\color{red}+0.7\%}$$)
std_unpack4xI8.ts	31.76 kB ($${\color{red}+0.7\%}$$)
std_pack4xU8.ts	31.76 kB ($${\color{red}+0.7\%}$$)
std_unpack4xU8.ts	31.76 kB ($${\color{red}+0.7\%}$$)
std_unpack2x16snorm.ts	31.76 kB ($${\color{red}+0.7\%}$$)
std_unpack4x8unorm.ts	31.78 kB ($${\color{red}+0.7\%}$$)
std_unpack4x8snorm.ts	31.83 kB ($${\color{red}+0.7\%}$$)
std_pack2x16snorm.ts	31.92 kB ($${\color{red}+0.7\%}$$)
std_pack2x16unorm.ts	31.92 kB ($${\color{red}+0.7\%}$$)
std_pack4xU8Clamp.ts	31.93 kB ($${\color{red}+0.7\%}$$)
std_pack4xI8Clamp.ts	31.94 kB ($${\color{red}+0.7\%}$$)
std_pack4x8snorm.ts	32.00 kB ($${\color{red}+0.7\%}$$)
std_pack4x8unorm.ts	32.00 kB ($${\color{red}+0.7\%}$$)
std_identity2.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_identity3.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_identity4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_rotationX4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_rotationY4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_rotationZ4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_scaling4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
std_translation4.ts	34.42 kB ($${\color{red}+0.6\%}$$)
d_matToArray.ts	34.55 kB ($${\color{red}+0.6\%}$$)
std_dot4I8Packed.ts	35.80 kB ($${\color{red}+0.6\%}$$)
std_dot4U8Packed.ts	35.80 kB ($${\color{red}+0.6\%}$$)
std_firstLeadingBit.ts	35.81 kB ($${\color{red}+0.6\%}$$)
std_determinant.ts	35.89 kB ($${\color{red}+0.6\%}$$)
std_dpdxFine.ts	35.89 kB ($${\color{red}+0.6\%}$$)
std_dpdyFine.ts	35.89 kB ($${\color{red}+0.6\%}$$)
std_dpdxCoarse.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_dpdyCoarse.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_fwidthFine.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_dpdx.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_dpdy.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_fwidthCoarse.ts	35.90 kB ($${\color{red}+0.6\%}$$)
std_fwidth.ts	35.91 kB ($${\color{red}+0.6\%}$$)
std_refract.ts	35.91 kB ($${\color{red}+0.6\%}$$)
std_trunc.ts	35.96 kB ($${\color{red}+0.6\%}$$)
std_reverseBits.ts	35.97 kB ($${\color{red}+0.6\%}$$)
std_quantizeToF16.ts	35.97 kB ($${\color{red}+0.6\%}$$)
std_countOneBits.ts	35.97 kB ($${\color{red}+0.6\%}$$)
std_extractBits.ts	35.98 kB ($${\color{red}+0.6\%}$$)
std_firstTrailingBit.ts	35.98 kB ($${\color{red}+0.6\%}$$)
std_countLeadingZeros.ts	35.99 kB ($${\color{red}+0.6\%}$$)
std_countTrailingZeros.ts	35.99 kB ($${\color{red}+0.6\%}$$)
std_insertBits.ts	35.99 kB ($${\color{red}+0.6\%}$$)
std_faceForward.ts	35.99 kB ($${\color{red}+0.6\%}$$)
std_intdiv.ts	36.00 kB ($${\color{red}+0.6\%}$$)
std_tan.ts	36.04 kB ($${\color{red}+0.6\%}$$)
std_degrees.ts	36.05 kB ($${\color{red}+0.6\%}$$)
std_radians.ts	36.05 kB ($${\color{red}+0.6\%}$$)
std_inverseSqrt.ts	36.08 kB ($${\color{red}+0.6\%}$$)
std_round.ts	36.09 kB ($${\color{red}+0.6\%}$$)
std_step.ts	36.22 kB ($${\color{red}+0.6\%}$$)
std_ldexp.ts	36.22 kB ($${\color{red}+0.6\%}$$)
std_fma.ts	36.22 kB ($${\color{red}+0.6\%}$$)
std_transpose.ts	36.29 kB ($${\color{red}+0.6\%}$$)
std_not.ts	36.45 kB ($${\color{red}+0.6\%}$$)
std_abs.ts	36.46 kB ($${\color{red}+0.6\%}$$)
std_gt.ts	36.83 kB ($${\color{red}+0.6\%}$$)
std_neg.ts	36.51 kB ($${\color{red}+0.6\%}$$)
std_or.ts	36.52 kB ($${\color{red}+0.6\%}$$)
std_and.ts	36.60 kB ($${\color{red}+0.6\%}$$)
