GPU for the Web Working Group - Activity in 2024
No matches.
TR publications
-
5 documents published.
-
WebGPU published as a Candidate Recommendation Snapshot. History of WebGPU
-
WebGPU Shading Language published as a Candidate Recommendation Snapshot. History of WebGPU Shading Language
-
WebGPU published twice as a Working Draft. History of WebGPU
-
WebGPU Shading Language published as a Working Draft. History of WebGPU Shading Language
-
Participants
-
5 participants joined.
-
Karen Ghavam (LunarG, Inc.) joined.
-
Shaobo Yan (Microsoft Corporation) joined.
-
Yang Gu (Microsoft Corporation) joined.
-
Mark Young (LunarG, Inc.) joined.
-
David Desormeaux (LunarG, Inc.) joined.
-
-
1 participant left.
-
Austin Eng (Google LLC) left.
-
-
1 organization joined.
-
LunarG, Inc. joined.
-
GitHub
gpuweb/cts
-
6 issues opened.
-
Deno 2.x can't run the test suite gpuweb/cts#4123, by @turbocrime.
-
Add options to force MaxLimitsTestMixin to default gpuweb/cts#4120, by @greggman.
-
Test CopyEI2T with Image has rotation metadata gpuweb/cts#4108, by @shaoboyan091.
-
Emit worker tests for WPT gpuweb/cts#4098, by @ErichDonGubler.
-
Add tests about creating pipeline layout with empty bind group layout gpuweb/cts#4075, by @Jiawei-Shao.
-
-
1 issue closed.
-
Emit worker tests for WPT gpuweb/cts#4098.
-
-
1 issue received 4 new comments.
-
51 pull requests opened.
-
Compat: refactor resource_usages tests for 0 storage buffer/textures gpuweb/cts#4126, by @greggman.
-
Compat: refactor setBindGroup for 0 storage buffers in frag stage. gpuweb/cts#4125, by @greggman.
-
Add "enforceDefaultLimits" flag gpuweb/cts#4124, by @greggman.
-
Compat: refactor maxStorage(Buffers/Textures)PerShaderStage test gpuweb/cts#4122, by @greggman.
-
Compat: createBGL, createPipelineLayout, etc... gpuweb/cts#4121, by @greggman.
-
Fix file path too long error gpuweb/cts#4119, by @greggman.
-
Compat: Add tests for maxStorage(Buffer/Texture)sIn(Vertex/Fragment)Stage gpuweb/cts#4118, by @greggman.
-
Create bind group gpuweb/cts#4117, by @greggman.
-
Compat: Fix texture-utils to for 0 storage buffers in frag stage gpuweb/cts#4115, by @greggman.
-
Test using a pipeline created with null bind group layout gpuweb/cts#4113, by @Jiawei-Shao.
-
Compat: refactor state_tracking test for 0 frag buffers. gpuweb/cts#4112, by @greggman.
-
CopyEI2T: Cases To Cover Rotated Images gpuweb/cts#4111, by @shaoboyan091.
-
Test creating pipeline with pipeline layout created with null bind gr… gpuweb/cts#4110, by @Jiawei-Shao.
-
Vertex state correctness gpuweb/cts#4109, by @greggman.
-
Reduce the number of subcases in null_bind_group_layouts gpuweb/cts#4107, by @kainino0x.
-
Loosen Viewport validation tests to match spec update gpuweb/cts#4106, by @toji.
-
Fragment shader tests for subgroup add and mul gpuweb/cts#4105, by @alan-baker.
-
Compat: Skip texture sync tests that use storage textures if 0 gpuweb/cts#4104, by @greggman.
-
Fix DeviceModifier gpuweb/cts#4103, by @greggman.
-
gen_wpt_cts_html: Add support for filtering the list of variants gpuweb/cts#4102, by @kainino0x.
-
Compat: skip memory_sync buffer tests when no support gpuweb/cts#4101, by @greggman.
-
Compat: make texture_view:write handle 0 storage textures gpuweb/cts#4100, by @greggman.
-
Compat: refactor storage_texture tests gpuweb/cts#4099, by @greggman.
-
Compat: refactor pipeline_bind_group_compat test for 0 storage buffers gpuweb/cts#4097, by @greggman.
-
Compat: refactor depth_clip_clamp for 0 storage buffers gpuweb/cts#4096, by @greggman.
-
Compat: update pipeline_bind_group_compat for 0 storage bufs/textures gpuweb/cts#4095, by @greggman.
-
Compat: skip in-pass-encoder tests if not enough storage textures gpuweb/cts#4094, by @greggman.
-
Refactor discard test for no storage buffers gpuweb/cts#4093, by @greggman.
-
Add some extra test cases for smoothstep low > high gpuweb/cts#4092, by @petermcneeleychromium.
-
Refactor derivative tests for 0 storage buffers. gpuweb/cts#4091, by @greggman.
-
Refactor texture builtin for 0 vert/frag storage bufs/tex. gpuweb/cts#4090, by @greggman.
-
Execution tests for subgroupElect gpuweb/cts#4089, by @alan-baker.
-
Compat: adjust for 0 vert/frag storage buffers/textures gpuweb/cts#4088, by @greggman.
-
Remove many test cases that never run because maxColorAttachments=8 gpuweb/cts#4087, by @kainino0x.
-
Execution tests for subgroupMin and subgroupMax gpuweb/cts#4086, by @alan-baker.
-
Move MaxLimitsTest to MaxLimitTestMixin gpuweb/cts#4085, by @greggman.
-
Expand valid input for smoothstep to include low > high gpuweb/cts#4084, by @petermcneeleychromium.
-
Remove subgroups f16 enable from cts ("subgroups_f16") gpuweb/cts#4083, by @petermcneeleychromium.
-
Set `featureLevel: "compatibility"` when compatibility=1 is enabled gpuweb/cts#4081, by @beaufortfrancois.
-
Reapply "Use subgroupMinSize, subgroupMaxSize from GPUAdapterInfo" (#… gpuweb/cts#4080, by @dneto0.
