GPU for the Web Working Group - Activity in 2025
No matches.
TR publications
-
5 documents published.
-
WebGPU Shading Language published 4 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published as a Candidate Recommendation Draft. History of WebGPU
-
GitHub
gpuweb/gpuweb
-
23 issues opened.
-
Compat: Should disallow textureBindingViewDimension = 'cube-array' on createTexture gpuweb/gpuweb#5496, by @greggman.
-
What can the WebGPU spec do to avoid inviting users into "spec lawyering" in the future? gpuweb/gpuweb#5495, by @juj.
-
Minor feature request: pushDebugGroup/insertMarker with color gpuweb/gpuweb#5493, by @magcius.
-
Functions should be able to accept pointers from any address space gpuweb/gpuweb#5492, by @Usnul.
-
[bindless] Shorter names for GPUResourceTable methods gpuweb/gpuweb#5491, by @Kangz.
-
`Queue.submit()` only checks if resources in bind groups have been destroyed gpuweb/gpuweb#5489, by @teoxoy.
-
Typos in maxStorageTexturesIn*Stage base values gpuweb/gpuweb#5488, by @kainino0x.
-
Restrict transient textures to 2D dimension gpuweb/gpuweb#5487, by @beaufortfrancois.
-
[buffer-view] Allowed size expressions for buffers gpuweb/gpuweb#5486, by @alan-baker.
-
Multisample Resolve issues gpuweb/gpuweb#5485, by @greggman.
-
Tracking issue: feature detection for new API changes gpuweb/gpuweb#5484, by @kainino0x.
-
f16 Convenience Polyfills for common gfx functions gpuweb/gpuweb#5482, by @petermcneeleychromium.
-
Detection of non optional new features gpuweb/gpuweb#5481, by @greggman.
-
CTS tests for proposed but not approved items gpuweb/gpuweb#5480, by @greggman.
-
Prevent canvas textures to be transient gpuweb/gpuweb#5479, by @beaufortfrancois.
-
allow user-defined functions to return a handle value (texture, sampler) gpuweb/gpuweb#5474, by @dneto0.
-
[immediates] Consider removing GPUPipelineLayoutDescriptor.immediateSize gpuweb/gpuweb#5473, by @Kangz.
-
Array of numeric scalars missed in list of allowed types for entry point formal parameters gpuweb/gpuweb#5472, by @zoddicus.
-
[bindless] Define the default types gpuweb/gpuweb#5471, by @dj2.
-
[bindless] Define the compatibiilty rules for WGSL types gpuweb/gpuweb#5470, by @dj2.
-
WGSL Proposal: StructView – Partial struct load/store for more efficient shader data access gpuweb/gpuweb#5468, by @tlecoz.
-
[bindless] Choose the order of slots returned by insertBinding gpuweb/gpuweb#5466, by @Kangz.
-
[bindless] What to do for oversized GPUResourceTable reflection. gpuweb/gpuweb#5465, by @Kangz.
-
-
8 issues closed.
-
[subgroup-id] uniformity of subgroup-id gpuweb/gpuweb#5368.
-
Restrict transient textures to 2D dimension gpuweb/gpuweb#5487.
-
Prevent canvas textures to be transient gpuweb/gpuweb#5479.
-
[immediates] Consider removing GPUPipelineLayoutDescriptor.immediateSize gpuweb/gpuweb#5473.
-
Out of bounds texture access with textureLoad with swizzle gpuweb/gpuweb#5437.
-
-
30 issues received 91 new comments.
-
Restrict transient textures to 2D dimension gpuweb/gpuweb#5487 (14 comments by @beaufortfrancois, @kainino0x, @mwyrzykowski)
-
CTS tests for proposed but not approved items gpuweb/gpuweb#5480 (12 comments by @andyleiserson, @beaufortfrancois, @ErichDonGubler, @greggman, @jimblandy, @kainino0x, @Kangz)
-
Prevent canvas textures to be transient gpuweb/gpuweb#5479 (8 comments by @beaufortfrancois, @greggman, @kainino0x, @mwyrzykowski)
-
How should implementations cope with Metal's 500k resources per-process soft-limit? gpuweb/gpuweb#5441 (5 comments by @mwyrzykowski, @teoxoy)
-
WGSL Proposal: StructView – Partial struct load/store for more efficient shader data access gpuweb/gpuweb#5468 (5 comments by @magcius, @petermcneeleychromium, @tlecoz)
-
Clarification of Sample Mask with respect to Sample Shading gpuweb/gpuweb#5457 (4 comments by @jimblandy, @mwyrzykowski, @petermcneeleychromium)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (4 comments by @kainino0x, @morti86)
-
Unified Shader Modules - link multiple GPUShaderModules into a pipeline shader stage gpuweb/gpuweb#5456 (4 comments by @bytesandwich, @dneto0, @mwyrzykowski)
-
f16 Convenience Polyfills for common gfx functions gpuweb/gpuweb#5482 (4 comments by @dneto0, @jimblandy)
-
[bindless] Shorter names for GPUResourceTable methods gpuweb/gpuweb#5491 (3 comments by @cwfitzgerald, @dj2, @JMS55)
-
`Queue.submit()` only checks if resources in bind groups have been destroyed gpuweb/gpuweb#5489 (3 comments by @kainino0x, @Kangz, @toji)
-
What can the WebGPU spec do to avoid inviting users into "spec lawyering" in the future? gpuweb/gpuweb#5495 (2 comments by @greggman, @magcius)
-
[immediates] Consider removing GPUPipelineLayoutDescriptor.immediateSize gpuweb/gpuweb#5473 (2 comments by @Kangz, @teoxoy)
-
[bindless] What to do for oversized GPUResourceTable reflection. gpuweb/gpuweb#5465 (2 comments by @kainino0x, @mwyrzykowski)
-
Detection of non optional new features gpuweb/gpuweb#5481 (2 comments by @beaufortfrancois, @kainino0x)
-
Error reporting for unbalanced compute/render passes does not match implementations gpuweb/gpuweb#5207 (1 comment by @greggman)
-
[bindless] Choose the order of slots returned by insertBinding gpuweb/gpuweb#5466 (1 comment by @mwyrzykowski)
-
Uniformity Analysis - Conservative Handling of Continues and a Possible Solution gpuweb/gpuweb#5363 (1 comment by @jimblandy)
-
[subgroup matrix] Allow subgroup_matrix types in private address space? gpuweb/gpuweb#5435 (1 comment by @jimblandy)
-
Array of numeric scalars missed in list of allowed types for entry point formal parameters gpuweb/gpuweb#5472 (1 comment by @dneto0)
-
[bindless] How should the resource table be set on the state? gpuweb/gpuweb#5463 (1 comment by @Kangz)
-
Uniformity Analysis - Conservative Handling of Loop Return and a Possible Solution gpuweb/gpuweb#5364 (1 comment by @dneto0)
-
-
8 pull requests opened.
-
Sync extension anchors gpuweb/gpuweb#5497, by @mehmetoguzderin.
-
Normalize proposal headers, generate proposal index gpuweb/gpuweb#5490, by @kainino0x.
-
Move accepted and inactive proposals into subdirectories. gpuweb/gpuweb#5483, by @jimblandy.
-
[github workflows] use ubuntu-24.04 gpuweb/gpuweb#5477, by @dneto0.
-
[wgsl] HTML rules: add <tr> inside <thead> gpuweb/gpuweb#5476, by @dneto0.
-
[wgsl] add third option for textureLoad OOB behavior gpuweb/gpuweb#5475, by @dneto0.
-
[bindless] Add discussion of MTL4ArgumentTable gpuweb/gpuweb#5469, by @Kangz.
-
Update bindless proposal with WGSL information. gpuweb/gpuweb#5467, by @dj2.
-
-
8 pull requests merged.
-
Add subgroup_uniformity language extension gpuweb/gpuweb#5431.
-
proposals/atomic-64-min-max.md: Removed duplicated link gpuweb/gpuweb#5426.
-
[wgsl] add third option for textureLoad OOB behavior gpuweb/gpuweb#5475.
-
[wgsl] HTML rules: add <tr> inside <thead> gpuweb/gpuweb#5476.
-
[bindless] Add discussion of MTL4ArgumentTable gpuweb/gpuweb#5469.
-
Update bindless proposal with WGSL information. gpuweb/gpuweb#5467.
-
[bindless] Introduce GPUResourceTable gpuweb/gpuweb#5464.
-
uniformity: relax rules when loop bodies always break or return gpuweb/gpuweb#5419.
-
-
1 pull request closed.
-
[github workflows] use ubuntu-24.04 gpuweb/gpuweb#5477.
-
-
13 pull requests received 64 new comments.
-
Specify immediate data API and WGSL <immediate> address space gpuweb/gpuweb#5423 (23 comments by @alan-baker, @dj2, @dneto0, @kainino0x, @Kangz, @shaoboyan091)
-
Move accepted and inactive proposals into subdirectories. gpuweb/gpuweb#5483 (8 comments by @beaufortfrancois, @dj2, @jimblandy, @kainino0x, @Kangz)
-
Add `TRANSIENT_ATTACHMENT` GPUTextureUsage gpuweb/gpuweb#5450 (7 comments by @beaufortfrancois, @kainino0x, @Kangz, @mwyrzykowski)
-
Update bindless proposal with WGSL information. gpuweb/gpuweb#5467 (5 comments by @beaufortfrancois, @dj2, @Kangz)
-
[bindless] Add discussion of MTL4ArgumentTable gpuweb/gpuweb#5469 (2 comments by @Kangz, @mwyrzykowski)
-
uniformity: relax rules when loop bodies always break or return gpuweb/gpuweb#5419 (1 comment by @dneto0)
-
gpuweb/cts
-
7 issues opened.
-
Add new fragment inputs to `maxInterStageShaderVariables:at_over` test gpuweb/cts#4540, by @ErichDonGubler.
-
Make Tests that check limits skip optional limits gpuweb/cts#4535, by @greggman.
-
SubgroupBroadcast test bug for subgroup sizes < 16 gpuweb/cts#4532, by @jenatali.
-
Require compatibility mode limits gpuweb/cts#4523, by @greggman.
-
1 issue closed.
-
5 issues received 17 new comments.
-
`maxInterStageShaderVariables` should be decremented for _each_ fragment input built-in, not just once gpuweb/cts#4538 (10 comments by @dneto0, @ErichDonGubler, @jimblandy, @kainino0x)
-
SubgroupBroadcast test bug for subgroup sizes < 16 gpuweb/cts#4532 (3 comments by @dneto0, @jenatali, @petermcneeleychromium)
-
Run tests for new/experimental features if ANY part of the feature is available gpuweb/cts#4525 (2 comments by @kainino0x)
-
Add more `copyExternalImageToTexture` tests as Chrome/Safari/Firefox are failing to agree how to load an image gpuweb/cts#4545 (1 comment by @greggman)
-
-
26 pull requests opened.
-
Validation: Add immediate data pipeline validation tests gpuweb/cts#4547, by @shaoboyan091.
-
Fix fragment input deductions in `max_variables_count,input` gpuweb/cts#4546, by @ErichDonGubler.
-
Cleanup some texture checks gpuweb/cts#4544, by @greggman.
-
Compat: Check that textureBindingViewDimension can not pass in cube-array gpuweb/cts#4543, by @greggman.
-
Compat: Add copy T2T, T2B, B2T tests for textureBindingViewDimesion gpuweb/cts#4542, by @greggman.
-
Test reflection of GPUTexture.textureBindingViewDimension. gpuweb/cts#4537, by @greggman.
-
Minor fix for operator desc gpuweb/cts#4536, by @petermcneeleychromium.
-
Bump @webgpu/types to 0.1.68 gpuweb/cts#4534, by @greggman.
-
Require depthOrArrayLayers=1 and mipLevelCount=1 for transient textures gpuweb/cts#4533, by @beaufortfrancois.
-
Restrict transient textures to 2d gpuweb/cts#4531, by @beaufortfrancois.
-
Update maxTextureDimesion3D to match spec. gpuweb/cts#4530, by @greggman.
-
feat: add `…validation,createBindGroupLayout:binding_types:*` gpuweb/cts#4529, by @ErichDonGubler.
-
Clean up old texture usage names in constants.ts gpuweb/cts#4528, by @kainino0x.
-
Test configure() with invalid usages, with some refactoring gpuweb/cts#4527, by @kainino0x.
-
Have resolve test test more formats. gpuweb/cts#4526, by @greggman.
-
Require compat mode limits gpuweb/cts#4524, by @greggman.
-
Revert "Rollin Immediates" gpuweb/cts#4522, by @greggman.
-
Add operation tests for transient attachment usage gpuweb/cts#4520, by @beaufortfrancois.
-
Fix more validation tests for transient attachment usage gpuweb/cts#4519, by @beaufortfrancois.
-
Fix transient attachment usage in validation tests gpuweb/cts#4518, by @beaufortfrancois.
-
Rollin Immediates gpuweb/cts#4517, by @shaoboyan091.
-
CTS for subgroup_uniformity language feature gpuweb/cts#4516, by @alan-baker.
-
Validate setImmediates API in different encoder types gpuweb/cts#4515, by @shaoboyan091.
-
-
20 pull requests merged.
-
Minor fix for operator desc gpuweb/cts#4536.
-
Bump @webgpu/types to 0.1.68 gpuweb/cts#4534.
-
Validate setImmediates API in different encoder types gpuweb/cts#4515.
-
Test configure() with invalid usages, with some refactoring gpuweb/cts#4527.
-
Restrict transient textures to 2d gpuweb/cts#4531.
-
CTS for subgroup_uniformity language feature gpuweb/cts#4516.
-
Update maxTextureDimesion3D to match spec. gpuweb/cts#4530.
-
Clean up old texture usage names in constants.ts gpuweb/cts#4528.
-
Fix more validation tests for transient attachment usage gpuweb/cts#4519.
-
Have resolve test test more formats. gpuweb/cts#4526.
-
Require compat mode limits gpuweb/cts#4524.
-
Revert "Rollin Immediates" gpuweb/cts#4522.
-
Add operation tests for transient attachment usage gpuweb/cts#4520.
-
Fix transient attachment usage in validation tests gpuweb/cts#4518.
-
Rollin Immediates gpuweb/cts#4517.
-
Add validation tests for transient attachment usage gpuweb/cts#4508.
-
uniformity: cover relaxed rules for loop with returns gpuweb/cts#4514.
-
1 pull request closed.
-
Add operation tests for transient attachment usage gpuweb/cts#4513.
-
-
17 pull requests received 109 new comments.
-
Validate setImmediates API in different encoder types gpuweb/cts#4515 (31 comments by @kainino0x, @shaoboyan091)
-
Fix more validation tests for transient attachment usage gpuweb/cts#4519 (17 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Test configure() with invalid usages, with some refactoring gpuweb/cts#4527 (7 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Test reflection of GPUTexture.textureBindingViewDimension. gpuweb/cts#4537 (6 comments by @greggman, @kainino0x)
-
Require depthOrArrayLayers=1 and mipLevelCount=1 for transient textures gpuweb/cts#4533 (6 comments by @beaufortfrancois, @kainino0x)
-
Minor change in sample mask behavior for the case of sample shading gpuweb/cts#4510 (5 comments by @kainino0x, @mwyrzykowski, @petermcneeleychromium)
-
Fix transient attachment usage in validation tests gpuweb/cts#4518 (4 comments by @beaufortfrancois, @greggman)
-
uniformity: cover relaxed rules for loop with returns gpuweb/cts#4514 (4 comments by @alan-baker, @dneto0)
-
`maxInterStageShaderVariables`: decrement max fragment input outputs per fragment input built-in gpuweb/cts#4539 (3 comments by @dneto0, @ErichDonGubler)
-
Add operation tests for transient attachment usage gpuweb/cts#4513 (3 comments by @beaufortfrancois)
-
Revert "'maxStorageBuffersInFragmentStage' and 'maxStorageBuffersInVertexStage' aren't (#4448)" gpuweb/cts#4521 (2 comments by @greggman, @kainino0x)
-
Add new fragment inputs to `maxInterStageShaderVariables:createRenderPipeline,at_over` test gpuweb/cts#4541 (1 comment by @ErichDonGubler)
-
TR publications
-
4 documents published.
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
WebGPU Shading Language published as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
Participants
-
1 participant joined.
-
Ruthvik Konda (Apple Inc.) joined.
-
GitHub
gpuweb/gpuweb
-
18 issues opened.
-
[bindless] How should the resource table be set on the state? gpuweb/gpuweb#5463, by @Kangz.
-
[bindless] OOM semantics for GPUResourceTable gpuweb/gpuweb#5462, by @Kangz.
-
maxColorAttachmentBytesPerSample on Metal in 2025 gpuweb/gpuweb#5460, by @kainino0x.
-
Clarification of Sample Mask with respect to Sample Shading gpuweb/gpuweb#5457, by @petermcneeleychromium.
-
Unified Shader Modules - link multiple GPUShaderModules into a pipeline shader stage gpuweb/gpuweb#5456, by @RobertBColton.
-
Language Server Roadmap gpuweb/gpuweb#5453, by @stefnotch.
-
How should implementations cope with Metal's 500k resources per-process soft-limit? gpuweb/gpuweb#5441, by @teoxoy.
-
Uniformity Analysis - conservative treatement of assignments gpuweb/gpuweb#5440, by @JamesLee-Jones.
-
Variable Value Analysis - Ambiguity in loop rule gpuweb/gpuweb#5439, by @JamesLee-Jones.
-
Out of bounds texture access with textureLoad with swizzle gpuweb/gpuweb#5437, by @greggman.
-
[subgroup matrix] Allow subgroup_matrix types in private address space? gpuweb/gpuweb#5435, by @alan-baker.
-
[subgroup matrix] Should types be parameterized with override-expressions? gpuweb/gpuweb#5434, by @alan-baker.
-
[Immediates] Should the limit be fixed at exactly 64? gpuweb/gpuweb#5430, by @kainino0x.
-
[WGSL fragment depth] Naming bikeshed gpuweb/gpuweb#5428, by @dj2.
-
Firefox shipping WebGPU MVP for macOS in Fx145 gpuweb/gpuweb#5427, by @b0bh00d.
-
-
6 issues closed.
-
[subgroup matrix] Should types be parameterized with override-expressions? gpuweb/gpuweb#5434.
-
[Immediates] Should the limit be fixed at exactly 64? gpuweb/gpuweb#5430.
-
[WGSL fragment depth] Naming bikeshed gpuweb/gpuweb#5428.
-
-
32 issues received 109 new comments.
-
Out of bounds texture access with textureLoad with swizzle gpuweb/gpuweb#5437 (10 comments by @greggman, @jimblandy, @kainino0x, @Kangz, @mwyrzykowski)
-
Reflecting GPUTexture.textureBindingViewDimension requires computing default in content timeline gpuweb/gpuweb#5432 (9 comments by @greggman, @kainino0x, @Kangz, @teoxoy)
-
Unified Shader Modules - link multiple GPUShaderModules into a pipeline shader stage gpuweb/gpuweb#5456 (8 comments by @dj2, @kainino0x, @magcius, @mighdoll, @RobertBColton)
-
[bindless] Add the proposal for how bindings are updated in resource tables. gpuweb/gpuweb#5379 (8 comments by @greggman, @Kangz, @mwyrzykowski)
-
Are "Fine" derivatives guaranteed to actually be fine? gpuweb/gpuweb#4325 (8 comments by @beaufortfrancois, @jrprice, @kainino0x, @mwyrzykowski, @teoxoy)
-
[WGSL fragment depth] (less, greater, any) Metal behavior gpuweb/gpuweb#5351 (7 comments by @beaufortfrancois, @Kangz, @magcius, @mwyrzykowski)
-
The behavior of the test “3d_texture_slices” does not seem to comply with the vulkan spec. gpuweb/gpuweb#5449 (4 comments by @kainino0x, @SlienceLambs)
-
How should implementations cope with Metal's 500k resources per-process soft-limit? gpuweb/gpuweb#5441 (4 comments by @greggman, @mwyrzykowski, @teoxoy)
-
Variable Value Analysis - Ambiguity in loop rule gpuweb/gpuweb#5439 (4 comments by @dneto0, @JamesLee-Jones)
-
WGSL Proposal for fragment depth (less, greater, any) gpuweb/gpuweb#5342 (4 comments by @beaufortfrancois, @jimblandy, @mwyrzykowski)
-
wgsl: support atomics on f32 (in workgroup and storage buffer) gpuweb/gpuweb#4894 (4 comments by @dneto0, @jimblandy, @jowens)
-
[bindless] Alternative to the GPUBindGroup addition of a variable size "dynamic binding array" gpuweb/gpuweb#5372 (3 comments by @cwfitzgerald, @Kangz)
-
[Immediates] Should the limit be fixed at exactly 64? gpuweb/gpuweb#5430 (3 comments by @kainino0x, @Kangz)
-
[subgroup matrix] Should types be parameterized with override-expressions? gpuweb/gpuweb#5434 (3 comments by @alan-baker, @jimblandy, @mwyrzykowski)
-
Add support for memoryless textures gpuweb/gpuweb#5396 (3 comments by @beaufortfrancois, @Kangz, @mwyrzykowski)
-
Uniformity Analysis - conservative treatement of assignments gpuweb/gpuweb#5440 (3 comments by @dneto0, @jimblandy)
-
[WGSL fragment depth] Naming bikeshed gpuweb/gpuweb#5428 (3 comments by @beaufortfrancois, @dj2, @Kangz)
-
Clarification of Sample Mask with respect to Sample Shading gpuweb/gpuweb#5457 (2 comments by @kainino0x, @magcius)
-
[subgroup matrix] Allow subgroup_matrix types in private address space? gpuweb/gpuweb#5435 (2 comments by @jimblandy)
-
uniformity analysis: desugar module-scope variables as if they are declared inside the entry point and passed down as parameters gpuweb/gpuweb#5445 (2 comments by @jimblandy)
-
Firefox shipping WebGPU MVP for macOS in Fx145 gpuweb/gpuweb#5427 (2 comments by @b0bh00d, @kainino0x)
-
[bindless] How should the resource table be set on the state? gpuweb/gpuweb#5463 (1 comment by @magcius)
-
Uniformity Analysis - Conservative Handling of Loop Return and a Possible Solution gpuweb/gpuweb#5364 (1 comment by @dneto0)
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236 (1 comment by @ccameron-chromium)
-
Allow storing handle types in lets, maybe returning them from some functions. gpuweb/gpuweb#5339 (1 comment by @jimblandy)
-
Spec doesn't allow local variables with the same names as parameters gpuweb/gpuweb#3502 (1 comment by @jimblandy)
-
-
22 pull requests opened.
-
[bindless] Introduce GPUResourceTable gpuweb/gpuweb#5464, by @Kangz.
-
WGSL grammar improvements gpuweb/gpuweb#5461, by @benbaarber.
-
Add missing `Render target pixel byte cost` for `8snorm` formats gpuweb/gpuweb#5459, by @Jiawei-Shao.
-
Subgroup matrix proposal updates gpuweb/gpuweb#5458, by @alan-baker.
-
Keep depth/stencil aspect checks gpuweb/gpuweb#5455, by @beaufortfrancois.
-
Update GPURenderPassDepthStencilAttachment Valid Usage algorithm gpuweb/gpuweb#5454, by @beaufortfrancois.
-
[bindless] don't allow updating before dynamic array gpuweb/gpuweb#5452, by @Kangz.
-
Fix typo in transient attachments proposal gpuweb/gpuweb#5451, by @beaufortfrancois.
-
Add `TRANSIENT_ATTACHMENT` GPUTextureUsage gpuweb/gpuweb#5450, by @beaufortfrancois.
-
[editorial] Tweak requestAdapter step ordering, feature level definitions (again) gpuweb/gpuweb#5448, by @kainino0x.
-
[reland as merge commit] [main branch] Specify Compatibility Mode gpuweb/gpuweb#5447, by @kainino0x.
-
Add Transient Attachments proposal gpuweb/gpuweb#5446, by @beaufortfrancois.
-
[bindless] Add proposal for updating dynamic binding arrays gpuweb/gpuweb#5444, by @Kangz.
-
[Compat] Fixes for Compat initialization and more precise prose gpuweb/gpuweb#5443, by @kainino0x.
-
Fix defaulting of textureBindingViewDimension gpuweb/gpuweb#5442, by @SenorBlanco.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5438, by @kainino0x.
-
Compat proposal: make all restriction headings H3 gpuweb/gpuweb#5436, by @SenorBlanco.
-
Disallow Fine variants of WGSL derivative fns gpuweb/gpuweb#5433, by @SenorBlanco.
-
Add subgroup_uniformity language extension gpuweb/gpuweb#5431, by @alan-baker.
-
Remove unused variable in proposal gpuweb/gpuweb#5429, by @SenorBlanco.
-
proposals/atomic-64-min-max.md: Removed duplicated link gpuweb/gpuweb#5426, by @jimblandy.
-
Rename atomic 64 to atomic vec2u gpuweb/gpuweb#5425, by @petermcneeleychromium.
-
-
19 pull requests merged.
-
[bindless] don't allow updating before dynamic array gpuweb/gpuweb#5452.
-
Add missing `Render target pixel byte cost` for `8snorm` formats gpuweb/gpuweb#5459.
-
Subgroup matrix proposal updates gpuweb/gpuweb#5458.
-
Keep depth/stencil aspect checks gpuweb/gpuweb#5455.
-
Update GPURenderPassDepthStencilAttachment Valid Usage algorithm gpuweb/gpuweb#5454.
-
Fix typo in transient attachments proposal gpuweb/gpuweb#5451.
-
Add Transient Attachments proposal gpuweb/gpuweb#5446.
-
[bindless] Add proposal for updating dynamic binding arrays gpuweb/gpuweb#5444.
-
[reland as merge commit] [main branch] Specify Compatibility Mode gpuweb/gpuweb#5447.
-
[main branch] Specify Compatibility Mode gpuweb/gpuweb#5402.
-
[Compat] Fixes for Compat initialization and more precise prose gpuweb/gpuweb#5443.
-
Fix defaulting of textureBindingViewDimension gpuweb/gpuweb#5442.
-
Disallow Fine variants of WGSL derivative fns gpuweb/gpuweb#5433.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5438.
-
Compat proposal: make all restriction headings H3 gpuweb/gpuweb#5436.
-
Remove unused variable in proposal gpuweb/gpuweb#5429.
-
Add subgroup_id language extension gpuweb/gpuweb#5416.
-
Rename atomic 64 to atomic vec2u gpuweb/gpuweb#5425.
-
-
20 pull requests received 98 new comments.
-
[main branch] Specify Compatibility Mode gpuweb/gpuweb#5402 (20 comments by @beaufortfrancois, @kainino0x, @Kangz, @SenorBlanco, @teoxoy)
-
Specify immediate data API and WGSL <immediate> address space gpuweb/gpuweb#5423 (16 comments by @alan-baker, @dneto0, @jimblandy, @Kangz, @mwyrzykowski, @shaoboyan091)
-
Add `TRANSIENT_ATTACHMENT` GPUTextureUsage gpuweb/gpuweb#5450 (11 comments by @beaufortfrancois, @kainino0x, @Kangz, @mwyrzykowski)
-
Fix defaulting of textureBindingViewDimension gpuweb/gpuweb#5442 (9 comments by @kainino0x, @SenorBlanco)
-
[bindless] Add proposal for updating dynamic binding arrays gpuweb/gpuweb#5444 (7 comments by @kainino0x, @Kangz)
-
[Compat] Fixes for Compat initialization and more precise prose gpuweb/gpuweb#5443 (5 comments by @kainino0x, @toji)
-
Add subgroup_uniformity language extension gpuweb/gpuweb#5431 (5 comments by @alan-baker, @dneto0, @Kangz)
-
Disallow Fine variants of WGSL derivative fns gpuweb/gpuweb#5433 (4 comments by @kainino0x, @SenorBlanco)
-
[bindless] don't allow updating before dynamic array gpuweb/gpuweb#5452 (3 comments by @kainino0x, @Kangz)
-
uniformity: relax rules when loop bodies always break or return gpuweb/gpuweb#5419 (3 comments by @dneto0)
-
Add missing `Render target pixel byte cost` for `8snorm` formats gpuweb/gpuweb#5459 (1 comment by @Kangz)
-
Update GPURenderPassDepthStencilAttachment Valid Usage algorithm gpuweb/gpuweb#5454 (1 comment by @beaufortfrancois)
-
[editorial] Tweak requestAdapter step ordering, feature level definitions (again) gpuweb/gpuweb#5448 (1 comment by @toji)
-
gpuweb/cts
-
5 issues opened.
-
Transient Attachments gpuweb/cts#4509, by @beaufortfrancois.
-
[Compat] Test Fine derivatives are disallowed by pipeline creation gpuweb/cts#4502, by @kainino0x.
-
Create render bundle tests gpuweb/cts#4492, by @greggman.
-
18 pull requests opened.
-
uniformity: cover relaxed rules for loop with returns gpuweb/cts#4514, by @dneto0.
-
Add operation tests for transient attachment usage gpuweb/cts#4513, by @beaufortfrancois.
-
Enable only results failures by default gpuweb/cts#4511, by @petermcneeleychromium.