std_lt.ts	36.60 kB ($${\color{red}+0.6\%}$$)
std_eq.ts	36.61 kB ($${\color{red}+0.6\%}$$)
std_cos.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_exp.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_log.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_sin.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_exp2.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_sign.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_acos.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_asin.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_atan.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_ceil.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_cosh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_log2.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_sinh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_sqrt.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_tanh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_acosh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_asinh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_atanh.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_floor.ts	36.63 kB ($${\color{red}+0.6\%}$$)
std_fract.ts	36.64 kB ($${\color{red}+0.6\%}$$)
std_ge.ts	36.67 kB ($${\color{red}+0.6\%}$$)
std_ne.ts	36.68 kB ($${\color{red}+0.6\%}$$)
std_le.ts	36.75 kB ($${\color{red}+0.6\%}$$)
std_saturate.ts	36.78 kB ($${\color{red}+0.6\%}$$)
std_pow.ts	36.81 kB ($${\color{red}+0.6\%}$$)
std_atan2.ts	36.82 kB ($${\color{red}+0.6\%}$$)
std_mix.ts	37.18 kB ($${\color{red}+0.6\%}$$)
std_clamp.ts	36.91 kB ($${\color{red}+0.6\%}$$)
std_max.ts	36.93 kB ($${\color{red}+0.6\%}$$)
std_min.ts	36.93 kB ($${\color{red}+0.6\%}$$)
std_smoothstep.ts	37.03 kB ($${\color{red}+0.6\%}$$)
std_select.ts	37.43 kB ($${\color{red}+0.6\%}$$)
std_mod.ts	37.46 kB ($${\color{red}+0.6\%}$$)
std_div.ts	37.50 kB ($${\color{red}+0.6\%}$$)
std_add.ts	37.58 kB ($${\color{red}+0.6\%}$$)
std_frexp.ts	38.54 kB ($${\color{red}+0.6\%}$$)
std_modf.ts	38.67 kB ($${\color{red}+0.6\%}$$)
std_all.ts	39.78 kB ($${\color{red}+0.5\%}$$)
std_bitShiftLeft.ts	39.76 kB ($${\color{red}+0.5\%}$$)
std_bitShiftRight.ts	39.77 kB ($${\color{red}+0.5\%}$$)
std_dot.ts	40.06 kB ($${\color{red}+0.5\%}$$)
std_length.ts	40.16 kB ($${\color{red}+0.5\%}$$)
std_any.ts	40.60 kB ($${\color{red}+0.5\%}$$)
std_cross.ts	40.27 kB ($${\color{red}+0.5\%}$$)
std_allEq.ts	40.81 kB ($${\color{red}+0.5\%}$$)
tgpu_lazy.ts	62.91 kB ($${\color{red}+0.5\%}$$)
tgpu_const.ts	63.11 kB ($${\color{red}+0.5\%}$$)
std_isCloseTo.ts	41.09 kB ($${\color{red}+0.5\%}$$)
tgpu_workgroupVar.ts	63.69 kB ($${\color{red}+0.5\%}$$)
tgpu_privateVar.ts	63.69 kB ($${\color{red}+0.5\%}$$)
std_normalize.ts	41.18 kB ($${\color{red}+0.5\%}$$)
std_bitcastF32toU32.ts	41.97 kB ($${\color{red}+0.5\%}$$)
std_bitcastU32toF32.ts	41.97 kB ($${\color{red}+0.5\%}$$)
std_bitcastU32toI32.ts	41.97 kB ($${\color{red}+0.5\%}$$)
std_mul.ts	42.20 kB ($${\color{red}+0.5\%}$$)
std_rotateZ4.ts	42.46 kB ($${\color{red}+0.5\%}$$)
std_translate4.ts	42.46 kB ($${\color{red}+0.5\%}$$)
std_scale4.ts	42.46 kB ($${\color{red}+0.5\%}$$)
std_rotateX4.ts	42.46 kB ($${\color{red}+0.5\%}$$)
std_rotateY4.ts	42.46 kB ($${\color{red}+0.5\%}$$)
std_sub.ts	42.58 kB ($${\color{red}+0.5\%}$$)
d_buffer.ts	1.03 kB
std_bufferArrayView.ts	41.26 kB
std_bufferLength.ts	42.40 kB
std_bufferView.ts	41.26 kB