-
Revert "Use subgroupMinSize, subgroupMaxSize from GPUAdapterInfo" gpuweb/cts#4079, by @dneto0.
-
Fix texture builtin weight reading gpuweb/cts#4078, by @greggman.
-
Add tests for GPURequestAdapterOptions featureLevel gpuweb/cts#4077, by @beaufortfrancois.
-
Add back unicode for texture diagrams gpuweb/cts#4074, by @greggman.
-
Texture Utils, log bad weights only if debugging gpuweb/cts#4073, by @greggman.
-
Move towards deprecating subgroups_f16 enable (feature gpuweb/cts#4072, by @petermcneeleychromium.
-
[wgsl] Add two more statement behavior test cases gpuweb/cts#4071, by @jrprice.
-
Improve fragment shader checks in subgroupBitwise gpuweb/cts#4070, by @alan-baker.
-
-
45 pull requests merged.
-
Compat: refactor setBindGroup for 0 storage buffers in frag stage. gpuweb/cts#4125.
-
Compat: refactor resource_usages tests for 0 storage buffer/textures gpuweb/cts#4126.
-
Compat: refactor maxStorage(Buffers/Textures)PerShaderStage test gpuweb/cts#4122.
-
Fix file path too long error gpuweb/cts#4119.
-
Compat: Fix texture-utils to for 0 storage buffers in frag stage gpuweb/cts#4115.
-
Create bind group gpuweb/cts#4117.
-
Test using a pipeline created with null bind group layout gpuweb/cts#4113.
-
Fragment shader tests for subgroup add and mul gpuweb/cts#4105.
-
Vertex state correctness gpuweb/cts#4109.
-
Reduce the number of subcases in null_bind_group_layouts gpuweb/cts#4107.
-
Compat: skip memory_sync buffer tests when no support gpuweb/cts#4101.
-
Compat: Skip texture sync tests that use storage textures if 0 gpuweb/cts#4104.
-
Compat: refactor depth_clip_clamp for 0 storage buffers gpuweb/cts#4096.
-
Fix DeviceModifier gpuweb/cts#4103.
-
gen_wpt_cts_html: Add support for filtering the list of variants gpuweb/cts#4102.
-
Refactor discard test for no storage buffers gpuweb/cts#4093.
-
Compat: make texture_view:write handle 0 storage textures gpuweb/cts#4100.
-
Compat: skip in-pass-encoder tests if not enough storage textures gpuweb/cts#4094.
-
Refactor texture builtin for 0 vert/frag storage bufs/tex. gpuweb/cts#4090.
-
Refactor derivative tests for 0 storage buffers. gpuweb/cts#4091.
-
Add some extra test cases for smoothstep low > high gpuweb/cts#4092.
-
Execution tests for subgroupElect gpuweb/cts#4089.
-
Execution tests for subgroupMin and subgroupMax gpuweb/cts#4086.
-
Compat: adjust for 0 vert/frag storage buffers/textures gpuweb/cts#4088.
-
Set `featureLevel: "compatibility"` when compatibility=1 is enabled gpuweb/cts#4081.
-
Remove subgroups f16 enable from cts ("subgroups_f16") gpuweb/cts#4083.
-
Remove many test cases that never run because maxColorAttachments=8 gpuweb/cts#4087.
-
Expand valid input for smoothstep to include low > high gpuweb/cts#4084.
-
Move MaxLimitsTest to MaxLimitTestMixin gpuweb/cts#4085.
-
Add back unicode for texture diagrams gpuweb/cts#4074.
-
Add tests for GPURequestAdapterOptions featureLevel gpuweb/cts#4077.
-
Move towards deprecating subgroups_f16 enable (feature gpuweb/cts#4072.
-
Revert "Use subgroupMinSize, subgroupMaxSize from GPUAdapterInfo" gpuweb/cts#4079.
-
Use minSubgroupSize, maxSubgroupSize from GPUAdapterInfo gpuweb/cts#4058.
-
Fix texture builtin weight reading gpuweb/cts#4078.
-
Texture Utils, log bad weights only if debugging gpuweb/cts#4073.
-
Improve fragment shader checks in subgroupBitwise gpuweb/cts#4070.
-
Add validation of subgroup_uniformity diagnostic gpuweb/cts#4066.
-
[wgsl] Add two more statement behavior test cases gpuweb/cts#4071.
-
-
27 pull requests received 83 new comments.
-
Loosen Viewport validation tests to match spec update gpuweb/cts#4106 (12 comments by @greggman, @kainino0x, @toji)
-
Set `featureLevel: "compatibility"` when compatibility=1 is enabled gpuweb/cts#4081 (11 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Compat: adjust for 0 vert/frag storage buffers/textures gpuweb/cts#4088 (8 comments by @greggman, @kainino0x)
-
Test using a pipeline created with null bind group layout gpuweb/cts#4113 (7 comments by @jiangzhaoming, @Jiawei-Shao, @kainino0x)
-
Remove subgroups f16 enable from cts ("subgroups_f16") gpuweb/cts#4083 (5 comments by @alan-baker, @petermcneeleychromium)
-
Add validation test about creating pipeline layout with null bind group layout gpuweb/cts#4076 (4 comments by @Jiawei-Shao, @kainino0x)
-
Refactor derivative tests for 0 storage buffers. gpuweb/cts#4091 (4 comments by @greggman, @jrprice)
-
Compat: refactor depth_clip_clamp for 0 storage buffers gpuweb/cts#4096 (2 comments by @greggman, @kainino0x)
-
Refactor texture builtin for 0 vert/frag storage bufs/tex. gpuweb/cts#4090 (2 comments by @shrekshao)
-
Compat: update pipeline_bind_group_compat for 0 storage bufs/textures gpuweb/cts#4095 (2 comments by @shrekshao)
-
Add operation test on the creation of pipeline layout with null bind group layout gpuweb/cts#4116 (1 comment by @jiangzhaoming)
-
Test creating pipeline with pipeline layout created with null bind gr… gpuweb/cts#4110 (1 comment by @jiangzhaoming)
-
Compat: refactor pipeline_bind_group_compat test for 0 storage buffers gpuweb/cts#4097 (1 comment by @shrekshao)
-
Add some extra test cases for smoothstep low > high gpuweb/cts#4092 (1 comment by @petermcneeleychromium)
-
Smoothstep validation changes : No compiler error for anything except low == high (for const/override) gpuweb/cts#4082 (1 comment by @petermcneeleychromium)
-
Reapply "Use subgroupMinSize, subgroupMaxSize from GPUAdapterInfo" (#… gpuweb/cts#4080 (1 comment by @dneto0)
-
Move towards deprecating subgroups_f16 enable (feature gpuweb/cts#4072 (1 comment by @petermcneeleychromium)
-
gpuweb/gpuweb
-
16 issues opened.