-
Add validation tests for transient attachment usage gpuweb/cts#4508, by @beaufortfrancois.
-
Add hasFeature function and disallow features.has gpuweb/cts#4506, by @greggman.
-
Fix swizzle operation test gpuweb/cts#4505, by @greggman.
-
Skip fine derivative tests in compatibility mode gpuweb/cts#4504, by @greggman.
-
Compat: Test that dpdxFine, dpdyFine, and fwidthFine are disallowed gpuweb/cts#4503, by @greggman.
-
Detect invalid end() in "pass_end_invalid_order" gpuweb/cts#4501, by @Kangz.
-
Software texture reader swizzle test gpuweb/cts#4500, by @greggman.
-
Correct an incorrectly referenced variable gpuweb/cts#4499, by @mei1127.
-
Fix texture-component-swizzle typo gpuweb/cts#4498, by @greggman.
-
const_array_elements limit test: split into MAX, MAX/8, MAX/64 cases gpuweb/cts#4497, by @dneto0.
-
Check that createView format is compatible with the view usage. gpuweb/cts#4495, by @Kangz.
-
Add a test for many timestamp query sets gpuweb/cts#4494, by @greggman.
-
Add Render Bundle operation tests gpuweb/cts#4493, by @greggman.
-
workgroup_size validation: support devices limited to 256 invocations gpuweb/cts#4491, by @dneto0.
-
-
15 pull requests merged.
-
uniformity: test uniformity when a loop body only returns gpuweb/cts#4477.
-
Fix swizzle operation test gpuweb/cts#4505.
-
Enable only results failures by default gpuweb/cts#4511.
-
Add hasFeature function and disallow features.has gpuweb/cts#4506.
-
Skip fine derivative tests in compatibility mode gpuweb/cts#4504.
-
Fix texture-component-swizzle typo gpuweb/cts#4498.
-
Compat: Test that dpdxFine, dpdyFine, and fwidthFine are disallowed gpuweb/cts#4503.
-
Detect invalid end() in "pass_end_invalid_order" gpuweb/cts#4501.
-
Correct an incorrectly referenced variable gpuweb/cts#4499.
-
Add a test for many timestamp query sets gpuweb/cts#4494.
-
Add Render Bundle operation tests gpuweb/cts#4493.
-
const_array_elements limit test: split into MAX, MAX/8, MAX/64 cases gpuweb/cts#4497.
-
Check that createView format is compatible with the view usage. gpuweb/cts#4495.
-
CTS for subgroup_id language feature gpuweb/cts#4487.
-
-
1 pull request closed.
-
Pass texture instead of texture view in operation tests gpuweb/cts#4429.
-
-
6 pull requests received 26 new comments.
-
Fix texture-component-swizzle typo gpuweb/cts#4498 (12 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Add a test for many timestamp query sets gpuweb/cts#4494 (6 comments by @greggman, @kainino0x, @mwyrzykowski)
-
Detect invalid end() in "pass_end_invalid_order" gpuweb/cts#4501 (3 comments by @andyleiserson, @Kangz)
-
TR publications
-
11 documents published.
-
WebGPU published 6 times as a Candidate Recommendation Draft. History of WebGPU
-
WebGPU Shading Language published 5 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
Participants
-
3 participants joined.
-
Mathis Brossier (W3C Invited Experts) joined.
-
lee mighdoll (W3C Invited Experts) joined.
-
Stefan Brandmair (W3C Invited Experts) joined.
-
-
3 participants left.
-
Karen Ghavam (LunarG, Inc.) left.
-
Mark Young (LunarG, Inc.) left.
-
David Desormeaux (LunarG, Inc.) left.
-
-
1 organization left.
-
LunarG, Inc. left.
-
GitHub
gpuweb/gpuweb
-
48 issues opened.
-
Clarification needed for vertex buffer offset alignment requirements gpuweb/gpuweb#5422, by @dj2.
-
[buffer-view] Unify to a single length function? gpuweb/gpuweb#5412, by @alan-baker.
-
[buffer-view] Function parameter matching relaxations gpuweb/gpuweb#5411, by @alan-baker.
-
[buffer-view] Settle on minimum binding size requirements gpuweb/gpuweb#5410, by @alan-baker.
-
[buffer-view] Extent of reinterpretation permitted gpuweb/gpuweb#5409, by @alan-baker.
-
wgsl: clarify desugaring of for loop without a condition clause gpuweb/gpuweb#5397, by @dneto0.
-
Add support for memoryless textures gpuweb/gpuweb#5396, by @alienself.
-
[proposals] Normalize roadmap headers and update completed proposals gpuweb/gpuweb#5385, by @kainino0x.
-
Clarify resource state validation on `submit()` gpuweb/gpuweb#5384, by @teoxoy.
-
[bindless] Add support for pass/command local unpinning of resources gpuweb/gpuweb#5381, by @Kangz.
-
[bindless] Add the proposed WGSL additions gpuweb/gpuweb#5380, by @Kangz.
-
[bindless] Add the proposal for how bindings are updated in resource tables. gpuweb/gpuweb#5379, by @Kangz.
-
[bindless] Stateless alternative to usage pinning? gpuweb/gpuweb#5378, by @Kangz.
-
[bindless] Add layout/shader compatibility and layout defaulting gpuweb/gpuweb#5377, by @Kangz.
-
[bindless] How to represent RO vs. RW usages for pinning? gpuweb/gpuweb#5376, by @Kangz.
-
[bindless] Add pin/unpin to GPUBuffer gpuweb/gpuweb#5375, by @Kangz.
-
[bindless] Determine the compatibility rules for `GPUDynamicBindingType` gpuweb/gpuweb#5374, by @Kangz.
-
[bindless] What limit for maxResourceTableSize? gpuweb/gpuweb#5373, by @Kangz.
-
[bindless] Alternative to the GPUBindGroup addition of a variable size "dynamic binding array" gpuweb/gpuweb#5372, by @Kangz.
-
[subgroup-id] consider adding a "num subgroups" builtin value gpuweb/gpuweb#5370, by @dneto0.
-
[subgroup-id] uniformity of subgroup-id gpuweb/gpuweb#5368, by @dneto0.
-
[subgroup-id] is there an acceptable polyfill on D3D? gpuweb/gpuweb#5367, by @dneto0.
-
subgroup-id proposal discussions gpuweb/gpuweb#5365, by @dneto0.
-
Uniformity Analysis - Conservative Handling of Loop Return and a Possible Solution gpuweb/gpuweb#5364, by @JamesLee-Jones.
-
Uniformity Analysis - Conservative Handling of Continues and a Possible Solution gpuweb/gpuweb#5363, by @JamesLee-Jones.
-
[Explicit BGL params] Should texture parameter names match the API spec? gpuweb/gpuweb#5359, by @dj2.
-
Verify the validation table gpuweb/gpuweb#5358, by @dj2.
-
[Explicit BGL params] Relax the function traceback rule? gpuweb/gpuweb#5357, by @dj2.
-
[Explicit BGL params] Requirement for the `unknown` token gpuweb/gpuweb#5356, by @dj2.
-
[Explicit BGL params] Verify the parameter conversions are valid gpuweb/gpuweb#5355, by @dj2.
-
Proposal for explicit binding layout parameters gpuweb/gpuweb#5353, by @dj2.
-
[WGSL fragment depth] (less, greater, any) Metal behavior gpuweb/gpuweb#5351, by @beaufortfrancois.
-
[WGSL fragment depth] (less, greater, any) syntax change gpuweb/gpuweb#5350, by @beaufortfrancois.
-
[WGSL fragment depth] (less, greater, any) portability gpuweb/gpuweb#5349, by @beaufortfrancois.
-
[WGSL fragment depth] (less, greater, any) validation gpuweb/gpuweb#5348, by @beaufortfrancois.
-
[subgroup-id] Specify range and density of the IDs gpuweb/gpuweb#5344, by @dneto0.
-
Consider adding some kind of draw-time validation of filtering / filterable compabitility. gpuweb/gpuweb#5343, by @Kangz.
-
WGSL Proposal for fragment depth (less, greater, any) gpuweb/gpuweb#5342, by @beaufortfrancois.
-
Validation fails to reject out-of-range fragment shader output locations gpuweb/gpuweb#5341, by @jimblandy.
-
Allow storing handle types in lets, maybe returning them from some functions. gpuweb/gpuweb#5339, by @Kangz.
-
[buffer-view] Viewing the content of a storage buffer as different types. gpuweb/gpuweb#5338, by @Kangz.
-
Add missing fragment input built-ins to "validating inter-stage interfaces" algorithm gpuweb/gpuweb#5337, by @teoxoy.
-
Does `GPUStorageTextureBindingLayout.format` serve a purpose? gpuweb/gpuweb#5336, by @cwfitzgerald.
-
Tracking issue to finish and land texel buffers in the spec. gpuweb/gpuweb#5334, by @Kangz.
-
Validated out more view usages early?? gpuweb/gpuweb#5331, by @greggman.
-
copyExternalImageToTexture is very slow gpuweb/gpuweb#5330, by @greggman.
-
-
51 issues closed.
-
[WGSL fragment depth] (less, greater, any) syntax change gpuweb/gpuweb#5350.
-
Clarify resource state validation on `submit()` gpuweb/gpuweb#5384.
-
[subgroup-id] Specify range and density of the IDs gpuweb/gpuweb#5344.
-
[subgroup-id] consider adding a "num subgroups" builtin value gpuweb/gpuweb#5370.
-
Validation fails to reject out-of-range fragment shader output locations gpuweb/gpuweb#5341.
-
[Immediate Data] Shall we support per-stage immediate data? gpuweb/gpuweb#5116.
-
[Immediate Data] Rename <immediate_data> to <immediate> gpuweb/gpuweb#5119.
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118.
-
[subgroup-id] is there an acceptable polyfill on D3D? gpuweb/gpuweb#5367.
-
wgsl: clarify desugaring of for loop without a condition clause gpuweb/gpuweb#5397.
-
Add function call statement and expression rules to uniformity analysis gpuweb/gpuweb#5128.
-
Desugaring assumptions in uniformity analysis gpuweb/gpuweb#5129.
-
Add a rule for empty statements to uniformity rules for statements gpuweb/gpuweb#5123.
-
Making uniformity analysis less conservative via statement behaviour analysis gpuweb/gpuweb#5100.
-
Add parenthesis rule to uniformity rules for expressions gpuweb/gpuweb#5121.
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987.
-
Limit the value passed into @align? gpuweb/gpuweb#5150.
-
Uniformity Analysis - Account for break if expression uniformity gpuweb/gpuweb#5277.
-
Clarify that `alias` names can be used as constructor function names gpuweb/gpuweb#5014.
-
add subgroups, and make them portable if possible gpuweb/gpuweb#4306.
-
Request for compute: anyInvocation() and allInvocation() gpuweb/gpuweb#2137.
-
[immediates] minimum maxImmediateSize should be 16? gpuweb/gpuweb#5320.
-
"shader-f16" requirements exclude all Qualcomm devices gpuweb/gpuweb#5006.
-
Bump macOS version to 10.15 gpuweb/gpuweb#3685.
-
Only 32 timestamp queries allowed on Metal MacOS? gpuweb/gpuweb#5261.
-
ErrorScopes are not useful for CommandEncoder methods gpuweb/gpuweb#4269.
-
Profiling copyTextureToTexture() etc. gpuweb/gpuweb#5225.
-
Implementation Status Page Status Clarification gpuweb/gpuweb#4745.
-
WebGPU Features + Limits Survey website, did you want to link to it? gpuweb/gpuweb#4074.
-
Verify the validation table gpuweb/gpuweb#5358.
-
What OSes and hardware are we targeting? gpuweb/gpuweb#1069.
-
Handling dynamic Metal maxThreadsPerThreadgroup gpuweb/gpuweb#5315.
-
Does `GPUStorageTextureBindingLayout.format` serve a purpose? gpuweb/gpuweb#5336.
-
What to do with webgpu.org gpuweb/gpuweb#4032.
-
When shall we check if feature is enabled? gpuweb/gpuweb#5329.
-
Validated out more view usages early?? gpuweb/gpuweb#5331.
-
copyExternalImageToTexture should allow copying to the new 16unorm formats gpuweb/gpuweb#5289.
-
[DISCUSS] Compute Shader non-buffer arguments gpuweb/gpuweb#762.
-
maxDynamicFoo should be required to be at most maxFoo in limits gpuweb/gpuweb#5069.
-
Should we disallow creating views that only includes `COPY_*` usages? gpuweb/gpuweb#5317.
-
Compat: Rename compat limits to clarify they're compat-only gpuweb/gpuweb#5322.
-
Allow copyExternalImageToTexture for all renderable formats gpuweb/gpuweb#5323.
-
copyExternalImageToTexture is very slow gpuweb/gpuweb#5330.
-
Maintainer access for the WebGPU landing page gpuweb/gpuweb#5324.
-
89 issues received 285 new comments.
-
Reintroduce source maps gpuweb/gpuweb#4844 (28 comments by @greggman, @jimblandy, @kainino0x, @Kangz, @mighdoll, @mwyrzykowski, @stefnotch)
-
Should texture swizzle move usage restrictions from createView to createBindGroup/beginRenderPass? gpuweb/gpuweb#5298 (21 comments by @beaufortfrancois, @kainino0x, @Kangz, @magcius, @mwyrzykowski)
-
WGSL Proposal for fragment depth (less, greater, any) gpuweb/gpuweb#5342 (14 comments by @beaufortfrancois, @dj2, @Kangz, @magcius)
-
Add support for memoryless textures gpuweb/gpuweb#5396 (11 comments by @AlbinBernhardssonARM, @alienself, @jimblandy, @kainino0x, @magcius, @mwyrzykowski)
-
Uniformity Analysis - Conservative Handling of Loop Return and a Possible Solution gpuweb/gpuweb#5364 (11 comments by @alan-baker, @dneto0, @JamesLee-Jones, @Kangz)
-
[subgroup-id] is there an acceptable polyfill on D3D? gpuweb/gpuweb#5367 (8 comments by @alan-baker, @Jiawei-Shao, @jimblandy, @petermcneeleychromium)
-
[bindless] Add the proposal for how bindings are updated in resource tables. gpuweb/gpuweb#5379 (7 comments by @greggman, @magcius)
-
[Explicit BGL params] Verify the parameter conversions are valid gpuweb/gpuweb#5355 (7 comments by @dneto0, @jimblandy, @kainino0x)
-
Handling dynamic Metal maxThreadsPerThreadgroup gpuweb/gpuweb#5315 (7 comments by @jimblandy, @jrprice, @mwyrzykowski)
-
Does `GPUStorageTextureBindingLayout.format` serve a purpose? gpuweb/gpuweb#5336 (7 comments by @cwfitzgerald, @kainino0x, @Kangz, @magcius)
-
Clarify resource state validation on `submit()` gpuweb/gpuweb#5384 (6 comments by @Kangz, @mwyrzykowski, @teoxoy)
-
[proposal] consider having compatible APIs for textureLoad on different types (texture_2d, texture_storage_2d, etc.) gpuweb/gpuweb#5387 (6 comments by @EriKWDev, @kainino0x)
-
[immediates] minimum maxImmediateSize should be 16? gpuweb/gpuweb#5320 (6 comments by @aleino-nv, @kainino0x, @Kangz)
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236 (5 comments by @ccameron-chromium, @jimblandy, @mwyrzykowski)
-
Add missing fragment input built-ins to "validating inter-stage interfaces" algorithm gpuweb/gpuweb#5337 (5 comments by @Kangz, @teoxoy, @toji)
-
[WGSL fragment depth] (less, greater, any) Metal behavior gpuweb/gpuweb#5351 (5 comments by @beaufortfrancois, @mwyrzykowski)
-
Maintainer access for the WebGPU landing page gpuweb/gpuweb#5324 (5 comments by @jimblandy, @kainino0x, @Kangz)
-
[Explicit BGL params] Requirement for the `unknown` token gpuweb/gpuweb#5356 (4 comments by @jimblandy, @kainino0x, @Kangz)
-
Proposal for explicit binding layout parameters gpuweb/gpuweb#5353 (4 comments by @dj2, @jimblandy, @kainino0x, @mwyrzykowski)
-
[subgroup-id] uniformity of subgroup-id gpuweb/gpuweb#5368 (4 comments by @alan-baker, @jimblandy, @Kangz)
-
What OSes and hardware are we targeting? gpuweb/gpuweb#1069 (4 comments by @dj2, @kainino0x, @Kangz)
-
When shall we check if feature is enabled? gpuweb/gpuweb#5329 (4 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Allow binding zero-sized GPUBufferBindings gpuweb/gpuweb#5312 (3 comments by @jimblandy, @magcius, @mwyrzykowski)
-
[subgroup-id] Specify range and density of the IDs gpuweb/gpuweb#5344 (3 comments by @alan-baker, @jimblandy)
-
Uniformity Analysis - Account for break if expression uniformity gpuweb/gpuweb#5277 (3 comments by @dneto0, @jimblandy)
-
Uniformity Analysis - Conservative Handling of Continues and a Possible Solution gpuweb/gpuweb#5363 (3 comments by @afd, @alan-baker)
-
Tracking issue to finish and land texel buffers in the spec. gpuweb/gpuweb#5334 (3 comments by @dneto0, @jrprice)
-
Remove the need to hard-code texel format for `texture_storage_2d` in shaders gpuweb/gpuweb#4298 (3 comments by @cwfitzgerald, @kainino0x, @Kangz)
-
Should we disallow creating views that only includes `COPY_*` usages? gpuweb/gpuweb#5317 (3 comments by @kainino0x, @Kangz)
-
[bindless] Alternative to the GPUBindGroup addition of a variable size "dynamic binding array" gpuweb/gpuweb#5372 (2 comments by @kainino0x, @magcius)
-
Clarification needed for vertex buffer offset alignment requirements gpuweb/gpuweb#5422 (2 comments by @kainino0x)
-
[Explicit BGL params] Should texture parameter names match the API spec? gpuweb/gpuweb#5359 (2 comments by @alan-baker, @jimblandy)
-
[Explicit BGL params] Relax the function traceback rule? gpuweb/gpuweb#5357 (2 comments by @jimblandy, @kainino0x)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (2 comments by @jspanchu, @kainino0x)
-
[Immediates] Should failing to set an immediate value used by the shader be a validation error? gpuweb/gpuweb#5318 (2 comments by @kainino0x, @shaoboyan091)
-
Counting of named variables allows programs which fail GPU[Render/Compute]Pipeline creation gpuweb/gpuweb#5401 (2 comments by @mwyrzykowski, @toji)
-
wgsl: clarify desugaring of for loop without a condition clause gpuweb/gpuweb#5397 (2 comments by @dneto0)
-
[bindless] What limit for maxResourceTableSize? gpuweb/gpuweb#5373 (2 comments by @Kangz, @mwyrzykowski)
-
Validation fails to reject out-of-range fragment shader output locations gpuweb/gpuweb#5341 (2 comments by @kainino0x, @Kangz)
-
what happens if vertex w position is zero, or negative? gpuweb/gpuweb#4129 (2 comments by @aleino-nv)
-
Allow copyExternalImageToTexture for all renderable formats gpuweb/gpuweb#5323 (2 comments by @Kangz)
-
Compat: Rename compat limits to clarify they're compat-only gpuweb/gpuweb#5322 (2 comments by @Kangz)
-
Should there be a way to prevent loop unrolling? gpuweb/gpuweb#4110 (2 comments by @kainino0x, @pverlind)
-
[bindless] How to represent RO vs. RW usages for pinning? gpuweb/gpuweb#5376 (1 comment by @magcius)
-
Change `layout: 'auto'` algorithm so it doesn't generate invalid bindGroupLayouts gpuweb/gpuweb#4952 (1 comment by @jimblandy)
-
Proposal: fully explicit, non-exclusive "auto" layouts in WGSL (without createPipelineLayout) gpuweb/gpuweb#4957 (1 comment by @jimblandy)
-
[editorial] Fix algorithms wording to handle textures/samplers passed as function args gpuweb/gpuweb#5328 (1 comment by @jimblandy)
-
setBindGroup should accept ArrayBuffer (AllowSharedBufferSource) for dynamicOffsetsData gpuweb/gpuweb#5212 (1 comment by @jimblandy)
-
Add function call statement and expression rules to uniformity analysis gpuweb/gpuweb#5128 (1 comment by @dneto0)
-
Add a rule for empty statements to uniformity rules for statements gpuweb/gpuweb#5123 (1 comment by @dneto0)
-
Explicitly state how to handle an if statement with an empty else in uniformity analysis gpuweb/gpuweb#5122 (1 comment by @dneto0)
-
[WGSL fragment depth] (less, greater, any) syntax change gpuweb/gpuweb#5350 (1 comment by @beaufortfrancois)
-
[Immediate Data] Shall we support per-stage immediate data? gpuweb/gpuweb#5116 (1 comment by @Kangz)
-
editorial: Add table of correspondences for builtin functions in common shader languages gpuweb/gpuweb#2502 (1 comment by @kainino0x)
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987 (1 comment by @Kangz)
-
[subgroup-id] consider adding a "num subgroups" builtin value gpuweb/gpuweb#5370 (1 comment by @alan-baker)
-
Finite Math Assumption makes WebGPU a difficult compilation target gpuweb/gpuweb#5109 (1 comment by @dneto0)
-
Error reporting for unbalanced compute/render passes does not match implementations gpuweb/gpuweb#5207 (1 comment by @Kangz)
-
WebGPU Features + Limits Survey website, did you want to link to it? gpuweb/gpuweb#4074 (1 comment by @Kangz)
-
`GPUBindGroupLayoutEntry` is accepting an empty dictionary for `externalTexture`, should accept boolean instead gpuweb/gpuweb#5249 (1 comment by @Kangz)
-
Allow storing handle types in lets, maybe returning them from some functions. gpuweb/gpuweb#5339 (1 comment by @alan-baker)
-
wgsl: add feature to enable more IEEE 754 floating point behaviours gpuweb/gpuweb#2259 (1 comment by @dneto0)
-
[buffer-view] Viewing the content of a storage buffer as different types. gpuweb/gpuweb#5338 (1 comment by @alan-baker)
-
-
45 pull requests opened.
-
[WGSL] Specify <immediate> address space gpuweb/gpuweb#5424, by @shaoboyan091.
-
Specify immediate data API and WGSL <immediate> address space gpuweb/gpuweb#5423, by @shaoboyan091.
-
[Explicit BGL params]: reorder rows columns in compatibility matrix gpuweb/gpuweb#5421, by @dneto0.
-
Fix typo: "bindings" -> "binding". gpuweb/gpuweb#5420, by @jimblandy.
-
uniformity: relax rules when loop bodies always break or return gpuweb/gpuweb#5419, by @dneto0.
-
Update immediate validation part gpuweb/gpuweb#5418, by @shaoboyan091.
-
Fix typo in uniformity function-scope variable analysis gpuweb/gpuweb#5417, by @dneto0.
-
Add subgroup_id language extension gpuweb/gpuweb#5416, by @alan-baker.
-
Compat: fix check for compatible texture views gpuweb/gpuweb#5415, by @kainino0x.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5414, by @kainino0x.
-
[editorial] Backport wording/formatting changes from Compat branch gpuweb/gpuweb#5408, by @kainino0x.
-
Compat: miscellaneous editorial nits gpuweb/gpuweb#5407, by @kainino0x.
-
[proposal] Rename push-constants.md to immediate-data.md gpuweb/gpuweb#5406, by @kainino0x.
-
Update validation of bind groups usage gpuweb/gpuweb#5405, by @toji.
-
Validate fragment shader output locations gpuweb/gpuweb#5404, by @toji.
-
Add new interstage builtins gpuweb/gpuweb#5403, by @toji.
-
[main branch] Specify Compatibility Mode gpuweb/gpuweb#5402, by @SenorBlanco.
-
Add buffer-view proposal gpuweb/gpuweb#5400, by @alan-baker.
-
Update immediate-data proposal gpuweb/gpuweb#5399, by @shaoboyan091.
-
wgsl: clarify desugaring of for loop without condition gpuweb/gpuweb#5398, by @dneto0.
-
uniformity: fix editorial issues gpuweb/gpuweb#5394, by @dneto0.
-
Switch to Built-in `depth_mode` parameter to `frag_depth` gpuweb/gpuweb#5393, by @beaufortfrancois.
-
uniformity: fix analysis for loop bodies that only return gpuweb/gpuweb#5392, by @dneto0.
-
Specify behaviour of core-features-and-limits. gpuweb/gpuweb#5391, by @SenorBlanco.
-
README: document Python venv for building WGSL spec gpuweb/gpuweb#5390, by @dneto0.
-
Allow lets of textures and samplers gpuweb/gpuweb#5389, by @alan-baker.
-
Remove roadmap note for texture component swizzle gpuweb/gpuweb#5388, by @beaufortfrancois.
-
[subgroup-id] Add issue resolutions gpuweb/gpuweb#5386, by @alan-baker.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5383, by @kainino0x.
-
Add a partial proposal for bindless. gpuweb/gpuweb#5382, by @Kangz.
-
wgsl:copyediting: type aliases can be used as value constructors gpuweb/gpuweb#5371, by @dneto0.
-
uniformity: fix analysis of break-if condition gpuweb/gpuweb#5369, by @dneto0.
-
subgroup-id proposal: link to umbrella bug gpuweb/gpuweb#5366, by @dneto0.
-
Compat: Add swizzle to texture views match gpuweb/gpuweb#5362, by @beaufortfrancois.
-
Add `"texture-component-swizzle"` feature gpuweb/gpuweb#5361, by @beaufortfrancois.
-
Move texture component swizzle usage restrictions gpuweb/gpuweb#5360, by @beaufortfrancois.
-
Add proposal for explicit binding layout parameters. gpuweb/gpuweb#5354, by @dj2.
-
Add WGSL proposal for fragment depth (less, greater, any) gpuweb/gpuweb#5352, by @beaufortfrancois.
-
Language feature for relaxing uniform address space layout gpuweb/gpuweb#5347, by @alan-baker.
-
subgroup-matrix proposal: capture discussion issues and feedback gpuweb/gpuweb#5346, by @dneto0.
-
Record issues for atomic-64-min-max proposal gpuweb/gpuweb#5345, by @dneto0.
-
move the webgpu logo below the W3C one gpuweb/gpuweb#5340, by @deniak.
-
Proposal for subgroup matrix feature gpuweb/gpuweb#5335, by @alan-baker.
-
copyExternalImageToTexture: allow all unorm/unorm-srgb/float/ufloat textures gpuweb/gpuweb#5333, by @kainino0x.
-
Proposal to add subgroup_id built-in value gpuweb/gpuweb#5332, by @alan-baker.
-
-
43 pull requests merged.
-
Update immediate validation part gpuweb/gpuweb#5418.
-
[Explicit BGL params]: reorder rows columns in compatibility matrix gpuweb/gpuweb#5421.
-
Fix typo: "bindings" -> "binding". gpuweb/gpuweb#5420.
-
Switch to Built-in `depth_mode` parameter to `frag_depth` gpuweb/gpuweb#5393.
-
Update validation of bind groups usage gpuweb/gpuweb#5405.
-
Fix typo in uniformity function-scope variable analysis gpuweb/gpuweb#5417.
-
Compat: fix check for compatible texture views gpuweb/gpuweb#5415.
-
Language feature for relaxing uniform address space layout gpuweb/gpuweb#5347.
-
[subgroup-id] Add issue resolutions gpuweb/gpuweb#5386.
-
Add buffer-view proposal gpuweb/gpuweb#5400.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5414.
-
[editorial] Backport wording/formatting changes from Compat branch gpuweb/gpuweb#5408.
-
Compat: miscellaneous editorial nits gpuweb/gpuweb#5407.
-
Validate fragment shader output locations gpuweb/gpuweb#5404.
-
Add new interstage builtins gpuweb/gpuweb#5403.
-
Update immediate-data proposal gpuweb/gpuweb#5399.
-
[proposal] Rename push-constants.md to immediate-data.md gpuweb/gpuweb#5406.
-
Specify behaviour of core-features-and-limits. gpuweb/gpuweb#5391.
-
wgsl: clarify desugaring of for loop without condition gpuweb/gpuweb#5398.
-
uniformity: fix analysis for loop bodies that only return gpuweb/gpuweb#5392.
-
uniformity: fix editorial issues gpuweb/gpuweb#5394.
-
Add a partial proposal for bindless. gpuweb/gpuweb#5382.
-
README: document Python venv for building WGSL spec gpuweb/gpuweb#5390.
-
Remove roadmap note for texture component swizzle gpuweb/gpuweb#5388.
-
Add proposal for explicit binding layout parameters. gpuweb/gpuweb#5354.
-
uniformity: fix analysis of break-if condition gpuweb/gpuweb#5369.
-
wgsl:copyediting: type aliases can be used as value constructors gpuweb/gpuweb#5371.
-
Compat: Add swizzle to texture views match gpuweb/gpuweb#5362.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5383.
-
Add `"texture-component-swizzle"` feature gpuweb/gpuweb#5361.
-
subgroup-id proposal: link to umbrella bug gpuweb/gpuweb#5366.