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

[reczkok]

@pullfrog do a preliminary review

[pullfrog]

ℹ️ No critical issues — one minor validation gap worth a look.

Reviewed changes

• d.bytes(n) schema (data/bytes.ts) — new WgslBuffer schema resolving to WGSL buffer<n> (or runtime-sized buffer for 0), with size/alignment/equality/layout/serialization wiring and struct/array/address-space guards.
• std.bufferView / std.bufferArrayView / std.bufferLength (std/bufferView.ts) — codegen-only builtins that reinterpret buffer bytes as host-shareable types, with alignment, fit, atomic, host-shareable and scope validation, plus std exports.
• buffer_view language-feature plumbing (resolutionCtx.ts, wgslExtensions.ts, resolveData.ts, types.ts) — requireLanguageFeature and a requires buffer_view; directive emitted ahead of the module.
• For-of over pointer iterables (tgsl/forOfUtils.ts) — getRangeSnippets now unwraps pointer data types (enables iterating a bufferView-returned array).
• Docs + example + tests — new buffer-views.mdx, the shape-stream example, and bytes/bufferView/serialization/generator tests.

I read the full diff, verified the relevant source paths (dualImpl, pinEref, ptr, array, resolveData, forOfUtils, dataIO, compiledIO, schemaMemoryLayout, schemasEqual), ran the new tests (bytes, std/bufferView, serializeDataSchema), the tgsl suite, and tsc --noEmit — all pass. I also checked the claims against the WGSL spec: the requires buffer_view; directive/feature name is correct, directive ordering relative to enable is unconstrained, and the alignment/MinTypeSize/invalid-pointer semantics line up with assertViewFits/minSizeOf.

The buffer<N> size rule is the one gap I found; detail inline.

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}

[Copilot]

🟡 Changes recommended

Pointer compatibility and decorated-schema validation gaps allow invalid WGSL to be generated.

4 open issues

• Location-decorated buffers bypass array element validation · New
• Decorated buffer properties bypass storage-schema validation · New
• Allowing incompatible buffer address spaces and access modes · New
• Decorated buffer schemas bypass private variable validation · New

What changed in this PR

Adds WGSL buffer views to TypeGPU, allowing shaders to interpret raw buffer bytes as typed data at runtime offsets.

Changes:

• Introduces d.buffer, buffer-view built-ins, and language-feature checks.
• Integrates raw buffers with bindings, memory layout, serialization, and CPU reads and writes.
• Adds tests, API documentation, and a packed-shapes example.

File	Description
packages/​typegpu/​tests/​wgslBuffer.test.ts	Tests buffer schemas, restrictions, and raw-byte operations.
packages/​typegpu/​tests/​tgsl/​wgslGenerator.test.ts	Tests iteration through array pointers.
packages/​typegpu/​tests/​std/​bufferView.test.ts	Tests view generation, validation, and mutability.
packages/​typegpu/​tests/​serializeDataSchema.test.ts	Adds buffer-schema serialization coverage.
packages/​typegpu/​tests/​bufferBindings.test.ts	Updates snapshots for reserved buffer names.
packages/​typegpu/​src/​wgslExtensions.ts	Defines the buffer-view language feature.
packages/​typegpu/​src/​types.ts	Adds language-feature requirements to resolution.
packages/​typegpu/​src/​tgsl/​wgslGenerator.ts	Validates buffer variable address spaces.
packages/​typegpu/​src/​tgsl/​forOfUtils.ts	Supports pointer-backed iteration ranges.
packages/​typegpu/​src/​tgsl/​conversion.ts	Adds size-based buffer-pointer compatibility.
packages/​typegpu/​src/​tgpuBindGroupLayout.ts	Allows typed buffers in raw-buffer bindings.
packages/​typegpu/​src/​std/​index.ts	Exports buffer-view built-ins.
packages/​typegpu/​src/​std/​bufferView.ts	Implements views, length queries, and validation.
packages/​typegpu/​src/​serial/​schema.ts	Serializes and restores buffer schemas.
packages/​typegpu/​src/​resolutionCtx.ts	Checks features and emits requires directives.
packages/​typegpu/​src/​nameUtils.ts	Reserves buffer-view built-in names.
packages/​typegpu/​src/​data/​wgslTypes.ts	Adds buffer types and GPU pointer representations.
packages/​typegpu/​src/​data/​struct.ts	Rejects buffer-typed struct members.
packages/​typegpu/​src/​data/​schemasEqual.ts	Compares buffer schema sizes.
packages/​typegpu/​src/​data/​schemaMemoryLayout.ts	Defines raw-buffer memory layouts.
packages/​typegpu/​src/​data/​ptr.ts	Allows buffer schemas in supported pointer types.
packages/​typegpu/​src/​data/​partialIO.ts	Supports raw-byte patches.
packages/​typegpu/​src/​data/​index.ts	Exports buffer schemas.
packages/​typegpu/​src/​data/​dataIO.ts	Reads and writes raw buffer bytes.
packages/​typegpu/​src/​data/​buffer.ts	Defines sized and runtime-sized buffer schemas.
packages/​typegpu/​src/​data/​array.ts	Rejects buffer-typed array elements.
packages/​typegpu/​src/​data/​alignmentOf.ts	Defines buffer alignment.
packages/​typegpu/​src/​core/​variable/​tgpuVariable.ts	Requires sizes for workgroup buffers.
packages/​typegpu/​src/​core/​resolve/​resolveData.ts	Resolves buffer schemas to WGSL.
packages/​typegpu-testing-utility/​src/​extendedIt.ts	Enables buffer views in mocked environments.
apps/​typegpu-docs/​tests/​individual-example-tests/​dispatch.test.ts	Updates reserved-name snapshots.
apps/​typegpu-docs/​src/​examples/​simple/​packed-shapes/​meta.json	Describes the packed-shapes example.
apps/​typegpu-docs/​src/​examples/​simple/​packed-shapes/​index.ts	Demonstrates runtime views over packed shapes.
apps/​typegpu-docs/​src/​examples/​simple/​packed-shapes/​index.html	Adds an unsupported-feature notice.
apps/​typegpu-docs/​src/​content/​docs/​apis/​buffer-views.mdx	Documents buffer-view APIs and usage.
apps/​typegpu-docs/​astro.config.mjs	Adds buffer views to documentation navigation.