-
Restrictions on pointers passed to functions are not documented where they should be gpuweb/gpuweb#5032, by @jimblandy.
-
Should texture_depth_multisampled_2d have a template parameter? gpuweb/gpuweb#5028, by @afd.
-
Scope of initializer vs. body in "for" statement gpuweb/gpuweb#5024, by @afd.
-
texel format `T` for `texture_2d<T>` should be of type `GPUTextureFormat` gpuweb/gpuweb#5023, by @gszep.
-
minutiae: What happens to GPUColors outside the range of f32? gpuweb/gpuweb#5020, by @kainino0x.
-
Allow aliasing of function declarations gpuweb/gpuweb#5016, by @kdashg.
-
Clarify that `alias` names can be used as constructor function names gpuweb/gpuweb#5014, by @k2d222.
-
Compat: Limits added for compat need to behave in a forward compatible way gpuweb/gpuweb#5013, by @greggman.
-
Incomplete information about varying behaviours gpuweb/gpuweb#5008, by @elliotsayes.
-
"shader-f16" requirements exclude all Qualcomm devices gpuweb/gpuweb#5006, by @toji.
-
Account for Vulkan's framebufferIntegerColorSampleCounts? gpuweb/gpuweb#5005, by @kainino0x.
-
Compatibility mode: support devices without integer texture format multisampling gpuweb/gpuweb#5004, by @SenorBlanco.
-
Compatibility Mode: GL_MAX_COMBINED_* limits in OpenGL ES gpuweb/gpuweb#5003, by @SenorBlanco.
-
1 issue closed.
-
Should texture_depth_multisampled_2d have a template parameter? gpuweb/gpuweb#5028.
-
-
33 issues received 78 new comments.
-
Compatibility mode: support devices without integer texture format multisampling gpuweb/gpuweb#5004 (11 comments by @greggman, @jimblandy, @kainino0x, @kdashg, @SenorBlanco)
-
Compatibility Mode: GL_MAX_COMBINED_* limits in OpenGL ES gpuweb/gpuweb#5003 (5 comments by @kainino0x, @SenorBlanco)
-
Compat: Limits added for compat need to behave in a forward compatible way gpuweb/gpuweb#5013 (4 comments by @greggman, @kainino0x, @teoxoy)
-
Account for Vulkan's framebufferIntegerColorSampleCounts? gpuweb/gpuweb#5005 (4 comments by @kainino0x, @teoxoy)
-
Compat: maxStorage(Buffers/Textures)In(Vertex/Fragment)Stage should be more like `maxStorage(Buffers/Textures)PerShaderStage`? gpuweb/gpuweb#5033 (3 comments by @greggman, @kainino0x, @teoxoy)
-
Proposal: "auto" layout algorithm should be able to generate any layout gpuweb/gpuweb#4956 (3 comments by @kainino0x, @kdashg)
-
Clarify that `alias` names can be used as constructor function names gpuweb/gpuweb#5014 (3 comments by @dneto0, @kdashg)
-
Proposal: fully explicit, non-exclusive "auto" layouts in WGSL (without createPipelineLayout) gpuweb/gpuweb#4957 (3 comments by @iwoplaza, @kdashg, @mighdoll)
-
Create timestamp-query2 for tiled architectures gpuweb/gpuweb#4995 (3 comments by @greggman, @mwyrzykowski)
-
"shader-f16" requirements exclude all Qualcomm devices gpuweb/gpuweb#5006 (3 comments by @dneto0, @jrprice, @kdashg)
-
No way to sample a `texture_1d` from a vertex shader gpuweb/gpuweb#5001 (3 comments by @jimblandy, @kainino0x, @kdashg)
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987 (3 comments by @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Should texture_depth_multisampled_2d have a template parameter? gpuweb/gpuweb#5028 (2 comments by @afd, @greggman)
-
texel format `T` for `texture_2d<T>` should be of type `GPUTextureFormat` gpuweb/gpuweb#5023 (2 comments by @greggman, @kainino0x)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (2 comments by @kainino0x, @SenorBlanco)
-
Allow Early Fragment Tests to be forced gpuweb/gpuweb#4891 (2 comments by @leopoldoizquierdoinvestors, @magcius)
-
Viewport rect must lie entirely within the current attachment size? gpuweb/gpuweb#373 (2 comments by @floooh, @kdashg)
-
Remove `maxFragmentCombinedOutputResources`? gpuweb/gpuweb#4018 (2 comments by @kainino0x, @SenorBlanco)
-
List of issues to discuss with TC39, WASM, JS engine teams gpuweb/gpuweb#747 (1 comment by @erights)
-
Restrictions on pointers passed to functions are not documented where they should be gpuweb/gpuweb#5032 (1 comment by @alan-baker)
-
minutiae: What happens to GPUColors outside the range of f32? gpuweb/gpuweb#5020 (1 comment by @kainino0x)
-
additional flags `--enable-features=Vulkan,VulkanFromANGLE,DefaultANGLEVulkan` required for chrome on linux gpuweb/gpuweb#5022 (1 comment by @kainino0x)
-
Validation requires vertex buffers to be set, even if they are unused gpuweb/gpuweb#4999 (1 comment by @jimblandy)
-
-
17 pull requests opened.
-
Compat: update new limits proposal gpuweb/gpuweb#5034, by @greggman.