-
Move texture component swizzle usage restrictions gpuweb/gpuweb#5360.
-
Add WGSL proposal for fragment depth (less, greater, any) gpuweb/gpuweb#5352.
-
subgroup-matrix proposal: capture discussion issues and feedback gpuweb/gpuweb#5346.
-
Proposal for subgroup matrix feature gpuweb/gpuweb#5335.
-
Record issues for atomic-64-min-max proposal gpuweb/gpuweb#5345.
-
Proposal to add subgroup_id built-in value gpuweb/gpuweb#5332.
-
move the webgpu logo below the W3C one gpuweb/gpuweb#5340.
-
Atomic 64 proposal gpuweb/gpuweb#5314.
-
Subtract position from Vulkan inter stage limits gpuweb/gpuweb#4518.
-
copyExternalImageToTexture: allow all unorm/unorm-srgb/float/ufloat textures gpuweb/gpuweb#5333.
-
[non-normative] Explain -srgb formats are not color managed/tagged gpuweb/gpuweb#4597.
-
Limit texture/sampler combinations gpuweb/gpuweb#5313.
-
-
5 pull requests closed.
-
[WGSL] Specify <immediate> address space gpuweb/gpuweb#5424.
-
Remove the greater-or-equal constraint on maxBindingsPerBindGroup gpuweb/gpuweb#4484.
-
Add GPUCommandEncoder.updateBuffer for embedding small buffer updates. gpuweb/gpuweb#650.
-
Specify [=debug label=]. gpuweb/gpuweb#1240.
-
Add additional dstformat in copyExternalImageToTexture with Tier1 gpuweb/gpuweb#5299.
-
-
42 pull requests received 174 new comments.
-
Specify behaviour of core-features-and-limits. gpuweb/gpuweb#5391 (15 comments by @kainino0x, @SenorBlanco, @teoxoy)
-
Add buffer-view proposal gpuweb/gpuweb#5400 (12 comments by @alan-baker, @dneto0, @jimblandy, @stefnotch)
-
Language feature for relaxing uniform address space layout gpuweb/gpuweb#5347 (11 comments by @alan-baker, @dneto0, @jimblandy)
-
Add subgroup_id language extension gpuweb/gpuweb#5416 (10 comments by @alan-baker, @dneto0, @jrprice)
-
Add proposal for explicit binding layout parameters. gpuweb/gpuweb#5354 (9 comments by @dj2, @dneto0, @jimblandy, @kainino0x, @Kangz)
-
Allow lets of textures and samplers gpuweb/gpuweb#5389 (8 comments by @alan-baker, @dneto0, @jimblandy, @mighdoll)
-
[WGSL] Specify <immediate> address space gpuweb/gpuweb#5424 (7 comments by @alan-baker, @jimblandy, @kainino0x, @shaoboyan091)
-
Proposal for subgroup matrix feature gpuweb/gpuweb#5335 (7 comments by @alan-baker, @dneto0, @jimblandy, @junjihashimoto, @mwyrzykowski)
-
[main branch] Specify Compatibility Mode gpuweb/gpuweb#5402 (6 comments by @jimblandy, @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Update validation of bind groups usage gpuweb/gpuweb#5405 (6 comments by @andyleiserson, @kainino0x, @toji)
-
Add a partial proposal for bindless. gpuweb/gpuweb#5382 (6 comments by @aleino-nv, @Kangz, @mwyrzykowski)
-
Atomic 64 proposal gpuweb/gpuweb#5314 (6 comments by @alan-baker, @mwyrzykowski, @petermcneeleychromium)
-
Specify immediate data API and WGSL <immediate> address space gpuweb/gpuweb#5423 (5 comments by @jimblandy, @kainino0x, @shaoboyan091)
-
uniformity: fix analysis of break-if condition gpuweb/gpuweb#5369 (5 comments by @alan-baker, @dneto0, @jrprice)
-
Switch to Built-in `depth_mode` parameter to `frag_depth` gpuweb/gpuweb#5393 (4 comments by @beaufortfrancois, @dj2, @kainino0x)
-
Subtract position from Vulkan inter stage limits gpuweb/gpuweb#4518 (4 comments by @kainino0x, @Kangz, @teoxoy)
-
Compat: fix check for compatible texture views gpuweb/gpuweb#5415 (3 comments by @kainino0x, @SenorBlanco)
-
Add `"texture-component-swizzle"` feature gpuweb/gpuweb#5361 (3 comments by @beaufortfrancois, @kainino0x, @Kangz)
-
uniformity: fix analysis for loop bodies that only return gpuweb/gpuweb#5392 (2 comments by @dneto0)
-
Compat: Add swizzle to texture views match gpuweb/gpuweb#5362 (2 comments by @beaufortfrancois, @kainino0x)
-
copyExternalImageToTexture: allow all unorm/unorm-srgb/float/ufloat textures gpuweb/gpuweb#5333 (2 comments by @kainino0x)
-
[non-normative] Explain -srgb formats are not color managed/tagged gpuweb/gpuweb#4597 (2 comments by @kainino0x, @Kangz)
-
[Explicit BGL params]: reorder rows columns in compatibility matrix gpuweb/gpuweb#5421 (1 comment by @kainino0x)
-
uniformity: relax rules when loop bodies always break or return gpuweb/gpuweb#5419 (1 comment by @dneto0)
-
subgroup-matrix proposal: capture discussion issues and feedback gpuweb/gpuweb#5346 (1 comment by @dneto0)
-
Remove the greater-or-equal constraint on maxBindingsPerBindGroup gpuweb/gpuweb#4484 (1 comment by @Kangz)
-
Add GPUCommandEncoder.updateBuffer for embedding small buffer updates. gpuweb/gpuweb#650 (1 comment by @Kangz)
-
Add additional dstformat in copyExternalImageToTexture with Tier1 gpuweb/gpuweb#5299 (1 comment by @kainino0x)
-
gpuweb/cts
-
8 issues opened.
-
OOM allocation in `src/webgpu/api/operation/buffers/map_oom.spec.ts` too small gpuweb/cts#4488, by @zoddicus.
-
Remove 16-byte alignment requirement from uniform buffers in helper functions gpuweb/cts#4485, by @jrprice.
-
[reftests] Use requestAnimationFrame instead of takeScreenshotDelayed gpuweb/cts#4482, by @kainino0x.
-
wgsl uniformity: test uniformity for loop where body has behaviour {Return} gpuweb/cts#4476, by @dneto0.
-
CTS not failing when it should? gpuweb/cts#4474, by @greggman.
-
wgsl: test uniformity of break-if value gpuweb/cts#4471, by @dneto0.
-
-
6 issues closed.
-
CTS not failing when it should? gpuweb/cts#4474.
-
Texture component swizzle 🥤 gpuweb/cts#4421.
-
wgsl: test uniformity of break-if value gpuweb/cts#4471.
-
4 issues received 6 new comments.
-
[reftests] Use requestAnimationFrame instead of takeScreenshotDelayed gpuweb/cts#4482 (3 comments by @kainino0x, @mwyrzykowski, @szager-chromium)
-
Test `webgpu:api,operation,adapter,requestAdapter:requestAdapter` fails with non expected fallback adapter gpuweb/cts#4480 (1 comment by @kainino0x)
-
-
18 pull requests opened.
-
Add CTS for texture_and_sampler_let language feature gpuweb/cts#4490, by @alan-baker.
-
Increase the OOM allocation size to max precise value gpuweb/cts#4489, by @zoddicus.
-
CTS for subgroup_id language feature gpuweb/cts#4487, by @alan-baker.
-
Adjust maxFractionalDiff for canvas color-space-conversion gpuweb/cts#4486, by @greggman.
-
wgsl: Add coverage for uniform_buffer_standard_layout gpuweb/cts#4484, by @jrprice.
-
Check GPUAdapterInfo.isFallbackAdapter on explicit `fallback` request gpuweb/cts#4483, by @tharkum.
-
Add validation test for semicolon before break-if. gpuweb/cts#4481, by @dj2.
-
Add web_platform reftest for copyElementImageToTexture gpuweb/cts#4479, by @szager-chromium.
-
remove use of 'subgroup_branching' diagnostic gpuweb/cts#4478, by @dneto0.
-
uniformity: test uniformity when a loop body only returns gpuweb/cts#4477, by @dneto0.
-
Catch unexpected WebGPU errors. gpuweb/cts#4475, by @greggman.
-
Clean unnecessary swizzle related casts gpuweb/cts#4473, by @beaufortfrancois.
-
wgsl: test uniformity analysis of break-if condition gpuweb/cts#4472, by @dneto0.
-
wgsl: Add validation tests for override-sized arrays gpuweb/cts#4470, by @jrprice.
-
Add logical operation override tests. gpuweb/cts#4469, by @dj2.
-
Cleanups in DevicePool.release gpuweb/cts#4467, by @kainino0x.
-
Add option to periodically replace device gpuweb/cts#4466, by @kainino0x.
-
Refactor copyExternalImageToTexture tested formats. gpuweb/cts#4465, by @greggman.
-
-
15 pull requests merged.
-
Increase the OOM allocation size to max precise value gpuweb/cts#4489.
-
Adjust maxFractionalDiff for canvas color-space-conversion gpuweb/cts#4486.
-
wgsl: Add coverage for uniform_buffer_standard_layout gpuweb/cts#4484.
-
Add validation test for semicolon before break-if. gpuweb/cts#4481.
-
remove use of 'subgroup_branching' diagnostic gpuweb/cts#4478.
-
Refactor copyExternalImageToTexture tested formats. gpuweb/cts#4465.
-
Clean unnecessary swizzle related casts gpuweb/cts#4473.
-
Catch unexpected WebGPU errors. gpuweb/cts#4475.
-
wgsl: test uniformity analysis of break-if condition gpuweb/cts#4472.
-
Add texture-component-swizzle tests gpuweb/cts#4427.
-
wgsl: Add validation tests for override-sized arrays gpuweb/cts#4470.
-
Add logical operation override tests. gpuweb/cts#4469.
-
Cleanups in DevicePool.release gpuweb/cts#4467.
-
Add option to periodically replace device gpuweb/cts#4466.
-
-
1 pull request closed.
-
Add web_platform reftest for copyElementImageToTexture gpuweb/cts#4479.
-
-
10 pull requests received 42 new comments.
-
Add texture-component-swizzle tests gpuweb/cts#4427 (11 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Add web_platform reftest for copyElementImageToTexture gpuweb/cts#4479 (7 comments by @kainino0x, @szager-chromium)
-
uniformity: test uniformity when a loop body only returns gpuweb/cts#4477 (6 comments by @dneto0, @jrprice)
-
Increase the OOM allocation size to max precise value gpuweb/cts#4489 (5 comments by @jrprice, @kainino0x, @zoddicus)
-
Check GPUAdapterInfo.isFallbackAdapter on explicit `fallback` request gpuweb/cts#4483 (2 comments by @kainino0x, @tharkum)
-
TR publications
-
7 documents published.
-
WebGPU Shading Language published 3 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published 4 times as a Candidate Recommendation Draft. History of WebGPU
-
Participants
-
1 chair appointed.
-
Jim Blandy became a chair.
-
-
1 chair stepped down.
-
Kelsey Gilbert stepped down as a chair.
-
-
4 participants joined.
-
Zheng Zhai (Ant Group Co., Ltd.) joined.
-
Jim Peng (Ant Group Co., Ltd.) joined.
-
Kelsey Gilbert (W3C Invited Experts) joined.
-
Fabio Bernardon (NVIDIA) joined.
-
-
3 participants left.
-
Kelsey Gilbert (W3C Invited Experts) left.
-
John Owens (Google LLC) left.
-
Shaobo Yan (Intel Corporation) left.
-
-
1 organization joined.
-
Ant Group Co., Ltd. joined.
-
GitHub
gpuweb/gpuweb
-
17 issues opened.
-
When shall we check if feature is enabled? gpuweb/gpuweb#5329, by @beaufortfrancois.
-
[editorial] Fix algorithms wording to handle textures/samplers passed as function args gpuweb/gpuweb#5328, by @kainino0x.
-
GPUDevice.lost missing from WebGPU draft specification gpuweb/gpuweb#5325, by @ayeteadoe.
-
Maintainer access for the WebGPU landing page gpuweb/gpuweb#5324, by @stefnotch.
-
Allow copyExternalImageToTexture for all renderable formats gpuweb/gpuweb#5323, by @greggman.
-
Compat: Rename compat limits to clarify they're compat-only gpuweb/gpuweb#5322, by @kainino0x.
-
[immediates] minimum maxImmediateSize should be 16? gpuweb/gpuweb#5320, by @kainino0x.
-
Should we disallow creating views that only includes `COPY_*` usages? gpuweb/gpuweb#5317, by @beaufortfrancois.
-
Compile time NaN support (Not a Number) gpuweb/gpuweb#5316, by @JD31.
-
Handling dynamic Metal maxThreadsPerThreadgroup gpuweb/gpuweb#5315, by @reeselevine.
-
Allow binding zero-sized GPUBufferBindings gpuweb/gpuweb#5312, by @magcius.
-
Clarify some documentation around configure()/getConfiguration() gpuweb/gpuweb#5307, by @kainino0x.
-
Doc discussing bitfield positions should clarify which position is least significant gpuweb/gpuweb#5303, by @aaeberharter.
-
Texture swizzle: syntax is verbose and doesn't match WGSL gpuweb/gpuweb#5296, by @kainino0x.
-
Add more metadata for HDR canvas output gpuweb/gpuweb#5294, by @kainino0x.
-
-
8 issues closed.
-
GPUDevice.lost missing from WebGPU draft specification gpuweb/gpuweb#5325.
-
Clarify some documentation around configure()/getConfiguration() gpuweb/gpuweb#5307.
-
Texture swizzle: syntax is verbose and doesn't match WGSL gpuweb/gpuweb#5296.
-
Extend WGSL to Support Additional Integer Types gpuweb/gpuweb#5152.
-
Why have literal suffixes at all? gpuweb/gpuweb#2664.
-
wgsl: add support for Prefix Operators i.e. ++i gpuweb/gpuweb#5181.
-
Feature detection for HDR canvas gpuweb/gpuweb#4828.
-
Should we disallow G,B,A swizzle with depth textures? gpuweb/gpuweb#5266.
-
-
41 issues received 137 new comments.
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236 (14 comments by @ccameron-chromium, @greggman, @kainino0x, @Kangz, @mwyrzykowski)
-
Should texture swizzle move usage restrictions from createView to createBindGroup/beginRenderPass? gpuweb/gpuweb#5298 (13 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz, @magcius)
-
Reintroduce source maps gpuweb/gpuweb#4844 (11 comments by @greggman, @jimblandy, @mighdoll, @stefnotch)
-
Texture swizzle: syntax is verbose and doesn't match WGSL gpuweb/gpuweb#5296 (10 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz, @magcius)
-
Investigation: Bindless resources gpuweb/gpuweb#380 (6 comments by @cwfitzgerald, @devshgraphicsprogramming, @mwyrzykowski)
-
Maintainer access for the WebGPU landing page gpuweb/gpuweb#5324 (5 comments by @greggman, @jimblandy, @kainino0x, @mehmetoguzderin, @stefnotch)
-
Should we disallow creating views that only includes `COPY_*` usages? gpuweb/gpuweb#5317 (5 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz)
-
copyExternalImageToTexture should allow copying to the new 16unorm formats gpuweb/gpuweb#5289 (5 comments by @Jiawei-Shao, @kainino0x)
-
Handling of NaNs with Clamp (max/min) gpuweb/gpuweb#5192 (4 comments by @jimblandy, @mwyrzykowski, @petermcneeleychromium)
-
Allow copyExternalImageToTexture for all renderable formats gpuweb/gpuweb#5323 (4 comments by @greggman, @kainino0x)
-
Consider lifting size%4 requirement of copyBufferToBuffer and writeBuffer gpuweb/gpuweb#5203 (4 comments by @kainino0x, @magcius, @mwyrzykowski)
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118 (3 comments by @kainino0x, @Kangz, @shaoboyan091)
-
[editorial] Fix algorithms wording to handle textures/samplers passed as function args gpuweb/gpuweb#5328 (2 comments by @kainino0x)
-
GPUDevice.lost missing from WebGPU draft specification gpuweb/gpuweb#5325 (2 comments by @ayeteadoe, @beaufortfrancois)
-
Compat: Rename compat limits to clarify they're compat-only gpuweb/gpuweb#5322 (2 comments by @kainino0x, @SenorBlanco)
-
[Immediates] Should failing to set an immediate value used by the shader be a validation error? gpuweb/gpuweb#5318 (2 comments by @Kangz, @shaoboyan091)
-
Handling dynamic Metal maxThreadsPerThreadgroup gpuweb/gpuweb#5315 (2 comments by @jrprice, @reeselevine)
-
Doc discussing bitfield positions should clarify which position is least significant gpuweb/gpuweb#5303 (2 comments by @aaeberharter, @alan-baker)
-
Dynamic texture selection is impossible for sample() calls gpuweb/gpuweb#2482 (2 comments by @magcius, @stefnotch)
-
function workgroup-memory args should be more flexible (WGSL proposed language change) gpuweb/gpuweb#5156 (1 comment by @dneto0)
-
Clarify some documentation around configure()/getConfiguration() gpuweb/gpuweb#5307 (1 comment by @kainino0x)
-
"Promise Ordering" section should mention device lost promise ordering gpuweb/gpuweb#5244 (1 comment by @kainino0x)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (1 comment by @kainino0x)
-
WGSL should forbid pointers to runtime-sized values as user-defined function parameters gpuweb/gpuweb#2268 (1 comment by @jrprice)
-
-
20 pull requests opened.
-
[wgsl] Add treesitter rule for reserved words gpuweb/gpuweb#5327, by @dneto0.
-
[infra] convert reserved word script to Python gpuweb/gpuweb#5326, by @dneto0.
-
[infra] Fail spec building if treesitter parser fails to build gpuweb/gpuweb#5321, by @dneto0.
-
Match WGSL syntax and clean open questions gpuweb/gpuweb#5319, by @beaufortfrancois.
-
Atomic 64 proposal gpuweb/gpuweb#5314, by @petermcneeleychromium.
-
Limit texture/sampler combinations gpuweb/gpuweb#5313, by @SenorBlanco.
-
[editorial] Clarify "equivalent texel representation" gpuweb/gpuweb#5311, by @kainino0x.
-
[bugfix] Fix copyExternalImageToTexture steps to handle *-srgb formats gpuweb/gpuweb#5310, by @kainino0x.
-
Improve docs around getConfiguration and feature detection in general gpuweb/gpuweb#5309, by @kainino0x.
-
[editorial] Reword instances of "may not" in API spec gpuweb/gpuweb#5308, by @kainino0x.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5306, by @kainino0x.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5305, by @kainino0x.
-
[editorial] Compat: add rigor to texture call validation gpuweb/gpuweb#5304, by @kainino0x.
-
Regenerate diagrams with latest image gpuweb/gpuweb#5302, by @mehmetoguzderin.
-
Validate textureBindingViewDimension gpuweb/gpuweb#5301, by @SenorBlanco.
-
Add missing word `be` gpuweb/gpuweb#5300, by @dj2.
-
Add additional dstformat in copyExternalImageToTexture with Tier1 gpuweb/gpuweb#5299, by @mei1127.
-
Implement depth texturing restrictions gpuweb/gpuweb#5297, by @SenorBlanco.
-
New per stage limits gpuweb/gpuweb#5295, by @SenorBlanco.
-
Added an explainer for primitive-index feature gpuweb/gpuweb#5293, by @toji.
-
-
18 pull requests merged.
-
[wgsl] Add treesitter rule for reserved words gpuweb/gpuweb#5327.
-
[infra] convert reserved word script to Python gpuweb/gpuweb#5326.
-
[editorial] Reword instances of "may not" in API spec gpuweb/gpuweb#5308.
-
[editorial] Clarify "equivalent texel representation" gpuweb/gpuweb#5311.
-
[infra] Fail spec building if treesitter parser fails to build gpuweb/gpuweb#5321.
-
[bugfix] Fix copyExternalImageToTexture steps to handle *-srgb formats gpuweb/gpuweb#5310.
-
New per stage limits gpuweb/gpuweb#5295.
-
Improve docs around getConfiguration and feature detection in general gpuweb/gpuweb#5309.
-
Match WGSL syntax and clean open questions gpuweb/gpuweb#5319.
-
Validate textureBindingViewDimension gpuweb/gpuweb#5301.
-
[editorial] Compat: add rigor to texture call validation gpuweb/gpuweb#5304.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5306.
-
Regenerate diagrams with latest image gpuweb/gpuweb#5302.
-
Implement depth texturing restrictions gpuweb/gpuweb#5297.
-
Add missing word `be` gpuweb/gpuweb#5300.
-
Added an explainer for primitive-index feature gpuweb/gpuweb#5293.
-
-
1 pull request closed.
-
Merge branch 'main' into featurebranch-compat gpuweb/gpuweb#5305.
-
-
18 pull requests received 112 new comments.
-
Limit texture/sampler combinations gpuweb/gpuweb#5313 (39 comments by @alan-baker, @beaufortfrancois, @greggman, @kainino0x, @SenorBlanco)
-
New per stage limits gpuweb/gpuweb#5295 (26 comments by @greggman, @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Implement depth texturing restrictions gpuweb/gpuweb#5297 (9 comments by @kainino0x, @SenorBlanco, @teoxoy)
-
Add additional dstformat in copyExternalImageToTexture with Tier1 gpuweb/gpuweb#5299 (8 comments by @greggman, @kainino0x)
-
Match WGSL syntax and clean open questions gpuweb/gpuweb#5319 (7 comments by @beaufortfrancois, @kainino0x)
-
[editorial] Reword instances of "may not" in API spec gpuweb/gpuweb#5308 (4 comments by @kainino0x, @toji)
-
[infra] convert reserved word script to Python gpuweb/gpuweb#5326 (3 comments by @dneto0, @mehmetoguzderin)
-
Remove redundant build files since latest generator for grammar can produce them succinctly gpuweb/gpuweb#5283 (2 comments by @dneto0, @mehmetoguzderin)
-
[infra] Fail spec building if treesitter parser fails to build gpuweb/gpuweb#5321 (1 comment by @mehmetoguzderin)
-
Improve docs around getConfiguration and feature detection in general gpuweb/gpuweb#5309 (1 comment by @kainino0x)
-
gpuweb/cts
-
4 issues opened.
-
Several limits related tests are broken gpuweb/cts#4462, by @greggman.
-
gen_wpt_cts_html.ts uses listing_meta.json which might be out of date gpuweb/cts#4461, by @greggman.
-
Don't use random data to test compressed texture copying? gpuweb/cts#4454, by @greggman.
-
CTS can be slow to look at results gpuweb/cts#4451, by @greggman.
-
-
1 issue closed.
-
Several limits related tests are broken gpuweb/cts#4462.
-
-
4 issues received 6 new comments.
-
gen_wpt_cts_html.ts uses listing_meta.json which might be out of date gpuweb/cts#4461 (2 comments by @greggman)
-
-
12 pull requests opened.
-
Show default device in standalone runner gpuweb/cts#4464, by @jrprice.
-
Fix issues wrt to compat only limits gpuweb/cts#4463, by @greggman.
-
Update copyExternalImageToTexture test formats gpuweb/cts#4460, by @greggman.
-
Fix usage of bgra8unorm when not-write only in createBindGroupLayout tests gpuweb/cts#4459, by @dj2.
-
Fix usage of bgra8unorm when non-write-only gpuweb/cts#4458, by @dj2.
-
Save failed results only gpuweb/cts#4457, by @petermcneeleychromium.
-
Add invariant test when not an entry point parameter. gpuweb/cts#4456, by @dj2.
-
Make gen_version fail gracefully if git is not available gpuweb/cts#4455, by @kainino0x.
-
Add tests for Tier2 read-write storage texture access gpuweb/cts#4453, by @mei1127.
-
Support git version from environment gpuweb/cts#4452, by @bsheedy-work.
-
Add checks for renderable formats with texture-format-tier1 enabled gpuweb/cts#4450, by @mei1127.
-
Add parsing configuration for Tier1 multisampled texture formats gpuweb/cts#4449, by @mei1127.
-
-
10 pull requests merged.
-
Show default device in standalone runner gpuweb/cts#4464.
-
Fix issues wrt to compat only limits gpuweb/cts#4463.
-
Update copyExternalImageToTexture test formats gpuweb/cts#4460.
-
Add invariant test when not an entry point parameter. gpuweb/cts#4456.
-
Make gen_version fail gracefully if git is not available gpuweb/cts#4455.
-
Save failed results only gpuweb/cts#4457.
-
Add tests for Tier2 read-write storage texture access gpuweb/cts#4453.
-
Add checks for renderable formats with texture-format-tier1 enabled gpuweb/cts#4450.
-
Add parsing configuration for Tier1 multisampled texture formats gpuweb/cts#4449.
-
-
3 pull requests closed.
-
Fix usage of bgra8unorm when non-write-only gpuweb/cts#4458.
-
Support git version from environment gpuweb/cts#4452.
-
7 pull requests received 45 new comments.
-
Add texture-component-swizzle tests gpuweb/cts#4427 (21 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Update copyExternalImageToTexture test formats gpuweb/cts#4460 (4 comments by @beaufortfrancois, @greggman, @kainino0x)
-
'maxStorageBuffersInFragmentStage' and 'maxStorageBuffersInVertexStag… gpuweb/cts#4448 (4 comments by @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Fix usage of bgra8unorm when not-write only in createBindGroupLayout tests gpuweb/cts#4459 (2 comments by @dj2, @greggman)
-
TR publications
-
8 documents published.
-
WebGPU published 6 times as a Candidate Recommendation Draft. History of WebGPU
-
WebGPU Shading Language published twice as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
Participants
-
2 participants joined.
-
Markus Tavenrath (NVIDIA) joined.
-
Daniel Crookston (Autodesk) joined.
-
GitHub
gpuweb/gpuweb
-
7 issues opened.
-
WebGPU inference is much slower than WASM for onnx-runtime-web gpuweb/gpuweb#5291, by @ShreshthSaxena.
-
copyExternalImageToTexture should allow copying to the new 16unorm formats gpuweb/gpuweb#5289, by @kainino0x.
-
WGSL: texture compare should specify the direction of comparison gpuweb/gpuweb#5285, by @greggman.
-
Lines in 3d extremely painful gpuweb/gpuweb#5282, by @petrasvestartas.
-
Support for YUV textures gpuweb/gpuweb#5281, by @jgh-twitch.
-
Uniformity Analysis - Account for break if expression uniformity gpuweb/gpuweb#5277, by @JamesLee-Jones.
-
-
12 issues closed.
-
WebGPU inference is much slower than WASM for onnx-runtime-web gpuweb/gpuweb#5291.
-
Lines in 3d extremely painful gpuweb/gpuweb#5282.
-
Support for gl_PrimitiveID / SV_PrimitiveID -like construct? gpuweb/gpuweb#1786.
-
Compatibility mode: support devices without integer texture format multisampling gpuweb/gpuweb#5004.
-
Compat: Disallow textureLoad() of depth textures in WGSL via validation. gpuweb/gpuweb#4512.
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489.
-
Tanh absolute precision gpuweb/gpuweb#5206.
-
docs: webgpu not supported by default on windows arm /w chrome/edge gpuweb/gpuweb#5272.
-
Should we raise an error if non-default swizzle is used without feature enabled? gpuweb/gpuweb#5264.
-
-
1 issue reopened.
-
docs: webgpu not supported by default on windows arm /w chrome/edge gpuweb/gpuweb#5272, by @kainino0x.
-
-
23 issues received 103 new comments.
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236 (27 comments by @ccameron-chromium, @greggman, @kainino0x, @mwyrzykowski, @reitowo)
-
Should we disallow G,B,A swizzle with depth textures? gpuweb/gpuweb#5266 (22 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (7 comments by @kainino0x, @yvvki)
-
Support Multi-GPUs gpuweb/gpuweb#995 (6 comments by @austinvhuang, @dead-claudia, @kainino0x, @mehmetoguzderin)
-
Lines in 3d extremely painful gpuweb/gpuweb#5282 (6 comments by @greggman, @kainino0x, @magcius, @petrasvestartas)
-
Only 32 timestamp queries allowed on Metal MacOS? gpuweb/gpuweb#5261 (5 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
Compat: Disallow `textureSampleLevel`/`textureSampleCompareLevel` with `texture_depth_2d_array` gpuweb/gpuweb#4519 (4 comments by @Kangz, @SenorBlanco)
-
Support for YUV textures gpuweb/gpuweb#5281 (4 comments by @jgh-twitch, @kainino0x, @Kangz, @mwyrzykowski)
-
docs: webgpu not supported by default on windows arm /w chrome/edge gpuweb/gpuweb#5272 (4 comments by @birkskyum, @dneto0, @kainino0x)
-
WGSL should forbid pointers to runtime-sized values as user-defined function parameters gpuweb/gpuweb#2268 (2 comments by @microaaron)
-
Linux (Ubuntu 20.04) + NVIDIA GTX 1660 Ti: WebGPU Vulkan adapter doesn’t expose shader-f16 although Vulkan reports shaderFloat16=true gpuweb/gpuweb#5290 (2 comments by @greggman, @jrprice)
-
Double precision floats (IEEE-754 binary64) gpuweb/gpuweb#2805 (2 comments by @BarthPaleologue, @Kangz)
-
Should we raise an error if non-default swizzle is used without feature enabled? gpuweb/gpuweb#5264 (2 comments by @beaufortfrancois, @Kangz)
-
copyExternalImageToTexture should allow copying to the new 16unorm formats gpuweb/gpuweb#5289 (1 comment by @kainino0x)
-
Compatibility mode: support devices without integer texture format multisampling gpuweb/gpuweb#5004 (1 comment by @Kangz)
-
Compat: Disallow textureLoad() of depth textures in WGSL via validation. gpuweb/gpuweb#4512 (1 comment by @Kangz)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (1 comment by @Kangz)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (1 comment by @Kangz)
-
`GPUBindGroupLayoutEntry` is accepting an empty dictionary for `externalTexture`, should accept boolean instead gpuweb/gpuweb#5249 (1 comment by @Kangz)
-
Add texture component mapping as an optional feature `"texture-component-swizzle"` gpuweb/gpuweb#5179 (1 comment by @jiangzhaoming)
-
"Promise Ordering" section should mention device lost promise ordering gpuweb/gpuweb#5244 (1 comment by @jimblandy)
-
-
9 pull requests opened.