_{Review effort: Balanced}

[pullfrog]

ℹ️ No critical issues — two rough edges worth a look.

Reviewed changes — the complete diff of this PR (9 commits, 37 files), which adds WGSL buffer<N>/buffer schemas (d.buffer), the std.bufferView / std.bufferArrayView / std.bufferLength builtins, requires buffer_view; emission, host-side buffer I/O, docs, and a packed-shapes example.

• d.buffer schema — new WgslBuffer type in data/buffer.ts (byteSize, 0 = runtime-sized) with size-multiple-of-4 validation and $validStorageSchema/$validUniformSchema tokens; threaded through resolveData, sizeOf/alignmentOf, schemasEqual, serial/schema, partialIO, and dataIO, and rejected as a struct property or array element.
• std.bufferView family — std/bufferView.ts emits the WGSL builtins, checks schema host-shareability and comptime alignment/fit, and preserves read-only/mutable view semantics.
• requires directives — resolutionCtx.ts/wgslExtensions.ts add requireLanguageFeature and requiredLanguageFeatures, emitted next to enable directives; resolution throws when root.enabledWgslLanguageFeatures lacks buffer_view.
• Type-level plumbing — ptr.ts accepts d.buffer in pointer constructors, tgsl/conversion.ts adds size-subtyping (fitsBufferParameter), tgpuBindGroupLayout.ts widens BoundSchema for d.buffer entries, and Ptr gains $gpuRepr.
• for...of over pointers — tgsl/forOfUtils.ts unwraps pointer iterables so buffer-view results iterate.
• Tests / docs / example — new tests/std/bufferView.test.ts and tests/wgslBuffer.test.ts, a buffer-views docs page, and the packed-shapes example.

I verified the emitted WGSL against W3C WGSL §6.3/§17.14/§4.1.2: the bufferView<T>(&buf, offset), bufferArrayView<T>(&buf, offset, size), and bufferLength(&buf) call shapes, the requires buffer_view; syntax/placement, and the buffer/buffer<N> types all conform. Host-side read/write paths are byte-correct (no endianness swap, exact region, partial-write semantics match the tests).

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}

[pullfrog]

✅ No new issues found.

Reviewed changes — the delta since the prior Pullfrog review (bab7afe → 318d02a). The branch was rebased, so the only net content change is the animate on gpu commit; every other file was already present at bab7afe.

• Reworked the packed-shapes example to animate on the GPU — shape motion moved into a Motion struct with a time uniform and a toLocal helper, replacing the previous per-frame CPU buffer rewrites; shape colors are now u32 palette indices resolved against a tgpu.const palette in the shader, and the CPU side only writes once per scene.
• Clarified bufferArrayView length in the docs — buffer-views.mdx now states the length is byteSize divided by the array stride (the element size rounded up to its alignment), matching WGSL StrideOf.

The example's manual offsets stay consistent with the CPU Scene struct layout: d.sizeOf(Circle)/d.sizeOf(Box) equal the corresponding array strides, and bufferArrayView sizes are exact stride multiples.

( ◦°^°◦) 🏴☠️

Ahoy, matey! I inspected the fresh commit — the shapes now dance on the GPU, and the view-length line in the docs be truthful. No leaks below the waterline, so I be givin' me approval.

  ^{｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}