-
Call out `texture_multisampled_2d` parameterization gpuweb/gpuweb#5030, by @greggman.
-
Compat: fix GL computation of storage limits gpuweb/gpuweb#5029, by @kainino0x.
-
Fix maxTextureDimension limits refs gpuweb/gpuweb#5027, by @beaufortfrancois.
-
Update subgroups CTS status gpuweb/gpuweb#5026, by @alan-baker.
-
Loosen viewport validation requirements gpuweb/gpuweb#5025, by @toji.
-
Add attributes and settings for consistent line endings gpuweb/gpuweb#5021, by @mehmetoguzderin.
-
Update Subgroup CTS status gpuweb/gpuweb#5019, by @alan-baker.
-
Add and name Features for all Compat Mode limitations. gpuweb/gpuweb#5018, by @SenorBlanco.
-
[M0 backport] Fix typos in refs to adapter capability guarantees section (#5015) gpuweb/gpuweb#5017, by @kainino0x.
-
Fix typos in refs to adapter capability guarantees section gpuweb/gpuweb#5015, by @tidoust.
-
Compatibility Mode: update to use featureLevel. gpuweb/gpuweb#5012, by @SenorBlanco.
-
Explainer: Update Adapter Selection and Device Init section gpuweb/gpuweb#5011, by @beaufortfrancois.
-
Add new fragment storage buffer, texture limits. gpuweb/gpuweb#5010, by @SenorBlanco.
-
Subgroups: clarify subgroups operations using f16 gpuweb/gpuweb#5009, by @jiangzhaoming.
-
Subgroups: add note for subgroupMinSize gpuweb/gpuweb#5007, by @jiangzhaoming.
-
Update CTS status for subgroups feature gpuweb/gpuweb#5002, by @alan-baker.
-
-
13 pull requests merged.
-
Call out `texture_multisampled_2d` parameterization gpuweb/gpuweb#5030.
-
texel buffer proposal: Update from review gpuweb/gpuweb#4985.
-
Update subgroups CTS status gpuweb/gpuweb#5026.
-
Fix maxTextureDimension limits refs gpuweb/gpuweb#5027.
-
Compatibility Mode: update to use featureLevel. gpuweb/gpuweb#5012.
-
Explainer: Update Adapter Selection and Device Init section gpuweb/gpuweb#5011.
-
Add attributes and settings for consistent line endings gpuweb/gpuweb#5021.
-
Update Subgroup CTS status gpuweb/gpuweb#5019.
-
[M0 backport] Fix typos in refs to adapter capability guarantees section (#5015) gpuweb/gpuweb#5017.
-
Fix typos in refs to adapter capability guarantees section gpuweb/gpuweb#5015.
-
Add new fragment storage buffer, texture limits. gpuweb/gpuweb#5010.
-
Compat mode: make f16 and f32 rendering optional. gpuweb/gpuweb#4983.
-
Update CTS status for subgroups feature gpuweb/gpuweb#5002.
-
-
7 pull requests received 79 new comments.
-
Add and name Features for all Compat Mode limitations. gpuweb/gpuweb#5018 (27 comments by @kainino0x, @mwyrzykowski, @SenorBlanco, @teoxoy)
-
Add subgroups feature gpuweb/gpuweb#4963 (23 comments by @alan-baker, @dneto0, @jimblandy, @kainino0x, @kdashg, @mwyrzykowski)
-
Compatibility Mode: update to use featureLevel. gpuweb/gpuweb#5012 (15 comments by @beaufortfrancois, @greggman, @kainino0x, @mwyrzykowski, @SenorBlanco)
-
Compat: fix GL computation of storage limits gpuweb/gpuweb#5029 (5 comments by @greggman, @kainino0x)
-
Add attributes and settings for consistent line endings gpuweb/gpuweb#5021 (2 comments by @kainino0x, @mehmetoguzderin)
-
Explainer: Update Adapter Selection and Device Init section gpuweb/gpuweb#5011 (2 comments by @kainino0x)
-
Lifecycle
-
1 charter extension.
-
Charter extended until 31 January 2025.
-
TR publications
-
14 documents published.
-
WebGPU Shading Language published 8 times as a Working Draft. History of WebGPU Shading Language
-
WebGPU published 6 times as a Working Draft. History of WebGPU
-
Participants
-
1 participant joined.
-
Rob Moran (ARM Limited) joined.
-
-
1 participant left.
-
Yang Gu (Microsoft Corporation) left.
-
GitHub
gpuweb/cts
-
7 issues opened.
-
wgsl: test edge0 > edge1 cases gpuweb/cts#4047, by @dneto0.
-
test duplicate (non-conflicting) diagnostics directives gpuweb/cts#4045, by @dneto0.
-
test WGSL builtins in all stages gpuweb/cts#4040, by @greggman.
-
fix vertex_state,correctness tests gpuweb/cts#4033, by @greggman.
-
Port WebGL Wasm conformance tests to WebGPU gpuweb/cts#4030, by @kenrussell.
-
4 issues closed.
-
test duplicate (non-conflicting) diagnostics directives gpuweb/cts#4045.
-
fix vertex_state,correctness tests gpuweb/cts#4033.
-
Fix `identifySamplePoints` for non-float formats gpuweb/cts#4017.
-
-
7 issues received 9 new comments.
-
test duplicate (non-conflicting) diagnostics directives gpuweb/cts#4045 (2 comments by @dneto0, @jrprice)
-
skipIfTextureFormatNotUsableAsStorageTexture does not handle bgra8unorm well gpuweb/cts#3847 (1 comment by @kainino0x)
-
timeout check triggers when expanding a set of tests in standalone gpuweb/cts#4019 (1 comment by @kainino0x)
-
-
37 pull requests opened.
-
Rework subgroup broadcast tests gpuweb/cts#4069, by @alan-baker.
-
subgroupBallot tests in fragment shaders gpuweb/cts#4068, by @alan-baker.
-
Compat: stop using storage buffer when reading texture mix weights gpuweb/cts#4067, by @greggman.