-
Fix all "Otherwise" to be in the same step with corresponding "If" gpuweb/gpuweb#5288, by @kainino0x.
-
Disallow invalid interpolation modes gpuweb/gpuweb#5287, by @SenorBlanco.
-
Disallow sample_mask and sample_index gpuweb/gpuweb#5286, by @SenorBlanco.
-
Fix typo in the main compute algorithm gpuweb/gpuweb#5278, by @jiangzhaoming.
-
Fix typo in #render-pass-encoder-finalization: loadOp -> storeOp gpuweb/gpuweb#5276, by @EbbDrop.
-
Fix table visual quirks gpuweb/gpuweb#5275, by @beaufortfrancois.
-
Use new generally available arm64 runners for building architecture image gpuweb/gpuweb#5274, by @mehmetoguzderin.
-
-
10 pull requests merged.
-
Disallow invalid interpolation modes gpuweb/gpuweb#5287.
-
Fix all "Otherwise" to be in the same step with corresponding "If" gpuweb/gpuweb#5288.
-
Implement textureBindingViewDimension gpuweb/gpuweb#5257.
-
Disallow sample_mask and sample_index gpuweb/gpuweb#5286.
-
Add primitive-index feature and associated builtin gpuweb/gpuweb#5273.
-
Fix typo in the main compute algorithm gpuweb/gpuweb#5278.
-
Fix table visual quirks gpuweb/gpuweb#5275.
-
Fix typo in #render-pass-encoder-finalization: loadOp -> storeOp gpuweb/gpuweb#5276.
-
Use new generally available arm64 runners for building architecture image gpuweb/gpuweb#5274.
-
Add validation changes to texture component swizzle proposal gpuweb/gpuweb#5252.
-
-
12 pull requests received 84 new comments.
-
Add primitive-index feature and associated builtin gpuweb/gpuweb#5273 (17 comments by @alan-baker, @kainino0x, @Kangz, @mwyrzykowski, @teoxoy, @toji)
-
Disallow sample_mask and sample_index gpuweb/gpuweb#5286 (7 comments by @greggman, @kainino0x, @SenorBlanco)
-
Fix typo in #render-pass-encoder-finalization: loadOp -> storeOp gpuweb/gpuweb#5276 (6 comments by @EbbDrop, @Kangz, @w3cbot)
-
Remove redundant build files since latest generator for grammar can produce them succinctly gpuweb/gpuweb#5283 (1 comment by @mehmetoguzderin)
-
Make G, B and A components of depth-stencil textures well-defined in texture component swizzle proposal gpuweb/gpuweb#5280 (1 comment by @kainino0x)
-
Use new generally available arm64 runners for building architecture image gpuweb/gpuweb#5274 (1 comment by @mehmetoguzderin)
gpuweb/cts
-
3 issues opened.
-
Test makes assumptions about expiry of current texture gpuweb/cts#4440, by @sagudev.
-
Malware in eslint-plugin-gpuweb-cts gpuweb/cts#4437, by @greggman.
-
2 issues received 4 new comments.
-
10 pull requests opened.
-
'maxStorageBuffersInFragmentStage' and 'maxStorageBuffersInVertexStag… gpuweb/cts#4448, by @mwyrzykowski.
-
Add test for GPUTextureFormats' resolve capability with tier1 gpuweb/cts#4446, by @mei1127.
-
Expectations in createBindGroup.spec.ts are invalid gpuweb/cts#4445, by @mwyrzykowski.
-
Correct vertex correctness tests gpuweb/cts#4444, by @mwyrzykowski.
-
Add tests for Tier1 write-only storage texture access gpuweb/cts#4442, by @mei1127.
-
Correct expected data in read-only related tests gpuweb/cts#4441, by @mei1127.
-
Correct expectation in texture_formats_tier1.spec.ts gpuweb/cts#4439, by @mwyrzykowski.
-
Add tests for Tier1 read-only storage texture access gpuweb/cts#4438, by @mei1127.
-
Test 16-bit UNORM/SNORM format capabilities with TextureFormatsTier1 gpuweb/cts#4436, by @mei1127.
-
-
10 pull requests merged.
-
Add tests for the primitive_index builtin gpuweb/cts#4435.
-
Expectations in createBindGroup.spec.ts are invalid gpuweb/cts#4445.
-
Add test for GPUTextureFormats' resolve capability with tier1 gpuweb/cts#4446.
-
Correct vertex correctness tests gpuweb/cts#4444.
-
Add tests for Tier1 write-only storage texture access gpuweb/cts#4442.
-
Correct expected data in read-only related tests gpuweb/cts#4441.
-
Correct expectation in texture_formats_tier1.spec.ts gpuweb/cts#4439.
-
Add tests for Tier1 read-only storage texture access gpuweb/cts#4438.
-
Test 16-bit UNORM/SNORM format capabilities with TextureFormatsTier1 gpuweb/cts#4436.
-
-
8 pull requests received 52 new comments.
-
Add texture-component-swizzle tests gpuweb/cts#4427 (37 comments by @beaufortfrancois, @greggman, @Kangz)
-
fix: properly apply `bindTextureResource` in `render_pass_descriptor:one_*_attachment` tests gpuweb/cts#4447 (2 comments by @ErichDonGubler)
-
Expectations in createBindGroup.spec.ts are invalid gpuweb/cts#4445 (2 comments by @greggman, @mwyrzykowski)
-
TR publications
-
10 documents published.
-
WebGPU Shading Language published 7 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
GitHub
gpuweb/gpuweb
-
11 issues opened.
-
docs: webgpu not supported by default on windows arm /w chrome/edge gpuweb/gpuweb#5272, by @birkskyum.
-
Non-normative grammar is incorrectly inlining const_assert gpuweb/gpuweb#5271, by @dj2.
-
Should we disallow G,B,A swizzle with depth textures? gpuweb/gpuweb#5266, by @beaufortfrancois.
-
Should we raise an error if non-default swizzle is used without feature enabled? gpuweb/gpuweb#5264, by @beaufortfrancois.
-
Only 32 timestamp queries allowed on Metal MacOS? gpuweb/gpuweb#5261, by @greggman.
-
Inconsistency in spec related to `diagnostic` token gpuweb/gpuweb#5258, by @0x2a-42.
-
[HIC] How to support HTML-in-Canvas in WebGPU gpuweb/gpuweb#5251, by @Kangz.
-
WGSL: Bug in template list discovery algorithm gpuweb/gpuweb#5246, by @k2d222.
-
"Promise Ordering" section should mention device lost promise ordering gpuweb/gpuweb#5244, by @teoxoy.
-
-
5 issues closed.
-
WGSL: Bug in template list discovery algorithm gpuweb/gpuweb#5246.
-
ARM64 DX Issues with Container gpuweb/gpuweb#4412.
-
-
1 issue reopened.
-
26 issues received 97 new comments.
-
mapSync on Workers - and possibly on the main thread gpuweb/gpuweb#2217 (15 comments by @ambroser53, @Dinnerbone, @kainino0x, @Kangz, @Morglod, @mwyrzykowski, @valadaptive)
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236 (11 comments by @ccameron-chromium, @kainino0x, @Kangz, @mwyrzykowski)
-
`GPUBindGroupLayoutEntry` is accepting an empty dictionary for `externalTexture`, should accept boolean instead gpuweb/gpuweb#5249 (10 comments by @greggman, @kainino0x, @Kangz, @mwyrzykowski, @saschanaz)
-
Predictable image formation for HDR color spaces gpuweb/gpuweb#4919 (8 comments by @ccameron-chromium, @donmccurdy, @kainino0x, @stephtr)
-
Only 32 timestamp queries allowed on Metal MacOS? gpuweb/gpuweb#5261 (7 comments by @greggman, @kainino0x, @mwyrzykowski, @teoxoy)
-
Should we disallow G,B,A swizzle with depth textures? gpuweb/gpuweb#5266 (6 comments by @beaufortfrancois, @greggman, @kainino0x, @magcius)
-
Should we raise an error if non-default swizzle is used without feature enabled? gpuweb/gpuweb#5264 (4 comments by @beaufortfrancois, @kainino0x)
-
Support for gl_PrimitiveID / SV_PrimitiveID -like construct? gpuweb/gpuweb#1786 (4 comments by @dj2, @Kangz, @mwyrzykowski)
-
wgsl: add support for Prefix Operators i.e. ++i gpuweb/gpuweb#5181 (4 comments by @dj2, @dneto0, @jimblandy, @kainino0x)
-
Should we raise an error if swizzle is used with `sampleType: "depth"` binding point? gpuweb/gpuweb#5265 (4 comments by @greggman, @kainino0x)
-
Investigation: Generate mipmaps gpuweb/gpuweb#386 (4 comments by @AshleyScirra, @devshgraphicsprogramming, @greggman)
-
WGSL: Bug in template list discovery algorithm gpuweb/gpuweb#5246 (2 comments by @dneto0, @jimblandy)
-
"Promise Ordering" section should mention device lost promise ordering gpuweb/gpuweb#5244 (2 comments by @jimblandy, @Kangz)
-
Feature Request: Let you pass a `GPUTexture` everywhere a `GPUTextureView` is currently used gpuweb/gpuweb#5215 (2 comments by @Kangz, @magcius)
-
docs: webgpu not supported by default on windows arm /w chrome/edge gpuweb/gpuweb#5272 (1 comment by @kainino0x)
-
Consider lifting size%4 requirement of copyBufferToBuffer and writeBuffer gpuweb/gpuweb#5203 (1 comment by @Kangz)
-
Error reporting for unbalanced compute/render passes does not match implementations gpuweb/gpuweb#5207 (1 comment by @Kangz)
-
setBindGroup should accept ArrayBuffer (AllowSharedBufferSource) for dynamicOffsetsData gpuweb/gpuweb#5212 (1 comment by @Kangz)
-
confirmation: bgra8unorm-storage enables reading a bgra8unorm texture as a storage texture gpuweb/gpuweb#5232 (1 comment by @greggman)
-
-
22 pull requests opened.
-
Add primitive-index feature and associated builtin gpuweb/gpuweb#5273, by @toji.
-
Fix template list discovery algorithm: missing continue gpuweb/gpuweb#5270, by @dneto0.
-
wgsl: support swizzle assignment gpuweb/gpuweb#5268, by @dneto0.
-
wgsl: rename "software_extension*" grammar element gpuweb/gpuweb#5263, by @dneto0.
-
Sync compat feature branch for CI gpuweb/gpuweb#5262, by @mehmetoguzderin.
-
Disallow multisampling of integer texture formats gpuweb/gpuweb#5260, by @SenorBlanco.
-
Disallow several texture capabilities in Compat mode. gpuweb/gpuweb#5259, by @SenorBlanco.
-
Implement textureBindingViewDimension gpuweb/gpuweb#5257, by @SenorBlanco.
-
Address message from CI about image related to tree-sitter upgrade gpuweb/gpuweb#5256, by @mehmetoguzderin.
-
wgsl/Makefile: make "clean" should remove grammar.js.pre gpuweb/gpuweb#5255, by @dneto0.
-
Update tree-sitter-cli and tree-sitter to 0.25.0 gpuweb/gpuweb#5254, by @dneto0.
-
Add validation changes to texture component swizzle proposal gpuweb/gpuweb#5252, by @beaufortfrancois.
-
Disallow texture format reinterpretation gpuweb/gpuweb#5250, by @SenorBlanco.
-
Include `externalTexture` in `createBindGroupLayout` validation. gpuweb/gpuweb#5248, by @jimblandy.
-
Disallow copying of multisampled textures. gpuweb/gpuweb#5247, by @SenorBlanco.
-
WGSL: Fix shader value of rg11b10ufloat texel format gpuweb/gpuweb#5243, by @jiangzhaoming.
-
Add Compat mode limits. gpuweb/gpuweb#5242, by @SenorBlanco.
-
Compat: depthBiasClamp must be zero. gpuweb/gpuweb#5241, by @SenorBlanco.
-
Regenerate diagrams through GitHub-hosted Codespaces gpuweb/gpuweb#5240, by @mehmetoguzderin.
-
Fix CI with maintenance gpuweb/gpuweb#5239, by @mehmetoguzderin.
-
CI maintenance to avoid CVEs gpuweb/gpuweb#5238, by @mehmetoguzderin.
-
Support Resolve for 16unorm and 16snorm in texture-formats-tier1 gpuweb/gpuweb#5237, by @Jiawei-Shao.
-
-
23 pull requests merged.
-
Widen tanh acceptable range (output values) gpuweb/gpuweb#5199.
-
Fix template list discovery algorithm: missing continue gpuweb/gpuweb#5270.
-
Clarify "wrapping behaviour" in `dot4I8Packed` gpuweb/gpuweb#5106.
-
Disallow several texture capabilities in Compat mode. gpuweb/gpuweb#5259.
-
wgsl: rename "software_extension*" grammar element gpuweb/gpuweb#5263.
-
Disallow multisampling of integer texture formats gpuweb/gpuweb#5260.
-
Sync compat feature branch for CI gpuweb/gpuweb#5262.
-
Address message from CI about image related to tree-sitter upgrade gpuweb/gpuweb#5256.
-
Disallow texture format reinterpretation gpuweb/gpuweb#5250.
-
Update tree-sitter-cli and tree-sitter to 0.25.0 gpuweb/gpuweb#5254.
-
wgsl/Makefile: make "clean" should remove grammar.js.pre gpuweb/gpuweb#5255.
-
Subgroups: clarify subgroups operations using f16 gpuweb/gpuweb#5009.
-
Add Compat mode limits. gpuweb/gpuweb#5242.
-
WGSL: Fix shader value of rg11b10ufloat texel format gpuweb/gpuweb#5243.
-
Disallow copying of multisampled textures. gpuweb/gpuweb#5247.
-
Compat: depthBiasClamp must be zero. gpuweb/gpuweb#5241.
-
Include `externalTexture` in `createBindGroupLayout` validation. gpuweb/gpuweb#5248.
-
Allow GPUTexture where GPUTextureView is used gpuweb/gpuweb#5228.
-
Compat: TextureViews of same Texture must match mips, aspect gpuweb/gpuweb#5224.
-
Compat: no two-component formats as storage gpuweb/gpuweb#5235.
-
Regenerate diagrams through GitHub-hosted Codespaces gpuweb/gpuweb#5240.
-
Fix CI with maintenance gpuweb/gpuweb#5239.
-
CI maintenance to avoid CVEs gpuweb/gpuweb#5238.
-
-
4 pull requests closed.
-
WGSL FAQ gpuweb/gpuweb#687.
-
Support Resolve for 16unorm and 16snorm in texture-formats-tier1 gpuweb/gpuweb#5237.
-
Add and name Features for all Compat Mode limitations. gpuweb/gpuweb#5018.
-
Drop ASTC requirement in compatibility mode. gpuweb/gpuweb#4566.
-
-
23 pull requests received 115 new comments.
-
Allow GPUTexture where GPUTextureView is used gpuweb/gpuweb#5228 (20 comments by @beaufortfrancois, @kainino0x, @rezadzed, @toji)
-
Add texture component swizzle proposal gpuweb/gpuweb#5217 (14 comments by @beaufortfrancois, @greggman, @Kangz)
-
Support Resolve for 16unorm and 16snorm in texture-formats-tier1 gpuweb/gpuweb#5237 (10 comments by @Jiawei-Shao, @kainino0x, @Kangz, @mwyrzykowski, @teoxoy)
-
Disallow texture format reinterpretation gpuweb/gpuweb#5250 (9 comments by @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Add validation changes to texture component swizzle proposal gpuweb/gpuweb#5252 (4 comments by @beaufortfrancois, @Kangz)
-
Address message from CI about image related to tree-sitter upgrade gpuweb/gpuweb#5256 (4 comments by @dneto0, @mehmetoguzderin)
-
Disallow several texture capabilities in Compat mode. gpuweb/gpuweb#5259 (3 comments by @kainino0x, @SenorBlanco, @teoxoy)
-
Subgroups: clarify subgroups operations using f16 gpuweb/gpuweb#5009 (3 comments by @jiangzhaoming, @mehmetoguzderin)
-
WGSL: Fix shader value of rg11b10ufloat texel format gpuweb/gpuweb#5243 (1 comment by @jiangzhaoming)
-
Compat: TextureViews of same Texture must match mips, aspect gpuweb/gpuweb#5224 (1 comment by @SenorBlanco)
gpuweb/cts
-
5 issues opened.
-
Uniformity tests generate invalid array index expressions gpuweb/cts#4430, by @JamesLee-Jones.
-
Subcase-skip log spam from some tests gpuweb/cts#4424, by @kainino0x.
-
Texture component swizzle 🥤 gpuweb/cts#4421, by @beaufortfrancois.
-
2 issues closed.
-
Uniformity tests generate invalid array index expressions gpuweb/cts#4430.
-
test that 'bgra8unorm' can not be used as read-only storage texture gpuweb/cts#4412.
-
-
5 issues received 16 new comments.
-
Test the view used is the default view if GPUTexture is used instead of GPUTextureView gpuweb/cts#4426 (2 comments by @beaufortfrancois, @greggman)
-
Add coverage asserting that `GPUBindingCommandsMixin.setBindGroup`'s `dynamicOffsetData` param. is `[AllowShared]` gpuweb/cts#4434 (1 comment by @kainino0x)
-
14 pull requests opened.
-
Add tests for the primitive_index builtin gpuweb/cts#4435, by @toji.
-
Additional coverage in copy_within_same_texture test gpuweb/cts#4433, by @andyleiserson.
-
Fix type of index in uniformity tests gpuweb/cts#4432, by @jrprice.
-
Allow skipIfTextureFormatNotSupported in beforeAllSubcases gpuweb/cts#4431, by @kainino0x.
-
Pass texture instead of texture view in operation tests gpuweb/cts#4429, by @beaufortfrancois.
-
Fix "resoruces" typo in resources.ts gpuweb/cts#4428, by @nt1m.
-
Add texture-component-swizzle tests gpuweb/cts#4427, by @greggman.
-
Remove resolve from tier1 formats gpuweb/cts#4425, by @greggman.
-
Tweak buffer_texture_copies:sample_count gpuweb/cts#4423, by @andyleiserson.
-
Fix subgroup shuffle (xor) access gpuweb/cts#4422, by @petermcneeleychromium.
-
Add copyBufferToTexture and copyTextureToBuffer validation tests gpuweb/cts#4420, by @greggman.
-
Refactor isTextureFormatUsableAsStorageFormatInCreateShaderModule gpuweb/cts#4419, by @greggman.
-
update shader validation types test for tier1 formats gpuweb/cts#4418, by @dneto0.
-
Fix api,operation,adapter,requestDevice:invalid test gpuweb/cts#4417, by @beaufortfrancois.
-
-
12 pull requests merged.
-
Expand acceptable range of tanh gpuweb/cts#4392.
-
Additional test cases for encoder states gpuweb/cts#4401.
-
Fix type of index in uniformity tests gpuweb/cts#4432.
-
Fix "resoruces" typo in resources.ts gpuweb/cts#4428.
-
Remove resolve from tier1 formats gpuweb/cts#4425.
-
Tweak buffer_texture_copies:sample_count gpuweb/cts#4423.
-
Fix subgroup shuffle (xor) access gpuweb/cts#4422.
-
Refactor isTextureFormatUsableAsStorageFormatInCreateShaderModule gpuweb/cts#4419.
-
Add copyBufferToTexture and copyTextureToBuffer validation tests gpuweb/cts#4420.
-
Fix api,operation,adapter,requestDevice:invalid test gpuweb/cts#4417.
-
Allow GPUTexture where GPUTextureView is used gpuweb/cts#4407.
-
update shader validation types test for tier1 formats gpuweb/cts#4418.
-
-
2 pull requests closed.
-
Make the CTS debuggable again gpuweb/cts#3142.
-
Update api tests for 3D textures with BC compression gpuweb/cts#3799.
-
-
10 pull requests received 33 new comments.
-
Add texture-component-swizzle tests gpuweb/cts#4427 (11 comments by @beaufortfrancois, @greggman, @Kangz)
-
Pass texture instead of texture view in operation tests gpuweb/cts#4429 (5 comments by @beaufortfrancois, @greggman, @Kangz)
-
Allow skipIfTextureFormatNotSupported in beforeAllSubcases gpuweb/cts#4431 (3 comments by @greggman, @kainino0x)
-
Allow GPUTexture where GPUTextureView is used gpuweb/cts#4407 (3 comments by @beaufortfrancois, @greggman)
-
Update api tests for 3D textures with BC compression gpuweb/cts#3799 (2 comments by @greggman, @mehmetoguzderin)
-
TR publications
-
7 documents published.
-
WebGPU Shading Language published 4 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
Participants
-
1 participant left.
-
Xiaojian Wang (Alibaba Group) left.
-
GitHub
gpuweb/gpuweb
-
8 issues opened.
-
copyExternalImageToTexture/importExternalTexture: Add HDR headroom parameter gpuweb/gpuweb#5236, by @ccameron-chromium.
-
confirmation: bgra8unorm-storage enables reading a bgra8unorm texture as a storage texture gpuweb/gpuweb#5232, by @greggman.
-
w3.org has different css than gpuweb.github.io/gpuweb gpuweb/gpuweb#5231, by @greggman.
-
Profiling copyTextureToTexture() etc. gpuweb/gpuweb#5225, by @munrocket.
-
Subgroup size not clearly specified to be a power of two gpuweb/gpuweb#5218, by @petermcneeleychromium.
-
Add normative references for ASTC and BC compressed formats gpuweb/gpuweb#5214, by @beaufortfrancois.
-
setBindGroup should accept ArrayBuffer (AllowSharedBufferSource) for dynamicOffsetsData gpuweb/gpuweb#5212, by @ottoville.
-
-
7 issues closed.
-
No way to sample a `texture_1d` from a vertex shader gpuweb/gpuweb#5001.
-
Proposal: read-write storage textures gpuweb/gpuweb#3838.
-
Add normative references for ASTC and BC compressed formats gpuweb/gpuweb#5214.
-
Subgroup size not clearly specified to be a power of two gpuweb/gpuweb#5218.
-
Auto-enable rg11b10ufloat-renderable with texture-formats-tier1 gpuweb/gpuweb#5200.
-
Determine whether we can drop MTLGPUFamilyMac1 support gpuweb/gpuweb#5187.
-
26 issues received 73 new comments.
-
Feature Request: Let you pass a `GPUTexture` everywhere a `GPUTextureView` is currently used gpuweb/gpuweb#5215 (11 comments by @greggman, @JMS55, @kainino0x, @Kangz, @magcius, @mwyrzykowski)
-
setBindGroup should accept ArrayBuffer (AllowSharedBufferSource) for dynamicOffsetsData gpuweb/gpuweb#5212 (8 comments by @greggman, @kainino0x, @ottoville)
-
Error reporting for unbalanced compute/render passes does not match implementations gpuweb/gpuweb#5207 (5 comments by @andyleiserson, @kainino0x, @Kangz, @teoxoy)
-
Investigation: Programmable Blending gpuweb/gpuweb#442 (5 comments by @autumnontape, @devshgraphicsprogramming)
-
confirmation: bgra8unorm-storage enables reading a bgra8unorm texture as a storage texture gpuweb/gpuweb#5232 (3 comments by @greggman, @kainino0x)
-
Proposal: read-write storage textures gpuweb/gpuweb#3838 (3 comments by @Jiawei-Shao, @kainino0x, @krauthaufen)
-
`GPUExternalTexture.[[expired]]` is described as immutable but is definitely not gpuweb/gpuweb#5196 (2 comments by @kainino0x, @Kangz)
-
w3.org has different css than gpuweb.github.io/gpuweb gpuweb/gpuweb#5231 (2 comments by @greggman, @kainino0x)
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043 (2 comments by @mwyrzykowski, @petermcneeleychromium)
-
Add normative references for ASTC and BC compressed formats gpuweb/gpuweb#5214 (2 comments by @beaufortfrancois, @kainino0x)
-
Extend WGSL to Support Additional Integer Types gpuweb/gpuweb#5152 (2 comments by @mehmetoguzderin, @Zosoled)
-
copyExternalImageToTexture/importExternalTexture: Need a way to opt-out of color space conversion gpuweb/gpuweb#4826 (1 comment by @ccameron-chromium)
-
Consider lifting size%4 requirement of copyBufferToBuffer and writeBuffer gpuweb/gpuweb#5203 (1 comment by @Kangz)
-
Auto-enable rg11b10ufloat-renderable with texture-formats-tier1 gpuweb/gpuweb#5200 (1 comment by @Kangz)
-
-
17 pull requests opened.
-
Compat: no two-component formats as storage gpuweb/gpuweb#5235, by @SenorBlanco.
-
Merge from 'main' at 20b5f6 gpuweb/gpuweb#5234, by @SenorBlanco.
-
Compat: texture view must contain all array layers gpuweb/gpuweb#5233, by @SenorBlanco.
-
add texture_1d support to textureSampleLevel gpuweb/gpuweb#5230, by @dneto0.
-
add texture_1d support to textureSampleLevel gpuweb/gpuweb#5229, by @dneto0.
-
Allow GPUTexture where GPUTextureView is used gpuweb/gpuweb#5228, by @beaufortfrancois.
-
Remove GPUComponentSwizzle identity from texture component swizzle proposal gpuweb/gpuweb#5227, by @beaufortfrancois.
-
Add optional feature `texture-formats-tier2` gpuweb/gpuweb#5226, by @Jiawei-Shao.
-
Compat: TextureViews of same Texture must match mips, aspect gpuweb/gpuweb#5224, by @SenorBlanco.
-
Compat: add limitation on cube-array. gpuweb/gpuweb#5223, by @SenorBlanco.
-
Add OpenGL and Compat to Texture Component Swizzle proposal gpuweb/gpuweb#5222, by @beaufortfrancois.
-
Add normative references for ASTC, BC, and ETC2 compressed formats gpuweb/gpuweb#5221, by @beaufortfrancois.
-
subgroups: subgroup size is uniform within a dispatch command gpuweb/gpuweb#5220, by @dneto0.
-
subgroups: Clarify subgroup size is a power of two gpuweb/gpuweb#5219, by @dneto0.
-
Add texture component swizzle proposal gpuweb/gpuweb#5217, by @beaufortfrancois.
-
Update install-dependencies.sh location gpuweb/gpuweb#5216, by @emmanuel-ferdman.
-
Auto-enable `rg11b10ufloat-renderable` with `texture-formats-tier1` gpuweb/gpuweb#5213, by @Jiawei-Shao.
-
-
16 pull requests merged.
-
Compat: texture view must contain all array layers gpuweb/gpuweb#5233.
-
Merge from 'main' at 20b5f6 gpuweb/gpuweb#5234.
-
add texture_1d support to textureSampleLevel gpuweb/gpuweb#5230.
-
Add optional feature `texture-formats-tier2` gpuweb/gpuweb#5226.
-
Remove GPUComponentSwizzle identity from texture component swizzle proposal gpuweb/gpuweb#5227.
-
Compat: add limitation on cube-array. gpuweb/gpuweb#5223.
-
Add normative references for ASTC, BC, and ETC2 compressed formats gpuweb/gpuweb#5221.
-
Add OpenGL and Compat to Texture Component Swizzle proposal gpuweb/gpuweb#5222.
-
Add texture component swizzle proposal gpuweb/gpuweb#5217.
-
subgroups: subgroup size is uniform within a dispatch command gpuweb/gpuweb#5220.
-
subgroups: Clarify subgroup size is a power of two gpuweb/gpuweb#5219.
-
Update install-dependencies.sh location gpuweb/gpuweb#5216.
-
Compat: Disallow copying of compressed textures gpuweb/gpuweb#5211.
-
Define featureLevels; introduce Compat Mode CSS; give an example gpuweb/gpuweb#5198.
-
Auto-enable `rg11b10ufloat-renderable` with `texture-formats-tier1` gpuweb/gpuweb#5213.
-
Minor fix note nan clamp min max gpuweb/gpuweb#5197.
-
-
1 pull request closed.
-
add texture_1d support to textureSampleLevel gpuweb/gpuweb#5229.
-
-
13 pull requests received 55 new comments.
-
Compat: TextureViews of same Texture must match mips, aspect gpuweb/gpuweb#5224 (12 comments by @kainino0x, @SenorBlanco, @teoxoy)
-
Allow GPUTexture where GPUTextureView is used gpuweb/gpuweb#5228 (10 comments by @beaufortfrancois, @kainino0x, @toji)
-
Compat: no two-component formats as storage gpuweb/gpuweb#5235 (8 comments by @kainino0x, @SenorBlanco)
-
Add texture component swizzle proposal gpuweb/gpuweb#5217 (6 comments by @beaufortfrancois, @greggman, @mwyrzykowski, @toji)
-
Add normative references for ASTC, BC, and ETC2 compressed formats gpuweb/gpuweb#5221 (6 comments by @beaufortfrancois, @toji)
-
Compat: Disallow copying of compressed textures gpuweb/gpuweb#5211 (3 comments by @kainino0x, @SenorBlanco)
-
Define featureLevels; introduce Compat Mode CSS; give an example gpuweb/gpuweb#5198 (3 comments by @SenorBlanco, @teoxoy)
-
Auto-enable `rg11b10ufloat-renderable` with `texture-formats-tier1` gpuweb/gpuweb#5213 (2 comments by @kainino0x)
-
subgroups: subgroup size is uniform within a dispatch command gpuweb/gpuweb#5220 (1 comment by @alan-baker)
-
gpuweb/cts
-
2 issues opened.
-
test that 'bgra8unorm' can not be used as read-only storage texture gpuweb/cts#4412, by @greggman.
-
-
5 issues closed.
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761.
-
Test `texture-compression-astc-sliced-3d` gpuweb/cts#3967.
-
5 issues received 7 new comments.
-
image_copy:mip_levels tests fail for bc compressed formats with AMD Radeon RX 5500 XT on macOS gpuweb/cts#4378 (3 comments by @beaufortfrancois, @greggman)
-
textureSampleLevel fails for bc7-rgba-unorm-srgb on macOS for AMD Radeon Pro 560X gpuweb/cts#4377 (1 comment by @beaufortfrancois)
-
-
16 pull requests opened.
-
tier1/2 add pipeline auto-layout tests. gpuweb/cts#4416, by @greggman.
-
Test layout auto/storage texture formats/access modes gpuweb/cts#4415, by @greggman.
-
Add support for texture-formats-tier2 gpuweb/cts#4414, by @greggman.
-
bump @webgpu/types to v0.1.63 gpuweb/cts#4413, by @greggman.
-
Refactor that not all storage texture support read-only gpuweb/cts#4411, by @greggman.
-
Add some texture builtin debug info. gpuweb/cts#4409, by @greggman.
-
Fix textureStore test gpuweb/cts#4408, by @greggman.
-
Allow GPUTexture where GPUTextureView is used gpuweb/cts#4407, by @beaufortfrancois.
-
Add texture_1d support to textureSampleLevel gpuweb/cts#4406, by @beaufortfrancois.
-
Update CTS to implement new f32 to int conversion saturation rules gpuweb/cts#4405, by @petermcneeleychromium.
-
Refactor depth_clamp_and_clip test gpuweb/cts#4404, by @greggman.
-
Add feature test for texture-formats-tier1 gpuweb/cts#4403, by @greggman.
-
Remove writeTimestamp tests gpuweb/cts#4402, by @greggman.
-
Additional test cases for encoder states gpuweb/cts#4401, by @andyleiserson.
-
Roll @webgpu/types gpuweb/cts#4400, by @greggman.
-
-
15 pull requests merged.
-
tier1/2 add pipeline auto-layout tests. gpuweb/cts#4416.
-
Test layout auto/storage texture formats/access modes gpuweb/cts#4415.
-
Add support for texture-formats-tier2 gpuweb/cts#4414.
-
bump @webgpu/types to v0.1.63 gpuweb/cts#4413.
-
Refactor that not all storage texture support read-only gpuweb/cts#4411.
-
Add some texture builtin debug info. gpuweb/cts#4409.
-
Fix textureStore test gpuweb/cts#4408.
-
Add texture_1d support to textureSampleLevel gpuweb/cts#4406.
-
Update CTS to implement new f32 to int conversion saturation rules gpuweb/cts#4405.
-
Refactor depth_clamp_and_clip test gpuweb/cts#4404.
-
Add feature test for texture-formats-tier1 gpuweb/cts#4403.
-
Remove writeTimestamp tests gpuweb/cts#4402.
-
Test GPUBuffer in GPUBindingResource gpuweb/cts#4390.
-
Roll @webgpu/types gpuweb/cts#4400.
-
-
8 pull requests received 38 new comments.
-
Add texture_1d support to textureSampleLevel gpuweb/cts#4406 (10 comments by @beaufortfrancois, @greggman, @jrprice)
-
Test GPUBuffer in GPUBindingResource gpuweb/cts#4390 (7 comments by @beaufortfrancois, @greggman, @Kangz)
-
Allow GPUTexture where GPUTextureView is used gpuweb/cts#4407 (4 comments by @beaufortfrancois, @greggman)
-
TR publications
-
7 documents published.
-
WebGPU Shading Language published 4 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
Participants
-
2 participants joined.
-
Peter Tanski (Capital One Discover Network) joined.
-
John Owens (Google LLC) joined.
-
-
1 organization joined.
-
Capital One Discover Network joined.
-
GitHub
gpuweb/gpuweb
-
12 issues opened.
-
Unrestricted Depth Range gpuweb/gpuweb#5209, by @kainino0x.
-
Error reporting for unbalanced compute/render passes does not match implementations gpuweb/gpuweb#5207, by @andyleiserson.
-
Tanh absolute precision gpuweb/gpuweb#5206, by @petermcneeleychromium.
-
Improve or remove `override`. gpuweb/gpuweb#5204, by @munrocket.
-
Consider lifting size%4 requirement of copyBufferToBuffer and writeBuffer gpuweb/gpuweb#5203, by @kainino0x.
-
Should setBindGroup allow for dynamic offsets from a SharedArrayBuffer? gpuweb/gpuweb#5201, by @mwyrzykowski.
-
Auto-enable rg11b10ufloat-renderable with texture-formats-tier1 gpuweb/gpuweb#5200, by @kainino0x.
-
`GPUExternalTexture.[[expired]]` is described as immutable but is definitely not gpuweb/gpuweb#5196, by @jimblandy.
-
Handling of NaNs with Clamp (max/min) gpuweb/gpuweb#5192, by @petermcneeleychromium.
-
API spec build error about `dynamic-range` gpuweb/gpuweb#5189, by @mehmetoguzderin.
-
Determine whether we can drop MTLGPUFamilyMac1 support gpuweb/gpuweb#5187, by @kainino0x.
-
Mutable bind groups gpuweb/gpuweb#5186, by @kainino0x.
-
-
5 issues closed.
-
Proposal: `formats-tier-1` extension gpuweb/gpuweb#3837.
-
Improve or remove `override`. gpuweb/gpuweb#5204.
-
Should setBindGroup allow for dynamic offsets from a SharedArrayBuffer? gpuweb/gpuweb#5201.
-
Feature request: Support GPUBuffer in GPUBindingResource gpuweb/gpuweb#5095.
-
API spec build error about `dynamic-range` gpuweb/gpuweb#5189.
-
-
32 issues received 72 new comments.
-
Allow render bundles to inherit state gpuweb/gpuweb#5171 (6 comments by @kainino0x, @Kangz, @mwyrzykowski, @toji)
-
Consider lifting size%4 requirement of copyBufferToBuffer and writeBuffer gpuweb/gpuweb#5203 (5 comments by @kainino0x, @mwyrzykowski, @pixelflinger)
-
Should setBindGroup allow for dynamic offsets from a SharedArrayBuffer? gpuweb/gpuweb#5201 (5 comments by @austinEng, @kainino0x, @mwyrzykowski)
-
Auto-enable rg11b10ufloat-renderable with texture-formats-tier1 gpuweb/gpuweb#5200 (3 comments by @Jiawei-Shao, @kainino0x, @mwyrzykowski)
-
Feature request: Support GPUBuffer in GPUBindingResource gpuweb/gpuweb#5095 (3 comments by @beaufortfrancois, @Kangz, @mwyrzykowski)
-
wgsl: add support for Prefix Operators i.e. ++i gpuweb/gpuweb#5181 (3 comments by @dj2, @Kangz, @mwyrzykowski)
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118 (2 comments by @Kangz)
-
Add texture component mapping as an optional feature `"texture-component-swizzle"` gpuweb/gpuweb#5179 (2 comments by @kainino0x, @Kangz)
-
allow declaring workgroup memory within function scope (WGSL proposed language change) gpuweb/gpuweb#5155 (2 comments by @jimblandy, @mwyrzykowski)
-
API spec build error about `dynamic-range` gpuweb/gpuweb#5189 (2 comments by @mehmetoguzderin, @tidoust)
-
Determine whether we can drop MTLGPUFamilyMac1 support gpuweb/gpuweb#5187 (1 comment by @mwyrzykowski)
-
Why do block-compressed textures have to be aligned to a multiple of the texel block size? gpuweb/gpuweb#2006 (1 comment by @magcius)
-
`GPUExternalTexture.[[expired]]` is described as immutable but is definitely not gpuweb/gpuweb#5196 (1 comment by @jimblandy)
-
Why GPUCommandBuffer and GPUCommandEncoder cannot be reused? gpuweb/gpuweb#4138 (1 comment by @Kangz)
-
builtin variables should be automatically available inside kernels (WGSL proposed language change) gpuweb/gpuweb#5153 (1 comment by @jimblandy)
-
13 pull requests opened.
-
Compat: Disallow copying of compressed textures gpuweb/gpuweb#5211, by @SenorBlanco.
-
Define feature levels; add CSS for compat FL; use CSS in one example gpuweb/gpuweb#5208, by @SenorBlanco.
-
Add [AllowShared] to setBindGroup gpuweb/gpuweb#5202, by @kainino0x.
-
Widen tanh acceptable range (output values) gpuweb/gpuweb#5199, by @petermcneeleychromium.
-
Define featureLevels; introduce Compat Mode CSS; give an example gpuweb/gpuweb#5198, by @SenorBlanco.
-
Minor fix note nan clamp min max gpuweb/gpuweb#5197, by @petermcneeleychromium.
-
wgsl: Fix the corresponding shader value of `rg32uint` and `rg32sint` gpuweb/gpuweb#5195, by @Jiawei-Shao.
-
In "Limits", remove duplicated note. gpuweb/gpuweb#5194, by @jimblandy.
-
Add GPUBuffer in GPUBindingResource gpuweb/gpuweb#5193, by @beaufortfrancois.
-
Resolve less/greater tests gpuweb/gpuweb#5191, by @emmanuel-ferdman.
-
Add definitions for valid featureLevels. gpuweb/gpuweb#5190, by @SenorBlanco.
-
Fix typo in WGSL spec, sec 13.3.4. gpuweb/gpuweb#5188, by @eliemichel.
-
Add inherited Bind Groups to Render Bundles. gpuweb/gpuweb#5185, by @toji.
-
-
9 pull requests merged.
-
Resolve less/greater tests gpuweb/gpuweb#5191.
-
Add [AllowShared] to setBindGroup gpuweb/gpuweb#5202.
-
Add GPUBuffer in GPUBindingResource gpuweb/gpuweb#5193.
-
Add optional feature `texture-formats-tier1` gpuweb/gpuweb#5160.
-
wgsl: Fix the corresponding shader value of `rg32uint` and `rg32sint` gpuweb/gpuweb#5195.
-
In "Limits", remove duplicated note. gpuweb/gpuweb#5194.
-
Fix typo in WGSL spec, sec 13.3.4. gpuweb/gpuweb#5188.
-
[binding array proposal] Choose a clearer name for binding array size, in BGL gpuweb/gpuweb#5174.
-
[immediates proposal] Rename to immediateSize and maxImmediateSize gpuweb/gpuweb#5173.
-
-
3 pull requests closed.
-
Add definitions for valid featureLevels. gpuweb/gpuweb#5190.
-
Define feature levels; add CSS for compat FL; use CSS in one example gpuweb/gpuweb#5208.
-
Add 16-bit normalized texture format feature gpuweb/gpuweb#4938.
-
-
14 pull requests received 72 new comments.
-
Add optional feature `texture-formats-tier1` gpuweb/gpuweb#5160 (21 comments by @Jiawei-Shao, @kainino0x, @Kangz, @teoxoy)
-
Define featureLevels; introduce Compat Mode CSS; give an example gpuweb/gpuweb#5198 (16 comments by @kainino0x, @SenorBlanco)
-
Add inherited Bind Groups to Render Bundles. gpuweb/gpuweb#5185 (11 comments by @jimblandy, @Kangz, @toji)
-
Add GPUBuffer in GPUBindingResource gpuweb/gpuweb#5193 (5 comments by @beaufortfrancois, @kainino0x)
-
Add definitions for valid featureLevels. gpuweb/gpuweb#5190 (4 comments by @kainino0x, @SenorBlanco)
-
Resolve less/greater tests gpuweb/gpuweb#5191 (4 comments by @emmanuel-ferdman, @mehmetoguzderin, @w3cbot)
-
Widen tanh acceptable range (output values) gpuweb/gpuweb#5199 (3 comments by @petermcneeleychromium)
-
Define feature levels; add CSS for compat FL; use CSS in one example gpuweb/gpuweb#5208 (2 comments by @SenorBlanco)
-
[binding array proposal] Choose a clearer name for binding array size, in BGL gpuweb/gpuweb#5174 (1 comment by @Kangz)
-
[immediates proposal] Rename to immediateSize and maxImmediateSize gpuweb/gpuweb#5173 (1 comment by @Kangz)
-
Compat: Add clarification on BC guarantees for sliced 3D feature gpuweb/gpuweb#5180 (1 comment by @Kangz)
-
gpuweb/cts
-
6 issues opened.
-
need more ImageBitmap copyExternalImageToTexture tests? gpuweb/cts#4391, by @greggman.
-
required_bytes_in_copy tests fail for ASTC formats with 3D dimension gpuweb/cts#4381, by @beaufortfrancois.
-
texture_size,3d_texture,compressed_format tests fail for ASTC formats gpuweb/cts#4380, by @beaufortfrancois.
-
6 issues closed.
-
Deno 2.x can't run the test suite gpuweb/cts#4123.
-
need more ImageBitmap copyExternalImageToTexture tests? gpuweb/cts#4391.
-
required_bytes_in_copy tests fail for ASTC formats with 3D dimension gpuweb/cts#4381.
-
-
10 issues received 38 new comments.
-
image_copy:mip_levels tests fail for bc compressed formats with AMD Radeon RX 5500 XT on macOS gpuweb/cts#4378 (12 comments by @beaufortfrancois, @greggman, @kainino0x, @mwyrzykowski)
-
need more ImageBitmap copyExternalImageToTexture tests? gpuweb/cts#4391 (10 comments by @greggman, @kainino0x)
-
Show some progress when running the CTS gpuweb/cts#2099 (4 comments by @greggman, @kainino0x, @vorg)
-
textureSampleLevel fails for bc7-rgba-unorm-srgb on macOS for AMD Radeon Pro 560X gpuweb/cts#4377 (4 comments by @beaufortfrancois, @mwyrzykowski)
-
JavaScript tests that test method overriding need to be limited to one concurrent test per device gpuweb/cts#4394 (2 comments by @greggman)
-
rowsPerImage_and_bytesPerRow tests fail for ASTC formats with 3D dimension gpuweb/cts#4382 (1 comment by @beaufortfrancois)
-
required_bytes_in_copy tests fail for ASTC formats with 3D dimension gpuweb/cts#4381 (1 comment by @beaufortfrancois)
-
texture_size,3d_texture,compressed_format tests fail for ASTC formats gpuweb/cts#4380 (1 comment by @beaufortfrancois)
-
-
14 pull requests opened.
-
In uncapturederror tests, don't log errors. gpuweb/cts#4398, by @greggman.
-
Test onuncapturederror order vs addEventListener gpuweb/cts#4397, by @greggman.
-
Fix out-of-order subcase logs gpuweb/cts#4396, by @kainino0x.
-
Follow common JavaScript style gpuweb/cts#4395, by @greggman.
-
Test addEventListener works with EventListener & EventHandler gpuweb/cts#4393, by @greggman.
-
Expand acceptable range of tanh gpuweb/cts#4392, by @petermcneeleychromium.
-
Test GPUBuffer in GPUBindingResource gpuweb/cts#4390, by @beaufortfrancois.
-
Add some tests related to GPUDevice extending EventTarget gpuweb/cts#4389, by @greggman.
-
Add some JavaScript inheritance tests. gpuweb/cts#4388, by @greggman.
-
Fix number_of_bind_group_layouts_exceeds_the_maximum_value test gpuweb/cts#4387, by @greggman.
-
Fix for running under deno gpuweb/cts#4385, by @andyleiserson.
-
Pipeline error for out of bounds array with dynamic and offset gpuweb/cts#4384, by @petermcneeleychromium.
-
Fix texture tests gpuweb/cts#4383, by @greggman.
-
If BC is supported, BC Sliced 3D must be supported gpuweb/cts#4379, by @beaufortfrancois.
-
-
13 pull requests merged.
-
In uncapturederror tests, don't log errors. gpuweb/cts#4398.
-
Test onuncapturederror order vs addEventListener gpuweb/cts#4397.
-
Test addEventListener works with EventListener & EventHandler gpuweb/cts#4393.
-
Fix out-of-order subcase logs gpuweb/cts#4396.
-
Follow common JavaScript style gpuweb/cts#4395.
-
Add some tests related to GPUDevice extending EventTarget gpuweb/cts#4389.
-
Add some JavaScript inheritance tests. gpuweb/cts#4388.
-
Fix number_of_bind_group_layouts_exceeds_the_maximum_value test gpuweb/cts#4387.
-
typo: add end quote to `texture format <X> requires feature: '<F>` gpuweb/cts#4369.
-
Fix for running under deno gpuweb/cts#4385.
-
Pipeline error for out of bounds array with dynamic and offset gpuweb/cts#4384.
-
If BC is supported, BC Sliced 3D must be supported gpuweb/cts#4379.
-
Fix texture tests gpuweb/cts#4383.
-
-
8 pull requests received 22 new comments.
-
Test GPUBuffer in GPUBindingResource gpuweb/cts#4390 (8 comments by @beaufortfrancois, @greggman, @Kangz)
-
Pipeline error for out of bounds array with dynamic and offset gpuweb/cts#4384 (2 comments by @jrprice, @petermcneeleychromium)
-
TR publications
-
8 documents published.
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
WebGPU Shading Language published 5 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
Participants
-
1 participant joined.
-
Emre Yalcinoglu (Autodesk) joined.
-
GitHub
gpuweb/gpuweb
-
27 issues opened.
-
wgsl: add support for Prefix Operators i.e. ++i gpuweb/gpuweb#5181, by @Dalayeth.
-
Add texture component mapping as an optional feature `"texture-component-swizzle"` gpuweb/gpuweb#5179, by @Kangz.
-
MultiDrawIndirect feature and maxDrawIndirectCount limit gpuweb/gpuweb#5175, by @fridenmf.
-
Can importExternalTexture requirements be relaxed? gpuweb/gpuweb#5172, by @mwyrzykowski.
-
Allow render bundles to inherit state gpuweb/gpuweb#5171, by @kainino0x.
-
Specification clarification regarding "discard" storeOp with MSAA? gpuweb/gpuweb#5168, by @mwyrzykowski.
-
Explainer text for mapping is confusing gpuweb/gpuweb#5165, by @kainino0x.
-
Allow creation of (single-layer) multisampled 2d-array views? gpuweb/gpuweb#5162, by @teoxoy.
-
Add `@builtin(renderbundle_index)` gpuweb/gpuweb#5158, by @greggman.
-
copyExternalImageToTexture with byte-exact `uint` format gpuweb/gpuweb#5157, by @unconed.
-
function workgroup-memory args should be more flexible (WGSL proposed language change) gpuweb/gpuweb#5156, by @jowens.
-
allow declaring workgroup memory within function scope (WGSL proposed language change) gpuweb/gpuweb#5155, by @jowens.
-
Extend WGSL to Support Additional Integer Types gpuweb/gpuweb#5152, by @MichealReed.
-
Limit the value passed into @align? gpuweb/gpuweb#5150, by @dj2.
-
Minor issue: Can we add definitions for true and false to the API gpuweb/gpuweb#5149, by @petermcneeleychromium.
-
Deliberately standardize the order of features in adapter.features and device.features gpuweb/gpuweb#5148, by @kainino0x.
-
HTML spec has a third "Update the rendering" step that we don't mention gpuweb/gpuweb#5146, by @kainino0x.
-
Clarification on scalar float interpolation? gpuweb/gpuweb#5144, by @mwyrzykowski.
-
"If configuration.device is found to be valid" step might be phrased as nondeterministic gpuweb/gpuweb#5143, by @kainino0x.
-
How to deal with limited nr of samplers allowed to be created on native APIs? gpuweb/gpuweb#5142, by @teoxoy.
-
@if for conditions - preview gpuweb/gpuweb#5140, by @mighdoll.
-
extend the grammar for more attributes gpuweb/gpuweb#5139, by @mighdoll.
-
alternative to const_assert in functions gpuweb/gpuweb#5138, by @mighdoll.
-
WGSL spec should note that overload resolution always yields a most preferable candidate gpuweb/gpuweb#5137, by @jimblandy.
-
-
9 issues closed.
-
Add `@builtin(renderbundle_index)` gpuweb/gpuweb#5158.
-
Can importExternalTexture requirements be relaxed? gpuweb/gpuweb#5172.
-
Allow creation of (single-layer) multisampled 2d-array views? gpuweb/gpuweb#5162.
-
Proposal: add WGSL instruction `workgroupUniformAtomicLoad` gpuweb/gpuweb#5124.
-
Minor issue: Can we add definitions for true and false to the API gpuweb/gpuweb#5149.
-
Compat: checking for compat is not backward compatible. gpuweb/gpuweb#5127.
-
Clarification on scalar float interpolation? gpuweb/gpuweb#5144.
-
A small subgroupInclusiveAdd documentation issue gpuweb/gpuweb#5134.
-
-
46 issues received 161 new comments.
-
Add `@builtin(renderbundle_index)` gpuweb/gpuweb#5158 (18 comments by @greggman, @kainino0x, @Kangz, @magcius, @mwyrzykowski, @toji)
-
How to deal with limited nr of samplers allowed to be created on native APIs? gpuweb/gpuweb#5142 (16 comments by @greggman, @kainino0x, @Kangz, @mwyrzykowski, @teoxoy)
-
builtin variables should be automatically available inside kernels (WGSL proposed language change) gpuweb/gpuweb#5153 (12 comments by @greggman, @jimblandy, @jowens, @magcius, @mighdoll)
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118 (10 comments by @kainino0x, @magcius, @mwyrzykowski, @shaoboyan091)
-
Allow render bundles to inherit state gpuweb/gpuweb#5171 (8 comments by @kainino0x, @mwyrzykowski, @toji)
-
Can importExternalTexture requirements be relaxed? gpuweb/gpuweb#5172 (8 comments by @kainino0x, @mwyrzykowski, @shaoboyan091)
-
Cannot upload/download to UMA storage buffer without an unnecessary copy and unnecessary memory use gpuweb/gpuweb#2388 (7 comments by @grovesNL, @Jiawei-Shao, @kainino0x, @Kangz)
-
copyExternalImageToTexture with byte-exact `uint` format gpuweb/gpuweb#5157 (5 comments by @beaufortfrancois, @kainino0x, @Kangz)
-
Why GPUCommandBuffer and GPUCommandEncoder cannot be reused? gpuweb/gpuweb#4138 (5 comments by @fridenmf, @mwyrzykowski)
-
Allow creation of (single-layer) multisampled 2d-array views? gpuweb/gpuweb#5162 (5 comments by @kainino0x, @Kangz, @teoxoy)
-
Compat: checking for compat is not backward compatible. gpuweb/gpuweb#5127 (4 comments by @greggman, @jimblandy, @kainino0x, @Kangz)
-
importExternalTexture: Support importing platform‑shared GPU textures (DXGI / IOSurface / dma‑buf) gpuweb/gpuweb#5167 (3 comments by @grovesNL, @kainino0x, @mehmetoguzderin)
-
builtins with a 3D variant should all have 1D variants (global_invocation_index, workgroup_index) (WGSL proposed language change) gpuweb/gpuweb#5154 (3 comments by @jimblandy, @jowens, @jrprice)
-
HTML spec has a third "Update the rendering" step that we don't mention gpuweb/gpuweb#5146 (3 comments by @kainino0x)
-
Finite Math Assumption makes WebGPU a difficult compilation target gpuweb/gpuweb#5109 (3 comments by @bjacob, @dj2, @kdashg)
-
allow declaring workgroup memory within function scope (WGSL proposed language change) gpuweb/gpuweb#5155 (2 comments by @jimblandy, @jowens)
-
MultiDrawIndirect feature and maxDrawIndirectCount limit gpuweb/gpuweb#5175 (2 comments by @fridenmf, @Jamesernator)
-
Deliberately standardize the order of features in adapter.features and device.features gpuweb/gpuweb#5148 (2 comments by @Kangz)
-
Minor issue: Can we add definitions for true and false to the API gpuweb/gpuweb#5149 (2 comments by @kainino0x, @petermcneeleychromium)
-
Proposal: add WGSL instruction `workgroupUniformAtomicLoad` gpuweb/gpuweb#5124 (2 comments by @kdashg, @mwyrzykowski)
-
Making uniformity analysis less conservative via statement behaviour analysis gpuweb/gpuweb#5100 (2 comments by @dneto0, @kdashg)
-
Add texture component mapping as an optional feature `"texture-component-swizzle"` gpuweb/gpuweb#5179 (1 comment by @mwyrzykowski)
-
function workgroup-memory args should be more flexible (WGSL proposed language change) gpuweb/gpuweb#5156 (1 comment by @jimblandy)
-
Specification clarification regarding "discard" storeOp with MSAA? gpuweb/gpuweb#5168 (1 comment by @kainino0x)
-
Explicitly state how to handle an if statement with an empty else in uniformity analysis gpuweb/gpuweb#5122 (1 comment by @kdashg)
-
Add function call statement and expression rules to uniformity analysis gpuweb/gpuweb#5128 (1 comment by @kdashg)
-
Add a rule for empty statements to uniformity rules for statements gpuweb/gpuweb#5123 (1 comment by @kdashg)
-
WorkgroupUniformLoad can return fixed-footprint arrays instead of creation-fixed-footprint arrays gpuweb/gpuweb#5093 (1 comment by @kdashg)
-
Sized Binding Arrays: Decide what can be arrayed and the syntax. gpuweb/gpuweb#5084 (1 comment by @kdashg)
-
-
16 pull requests opened.
-
BC does not guarantee Sliced 3D gpuweb/gpuweb#5184, by @mehmetoguzderin.
-
Fix custom dark mode styles for Bikeshed switcher gpuweb/gpuweb#5183, by @kainino0x.
-
Copy license to out/ for gh-pages gpuweb/gpuweb#5182, by @kainino0x.
-
Compat: Add clarification on BC guarantees for sliced 3D feature gpuweb/gpuweb#5180, by @mehmetoguzderin.
-
Notes on float for unbounded gpuweb/gpuweb#5176, by @petermcneeleychromium.
-
[binding array proposal] Choose a clearer name for binding array size, in BGL gpuweb/gpuweb#5174, by @kainino0x.
-
[immediates proposal] Rename to immediateSize and maxImmediateSize gpuweb/gpuweb#5173, by @kainino0x.
-
Add FSA to LICENSE, and inline license text gpuweb/gpuweb#5170, by @amaiorano.
-
Specify that ASTC Sliced 3D feature requires ASTC feature gpuweb/gpuweb#5169, by @beaufortfrancois.
-
Add guidance to build the image with devcontainer gpuweb/gpuweb#5166, by @mehmetoguzderin.
-
Make quotes part of the link for consistency gpuweb/gpuweb#5163, by @mehmetoguzderin.
-
Add optional feature `texture-formats-tier1` gpuweb/gpuweb#5160, by @Jiawei-Shao.
-
Add missing "subgroups" in the list of valid extension names gpuweb/gpuweb#5159, by @Jiawei-Shao.
-
Add "core-features-and-limits" to spec gpuweb/gpuweb#5147, by @kainino0x.
-
Add workgroupUniformLoad overload gpuweb/gpuweb#5141, by @beaufortfrancois.
-
[editorial] Fix note in subgroupInclusiveAdd gpuweb/gpuweb#5136, by @alan-baker.
-
-
13 pull requests merged.
-
Fix custom dark mode styles for Bikeshed switcher gpuweb/gpuweb#5183.
-
BC does not guarantee Sliced 3D gpuweb/gpuweb#5184.