-
Add validation of subgroup_uniformity diagnostic gpuweb/cts#4066, by @alan-baker.
-
Validation tests for early eval errors in subgroup builtins gpuweb/cts#4065, by @alan-baker.
-
Uniformity tests for subgroup built-in functions gpuweb/cts#4064, by @alan-baker.
-
Test texture+shader combo usage. gpuweb/cts#4063, by @greggman.
-
Compat: Test texture+sampler limits are enforced gpuweb/cts#4062, by @greggman.
-
Add tests for subgroupShuffle variants gpuweb/cts#4061, by @alan-baker.
-
Special case `minSubgroupSize` and `maxSubgroupSize` gpuweb/cts#4060, by @greggman.
-
Add MaxLimitsTest that sets all limits to the adapter's gpuweb/cts#4059, by @greggman.
-
Use minSubgroupSize, maxSubgroupSize from GPUAdapterInfo gpuweb/cts#4058, by @dneto0.
-
Allow unknown limits to be requested with value undefined gpuweb/cts#4057, by @beaufortfrancois.
-
Compat: Validate non-comparsion sampler + depth texture fails. gpuweb/cts#4055, by @greggman.
-
Compat: Don't test depth textures with non-comparison samplers gpuweb/cts#4054, by @greggman.
-
Add warning in checkCallResults for compressed textures gpuweb/cts#4053, by @greggman.
-
wip: textureLoad + aspect for importExternalTexture gpuweb/cts#4052, by @Kangz.
-
Test texture mix weights separately gpuweb/cts#4051, by @greggman.
-
Do not query weights for `textureSampleBaseClampToEdge` gpuweb/cts#4050, by @greggman.
-
Fix readTextureToTextureViews for multisampled textures gpuweb/cts#4049, by @greggman.
-
Fix makeRandomDepthComparisonTexelGenerator gpuweb/cts#4048, by @greggman.
-
Fix bgra8unorm storage handling gpuweb/cts#4046, by @wailin247.
-
Try to relax textureSampleBias gpuweb/cts#4044, by @greggman.
-
[Compat] Skip depthBiasClamp !== 0 cases gpuweb/cts#4043, by @shrekshao.
-
Expand textureStorage tests gpuweb/cts#4039, by @greggman.
-
Fix tests related to subgroup uniformity diagnostics gpuweb/cts#4038, by @alan-baker.
-
Add Stages and ExternalTexture to textureDimension tests gpuweb/cts#4037, by @greggman.
-
test textureNumLevels/Layers/Samples with stages gpuweb/cts#4036, by @greggman.
-
Fragment subgroup_size test: Don't use broadcasting gpuweb/cts#4035, by @jiangzhaoming.
-
Assert GPUAdapterInfo instances gpuweb/cts#4034, by @beaufortfrancois.
-
Compat: skip 'cube-array' validation tests. gpuweb/cts#4032, by @SenorBlanco.
-
Test non-filterable textures can't be used with filtering sampler. gpuweb/cts#4031, by @greggman.
-
Update webgpu/types to 0.1.50 gpuweb/cts#4029, by @Kangz.
-
Don't test texture_depth_xxx with linear gpuweb/cts#4028, by @greggman.
-
DO NOT SUBMIT: Add coverage of 1-byte component vertex formats and un… gpuweb/cts#4027, by @Kangz.
-
Re-enable linear with comparisons. gpuweb/cts#4026, by @greggman.
-
-
36 pull requests merged.
-
Rework subgroup broadcast tests gpuweb/cts#4069.
-
subgroupBallot tests in fragment shaders gpuweb/cts#4068.
-
Validation tests for early eval errors in subgroup builtins gpuweb/cts#4065.
-
Uniformity tests for subgroup built-in functions gpuweb/cts#4064.
-
Add tests for subgroupShuffle variants gpuweb/cts#4061.
-
Compat: stop using storage buffer when reading texture mix weights gpuweb/cts#4067.
-
Test texture+shader combo usage. gpuweb/cts#4063.
-
Compat: Test texture+sampler limits are enforced gpuweb/cts#4062.
-
Special case `minSubgroupSize` and `maxSubgroupSize` gpuweb/cts#4060.
-
Add MaxLimitsTest that sets all limits to the adapter's gpuweb/cts#4059.
-
Add validation tests for Multi-Draw Indirect gpuweb/cts#3962.
-
Allow unknown limits to be requested with value undefined gpuweb/cts#4057.
-
Compat: Validate non-comparsion sampler + depth texture fails. gpuweb/cts#4055.
-
Compat: Don't test depth textures with non-comparison samplers gpuweb/cts#4054.
-
Add warning in checkCallResults for compressed textures gpuweb/cts#4053.
-
Test texture mix weights separately gpuweb/cts#4051.
-
Do not query weights for `textureSampleBaseClampToEdge` gpuweb/cts#4050.
-
Fix readTextureToTextureViews for multisampled textures gpuweb/cts#4049.
-
Fix makeRandomDepthComparisonTexelGenerator gpuweb/cts#4048.
-
Test fixes for alignment validation landing gpuweb/cts#4016.
-
Try to relax textureSampleBias gpuweb/cts#4044.
-
Fix tests related to subgroup uniformity diagnostics gpuweb/cts#4038.
-
[Compat] Skip depthBiasClamp !== 0 cases gpuweb/cts#4043.
-
Expand textureStorage tests gpuweb/cts#4039.
-
Add Stages and ExternalTexture to textureDimension tests gpuweb/cts#4037.
-
test textureNumLevels/Layers/Samples with stages gpuweb/cts#4036.
-
Fragment subgroup_size test: Don't use broadcasting gpuweb/cts#4035.
-
Assert GPUAdapterInfo instances gpuweb/cts#4034.
-
Add tests for GPUDevice.adapterInfo gpuweb/cts#4023.
-
Compat: skip 'cube-array' validation tests. gpuweb/cts#4032.
-
Test non-filterable textures can't be used with filtering sampler. gpuweb/cts#4031.
-
Update webgpu/types to 0.1.50 gpuweb/cts#4029.