-
Notes on float for unbounded gpuweb/gpuweb#5176.
-
Copy license to out/ for gh-pages gpuweb/gpuweb#5182.
-
Add FSA to LICENSE, and inline license text gpuweb/gpuweb#5170.
-
Subgroups: add note for subgroupMinSize gpuweb/gpuweb#5007.
-
Specify that ASTC Sliced 3D feature requires ASTC feature gpuweb/gpuweb#5169.
-
Make quotes part of the link for consistency gpuweb/gpuweb#5163.
-
Add guidance to build the image with devcontainer gpuweb/gpuweb#5166.
-
Add "core-features-and-limits" to spec gpuweb/gpuweb#5147.
-
Add missing "subgroups" in the list of valid extension names gpuweb/gpuweb#5159.
-
Add workgroupUniformLoad overload gpuweb/gpuweb#5141.
-
[editorial] Fix note in subgroupInclusiveAdd gpuweb/gpuweb#5136.
-
-
2 pull requests closed.
-
Compat: Add clarification on BC guarantees for sliced 3D feature gpuweb/gpuweb#5180.
-
Add clamp-to-border-color sampler address mode gpuweb/gpuweb#5107.
-
-
14 pull requests received 74 new comments.
-
Add optional feature `texture-formats-tier1` gpuweb/gpuweb#5160 (21 comments by @Jiawei-Shao, @kainino0x, @Kangz, @mwyrzykowski, @teoxoy)
-
Add workgroupUniformLoad overload gpuweb/gpuweb#5141 (14 comments by @alan-baker, @beaufortfrancois, @dneto0)
-
Notes on float for unbounded gpuweb/gpuweb#5176 (8 comments by @dneto0, @Kangz, @petermcneeleychromium)
-
Compat: Add clarification on BC guarantees for sliced 3D feature gpuweb/gpuweb#5180 (7 comments by @kainino0x, @mehmetoguzderin, @SenorBlanco)
-
Add FSA to LICENSE, and inline license text gpuweb/gpuweb#5170 (7 comments by @amaiorano, @kainino0x, @tidoust)
-
Add guidance to build the image with devcontainer gpuweb/gpuweb#5166 (4 comments by @kainino0x, @mehmetoguzderin)
-
[binding array proposal] Choose a clearer name for binding array size, in BGL gpuweb/gpuweb#5174 (2 comments by @kainino0x, @Kangz)
-
[immediates proposal] Rename to immediateSize and maxImmediateSize gpuweb/gpuweb#5173 (2 comments by @Kangz, @teoxoy)
-
Allow creating mappable buffer with more usages as optional features gpuweb/gpuweb#5108 (1 comment by @Jiawei-Shao)
-
gpuweb/cts
-
15 issues opened.
-
requestAdapter_invalid_featureLevel fails on compat-only implementation gpuweb/cts#4372, by @kainino0x.
-
op: Test camera capture using copyExternalImageToTexture gpuweb/cts#4364, by @kainino0x.
-
cameraCapture test doesn't reliably get a real frame gpuweb/cts#4363, by @kainino0x.
-
Add test for device.onuncapturederror gpuweb/cts#4361, by @kainino0x.
-
workgroupUniformLoad: test pointer-to-atomic variant gpuweb/cts#4341, by @dneto0.
-
Ensure devicePool.release is called for both providers even if one rejects gpuweb/cts#4339, by @greggman.
-
cmdline runner not exiting gpuweb/cts#4325, by @amaiorano.
-
Add coverage of copying a canvas drawing_buffer after the end of the RAF gpuweb/cts#4323, by @Kangz.
-
Consider adding the current "test" in the device label. gpuweb/cts#4322, by @Kangz.
-
19 issues closed.
-
cameraCapture test doesn't reliably get a real frame gpuweb/cts#4363.
-
Add test for device.onuncapturederror gpuweb/cts#4361.
-
op: test writeBuffer and writeTexture with all possible sources gpuweb/cts#2488.
-
refactor textureSampleBias tests gpuweb/cts#4153.
-
val: queue,submit gpuweb/cts#2376.
-
val: queries,general gpuweb/cts#2375.
-
val: queries,begin_end gpuweb/cts#2374.
-
Test restrictions on limits that can be exposed by adapters gpuweb/cts#2277.
-
wgsl: execution test for textureLoad gpuweb/cts#1262.
-
val: capability_checks,limits gpuweb/cts#2195.
-
texture_zero.spec.ts makes compat incompatible texture views gpuweb/cts#3145.
-
Change compat shader module errors to pipeline errors gpuweb/cts#3893.
-
Test that bias in textureSampleBias is clamped to [-16.0, 15.99] gpuweb/cts#3883.
-
cmdline runner not exiting gpuweb/cts#4325.
-
13 issues received 45 new comments.
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761 (12 comments by @beaufortfrancois, @greggman, @mehmetoguzderin)
-
workgroupUniformLoad: test pointer-to-atomic variant gpuweb/cts#4341 (7 comments by @beaufortfrancois, @dj2, @dneto0)
-
image_copy:mip_levels tests fail for bc compressed formats with AMD Radeon RX 5500 XT on macOS gpuweb/cts#4378 (5 comments by @beaufortfrancois, @greggman)
-
Add coverage for `SharedArrayBuffer` with `GPUDevice.{writeTexture,writeBuffer}` gpuweb/cts#4345 (5 comments by @ErichDonGubler, @kainino0x)
-
textureSampleLevel fails for bc7-rgba-unorm-srgb on macOS for AMD Radeon Pro 560X gpuweb/cts#4377 (3 comments by @beaufortfrancois, @greggman)
-
cameraCapture test doesn't reliably get a real frame gpuweb/cts#4363 (3 comments by @kainino0x, @shaoboyan091)
-
Ensure devicePool.release is called for both providers even if one rejects gpuweb/cts#4339 (3 comments by @greggman, @kainino0x)
-
Consider adding the current "test" in the device label. gpuweb/cts#4322 (2 comments by @greggman, @kainino0x)
-
Add a second call to GPUCanvasContext.getCurrentTexture() in viewFormats tests. gpuweb/cts#4337 (1 comment by @SenorBlanco)
-
Test that bias in textureSampleBias is clamped to [-16.0, 15.99] gpuweb/cts#3883 (1 comment by @greggman)
-
-
46 pull requests opened.
-
Fix usages of textureDimensionAndFormatCompatible gpuweb/cts#4376, by @greggman.
-
If ASTC Sliced 3D is supported, ASTC must be supported gpuweb/cts#4375, by @beaufortfrancois.
-
Use 0.4 for depth unorm value in depth_compare_func rendering test gpuweb/cts#4374, by @jiangzhaoming.
-
Skip some requestAdapter_invalid_featureLevel tests in compat-only gpuweb/cts#4373, by @beaufortfrancois.
-
Test 3D compressed formats in copyTextureToTexture gpuweb/cts#4371, by @beaufortfrancois.
-
New buffer,create:size mappedAtCreation behavior gpuweb/cts#4370, by @toji.
-
typo: add end quote to `texture format <X> requires feature: '<F>` gpuweb/cts#4369, by @ErichDonGubler.
-
Basic workgroup override expression test gpuweb/cts#4368, by @petermcneeleychromium.
-
Test onuncapturederror gpuweb/cts#4367, by @kainino0x.
-
Remove unload handler gpuweb/cts#4366, by @kainino0x.
-
cameraCapture: Simplify codepaths, try to wait for a real frame gpuweb/cts#4365, by @kainino0x.
-
Add feature tests for 3D texture with ASTC compressed formats gpuweb/cts#4362, by @beaufortfrancois.
-
Test an empty resolve pass gpuweb/cts#4360, by @kainino0x.
-
Prep for testing compressed 3d sliced texture formats gpuweb/cts#4359, by @greggman.
-
Expression in workgroup size gpuweb/cts#4358, by @petermcneeleychromium.
-
Add feature tests for 3D texture with BC compressed formats gpuweb/cts#4357, by @beaufortfrancois.
-
Add BC sliced 3D adapter capability checks gpuweb/cts#4356, by @beaufortfrancois.
-
Wait for the device pools to release gpuweb/cts#4355, by @greggman.
-
Remove some out of date comments gpuweb/cts#4354, by @greggman.
-
Remove WGSLTextureSampleTest and WGSLTextureQueryTest gpuweb/cts#4353, by @greggman.
-
Check packed texture formats features for 3d texture creations gpuweb/cts#4352, by @beaufortfrancois.
-
Update fp primer unbounded precision gpuweb/cts#4351, by @petermcneeleychromium.
-
Refactor validation tests gpuweb/cts#4350, by @greggman.
-
Remove some unimplemented from clear_value.spec.ts gpuweb/cts#4349, by @greggman.
-
Add uniformity test for workgroupUniformLoad overload gpuweb/cts#4348, by @beaufortfrancois.
-
Add tests for loading an atomic inside a struct gpuweb/cts#4347, by @beaufortfrancois.
-
Remove TextureTestMixin gpuweb/cts#4346, by @greggman.
-
Fix block/enforce/compat option interaction gpuweb/cts#4344, by @greggman.
-
Remove ValidationTest gpuweb/cts#4343, by @greggman.
-
remove MaxLimitsTestMixin gpuweb/cts#4342, by @greggman.
-
Fix shader,execution,expression,call,builtin,subgroupShuffle:compute,split gpuweb/cts#4340, by @haoxli.
-
web_platform,canvas,configure:viewFormats: call getCurrentTexture() twice. gpuweb/cts#4338, by @SenorBlanco.
-
Add tests for consumed adapter in requestDevice gpuweb/cts#4336, by @beaufortfrancois.
-
Unbounded precision intervals for addition and subtraction gpuweb/cts#4335, by @petermcneeleychromium.
-
Save cts results as json gpuweb/cts#4334, by @petermcneeleychromium.
-
Use featureLevel instead of compatibilityMode gpuweb/cts#4333, by @beaufortfrancois.
-
Add tests for workgroupUniformLoad overload gpuweb/cts#4332, by @beaufortfrancois.
-
Add known failure to texture_builtin_notes.md gpuweb/cts#4331, by @greggman.
-
Stencil copy test updates gpuweb/cts#4330, by @toji.
-
Try to destroy all devices on page/worker teardown gpuweb/cts#4329, by @kainino0x.
-
Fix assertReject gpuweb/cts#4328, by @kainino0x.
-
Destroy devices when cleaning the device pool gpuweb/cts#4327, by @kainino0x.
-
Fix cmdline runner not exiting gpuweb/cts#4326, by @amaiorano.
-
error_scope tests: break up awaits into chunks gpuweb/cts#4324, by @kainino0x.
-
Relax tolerance of readbackFromWebGPUCanvas WPT test to 3 ULP gpuweb/cts#4321, by @jiangzhaoming.
-
Fix subgroupMul execution tests for subgroup size > 32 gpuweb/cts#4319, by @jiangzhaoming.
-
-
46 pull requests merged.
-
Fix usages of textureDimensionAndFormatCompatible gpuweb/cts#4376.
-
If ASTC Sliced 3D is supported, ASTC must be supported gpuweb/cts#4375.
-
Use 0.4 for depth unorm value in depth_compare_func rendering test gpuweb/cts#4374.
-
Skip some requestAdapter_invalid_featureLevel tests in compat-only gpuweb/cts#4373.
-
Test 3D compressed formats in copyTextureToTexture gpuweb/cts#4371.
-
cameraCapture: Simplify codepaths, try to wait for a real frame gpuweb/cts#4365.
-
Add tests for consumed adapter in requestDevice gpuweb/cts#4336.
-
Basic workgroup override expression test gpuweb/cts#4368.
-
New buffer,create:size mappedAtCreation behavior gpuweb/cts#4370.
-
Test onuncapturederror gpuweb/cts#4367.
-
Remove unload handler gpuweb/cts#4366.
-
Prep for testing compressed 3d sliced texture formats gpuweb/cts#4359.
-
Add feature tests for 3D texture with ASTC compressed formats gpuweb/cts#4362.
-
Test an empty resolve pass gpuweb/cts#4360.
-
Expression in workgroup size gpuweb/cts#4358.
-
Add feature tests for 3D texture with BC compressed formats gpuweb/cts#4357.
-
Add BC sliced 3D adapter capability checks gpuweb/cts#4356.
-
Wait for the device pools to release gpuweb/cts#4355.
-
Check packed texture formats features for 3d texture creations gpuweb/cts#4352.
-
Fix block/enforce/compat option interaction gpuweb/cts#4344.
-
Remove WGSLTextureSampleTest and WGSLTextureQueryTest gpuweb/cts#4353.
-
Remove some out of date comments gpuweb/cts#4354.
-
Add tests for loading an atomic inside a struct gpuweb/cts#4347.
-
Add uniformity test for workgroupUniformLoad overload gpuweb/cts#4348.
-
Update fp primer unbounded precision gpuweb/cts#4351.
-
Refactor validation tests gpuweb/cts#4350.
-
Remove some unimplemented from clear_value.spec.ts gpuweb/cts#4349.
-
Remove TextureTestMixin gpuweb/cts#4346.
-
Remove ValidationTest gpuweb/cts#4343.
-
remove MaxLimitsTestMixin gpuweb/cts#4342.
-
Unbounded precision intervals for addition and subtraction gpuweb/cts#4335.
-
Add tests for workgroupUniformLoad overload gpuweb/cts#4332.
-
Use featureLevel instead of compatibilityMode gpuweb/cts#4333.
-
Save cts results as json gpuweb/cts#4334.
-
Try to destroy all devices on page/worker teardown gpuweb/cts#4329.
-
Add known failure to texture_builtin_notes.md gpuweb/cts#4331.
-
Stencil copy test updates gpuweb/cts#4330.
-
Destroy devices when cleaning the device pool gpuweb/cts#4327.
-
Fix assertReject gpuweb/cts#4328.
-
Fix cmdline runner not exiting gpuweb/cts#4326.
-
Add tests for copyBufferToBuffer overloads gpuweb/cts#4310.
-
error_scope tests: break up awaits into chunks gpuweb/cts#4324.
-
Fix subgroupMul execution tests for subgroup size > 32 gpuweb/cts#4319.
-
Relax tolerance of readbackFromWebGPUCanvas WPT test to 3 ULP gpuweb/cts#4321.
-
-
1 pull request reopened.
-
Fix shader,execution,expression,call,builtin,subgroupShuffle:compute,split gpuweb/cts#4340, by @haoxli.
-
-
27 pull requests received 90 new comments.
-
Unbounded precision intervals for addition and subtraction gpuweb/cts#4335 (17 comments by @dneto0, @petermcneeleychromium, @zoddicus)
-
cameraCapture: Simplify codepaths, try to wait for a real frame gpuweb/cts#4365 (15 comments by @kainino0x, @shaoboyan091)
-
Add tests for consumed adapter in requestDevice gpuweb/cts#4336 (13 comments by @beaufortfrancois, @kainino0x)
-
Add tests for copyBufferToBuffer overloads gpuweb/cts#4310 (5 comments by @beaufortfrancois, @kainino0x)
-
Fix usages of textureDimensionAndFormatCompatible gpuweb/cts#4376 (3 comments by @beaufortfrancois, @greggman)
-
Test 3D compressed formats in copyTextureToTexture gpuweb/cts#4371 (3 comments by @beaufortfrancois, @greggman)
-
Prep for testing compressed 3d sliced texture formats gpuweb/cts#4359 (3 comments by @beaufortfrancois, @greggman)
-
Try to destroy all devices on page/worker teardown gpuweb/cts#4329 (3 comments by @beaufortfrancois, @kainino0x)
-
Use 0.4 for depth unorm value in depth_compare_func rendering test gpuweb/cts#4374 (2 comments by @jiangzhaoming, @kainino0x)
-
Add feature tests for 3D texture with BC compressed formats gpuweb/cts#4357 (2 comments by @beaufortfrancois, @kainino0x)
-
Fix shader,execution,expression,call,builtin,subgroupShuffle:compute,split gpuweb/cts#4340 (2 comments by @haoxli)
-
If ASTC Sliced 3D is supported, ASTC must be supported gpuweb/cts#4375 (1 comment by @beaufortfrancois)
-
Add feature tests for 3D texture with ASTC compressed formats gpuweb/cts#4362 (1 comment by @beaufortfrancois)
-
Check packed texture formats features for 3d texture creations gpuweb/cts#4352 (1 comment by @kainino0x)
-
Relax tolerance of readbackFromWebGPUCanvas WPT test to 3 ULP gpuweb/cts#4321 (1 comment by @jiangzhaoming)
-
Fix subgroupMul execution tests for subgroup size > 32 gpuweb/cts#4319 (1 comment by @jiangzhaoming)
-
TR publications
-
5 documents published.
-
WebGPU Shading Language published twice as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published 3 times as a Candidate Recommendation Draft. History of WebGPU
-
GitHub
gpuweb/cts
-
7 issues opened.
-
workgroupUniformLoad: T (return type) must be concrete constructible gpuweb/cts#4301, by @dneto0.
-
Auto-generate tests using mutation testing gpuweb/cts#4299, by @ambergorzynski.
-
Remove Float16Array polyfill gpuweb/cts#4297, by @kainino0x.
-
api,operation,adapter,requestDevice* expectations are incorrect gpuweb/cts#4288, by @mwyrzykowski.
-
wgsl: binding_array is no longer a keyword gpuweb/cts#4256, by @dneto0.
-
-
5 issues closed.
-
Check GPUAdapterInfo.isFallbackAdapter is present and a boolean gpuweb/cts#2356.
-
`logResult[n].extra` fails to serialize in worker tests gpuweb/cts#4172.
-
api,operation,adapter,requestDevice* expectations are incorrect gpuweb/cts#4288.
-
wgsl: binding_array is no longer a keyword gpuweb/cts#4256.
-
-
3 issues reopened.
-
api,operation,adapter,requestDevice* expectations are incorrect gpuweb/cts#4288, by @mwyrzykowski.
-
9 issues received 24 new comments.
-
`logResult[n].extra` fails to serialize in worker tests gpuweb/cts#4172 (6 comments by @ErichDonGubler, @kainino0x)
-
api,operation,adapter,requestDevice* expectations are incorrect gpuweb/cts#4288 (4 comments by @greggman, @kainino0x, @mwyrzykowski)
-
Remove shaderModule and compilationInfo from ErrorWithExtra in shader_validation_test.ts? gpuweb/cts#4248 (4 comments by @greggman, @kainino0x)
-
Check GPUAdapterInfo.isFallbackAdapter is present and a boolean gpuweb/cts#2356 (3 comments by @beaufortfrancois, @kainino0x)
-
Auto-generate tests using mutation testing gpuweb/cts#4299 (3 comments by @afd, @ambergorzynski, @dneto0)
-
[wgsl] Validate errors are not generated for overrides which aren't present in the module. gpuweb/cts#3747 (1 comment by @dj2)
-
-
68 pull requests opened.
-
Gate use of OffscreenCanvas on it being available gpuweb/cts#4317, by @Kangz.
-
Add testing for user errors for workgroup size gpuweb/cts#4316, by @petermcneeleychromium.
-
Remove lod_clamp.spec.ts and add cases to texture builtin tests gpuweb/cts#4315, by @greggman.
-
Remove rg11b10ufloat exceptions from format_info.ts gpuweb/cts#4314, by @greggman.
-
Add block-all-features flag gpuweb/cts#4313, by @greggman.
-
Cleanup storage texture tests gpuweb/cts#4312, by @greggman.
-
Remove lod_clamp.spec.ts gpuweb/cts#4311, by @greggman.
-
Add tests for copyBufferToBuffer overloads gpuweb/cts#4310, by @beaufortfrancois.
-
Compat: unlower limits gpuweb/cts#4308, by @greggman.
-
Workgroup uniform load constructible type only gpuweb/cts#4307, by @petermcneeleychromium.
-
Add shader tests for using texture view for externalTexture binding gpuweb/cts#4306, by @beaufortfrancois.
-
Add some skipIfTextureFormatNotUsableAsRenderAttachment to skip rg11b… gpuweb/cts#4305, by @shrekshao.
-
Add validation tests for using texture view for externalTexture binding gpuweb/cts#4304, by @beaufortfrancois.
-
Add tests for GPUAdapterInfo.isFallbackAdapter gpuweb/cts#4303, by @beaufortfrancois.
-
Roll types 0.1.52 -> 0.1.59 gpuweb/cts#4302, by @kainino0x.
-
Use the texture view in the texture builtin tests. gpuweb/cts#4298, by @greggman.
-
Minor fix for derivative tests for incorrect indexing gpuweb/cts#4296, by @petermcneeleychromium.
-
Add tests for mapping behavior around device.destroy() gpuweb/cts#4295, by @kainino0x.
-
Expand the sample_texture_combo test to include storage textures. gpuweb/cts#4294, by @greggman.
-
Fix rg11b10ufloat filter_mode tests gpuweb/cts#4293, by @shrekshao.
-
Update requestDevice:always_return_device gpuweb/cts#4292, by @shrekshao.
-
Fix limits issues. gpuweb/cts#4291, by @greggman.
-
Testing clamping with textureSampleBias is prone to precision issues.… gpuweb/cts#4290, by @greggman.
-
Change kDefaultCTSOptions.debug to false gpuweb/cts#4289, by @kainino0x.
-
Remove deprecated from gpu_test.ts/format_info.ts gpuweb/cts#4286, by @greggman.
-
Remove kPossiblyRenderableColorTextureFormats gpuweb/cts#4285, by @greggman.
-
Refactor maxColorAttachmentBytesPerSample.spec.ts gpuweb/cts#4284, by @greggman.
-
Refactor float32_blendable.spec.ts gpuweb/cts#4283, by @greggman.
-
Refactor api/validation/texture/* gpuweb/cts#4282, by @greggman.
-
Refactor api/validation/resource_usages/* gpuweb/cts#4281, by @greggman.
-
Refactor api/validation/render_pipeline/* gpuweb/cts#4280, by @greggman.
-
Refactor api/validation/queue/* gpuweb/cts#4279, by @greggman.
-
Refactor api/validation/* gpuweb/cts#4278, by @greggman.
-
Refactor api/validation/* with format_info.ts changes gpuweb/cts#4277, by @greggman.
-
Refactor api/validation/(encoding|query_set|render_pass|shader_module) gpuweb/cts#4276, by @greggman.
-
Refactor api/validation/create* gpuweb/cts#4275, by @greggman.
-
Bump tsconfig.json lib to es2023 gpuweb/cts#4274, by @greggman.
-
Refactor src/webpgu/api/validation/encoding and a few extra gpuweb/cts#4273, by @greggman.
-
Fix texture_intra_invocation_coherence.spec.ts gpuweb/cts#4272, by @greggman.
-
Add labels to command encoders gpuweb/cts#4271, by @greggman.
-
Make selectDeviceXXX unreachable on AllFeaturesMaxLimitsGPUTest gpuweb/cts#4270, by @greggman.
-
Refactor shader/validation/expression/overload_resolution.spec.ts gpuweb/cts#4269, by @greggman.
-
Refactor webgpu/examples gpuweb/cts#4268, by @greggman.
-
Refactor shader/execution/expression/call/builtin/texture* gpuweb/cts#4267, by @greggman.
-
Refactor shader/execution/* gpuweb/cts#4266, by @greggman.
-
Refactor api/operation gpuweb/cts#4265, by @greggman.
-
Return mismatched device holder to the mismatched device pool gpuweb/cts#4264, by @greggman.
-
Fix shader/execution/expression/call/builtin/textureSampleBias gpuweb/cts#4263, by @greggman.
-
Add comment case to WGSL enable test. gpuweb/cts#4262, by @greggman.
-
Fix shader/validation/types/textures gpuweb/cts#4261, by @greggman.
-
Fix enable regular expression. gpuweb/cts#4260, by @greggman.
-
Refactor shader/validation/expression/matrix/* gpuweb/cts#4258, by @greggman.
-
wgsl: test declaring a name 'binding_array' gpuweb/cts#4257, by @dneto0.
-
Make ShaderValidationTest auto skip if required features missing. gpuweb/cts#4255, by @greggman.
-
Refactor shader/validation/shader_io/* gpuweb/cts#4254, by @greggman.
-
Refactor shader/validation/statement/* gpuweb/cts#4253, by @greggman.
-
Refactor shader/validation/types/* + 1 gpuweb/cts#4252, by @greggman.
-
Refactor shader/validtion/expression/(access|binary|unary) gpuweb/cts#4251, by @greggman.
-
Refactor shader/validation/(function|parse)/* gpuweb/cts#4250, by @greggman.
-
Remove shaderModule and compilationInfo from error.extra gpuweb/cts#4249, by @greggman.
-
Refactor shader/validation/expression/call/builtin/* gpuweb/cts#4247, by @greggman.
-
Minor fix to nested array tests gpuweb/cts#4246, by @petermcneeleychromium.
-
Unreserve the `binding_array` keyword. gpuweb/cts#4245, by @Kangz.
-
Refactor shader/validtion/expression/(access|binary|unary) gpuweb/cts#4244, by @greggman.
-
Refactor shader/validation/types/* gpuweb/cts#4243, by @greggman.
-
-
66 pull requests merged.
-
Remove lod_clamp.spec.ts and add cases to texture builtin tests gpuweb/cts#4315.
-
Gate use of OffscreenCanvas on it being available gpuweb/cts#4317.
-
Add testing for user errors for workgroup size gpuweb/cts#4316.
-
Remove rg11b10ufloat exceptions from format_info.ts gpuweb/cts#4314.
-
Add block-all-features flag gpuweb/cts#4313.
-
Cleanup storage texture tests gpuweb/cts#4312.
-
fix typos gpuweb/cts#4309.
-
Compat: unlower limits gpuweb/cts#4308.
-
Workgroup uniform load constructible type only gpuweb/cts#4307.
-
Add shader tests for using texture view for externalTexture binding gpuweb/cts#4306.
-
Add tests for GPUAdapterInfo.isFallbackAdapter gpuweb/cts#4303.
-
Roll types 0.1.52 -> 0.1.59 gpuweb/cts#4302.
-
Use the texture view in the texture builtin tests. gpuweb/cts#4298.
-
Minor fix for derivative tests for incorrect indexing gpuweb/cts#4296.
-
Expand the sample_texture_combo test to include storage textures. gpuweb/cts#4294.
-
Fix rg11b10ufloat filter_mode tests gpuweb/cts#4293.
-
Update requestDevice:always_return_device gpuweb/cts#4292.
-
Fix limits issues. gpuweb/cts#4291.
-
Change kDefaultCTSOptions.debug to false gpuweb/cts#4289.
-
Remove deprecated from gpu_test.ts/format_info.ts gpuweb/cts#4286.
-
Remove kPossiblyRenderableColorTextureFormats gpuweb/cts#4285.
-
Bump tsconfig.json lib to es2023 gpuweb/cts#4274.
-
Refactor api/validation/* with format_info.ts changes gpuweb/cts#4277.
-
Refactor api/validation/resource_usages/* gpuweb/cts#4281.
-
Refactor maxColorAttachmentBytesPerSample.spec.ts gpuweb/cts#4284.
-
Refactor api/validation/texture/* gpuweb/cts#4282.
-
Refactor float32_blendable.spec.ts gpuweb/cts#4283.
-
Refactor api/validation/render_pipeline/* gpuweb/cts#4280.
-
Refactor api/validation/queue/* gpuweb/cts#4279.
-
Refactor api/validation/* gpuweb/cts#4278.
-
Refactor api/validation/create* gpuweb/cts#4275.
-
Refactor src/webpgu/api/validation/encoding and a few extra gpuweb/cts#4273.
-
Fix texture_intra_invocation_coherence.spec.ts gpuweb/cts#4272.
-
Make selectDeviceXXX unreachable on AllFeaturesMaxLimitsGPUTest gpuweb/cts#4270.
-
Add labels to command encoders gpuweb/cts#4271.
-
Refactor shader/validation/expression/overload_resolution.spec.ts gpuweb/cts#4269.
-
Refactor webgpu/examples gpuweb/cts#4268.
-
Refactor shader/execution/* gpuweb/cts#4266.
-
Refactor shader/execution/expression/call/builtin/texture* gpuweb/cts#4267.
-
Refactor api/operation gpuweb/cts#4265.
-
Return mismatched device holder to the mismatched device pool gpuweb/cts#4264.
-
Add comment case to WGSL enable test. gpuweb/cts#4262.
-
Fix shader/validation/types/textures gpuweb/cts#4261.
-
Fix shader/execution/expression/call/builtin/textureSampleBias gpuweb/cts#4263.
-
Remove shaderModule and compilationInfo from error.extra gpuweb/cts#4249.
-
Fix enable regular expression. gpuweb/cts#4260.
-
wgsl: test declaring a name 'binding_array' gpuweb/cts#4257.
-
Refactor shader/validation/(function|parse)/* gpuweb/cts#4250.
-
Refactor shader/validation/expression/call/builtin/* gpuweb/cts#4247.
-
Refactor shader/validtion/expression/(access|binary|unary) gpuweb/cts#4251.
-
Refactor shader/validation/types/* + 1 gpuweb/cts#4252.
-
Refactor shader/validation/statement/* gpuweb/cts#4253.