-
Don't test texture_depth_xxx with linear gpuweb/cts#4028.
-
Fix fragment shader subgroup builtin io test gpuweb/cts#4024.
-
Re-enable linear with comparisons. gpuweb/cts#4026.
-
-
2 pull requests closed.
-
Add Rendering tests for Multi Draw Indirect gpuweb/cts#3970.
-
32 pull requests received 153 new comments.
-
Add validation tests for Multi-Draw Indirect gpuweb/cts#3962 (39 comments by @beaufortfrancois, @kainino0x, @Sirtsu55)
-
Test fixes for alignment validation landing gpuweb/cts#4016 (9 comments by @dneto0, @petermcneeleychromium)
-
Fix readTextureToTexelViews to handle all types of formats and viewDimensions gpuweb/cts#4025 (7 comments by @greggman, @shrekshao)
-
Compat: Test texture+sampler limits are enforced gpuweb/cts#4062 (6 comments by @greggman, @shrekshao)
-
Add MaxLimitsTest that sets all limits to the adapter's gpuweb/cts#4059 (5 comments by @greggman, @kainino0x)
-
test textureNumLevels/Layers/Samples with stages gpuweb/cts#4036 (4 comments by @greggman, @shrekshao)
-
Validation tests for early eval errors in subgroup builtins gpuweb/cts#4065 (3 comments by @alan-baker, @petermcneeleychromium)
-
Use minSubgroupSize, maxSubgroupSize from GPUAdapterInfo gpuweb/cts#4058 (3 comments by @alan-baker, @dneto0)
-
Fragment subgroup_size test: Don't use broadcasting gpuweb/cts#4035 (3 comments by @alan-baker, @jiangzhaoming)
-
Allow unknown limits to be requested with value undefined gpuweb/cts#4057 (2 comments by @kainino0x)
-
Compat: Validate non-comparsion sampler + depth texture fails. gpuweb/cts#4055 (2 comments by @greggman, @shrekshao)
-
Compat: Don't test depth textures with non-comparison samplers gpuweb/cts#4054 (2 comments by @greggman, @shrekshao)
-
Do not query weights for `textureSampleBaseClampToEdge` gpuweb/cts#4050 (2 comments by @greggman, @kainino0x)
-
fix(diagnostic): expect success on duplicate validation if severities are the same gpuweb/cts#4042 (2 comments by @alan-baker, @ErichDonGubler)
-
Test non-filterable textures can't be used with filtering sampler. gpuweb/cts#4031 (2 comments by @greggman, @kainino0x)
-
gpuweb/gpuweb
-
22 issues opened.
-
No way to sample a `texture_1d` from a vertex shader gpuweb/gpuweb#5001, by @Tim-Evans-Seequent.
-
Validation requires vertex buffers to be set, even if they are unused gpuweb/gpuweb#4999, by @kainino0x.
-
WGSL loop examples are out of date gpuweb/gpuweb#4996, by @greggman.
-
Create timestamp-query2 for tiled architectures gpuweb/gpuweb#4995, by @greggman.
-
Validation that vertex attributes exist is in the wrong place gpuweb/gpuweb#4994, by @kainino0x.
-
Styling of adapter immutable properties is wrong gpuweb/gpuweb#4990, by @heycam.
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987, by @kainino0x.
-
Compat: Limit the number of sample/texture combinations gpuweb/gpuweb#4986, by @greggman.
-
is `x -- 7` valid? gpuweb/gpuweb#4984, by @k2d222.
-
3+ errors are produced if 'auto' layout algorithm fails gpuweb/gpuweb#4977, by @kainino0x.
-
Does WGSL intentionally permit duplicate global diagnostic filters? gpuweb/gpuweb#4976, by @jimblandy.
-
enhancement: uniform buffer can use standard layout gpuweb/gpuweb#4973, by @dneto0.
-
bool has 4 byte size and alignment gpuweb/gpuweb#4972, by @dneto0.
-
Move isFallbackAdapter into GPUAdapterInfo gpuweb/gpuweb#4971, by @beaufortfrancois.
-
Compat: Disallow usage of depth textures with non-compare sampling functions. gpuweb/gpuweb#4970, by @teoxoy.
-
GPUCommandEncoder.beginRenderPass() allows for empty list of attachments? gpuweb/gpuweb#4968, by @mwyrzykowski.
-
Issue: Data Alignment for Uint16Array in WebGPU gpuweb/gpuweb#4966, by @tianfakuang.
-
offsetOfMember rule forbids an offset of 0. gpuweb/gpuweb#4959, by @dneto0.
-
Proposal: "auto" layout algorithm should be able to generate any layout gpuweb/gpuweb#4956, by @kainino0x.
-
publish-TR-webgpu failing due to bikeshed change gpuweb/gpuweb#4954, by @kainino0x.
-
Change `layout: 'auto'` algorithm so it doesn't generate invalid bindGroupLayouts gpuweb/gpuweb#4952, by @greggman.
-
-
24 issues closed.
-
Compat: Limit the number of sample/texture combinations gpuweb/gpuweb#4986.
-
WGSL loop examples are out of date gpuweb/gpuweb#4996.
-
Validation that vertex attributes exist is in the wrong place gpuweb/gpuweb#4994.
-
Styling of adapter immutable properties is wrong gpuweb/gpuweb#4990.
-
Compat: Disallow usage of depth textures with non-compare sampling functions. gpuweb/gpuweb#4970.
-
Add `subgroups-f16` feature dependency validation to subgroups proposal? gpuweb/gpuweb#4734.
-
Feature request: support one-component vertex attribute format gpuweb/gpuweb#4340.
-
Compat: sample locations don't match the spec in OpenGL gpuweb/gpuweb#4804.
-
Improve documentation for setBindGroup about order of dynamic offsets. gpuweb/gpuweb#4921.
-
bool has 4 byte size and alignment gpuweb/gpuweb#4972.
-
Does WGSL intentionally permit duplicate global diagnostic filters? gpuweb/gpuweb#4976.