-
Refactor shader/validation/expression/matrix/* gpuweb/cts#4258.
-
Refactor shader/validation/shader_io/* gpuweb/cts#4254.
-
Make ShaderValidationTest auto skip if required features missing. gpuweb/cts#4255.
-
Unreserve the `binding_array` keyword. gpuweb/cts#4245.
-
Minor fix to nested array tests gpuweb/cts#4246.
-
Refactor idl/* gpuweb/cts#4239.
-
Refactor shader/validation/shader_io/* gpuweb/cts#4241.
-
Refactor shader/validation/(funcitons|parse) gpuweb/cts#4240.
-
-
4 pull requests closed.
-
Remove lod_clamp.spec.ts gpuweb/cts#4311.
-
Refactor shader/validation/statement/* gpuweb/cts#4242.
-
Refactor shader/validtion/expression/(access|binary|unary) gpuweb/cts#4244.
-
Refactor shader/validation/types/* gpuweb/cts#4243.
-
-
18 pull requests received 39 new comments.
-
Add tests for GPUAdapterInfo.isFallbackAdapter gpuweb/cts#4303 (4 comments by @beaufortfrancois, @kainino0x)
-
Add tests for mapping behavior around device.destroy() gpuweb/cts#4295 (4 comments by @kainino0x, @lokokung)
-
Expand the sample_texture_combo test to include storage textures. gpuweb/cts#4294 (4 comments by @greggman, @shrekshao)
-
Make ShaderValidationTest auto skip if required features missing. gpuweb/cts#4255 (2 comments by @greggman, @shrekshao)
-
Add coverage for overrides in nested arrays and overrides in array members of structures gpuweb/cts#4218 (2 comments by @greggman, @petermcneeleychromium)
gpuweb/gpuweb
-
25 issues opened.
-
A small subgroupInclusiveAdd documentation issue gpuweb/gpuweb#5134, by @mikialex.
-
Integrate with proposed ECMAScript Explicit Resource Management gpuweb/gpuweb#5132, by @kainino0x.
-
Desugaring assumptions in uniformity analysis gpuweb/gpuweb#5129, by @JamesLee-Jones.
-
Add function call statement and expression rules to uniformity analysis gpuweb/gpuweb#5128, by @JamesLee-Jones.
-
Compat: checking for compat is not backward compatible. gpuweb/gpuweb#5127, by @greggman.
-
Compat: raise maxStorage(Buffers/Texture)InFragmentStage from 0 to 4 gpuweb/gpuweb#5125, by @greggman.
-
Proposal: add WGSL instruction `workgroupUniformAtomicLoad` gpuweb/gpuweb#5124, by @jowens.
-
Add a rule for empty statements to uniformity rules for statements gpuweb/gpuweb#5123, by @JamesLee-Jones.
-
Add parenthesis rule to uniformity rules for expressions gpuweb/gpuweb#5121, by @JamesLee-Jones.
-
[Immediate Data] Rename <immediate_data> to <immediate> gpuweb/gpuweb#5119, by @shaoboyan091.
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118, by @shaoboyan091.
-
[Immediate Data] Add internal slot [[immediate data]] in GPUBindingCommandsMixin gpuweb/gpuweb#5117, by @shaoboyan091.
-
[Immediate Data] Shall we support per-stage immediate data? gpuweb/gpuweb#5116, by @shaoboyan091.
-
Finite Math Assumption makes WebGPU a difficult compilation target gpuweb/gpuweb#5109, by @bjacob.
-
mappedAtCreation size alignment is checked after mapping already happened gpuweb/gpuweb#5105, by @kainino0x.
-
Compat: maxStorageBuffersPerShaderStage doesn't need to be lowered gpuweb/gpuweb#5103, by @kainino0x.
-
Spec requires mappedAtCreation to still work after device is destroyed gpuweb/gpuweb#5102, by @kainino0x.
-
mapAsync() algorithm's `deviceLost` variable is uninitialized on validation error gpuweb/gpuweb#5101, by @kainino0x.
-
Making uniformity analysis less conservative via statement behaviour analysis gpuweb/gpuweb#5100, by @JamesLee-Jones.
-
Improve source size calculation used by `copyExternalImageToTexture` with `HTMLImageElement` gpuweb/gpuweb#5097, by @teoxoy.
-
Feature request: Support GPUBuffer in GPUBindingResource gpuweb/gpuweb#5095, by @beaufortfrancois.
-
-
18 issues closed.
-
Compat: raise maxStorage(Buffers/Texture)InFragmentStage from 0 to 4 gpuweb/gpuweb#5125.
-
Compat: maxStorageBuffersPerShaderStage doesn't need to be lowered gpuweb/gpuweb#5103.
-
[Immediate Data] Add internal slot [[immediate data]] in GPUBindingCommandsMixin gpuweb/gpuweb#5117.
-
mappedAtCreation size alignment is checked after mapping already happened gpuweb/gpuweb#5105.
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513.
-
Make the last 2 arguments to `copyBufferToBuffer` optional gpuweb/gpuweb#4807.
-
Move isFallbackAdapter into GPUAdapterInfo gpuweb/gpuweb#4971.
-
Compatibility Mode: GL_MAX_COMBINED_* limits in OpenGL ES gpuweb/gpuweb#5003.
-
binding_array should not be a reserved word gpuweb/gpuweb#5088.
-
Hardware feature levels (and compat) gpuweb/gpuweb#4656.
-
Limit on number of active occlusion queries? gpuweb/gpuweb#4603.
-
Compat: Limits added for compat need to behave in a forward compatible way gpuweb/gpuweb#5013.
-
-
1 issue reopened.
-
Sized Binding Arrays: Uniformity of accesses gpuweb/gpuweb#5085, by @Kangz.
-
-
55 issues received 146 new comments.
-
Cannot upload/download to UMA storage buffer without an unnecessary copy and unnecessary memory use gpuweb/gpuweb#2388 (13 comments by @Jiawei-Shao, @JMS55, @kainino0x, @Kangz, @mwyrzykowski, @sebbbi)
-
Sized Binding Arrays: Uniformity of accesses gpuweb/gpuweb#5085 (12 comments by @dneto0, @jimblandy, @Kangz, @kdashg, @magcius)
-
Resurrect clamp-to-border gpuweb/gpuweb#1305 (11 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz, @mwyrzykowski)
-
Spec requires mappedAtCreation to still work after device is destroyed gpuweb/gpuweb#5102 (9 comments by @kainino0x, @Kangz, @kdashg, @mwyrzykowski)
-
Finite Math Assumption makes WebGPU a difficult compilation target gpuweb/gpuweb#5109 (6 comments by @bjacob, @dj2)
-
Add limited support for 64 bit atomics gpuweb/gpuweb#5071 (6 comments by @JMS55, @kdashg, @petermcneeleychromium)
-
Sized Binding Arrays: Decide what can be arrayed and the syntax. gpuweb/gpuweb#5084 (5 comments by @dneto0, @jimblandy, @magcius, @mwyrzykowski)
-
Consider allowing overridden pipeline constants to refer to a non-existent identifier string gpuweb/gpuweb#5112 (4 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
[Immediate Data] Add internal slot [[immediate data]] in GPUBindingCommandsMixin gpuweb/gpuweb#5117 (4 comments by @kainino0x, @Kangz, @shaoboyan091)
-
mapAsync() algorithm's `deviceLost` variable is uninitialized on validation error gpuweb/gpuweb#5101 (4 comments by @kainino0x, @Kangz)
-
Move isFallbackAdapter into GPUAdapterInfo gpuweb/gpuweb#4971 (4 comments by @beaufortfrancois, @Kangz)
-
[Immediates] Support `SetImmediateData()` in `RenderBundle` gpuweb/gpuweb#5118 (3 comments by @kainino0x, @Kangz)
-
mapSync on Workers - and possibly on the main thread gpuweb/gpuweb#2217 (3 comments by @aringleinfigma, @Morglod, @toji)
-
Make the last 2 arguments to `copyBufferToBuffer` optional gpuweb/gpuweb#4807 (3 comments by @kainino0x, @Kangz)
-
WGSL textureDimensions allows lod to be passed to 1D texture, but WebGPU doesn't support mipmaps on 1D textures gpuweb/gpuweb#5110 (3 comments by @jrprice, @mwyrzykowski)
-
Feature request: Support GPUBuffer in GPUBindingResource gpuweb/gpuweb#5095 (3 comments by @beaufortfrancois, @magcius)
-
Make template enumerants context-dependent names gpuweb/gpuweb#5082 (3 comments by @dneto0, @jimblandy)
-
"shader-f16" requirements exclude all Qualcomm devices gpuweb/gpuweb#5006 (3 comments by @dneto0, @Kangz, @kdashg)
-
Compat: maxStorage(Buffers/Textures)In(Vertex/Fragment)Stage should be more like `maxStorage(Buffers/Textures)PerShaderStage`? gpuweb/gpuweb#5033 (3 comments by @kainino0x, @Kangz, @teoxoy)
-
[Immediate Data] Rename <immediate_data> to <immediate> gpuweb/gpuweb#5119 (2 comments by @Kangz, @mwyrzykowski)
-
[Immediate Data] Shall we support per-stage immediate data? gpuweb/gpuweb#5116 (2 comments by @Kangz, @kdashg)
-
Proposal: add WGSL instruction `workgroupUniformAtomicLoad` gpuweb/gpuweb#5124 (2 comments by @dj2, @jowens)
-
Improve source size calculation used by `copyExternalImageToTexture` with `HTMLImageElement` gpuweb/gpuweb#5097 (2 comments by @emilio, @Kangz)
-
mappedAtCreation size alignment is checked after mapping already happened gpuweb/gpuweb#5105 (2 comments by @kainino0x, @Kangz)
-
WorkgroupUniformLoad can return fixed-footprint arrays instead of creation-fixed-footprint arrays gpuweb/gpuweb#5093 (2 comments by @dneto0, @petermcneeleychromium)
-
Integrate with proposed ECMAScript Explicit Resource Management gpuweb/gpuweb#5132 (1 comment by @kainino0x)
-
Compat: raise maxStorage(Buffers/Texture)InFragmentStage from 0 to 4 gpuweb/gpuweb#5125 (1 comment by @Kangz)
-
texture builtins: For most texture builtins, Vulkan only allows compile-time-constant offset, and with with bounded range gpuweb/gpuweb#1235 (1 comment by @dneto0)
-
Compat: maxStorageBuffersPerShaderStage doesn't need to be lowered gpuweb/gpuweb#5103 (1 comment by @Kangz)
-
Explicitly state how to handle an if statement with an empty else in uniformity analysis gpuweb/gpuweb#5122 (1 comment by @JamesLee-Jones)
-
wgsl: add way to get special values for a given numeric type (e.g. max, min, lowest) gpuweb/gpuweb#3431 (1 comment by @alienself)
-
Compatibility Mode: GL_MAX_COMBINED_* limits in OpenGL ES gpuweb/gpuweb#5003 (1 comment by @SenorBlanco)
-
Making uniformity analysis less conservative via statement behaviour analysis gpuweb/gpuweb#5100 (1 comment by @afd)
-
minutiae: What happens to GPUColors outside the range of f32? gpuweb/gpuweb#5020 (1 comment by @Kangz)
-
Compat: Limits added for compat need to behave in a forward compatible way gpuweb/gpuweb#5013 (1 comment by @Kangz)
-
Proposal: fully explicit, non-exclusive "auto" layouts in WGSL (without createPipelineLayout) gpuweb/gpuweb#4957 (1 comment by @kdashg)
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043 (1 comment by @kdashg)
-
Proposal: "auto" layout algorithm should be able to generate any layout gpuweb/gpuweb#4956 (1 comment by @kdashg)
-
Clarify (lack of) guarantees provided by timestamp queries (was: Define the clock and time origin of timestamp queries) gpuweb/gpuweb#4361 (1 comment by @kainino0x)
-
What should limits related to features repo on the device when the feature is not enabled? gpuweb/gpuweb#4776 (1 comment by @kainino0x)
-
Allow `GPUBufferUsage.QUERY_RESOLVE | GPUBufferUsage.MAP_READ` buffer usage gpuweb/gpuweb#4383 (1 comment by @kdashg)
-
maxDynamicFoo should be required to be at most maxFoo in limits gpuweb/gpuweb#5069 (1 comment by @Kangz)
-
-
12 pull requests opened.
-
Compat: Raise maxStorageTexturesInFragment and maxStorageBuffersInFragmentStage to 4 gpuweb/gpuweb#5131, by @greggman.
-
workgroupUniformLoad: returns concrete constructible type gpuweb/gpuweb#5120, by @dneto0.
-
Add device.simulateLoss(), and prevent mappedAtCreation on destroyed devices gpuweb/gpuweb#5115, by @kainino0x.
-
mapAsync always AbortErrors after device loss gpuweb/gpuweb#5114, by @kainino0x.
-
Synchronously validate size % 4 == 0 on mappedAtCreation gpuweb/gpuweb#5113, by @kainino0x.
-
Allow creating mappable buffer with more usages as optional features gpuweb/gpuweb#5108, by @Jiawei-Shao.
-
Add clamp-to-border-color sampler address mode gpuweb/gpuweb#5107, by @beaufortfrancois.
-
Clarify "wrapping behaviour" in `dot4I8Packed` gpuweb/gpuweb#5106, by @robamler.
-
Compat: Un-lower maxStorageBuffersPerShaderStage gpuweb/gpuweb#5104, by @kainino0x.
-
Move GPUAdapter isFallbackAdapter attribute to GPUAdapterInfo gpuweb/gpuweb#5099, by @beaufortfrancois.
-
copyBufferToBuffer: make offsets and sizes optional gpuweb/gpuweb#5098, by @kainino0x.
-
WGSL: fix typo gpuweb/gpuweb#5094, by @bbbbx.
-
-
10 pull requests merged.
-
Compat: Un-lower maxStorageBuffersPerShaderStage gpuweb/gpuweb#5104.
-
workgroupUniformLoad: returns concrete constructible type gpuweb/gpuweb#5120.
-
Synchronously validate size % 4 == 0 on mappedAtCreation gpuweb/gpuweb#5113.
-
mapAsync always AbortErrors after device loss gpuweb/gpuweb#5114.
-
Allow using a texture view for an externalTexture binding gpuweb/gpuweb#5079.
-
copyBufferToBuffer: make offsets and sizes optional gpuweb/gpuweb#5098.
-
Move GPUAdapter isFallbackAdapter attribute to GPUAdapterInfo gpuweb/gpuweb#5099.
-
wgsl: Remove binding_array from reserved keywords gpuweb/gpuweb#5091.
-
WGSL: fix typo gpuweb/gpuweb#5094.
-
1 pull request closed.
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613.
-
-
15 pull requests received 49 new comments.
-
Allow using a texture view for an externalTexture binding gpuweb/gpuweb#5079 (14 comments by @beaufortfrancois, @kainino0x, @Kangz)
-
Allow creating mappable buffer with more usages as optional features gpuweb/gpuweb#5108 (12 comments by @Jiawei-Shao, @jimblandy, @kainino0x, @Kangz, @mwyrzykowski)
-
Add device.simulateLoss(), and prevent mappedAtCreation on destroyed devices gpuweb/gpuweb#5115 (5 comments by @kainino0x, @Kangz)
-
Move GPUAdapter isFallbackAdapter attribute to GPUAdapterInfo gpuweb/gpuweb#5099 (4 comments by @kainino0x, @Kangz)
-
wgsl: Remove binding_array from reserved keywords gpuweb/gpuweb#5091 (2 comments by @dneto0, @Kangz)
-
Compat: Raise maxStorageTexturesInFragment and maxStorageBuffersInFragmentStage to 4 gpuweb/gpuweb#5131 (1 comment by @kainino0x)
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613 (1 comment by @kainino0x)
-
Compat: remove adapter.featureLevel; add "core-features-and-limits" gpuweb/gpuweb#5036 (1 comment by @Kangz)
-
TR publications
-
4 documents published.
-
WebGPU Shading Language published 3 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
WebGPU published as a Candidate Recommendation Draft. History of WebGPU
-
GitHub
gpuweb/cts
-
7 issues opened.
-
Refactor Mixins and other inheritance gpuweb/cts#4226, by @greggman.
-
Add tests for local constants in array size gpuweb/cts#4190, by @sagudev.
-
wgsl: float to int conversion: now maps to a value in original type gpuweb/cts#4189, by @dneto0.
-
Compat: Remove testing of integer format multi-sampling gpuweb/cts#4187, by @vonture.
-
Refactor format_info.ts and related texture functions gpuweb/cts#4181, by @greggman.
-
Switch most of the CTS to ask for all features gpuweb/cts#4178, by @greggman.
-
`logResult[n].extra` fails to serialize in worker tests gpuweb/cts#4172, by @ErichDonGubler.
-
-
2 issues closed.
-
Compat: Remove testing of integer format multi-sampling gpuweb/cts#4187.
-
Add wgslLanguageFeatures test gpuweb/cts#4139.
-
-
1 issue reopened.
-
Compat: Remove testing of integer format multi-sampling gpuweb/cts#4187, by @vonture.
-
-
5 issues received 8 new comments.
-
Refactor format_info.ts and related texture functions gpuweb/cts#4181 (3 comments by @greggman, @kainino0x)
-
`logResult[n].extra` fails to serialize in worker tests gpuweb/cts#4172 (1 comment by @ErichDonGubler)
-
-
66 pull requests opened.
-
Refactor shader/validation/statement/* gpuweb/cts#4242, by @greggman.
-
Refactor shader/validation/shader_io/* gpuweb/cts#4241, by @greggman.
-
Refactor shader/validation/(funcitons|parse) gpuweb/cts#4240, by @greggman.
-
Refactor idl/* gpuweb/cts#4239, by @greggman.
-
Refactor web_platform/* gpuweb/cts#4238, by @greggman.
-
Refactor util/* gpuweb/cts#4237, by @greggman.
-
Refactor shader/validation/* s///2 gpuweb/cts#4236, by @greggman.
-
Refactor shader/validation/* s///3 gpuweb/cts#4235, by @greggman.
-
Refactor shader/validation/* gpuweb/cts#4234, by @greggman.
-
Refactor shader/execution/expression/call/builtin/[q-z]* gpuweb/cts#4233, by @greggman.
-
Fix maxDynamic(Storage|Uniform)BuffersPerPipelineLayout tests gpuweb/cts#4232, by @greggman.
-
Fix maxDynamicStorageBuffersPerPipelineLayout test gpuweb/cts#4231, by @greggman.
-
Refactor shader/execution/expression/call/builtin/[e-p]* gpuweb/cts#4230, by @greggman.
-
Refactor shader/exesrc/webgpu/shader/execution/expression/call/builtin/[a-d]* gpuweb/cts#4229, by @greggman.
-
core-features-and-limits update gpuweb/cts#4228, by @shrekshao.
-
Fixes for refactor gpuweb/cts#4227, by @greggman.
-
Refactor shader/execution/expression/call/builtin/texture* gpuweb/cts#4225, by @greggman.
-
Make ShaderValidationTest inherit from AllFeaturesMaxLimitsGPUTest gpuweb/cts#4224, by @greggman.
-
Refactor shader/execution/* gpuweb/cts#4223, by @greggman.
-
Refactor shader/execution/statement gpuweb/cts#4222, by @greggman.
-
fix some spelling errors gpuweb/cts#4221, by @greggman.
-
Refactor shader/execution/shader_io gpuweb/cts#4220, by @greggman.
-
Refactor shader/execution/flow_control gpuweb/cts#4219, by @greggman.
-
Refactor src/webpgu/api/validation/encoding and a few extra gpuweb/cts#4217, by @greggman.
-
Refactor shader/execution/expression/unary gpuweb/cts#4216, by @greggman.
-
Fix refactors gpuweb/cts#4215, by @greggman.
-
Refactor shader/execution/expression/constructor gpuweb/cts#4214, by @greggman.
-
Refactor shader/execution/expression/binary/* gpuweb/cts#4213, by @greggman.
-
Refactor shader/execution/expression/access/* gpuweb/cts#4212, by @greggman.
-
Refactor src/webgpu/shader/validation/decl/* gpuweb/cts#4211, by @greggman.
-
Refactor shader/execution/expression/call/builtin/atomics/* gpuweb/cts#4210, by @greggman.
-
Refactor api/validation/buffer gpuweb/cts#4209, by @greggman.
-
Refactor api/operation/sampling gpuweb/cts#4208, by @greggman.
-
Refact api/operation/command_buffer/* for texture formats gpuweb/cts#4207, by @greggman.
-
Refactor api/operations/resource_init gpuweb/cts#4206, by @greggman.
-
Refactor api/operation/rendering/* gpuweb/cts#4205, by @greggman.
-
Refactor api/operation/render_pipeline for texture formats gpuweb/cts#4204, by @greggman.
-
Change tests to AllFeaturesAndMaxLimitsGPUTest gpuweb/cts#4203, by @greggman.
-
Refactor feature tests for texture formats. gpuweb/cts#4202, by @greggman.
-
Refactor api/operation/texture_view/* for texture formats gpuweb/cts#4201, by @greggman.
-
Refactor opreation/render_pass/* for texture_formats gpuweb/cts#4200, by @greggman.
-
Refactor textureDimensions and textureLoad tests for texture formats gpuweb/cts#4199, by @greggman.
-
Refactor compat tests for texture formats gpuweb/cts#4198, by @greggman.
-
Refactor render_pass_descriptor tests for texture formats. gpuweb/cts#4197, by @greggman.
-
Refactor create view for texture formats gpuweb/cts#4196, by @greggman.
-
Refactor copyTextureToTexture test for texture formats gpuweb/cts#4195, by @greggman.
-
Fix textureSampleCompareLevel sample point code. gpuweb/cts#4194, by @greggman.
-
Refactor createBindGroup validation test for texture formats gpuweb/cts#4193, by @greggman.
-
Rename deprecated texture info functions and add new ones. gpuweb/cts#4192, by @greggman.
-
Skip textureNumSamples tests on non-multisampled formats gpuweb/cts#4191, by @vonture.
-
Compat: Disable multisampling tests on integer formats gpuweb/cts#4188, by @vonture.
-
Compat: Fix pipeline creation test gpuweb/cts#4186, by @greggman.
-
Skip color-space-conversion test if no ImageData gpuweb/cts#4185, by @greggman.
-
Add `AllFeaturesMaxLimitsGPUTest` gpuweb/cts#4184, by @greggman.
-
Add override test for short circuiting in arrays gpuweb/cts#4183, by @petermcneeleychromium.
-
Fix validation tests on the index range of `getBindGroupLayout()` gpuweb/cts#4182, by @Jiawei-Shao.
-
Fix rgba32float on compat related comment gpuweb/cts#4180, by @greggman.
-
Remove checks for skipIfColorRenderableNotSupportedForFormat gpuweb/cts#4179, by @greggman.
-
Fix readTextureToTexelViews for cubemaps gpuweb/cts#4177, by @greggman.
-
Test mip levels with textureStore gpuweb/cts#4176, by @greggman.
-
Test SameObject properties gpuweb/cts#4175, by @greggman.
-
Compat: adjust texture-compression capabilities gpuweb/cts#4174, by @shrekshao.
-
Move skip and skipIf to SubcaseBatchState gpuweb/cts#4173, by @greggman.
-
Use getGPU instead of navigator.gpu in maxStorageTexturesPerShaderSta… gpuweb/cts#4171, by @Kangz.
-
Override short circuiting in functions gpuweb/cts#4170, by @petermcneeleychromium.
-
-
62 pull requests merged.
-
Refactor util/* gpuweb/cts#4237.
-
Refactor shader/execution/expression/call/builtin/texture* gpuweb/cts#4225.
-
Refactor web_platform/* gpuweb/cts#4238.
-
core-features-and-limits update gpuweb/cts#4228.
-
Refactor shader/execution/expression/call/builtin/[e-p]* gpuweb/cts#4230.
-
Refactor shader/validation/* s///3 gpuweb/cts#4235.
-
Refactor shader/validation/* s///2 gpuweb/cts#4236.
-
Fix maxDynamic(Storage|Uniform)BuffersPerPipelineLayout tests gpuweb/cts#4232.
-
Refactor shader/execution/expression/call/builtin/[q-z]* gpuweb/cts#4233.
-
Refactor shader/validation/* gpuweb/cts#4234.
-
Fix maxDynamicStorageBuffersPerPipelineLayout test gpuweb/cts#4231.
-
Fixes for refactor gpuweb/cts#4227.
-
fix some spelling errors gpuweb/cts#4221.
-
Refactor shader/execution/expression/call/builtin/atomics/* gpuweb/cts#4210.
-
Refactor shader/execution/* gpuweb/cts#4223.
-
Refactor shader/execution/statement gpuweb/cts#4222.
-
Refactor shader/execution/shader_io gpuweb/cts#4220.
-
Refactor shader/execution/flow_control gpuweb/cts#4219.
-
Make ShaderValidationTest inherit from AllFeaturesMaxLimitsGPUTest gpuweb/cts#4224.
-
Refactor shader/execution/expression/unary gpuweb/cts#4216.
-
Refactor shader/execution/expression/constructor gpuweb/cts#4214.
-
Refactor shader/execution/expression/binary/* gpuweb/cts#4213.
-
Refactor shader/execution/expression/access/* gpuweb/cts#4212.
-
Refactor src/webgpu/shader/validation/decl/* gpuweb/cts#4211.
-
Refactor api/validation/buffer gpuweb/cts#4209.
-
Refactor api/operation/sampling gpuweb/cts#4208.
-
Refact api/operation/command_buffer/* for texture formats gpuweb/cts#4207.
-
Fix refactors gpuweb/cts#4215.
-
Refactor api/operation/texture_view/* for texture formats gpuweb/cts#4201.
-
Refactor api/operations/resource_init gpuweb/cts#4206.
-
Refactor api/operation/rendering/* gpuweb/cts#4205.
-
Refactor api/operation/render_pipeline for texture formats gpuweb/cts#4204.
-
Change tests to AllFeaturesAndMaxLimitsGPUTest gpuweb/cts#4203.
-
Refactor feature tests for texture formats. gpuweb/cts#4202.
-
Skip textureNumSamples tests on non-multisampled formats gpuweb/cts#4191.
-
Refactor opreation/render_pass/* for texture_formats gpuweb/cts#4200.
-
Refactor textureDimensions and textureLoad tests for texture formats gpuweb/cts#4199.
-
Refactor compat tests for texture formats gpuweb/cts#4198.
-
Refactor render_pass_descriptor tests for texture formats. gpuweb/cts#4197.
-
Refactor createBindGroup validation test for texture formats gpuweb/cts#4193.
-
Refactor create view for texture formats gpuweb/cts#4196.
-
Refactor copyTextureToTexture test for texture formats gpuweb/cts#4195.
-
Fix textureSampleCompareLevel sample point code. gpuweb/cts#4194.
-
Rename deprecated texture info functions and add new ones. gpuweb/cts#4192.
-
Compat: Disable multisampling tests on integer formats gpuweb/cts#4188.
-
Compat: Fix pipeline creation test gpuweb/cts#4186.
-
Add `AllFeaturesMaxLimitsGPUTest` gpuweb/cts#4184.
-
Skip color-space-conversion test if no ImageData gpuweb/cts#4185.
-
Add override test for short circuiting in arrays gpuweb/cts#4183.
-
Remove checks for skipIfColorRenderableNotSupportedForFormat gpuweb/cts#4179.
-
Fix validation tests on the index range of `getBindGroupLayout()` gpuweb/cts#4182.
-
Fix rgba32float on compat related comment gpuweb/cts#4180.
-
Loosen Viewport validation tests to match spec update gpuweb/cts#4106.
-
Fix readTextureToTexelViews for cubemaps gpuweb/cts#4177.
-
Test mip levels with textureStore gpuweb/cts#4176.
-
Test SameObject properties gpuweb/cts#4175.
-
Move skip and skipIf to SubcaseBatchState gpuweb/cts#4173.
-
Override short circuiting in functions gpuweb/cts#4170.
-
Test properties that are supposed to be readonly are. gpuweb/cts#4169.
-
-
2 pull requests closed.
-
Refactor src/webpgu/api/validation/encoding and a few extra gpuweb/cts#4217.
-
-
17 pull requests received 39 new comments.
-
Remove checks for skipIfColorRenderableNotSupportedForFormat gpuweb/cts#4179 (6 comments by @greggman, @shrekshao)
-
Skip textureNumSamples tests on non-multisampled formats gpuweb/cts#4191 (5 comments by @greggman, @kainino0x, @vonture)
-
Refactor copyTextureToTexture test for texture formats gpuweb/cts#4195 (4 comments by @greggman, @shrekshao)
-
Add coverage for overrides in nested arrays and overrides in array members of structures gpuweb/cts#4218 (3 comments by @jrprice, @petermcneeleychromium)
-
Compat: remove select device for float16(32)-renderable features gpuweb/cts#4166 (3 comments by @greggman, @shrekshao)
-
Refact api/operation/command_buffer/* for texture formats gpuweb/cts#4207 (2 comments by @greggman, @shrekshao)
-
Rename deprecated texture info functions and add new ones. gpuweb/cts#4192 (2 comments by @greggman, @kainino0x)
-
Skip color-space-conversion test if no ImageData gpuweb/cts#4185 (2 comments by @greggman, @shrekshao)
-
Compat: adjust texture-compression capabilities gpuweb/cts#4174 (2 comments by @greggman, @shrekshao)
-
Fix maxDynamic(Storage|Uniform)BuffersPerPipelineLayout tests gpuweb/cts#4232 (1 comment by @shrekshao)
-
Refactor src/webpgu/api/validation/encoding and a few extra gpuweb/cts#4217 (1 comment by @greggman)
-
gpuweb/gpuweb
-
19 issues opened.
-
binding_array should not be a reserved word gpuweb/gpuweb#5088, by @jrprice.
-
What happens to canvas with unpremultplied alpha when you call toDataURL or toBlob gpuweb/gpuweb#5087, by @greggman.
-
WGSL example "component reference from a composite reference" has a typo gpuweb/gpuweb#5086, by @brendan-duncan.
-
Sized Binding Arrays: Uniformity of accesses gpuweb/gpuweb#5085, by @Kangz.
-
Sized Binding Arrays: Decide what can be arrayed and the syntax. gpuweb/gpuweb#5084, by @Kangz.
-
WGSL Pointer example has invalid code gpuweb/gpuweb#5083, by @brendan-duncan.
-
Make template enumerants context-dependent names gpuweb/gpuweb#5082, by @k2d222.
-
Refactor @builtin docs? gpuweb/gpuweb#5081, by @greggman.
-
Add localization support for GPUCompilationMessage.message and GPUError.message gpuweb/gpuweb#5078, by @kainino0x.
-
rg11b10ufloat-renderable makes rg11b10ufloat resolvable? gpuweb/gpuweb#5075, by @greggman.
-
Actually explain that WebGPU is automatically synchronized gpuweb/gpuweb#5074, by @kainino0x.
-
The documentation for the `refract` intrinsic does not mention normalization gpuweb/gpuweb#5072, by @armansito.
-
Add limited support for 64 bit atomics gpuweb/gpuweb#5071, by @petermcneeleychromium.
-
[WESL] Host-visible names that shadow predeclared names gpuweb/gpuweb#5070, by @stefnotch.
-
maxDynamicFoo should be required to be at most maxFoo in limits gpuweb/gpuweb#5069, by @Kangz.
-
Cubemap local coordinate selection is not in the specification gpuweb/gpuweb#5065, by @magcius.
-
[WebGPU] Index resolution is potentially underspecified gpuweb/gpuweb#5064, by @mwyrzykowski.
-
Can not pass a single f16 scalar to shader due to buffer size must be aligned to 4 bytes gpuweb/gpuweb#5062, by @zcbenz.
-
Add builtin swap function gpuweb/gpuweb#5060, by @zcbenz.
-
-
20 issues closed.
-
WGSL Pointer example has invalid code gpuweb/gpuweb#5083.
-
WGSL example "component reference from a composite reference" has a typo gpuweb/gpuweb#5086.
-
The documentation for the `refract` intrinsic does not mention normalization gpuweb/gpuweb#5072.
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043.
-
Internationalization Metadata for Human-readable Messages gpuweb/gpuweb#2780.
-
rg11b10ufloat-renderable makes rg11b10ufloat resolvable? gpuweb/gpuweb#5075.
-
Problem measuring time deltas with writeTimestamp gpuweb/gpuweb#3952.
-
Expose optional texture capabilities in WebGPU gpuweb/gpuweb#2630.
-
16-bit normalized texture support gpuweb/gpuweb#3001.
-
Collaborate with the immersive web groups for WebXR interop gpuweb/gpuweb#2778.
-
Multisample Coverage on Metal gpuweb/gpuweb#959.
-
Allow getting adapter information from GPUDevice gpuweb/gpuweb#4810.
-
Resurrect vertex formats dropped for macOS < 10.13 gpuweb/gpuweb#4549.
-
Compat: renderable float16, float32 textures gpuweb/gpuweb#4417.
-
Incomplete information about varying behaviours gpuweb/gpuweb#5008.
-
copyBufferToBuffer : incorrect warning when source == target gpuweb/gpuweb#5045.
-
Add missing enums for WGSLLanguageFeatureName gpuweb/gpuweb#5057.
-
-
28 issues received 70 new comments.
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043 (7 comments by @aleino-nv, @dneto0, @mwyrzykowski, @petermcneeleychromium)
-
Internationalization Metadata for Human-readable Messages gpuweb/gpuweb#2780 (5 comments by @aphillips, @kainino0x)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (5 comments by @greggman, @gszep, @kainino0x, @T3sT3ro)
-
Collaborate with the immersive web groups for WebXR interop gpuweb/gpuweb#2778 (5 comments by @alienself, @kainino0x, @Kangz, @mikemainguy)
-
Can not pass a single f16 scalar to shader due to buffer size must be aligned to 4 bytes gpuweb/gpuweb#5062 (5 comments by @Kangz, @magcius, @shaoboyan091, @zcbenz)
-
Cubemap local coordinate selection is not in the specification gpuweb/gpuweb#5065 (4 comments by @kainino0x, @magcius)
-
Sized Binding Arrays: Decide what can be arrayed and the syntax. gpuweb/gpuweb#5084 (2 comments by @alan-baker, @Kangz)
-
Make the last 2 arguments to `copyBufferToBuffer` optional gpuweb/gpuweb#4807 (2 comments by @kainino0x)
-
[WebGPU] Index resolution is potentially underspecified gpuweb/gpuweb#5064 (2 comments by @kainino0x)
-
maxDynamicFoo should be required to be at most maxFoo in limits gpuweb/gpuweb#5069 (2 comments by @greggman, @kainino0x)
-
WGSL example "component reference from a composite reference" has a typo gpuweb/gpuweb#5086 (1 comment by @mehmetoguzderin)
-
What happens to canvas with unpremultplied alpha when you call toDataURL or toBlob gpuweb/gpuweb#5087 (1 comment by @kainino0x)
-
rg11b10ufloat-renderable makes rg11b10ufloat resolvable? gpuweb/gpuweb#5075 (1 comment by @kainino0x)
-
Allow using a (float32-filterable 2d single level) texture view for an externalTexture binding. gpuweb/gpuweb#4504 (1 comment by @kainino0x)
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987 (1 comment by @Kangz)
-
wgsl: support atomics on f32 (in workgroup and storage buffer) gpuweb/gpuweb#4894 (1 comment by @petermcneeleychromium)
-
10 pull requests opened.
-
Reformat builtins table gpuweb/gpuweb#5092, by @greggman.
-
wgsl: Remove binding_array from reserved keywords gpuweb/gpuweb#5091, by @jrprice.
-
Fix code in pointer-as-short-name example gpuweb/gpuweb#5090, by @dneto0.
-
Fix variable naming in WGSL example gpuweb/gpuweb#5089, by @mehmetoguzderin.
-
Mention normalization in `refract` gpuweb/gpuweb#5080, by @mehmetoguzderin.
-
Allow using a texture view for an externalTexture binding gpuweb/gpuweb#5079, by @beaufortfrancois.
-
Fix prose for rg11b10ufloat-renderable to allow resolving gpuweb/gpuweb#5077, by @kainino0x.
-
Handle negative index in vertex processing gpuweb/gpuweb#5076, by @kainino0x.
-
float to int conversion maps to value in original type gpuweb/gpuweb#5073, by @dneto0.
-
Add GPUTextureView support to importExternalTexture() gpuweb/gpuweb#5068, by @beaufortfrancois.
-
-
8 pull requests merged.
-
Fix code in pointer-as-short-name example gpuweb/gpuweb#5090.
-
Fix variable naming in WGSL example gpuweb/gpuweb#5089.
-
Mention normalization in `refract` gpuweb/gpuweb#5080.
-
float to int conversion maps to value in original type gpuweb/gpuweb#5073.
-
Fix prose for rg11b10ufloat-renderable to allow resolving gpuweb/gpuweb#5077.
-
Compat: remove adapter.featureLevel; add "core-features-and-limits" gpuweb/gpuweb#5036.
-
Compat: Require capabilities provided by GL_EXT_color_buffer_float gpuweb/gpuweb#5055.
-
Define WGSL assertions as their own kind gpuweb/gpuweb#5054.
-
-
1 pull request closed.
-
Add GPUTextureView support to importExternalTexture() gpuweb/gpuweb#5068.
-
-
10 pull requests received 29 new comments.
-
Allow using a texture view for an externalTexture binding gpuweb/gpuweb#5079 (9 comments by @beaufortfrancois, @kainino0x)
-
Add GPUTextureView support to importExternalTexture() gpuweb/gpuweb#5068 (7 comments by @beaufortfrancois, @kainino0x, @Kangz)
-
Define WGSL assertions as their own kind gpuweb/gpuweb#5054 (5 comments by @alan-baker, @mehmetoguzderin)
-
Compat: remove adapter.featureLevel; add "core-features-and-limits" gpuweb/gpuweb#5036 (2 comments by @greggman, @kainino0x)
-
Lifecycle
TR publications
-
5 documents published.
-
WebGPU published twice as a Candidate Recommendation Draft. History of WebGPU
-
WebGPU Shading Language published 3 times as a Candidate Recommendation Draft. History of WebGPU Shading Language
-
Participants
-
4 participants left.
-
Zeyu Guo (Agora.io) left.
-
yang lee (Agora.io) left.
-
Chun Gao (Agora.io) left.
-
Zhaoming Jiang (Intel Corporation) left.
-
-
1 organization left.
-
Agora.io left.
-
GitHub
gpuweb/cts
-
4 issues opened.
-
refactor textureSampleBias tests gpuweb/cts#4153, by @greggman.
-
Add more JS/IDL tests gpuweb/cts#4142, by @greggman.
-
Add wgslLanguageFeatures test gpuweb/cts#4139, by @sagudev.
-
Switch enforce-default-limits flag to use limits table gpuweb/cts#4132, by @greggman.
-
-
2 issues closed.
-
Add options to force MaxLimitsTestMixin to default gpuweb/cts#4120.
-
5 issues received 5 new comments.
-
39 pull requests opened.
-
Test properties that are supposed to be readonly are. gpuweb/cts#4169, by @greggman.
-
Compat: update maxXXXInYYYStage tests gpuweb/cts#4168, by @greggman.
-
Compat: is multisampled skips for rgba16float and r32float fixes gpuweb/cts#4167, by @shrekshao.
-
Compat: remove select device for float16(32)-renderable features gpuweb/cts#4166, by @shrekshao.
-
Test 1 layer 2d-array textures. gpuweb/cts#4165, by @greggman.
-
Check more JavaScript features. gpuweb/cts#4164, by @greggman.
-
Compat: fix float16(32)-renderable tests (more in validation, shader,… gpuweb/cts#4163, by @shrekshao.
-
Compat: fix float16(32)-renderable tests (compat) gpuweb/cts#4162, by @shrekshao.
-
Compat: fix float16(32)-renderable tests (web_platform) gpuweb/cts#4161, by @shrekshao.
-
Compat: fix float16(32)-renderable tests (shader,execution) gpuweb/cts#4160, by @shrekshao.
-
Compat: fix more float16(32)-renderable tests gpuweb/cts#4159, by @shrekshao.
-
Fix bug: fragment output type mismatch gpuweb/cts#4158, by @shrekshao.
-
Test reading mip levels with textureLoad and 2d textures. gpuweb/cts#4157, by @greggman.
-
Skip web_platform tests if API does not exist gpuweb/cts#4156, by @greggman.
-
Compat: Fix pipeline_bind_group_compat.spec.ts for 0 storage bufs/texs gpuweb/cts#4155, by @greggman.
-
Compat: fix texture,in_pass_encoder for 0 storage textures gpuweb/cts#4154, by @greggman.
-
Compat: fix float16(32)-renderable tests (validation) gpuweb/cts#4152, by @shrekshao.
-
Test a bunch of JavaScript stuff. gpuweb/cts#4151, by @greggman.
-
Refactor selectDeviceOrSkipTest system gpuweb/cts#4150, by @greggman.
-
Compat: Refactor draw test for 0 storage buffers gpuweb/cts#4149, by @greggman.
-
Compat: fix float16(32)-renderable tests gpuweb/cts#4148, by @shrekshao.
-
Fix mem-sync same_subresource tests for 0 storage buffers. gpuweb/cts#4147, by @greggman.
-
Compat: Fix canvas configure test for 0 storagebuffers gpuweb/cts#4146, by @greggman.
-
Remove unsupoorted alignment in execution tests gpuweb/cts#4145, by @petermcneeleychromium.
-
Add max_combined_limits operation test. gpuweb/cts#4144, by @greggman.
-
Remove api/operation/texture_view/read.spec.ts gpuweb/cts#4143, by @greggman.
-
Adjust print_environment for node gpuweb/cts#4141, by @greggman.
-
Update cmdline.ts and server.ts help strings. gpuweb/cts#4140, by @greggman.
-
Make it easier to click options. gpuweb/cts#4138, by @greggman.
-
Refactor maxStorage(Buffers/Textures)PerShaderStage tests gpuweb/cts#4137, by @greggman.
-
Fix limits tests with new extra limits gpuweb/cts#4136, by @greggman.
-
Add multiple format image files for copyToTexture,image_file cases gpuweb/cts#4135, by @shaoboyan091.
-
Ignore empty bind group layout when checking setBindGroup against the… gpuweb/cts#4134, by @Jiawei-Shao.
-
Test maxStorage(Buffers/Textures)In(Fragment/Vertex)Stage gpuweb/cts#4133, by @greggman.
-
Update code that uses adapter.isCompatibilityMode gpuweb/cts#4131, by @kainino0x.
-
Update print_environment to show featureLevel and defaultDevice gpuweb/cts#4130, by @kainino0x.
-
Compat: Refactor subgroup tests for 0 storage buffers in frag gpuweb/cts#4129, by @greggman.
-
Compat: Refactor fwidth/Fine/Coarse for 0 storage buffers. gpuweb/cts#4128, by @greggman.
-
Add texture_builtin_notes.md gpuweb/cts#4127, by @greggman.
-
-
45 pull requests merged.
-
Compat: update maxXXXInYYYStage tests gpuweb/cts#4168.
-
Compat: is multisampled skips for rgba16float and r32float fixes gpuweb/cts#4167.
-
Check more JavaScript features. gpuweb/cts#4164.
-
Compat: remove select device for float16(32)-renderable features gpuweb/cts#4166.
-
Test 1 layer 2d-array textures. gpuweb/cts#4165.
-
Compat: fix more float16(32)-renderable tests gpuweb/cts#4159.
-
Compat: fix float16(32)-renderable tests (shader,execution) gpuweb/cts#4160.
-
Compat: fix float16(32)-renderable tests (compat) gpuweb/cts#4162.
-
Compat: fix float16(32)-renderable tests (web_platform) gpuweb/cts#4161.
-
Fix bug: fragment output type mismatch gpuweb/cts#4158.
-
Test reading mip levels with textureLoad and 2d textures. gpuweb/cts#4157.
-
Skip web_platform tests if API does not exist gpuweb/cts#4156.
-
Compat: fix texture,in_pass_encoder for 0 storage textures gpuweb/cts#4154.
-
Compat: fix float16(32)-renderable tests (validation) gpuweb/cts#4152.
-
Test a bunch of JavaScript stuff. gpuweb/cts#4151.
-
Compat: fix float16(32)-renderable tests gpuweb/cts#4148.
-
Refactor selectDeviceOrSkipTest system gpuweb/cts#4150.
-
Compat: Refactor draw test for 0 storage buffers gpuweb/cts#4149.
-
Compat: Fix canvas configure test for 0 storagebuffers gpuweb/cts#4146.
-
Fix mem-sync same_subresource tests for 0 storage buffers. gpuweb/cts#4147.
-
Add max_combined_limits operation test. gpuweb/cts#4144.
-
Add multiple format image files for copyToTexture,image_file cases gpuweb/cts#4135.
-
Remove unsupoorted alignment in execution tests gpuweb/cts#4145.
-
Test maxStorage(Buffers/Textures)In(Fragment/Vertex)Stage gpuweb/cts#4133.
-
Remove api/operation/texture_view/read.spec.ts gpuweb/cts#4143.
-
Refactor maxStorage(Buffers/Textures)PerShaderStage tests gpuweb/cts#4137.
-
Adjust print_environment for node gpuweb/cts#4141.
-
Update cmdline.ts and server.ts help strings. gpuweb/cts#4140.
-
Make it easier to click options. gpuweb/cts#4138.
-
Update print_environment to show featureLevel and defaultDevice gpuweb/cts#4130.
-
Compat: refactor state_tracking test for 0 frag buffers. gpuweb/cts#4112.
-
Fix limits tests with new extra limits gpuweb/cts#4136.
-
Compat: Refactor fwidth/Fine/Coarse for 0 storage buffers. gpuweb/cts#4128.
-
Update code that uses adapter.isCompatibilityMode gpuweb/cts#4131.
-
Compat: refactor storage_texture tests gpuweb/cts#4099.
-
Add "enforceDefaultLimits" flag gpuweb/cts#4124.
-
CopyEI2T: Cases To Cover Rotated Images gpuweb/cts#4111.
-
Compat: Refactor subgroup tests for 0 storage buffers in frag gpuweb/cts#4129.
-
Add texture_builtin_notes.md gpuweb/cts#4127.
-
Compat: createBGL, createPipelineLayout, etc... gpuweb/cts#4121.
-
-
28 pull requests received 154 new comments.
-
Compat: fix float16(32)-renderable tests (more in validation, shader,… gpuweb/cts#4163 (12 comments by @greggman, @shrekshao)
-
Compat: fix float16(32)-renderable tests (validation) gpuweb/cts#4152 (12 comments by @greggman, @shrekshao)
-
Compat: Add tests for maxStorage(Buffer/Texture)sIn(Vertex/Fragment)Stage gpuweb/cts#4118 (10 comments by @greggman, @kainino0x)
-
Add multiple format image files for copyToTexture,image_file cases gpuweb/cts#4135 (7 comments by @kainino0x, @shaoboyan091)
-
Compat: fix float16(32)-renderable tests (shader,execution) gpuweb/cts#4160 (6 comments by @greggman, @shrekshao)
-
Add operation test on the creation of pipeline layout with null bind group layout gpuweb/cts#4116 (5 comments by @Jiawei-Shao, @kainino0x)
-
Compat: refactor state_tracking test for 0 frag buffers. gpuweb/cts#4112 (4 comments by @greggman, @shrekshao)
-
Update print_environment to show featureLevel and defaultDevice gpuweb/cts#4130 (4 comments by @beaufortfrancois, @kainino0x)
-
Compat: Fix pipeline_bind_group_compat.spec.ts for 0 storage bufs/texs gpuweb/cts#4155 (3 comments by @greggman, @shrekshao)
-
Compat: Refactor fwidth/Fine/Coarse for 0 storage buffers. gpuweb/cts#4128 (3 comments by @greggman, @jrprice)
-
Test maxStorage(Buffers/Textures)In(Fragment/Vertex)Stage gpuweb/cts#4133 (2 comments by @greggman, @shrekshao)
-
Compat: remove select device for float16(32)-renderable features gpuweb/cts#4166 (1 comment by @shrekshao)
-
Ignore empty bind group layout when checking setBindGroup against the… gpuweb/cts#4134 (1 comment by @Jiawei-Shao)
gpuweb/gpuweb
-
6 issues opened.
-
API spec build broken on Bikeshed 5.0.0 gpuweb/gpuweb#5058, by @kainino0x.
-
Add missing enums for WGSLLanguageFeatureName gpuweb/gpuweb#5057, by @beaufortfrancois.
-
EXT_color_buffer_float provides less support than WebGPU Core gpuweb/gpuweb#5049, by @kainino0x.
-
copyBufferToBuffer : incorrect warning when source == target gpuweb/gpuweb#5045, by @tlecoz.
-
[editorial] subgroupMin/Max nan behavior gpuweb/gpuweb#5044, by @alan-baker.
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043, by @mwyrzykowski.
-
-
4 issues closed.
-
API spec build broken on Bikeshed 5.0.0 gpuweb/gpuweb#5058.
-
Viewport rect must lie entirely within the current attachment size? gpuweb/gpuweb#373.
-
EXT_color_buffer_float provides less support than WebGPU Core gpuweb/gpuweb#5049.
-
Validation requires vertex buffers to be set, even if they are unused gpuweb/gpuweb#4999.
-
-
2 issues reopened.
-
API spec build broken on Bikeshed 5.0.0 gpuweb/gpuweb#5058, by @mehmetoguzderin.
-
Compat: renderable float16, float32 textures gpuweb/gpuweb#4417, by @kainino0x.
-
-
29 issues received 65 new comments.
-
Can float -> int conversions be relaxed at upper/lower bounds? gpuweb/gpuweb#5043 (13 comments by @aleino-nv, @jimblandy, @kdashg, @mwyrzykowski, @petermcneeleychromium)
-
EXT_color_buffer_float provides less support than WebGPU Core gpuweb/gpuweb#5049 (6 comments by @kainino0x, @shrekshao)
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987 (5 comments by @kainino0x, @Kangz)
-
API spec build broken on Bikeshed 5.0.0 gpuweb/gpuweb#5058 (4 comments by @kainino0x, @mehmetoguzderin)
-
copyBufferToBuffer : incorrect warning when source == target gpuweb/gpuweb#5045 (4 comments by @greggman, @kainino0x, @tlecoz)
-
Proposal: fully explicit, non-exclusive "auto" layouts in WGSL (without createPipelineLayout) gpuweb/gpuweb#4957 (3 comments by @jimblandy, @Kangz, @kdashg)
-
Restrictions on pointers passed to functions are not documented where they should be gpuweb/gpuweb#5032 (3 comments by @kdashg, @rcoreilly)
-
Compat: renderable float16, float32 textures gpuweb/gpuweb#4417 (3 comments by @kainino0x, @SenorBlanco)
-
Viewport rect must lie entirely within the current attachment size? gpuweb/gpuweb#373 (2 comments by @floooh, @Kangz)
-
Compatibility mode: support devices without integer texture format multisampling gpuweb/gpuweb#5004 (2 comments by @Kangz)
-
texel format `T` for `texture_2d<T>` should be of type `GPUTextureFormat` gpuweb/gpuweb#5023 (1 comment by @kdashg)
-
Validation requires vertex buffers to be set, even if they are unused gpuweb/gpuweb#4999 (1 comment by @Kangz)
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513 (1 comment by @Kangz)
-
Compat: Disallow usage of depth textures with non-compare sampling functions. gpuweb/gpuweb#4970 (1 comment by @Kangz)
-
Compat: Should requesting compat mode on a device that supports core, return core or compat? gpuweb/gpuweb#4731 (1 comment by @Kangz)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (1 comment by @Kangz)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (1 comment by @Kangz)
-
wgsl: Sketch design for arrays of buffer-bindings gpuweb/gpuweb#2105 (1 comment by @devshgraphicsprogramming)
-
Support for gl_PrimitiveID / SV_PrimitiveID -like construct? gpuweb/gpuweb#1786 (1 comment by @coffeenotfound)
-
Fake transfers for getMappedRange-created ArrayBuffers? gpuweb/gpuweb#2072 (1 comment by @kainino0x)
-
List of issues to discuss with TC39, WASM, JS engine teams gpuweb/gpuweb#747 (1 comment by @kainino0x)
-
-
18 pull requests opened.
-
Update Bikeshed to 5.0.2 gpuweb/gpuweb#5059, by @mehmetoguzderin.
-
[editorial] Compat: Move IDL for textureBindingViewDimension into the appropriate section gpuweb/gpuweb#5056, by @kainino0x.
-
Compat: Require capabilities provided by GL_EXT_color_buffer_float gpuweb/gpuweb#5055, by @kainino0x.
-
Define WGSL assertions as their own kind gpuweb/gpuweb#5054, by @mehmetoguzderin.
-
Compat: disallow multisampled integer textures gpuweb/gpuweb#5053, by @kainino0x.
-
Compat: disallow multisampled f16/f32 textures gpuweb/gpuweb#5052, by @kainino0x.
-
Re-pin Bikeshed version to make build-validate-publish work gpuweb/gpuweb#5051, by @kainino0x.
-
Fix API spec build errors gpuweb/gpuweb#5050, by @mehmetoguzderin.
-
wgsl: further clarify override expr evaluation only if in the shader gpuweb/gpuweb#5048, by @dneto0.
-
Fix incorrect end tags gpuweb/gpuweb#5047, by @kainino0x.
-
Fix tags that cause build errors in the WGSL spec gpuweb/gpuweb#5046, by @mehmetoguzderin.
-
Allow `index < limits.maxBindGroups` in `getBindGroupLayout()` gpuweb/gpuweb#5042, by @Jiawei-Shao.
-
Adjust metadata and workflow to resume publication to /TR gpuweb/gpuweb#5041, by @tidoust.
-
Add placeholder Deadlines to specs as a workaround for Bikeshed problem gpuweb/gpuweb#5040, by @kainino0x.
-
Correspondence: document Vulkan's `maxFragmentCombinedOutputResources` gpuweb/gpuweb#5039, by @kainino0x.
-
Move Correspondence Reference from LD to UD, to fix CSS gpuweb/gpuweb#5038, by @kainino0x.
-
Compat: remove GPUAdapter.featureLevel gpuweb/gpuweb#5037, by @kainino0x.
-
Compat: remove adapter.featureLevel; add "core-features-and-limits" gpuweb/gpuweb#5036, by @kainino0x.
-
-
17 pull requests merged.
-
Update Bikeshed to 5.0.2 gpuweb/gpuweb#5059.
-
Loosen viewport validation requirements gpuweb/gpuweb#5025.
-
Compat: disallow multisampled f16/f32 textures gpuweb/gpuweb#5052.
-
Compat: disallow multisampled integer textures gpuweb/gpuweb#5053.
-
Add a proposal for sized-binding-arrays gpuweb/gpuweb#4940.
-
Compat: update new limits proposal gpuweb/gpuweb#5034.
-
Re-pin Bikeshed version to make build-validate-publish work gpuweb/gpuweb#5051.
-
wgsl: further clarify override expr evaluation only if in the shader gpuweb/gpuweb#5048.
-
Allow `index < limits.maxBindGroups` in `getBindGroupLayout()` gpuweb/gpuweb#5042.
-
Fix incorrect end tags gpuweb/gpuweb#5047.
-
Fix tags that cause build errors in the WGSL spec gpuweb/gpuweb#5046.
-
Add subgroups feature gpuweb/gpuweb#4963.
-
Adjust metadata and workflow to resume publication to /TR gpuweb/gpuweb#5041.
-
Correspondence: document Vulkan's `maxFragmentCombinedOutputResources` gpuweb/gpuweb#5039.
-
Move Correspondence Reference from LD to UD, to fix CSS gpuweb/gpuweb#5038.
-
Add placeholder Deadlines to specs as a workaround for Bikeshed problem gpuweb/gpuweb#5040.
-
-
3 pull requests closed.
-
Compat: remove GPUAdapter.featureLevel gpuweb/gpuweb#5037.
-
Fix API spec build errors gpuweb/gpuweb#5050.
-
Compat: describe as a featureLevel below core, upgradeable to core gpuweb/gpuweb#4898.
-
-
20 pull requests received 94 new comments.
-
Compat: update new limits proposal gpuweb/gpuweb#5034 (21 comments by @greggman, @jimblandy, @kainino0x, @Kangz)
-
Compat: remove adapter.featureLevel; add "core-features-and-limits" gpuweb/gpuweb#5036 (9 comments by @beaufortfrancois, @greggman, @kainino0x, @Kangz)
-
Compat: remove GPUAdapter.featureLevel gpuweb/gpuweb#5037 (8 comments by @greggman, @kainino0x, @Kangz)
-
Loosen viewport validation requirements gpuweb/gpuweb#5025 (7 comments by @greggman, @kainino0x, @mwyrzykowski, @toji)
-
Fix tags that cause build errors in the WGSL spec gpuweb/gpuweb#5046 (7 comments by @kainino0x, @mehmetoguzderin)
-
Compat: disallow multisampled f16/f32 textures gpuweb/gpuweb#5052 (6 comments by @greggman, @kainino0x, @SenorBlanco, @shrekshao)
-
Allow `index < limits.maxBindGroups` in `getBindGroupLayout()` gpuweb/gpuweb#5042 (6 comments by @Jiawei-Shao, @kainino0x, @Kangz, @mehmetoguzderin)
-
Compat mode: make f16 and f32 rendering optional. gpuweb/gpuweb#4983 (4 comments by @kainino0x, @Kangz, @kdashg, @shrekshao)
-
Re-pin Bikeshed version to make build-validate-publish work gpuweb/gpuweb#5051 (3 comments by @kainino0x, @mehmetoguzderin)
-
Compat: disallow multisampled integer textures gpuweb/gpuweb#5053 (2 comments by @kainino0x, @kdashg)
-
Add placeholder Deadlines to specs as a workaround for Bikeshed problem gpuweb/gpuweb#5040 (1 comment by @kainino0x)
-
Compat: describe as a featureLevel below core, upgradeable to core gpuweb/gpuweb#4898 (1 comment by @kainino0x)
-