-
Fingerprinting Surface gpuweb/gpuweb#3101.
-
Investigation: Querying Subgroup Support gpuweb/gpuweb#78.
-
GPUCommandEncoder.beginRenderPass() allows for empty list of attachments? gpuweb/gpuweb#4968.
-
Issue: Data Alignment for Uint16Array in WebGPU gpuweb/gpuweb#4966.
-
offsetOfMember rule forbids an offset of 0. gpuweb/gpuweb#4959.
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841.
-
publish-TR-webgpu failing due to bikeshed change gpuweb/gpuweb#4954.
-
Dealing with holes in the pipeline layout. gpuweb/gpuweb#2043.
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573.
-
-
2 issues reopened.
-
Resurrect vertex formats dropped for macOS < 10.13 gpuweb/gpuweb#4549, by @kainino0x.
-
Compat: renderable float16, float32 textures gpuweb/gpuweb#4417, by @teoxoy.
-
-
45 issues received 131 new comments.
-
Compat: Disallow usage of depth textures with non-compare sampling functions. gpuweb/gpuweb#4970 (14 comments by @greggman, @kainino0x, @teoxoy)
-
Proposal: fully explicit, non-exclusive "auto" layouts in WGSL (without createPipelineLayout) gpuweb/gpuweb#4957 (8 comments by @alan-baker, @jimblandy, @kainino0x, @kdashg, @magcius, @mighdoll)
-
How to add features and limits for Compat after Core ships? gpuweb/gpuweb#4987 (7 comments by @greggman, @kainino0x, @mwyrzykowski, @SenorBlanco, @teoxoy)
-
enhancement: uniform buffer can use standard layout gpuweb/gpuweb#4973 (7 comments by @dneto0, @kdashg, @magcius)
-
Does WGSL intentionally permit duplicate global diagnostic filters? gpuweb/gpuweb#4976 (7 comments by @alan-baker, @dneto0, @jimblandy, @kdashg)
-
bool has 4 byte size and alignment gpuweb/gpuweb#4972 (7 comments by @dneto0, @greggman, @jimblandy, @kdashg)
-
Asynchronous shader reflection information gpuweb/gpuweb#624 (7 comments by @7ombie, @eliemichel, @greggman, @kainino0x)
-
Compat: sample locations don't match the spec in OpenGL gpuweb/gpuweb#4804 (5 comments by @greggman, @kainino0x, @kdashg)
-
DrawIndirectCount gpuweb/gpuweb#1354 (5 comments by @beaufortfrancois, @kainino0x, @Kangz, @RenaudRohlinger, @Sirtsu55)
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841 (5 comments by @dneto0, @greggman, @kainino0x)
-
smoothstep: please remove error check when edge0 > edge1 for compile-time constants gpuweb/gpuweb#4900 (4 comments by @dneto0, @greggman, @jimblandy, @kdashg)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (4 comments by @greggman, @magcius, @mwyrzykowski)
-
Resurrect vertex formats dropped for macOS < 10.13 gpuweb/gpuweb#4549 (3 comments by @kainino0x, @Kangz)
-
Compat: Should requesting compat mode on a device that supports core, return core or compat? gpuweb/gpuweb#4731 (3 comments by @greggman, @Kangz, @kdashg)
-
GPUCommandEncoder.beginRenderPass() allows for empty list of attachments? gpuweb/gpuweb#4968 (3 comments by @kainino0x, @mwyrzykowski)
-
Collaborate with the immersive web groups for WebXR interop gpuweb/gpuweb#2778 (3 comments by @forEachWhileTrue, @jespertheend, @toji)
-
Compat: Limit the number of sample/texture combinations gpuweb/gpuweb#4986 (2 comments by @kainino0x, @teoxoy)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (2 comments by @kdashg, @teoxoy)
-
Proposal: "auto" layout algorithm should be able to generate any layout gpuweb/gpuweb#4956 (2 comments by @kainino0x, @kdashg)
-
3+ errors are produced if 'auto' layout algorithm fails gpuweb/gpuweb#4977 (2 comments by @greggman, @kainino0x)
-
Issue: Data Alignment for Uint16Array in WebGPU gpuweb/gpuweb#4966 (2 comments by @Kangz, @tianfakuang)
-
Validation requires vertex buffers to be set, even if they are unused gpuweb/gpuweb#4999 (1 comment by @kainino0x)
-
Viewport rect must lie entirely within the current attachment size? gpuweb/gpuweb#373 (1 comment by @kainino0x)
-
Adding DataView to WGSL to read various types of 1-byte-aligned values from a storage buffer. gpuweb/gpuweb#2429 (1 comment by @lifaon74)
-
Styling of adapter immutable properties is wrong gpuweb/gpuweb#4990 (1 comment by @beaufortfrancois)
-
Add `subgroups-f16` feature dependency validation to subgroups proposal? gpuweb/gpuweb#4734 (1 comment by @dneto0)
-
Feature request: support one-component vertex attribute format gpuweb/gpuweb#4340 (1 comment by @Kangz)
-
Investigation: how WebNN / WebGPU interop could be happening gpuweb/gpuweb#2500 (1 comment by @anssiko)
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513 (1 comment by @SenorBlanco)
-
Cannot upload/download to UMA storage buffer without an unnecessary copy and unnecessary memory use gpuweb/gpuweb#2388 (1 comment by @Kangz)
-
Criteria for pushing the WebGPU and WGSL specification to Candidate Recommendation gpuweb/gpuweb#3107 (1 comment by @Kangz)
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426 (1 comment by @Jaisiero)
-
-
28 pull requests opened.
-
[M0 backport] Fix attribute validation (#4998) gpuweb/gpuweb#5000, by @kainino0x.
-
Fix attribute validation gpuweb/gpuweb#4998, by @kainino0x.
-
Change GLSL loop examples to WGSL examples. gpuweb/gpuweb#4997, by @greggman.
-
[M0 backport] Fix styling of adapter immutable properties (#4991) gpuweb/gpuweb#4993, by @kainino0x.
-
Fix host-shareable types note gpuweb/gpuweb#4992, by @alan-baker.
-
Fix styling of adapter immutable properties gpuweb/gpuweb#4991, by @beaufortfrancois.
-
Compat: Limit the number of texture+sampler combinations gpuweb/gpuweb#4989, by @greggman.
-
Compat: Disallow using a depth texture with a non-comparison sampler gpuweb/gpuweb#4988, by @greggman.
-
texel buffer proposal: Update from review gpuweb/gpuweb#4985, by @dneto0.
-
Compat mode: make f16 and f32 rendering optional. gpuweb/gpuweb#4983, by @SenorBlanco.
-
Adding clearer notes and examples for dynamic offset ordering gpuweb/gpuweb#4982, by @toji.
-
wgsl: smoothstep is well defined for edge0 > edge1 gpuweb/gpuweb#4981, by @dneto0.
-
Clarify uniqueness of nearest enclosing diagnostic gpuweb/gpuweb#4980, by @dneto0.
-
wgsl: @align(n) must divide required-align-of, for all structs gpuweb/gpuweb#4978, by @dneto0.
-
Update subgroups proposal gpuweb/gpuweb#4975, by @alan-baker.
-
wgsl: bool has size and alignment of 4 bytes gpuweb/gpuweb#4974, by @dneto0.
-
Plainly state alignment requirement for scalars and vectors gpuweb/gpuweb#4965, by @dneto0.
-
wgsl: fix typo gpuweb/gpuweb#4964, by @dneto0.
-
Add subgroups feature gpuweb/gpuweb#4963, by @alan-baker.
-
WIP: Add subgroups feature gpuweb/gpuweb#4962, by @alan-baker.
-
[M0 backport] wgsl: allow OffsetOfMember to be 0 gpuweb/gpuweb#4961, by @dneto0.
-
wgsl: allow OffsetOfMember to be 0 gpuweb/gpuweb#4960, by @dneto0.
-
Check bindGroupLayout entries are empty gpuweb/gpuweb#4958, by @beaufortfrancois.
-
[M0 backport] Allow `featureLevel: "compatibility"`, which does nothing (#4897) gpuweb/gpuweb#4955, by @kainino0x.
-
[M0 backport] Clarify that filtering sampler_comparison is allowed with depth texture gpuweb/gpuweb#4953, by @kainino0x.
-
Add 1-component vertex formats and unorm8x4-bgra gpuweb/gpuweb#4951, by @Kangz.
-
Clarify that filtering sampler_comparison is allowed with depth texture gpuweb/gpuweb#4950, by @kainino0x.
-
Fix default pipeline layout determination of sampling type for f32 textures gpuweb/gpuweb#4949, by @dneto0.
-
-
33 pull requests merged.
-
Compat: Limit the number of texture+sampler combinations gpuweb/gpuweb#4989.
-
Change GLSL loop examples to WGSL examples. gpuweb/gpuweb#4997.
-
[M0 backport] Fix attribute validation (#4998) gpuweb/gpuweb#5000.
-
Fix attribute validation gpuweb/gpuweb#4998.
-
[M0 backport] Fix styling of adapter immutable properties (#4991) gpuweb/gpuweb#4993.
-
Fix styling of adapter immutable properties gpuweb/gpuweb#4991.
-
Fix host-shareable types note gpuweb/gpuweb#4992.
-
Compat: Disallow using a depth texture with a non-comparison sampler gpuweb/gpuweb#4988.
-
wgsl: @align(n) must divide required-align-of, for all structs gpuweb/gpuweb#4978.
-
Texel buffer proposal gpuweb/gpuweb#4912.
-
Adding clearer notes and examples for dynamic offset ordering gpuweb/gpuweb#4982.
-
Clarify uniqueness of nearest enclosing diagnostic gpuweb/gpuweb#4980.
-
wgsl: smoothstep is well defined for edge0 > edge1 gpuweb/gpuweb#4981.
-
wgsl: bool has size and alignment of 4 bytes gpuweb/gpuweb#4974.
-
Add xrCompatible adapter flag gpuweb/gpuweb#4942.
-
Add note about toneMapping mode in configure gpuweb/gpuweb#4922.
-
Update subgroups proposal gpuweb/gpuweb#4975.
-
wgsl: fix typo gpuweb/gpuweb#4964.
-
[M0 backport] wgsl: allow OffsetOfMember to be 0 gpuweb/gpuweb#4961.
-
wgsl: allow OffsetOfMember to be 0 gpuweb/gpuweb#4960.
-
Check bindGroupLayout entries are empty gpuweb/gpuweb#4958.
-
Move specs from webgpu (community group) to gpuwg (working group) gpuweb/gpuweb#4925.
-
Proposal: expose GPUAdapterInfo from GPUDevice gpuweb/gpuweb#4809.
-
[M0 backport] Allow `featureLevel: "compatibility"`, which does nothing (#4897) gpuweb/gpuweb#4955.
-
Remove the [[destroy started]] state, go straight to 'lost' gpuweb/gpuweb#4903.
-
Allow `featureLevel: "compatibility"`, which does nothing gpuweb/gpuweb#4897.
-
Clarify that filtering sampler_comparison is allowed with depth texture gpuweb/gpuweb#4950.
-
Add 1-component vertex formats and unorm8x4-bgra gpuweb/gpuweb#4951.
-
Allow empty bindGroupLayouts in GPUPipelineLayoutDescriptor gpuweb/gpuweb#4946.
-
Rename image copies -> texel copies gpuweb/gpuweb#4838.
-
Compatibility Mode: add vertex storage buffer and texture limits gpuweb/gpuweb#4927.
-
Fix default pipeline layout determination of sampling type for f32 textures gpuweb/gpuweb#4949.
-
-
3 pull requests closed.
-
Plainly state alignment requirement for scalars and vectors gpuweb/gpuweb#4965.
-
WIP: Add subgroups feature gpuweb/gpuweb#4962.
-
Remove textureGather on integer textures gpuweb/gpuweb#4904.
-
-
23 pull requests received 174 new comments.
-
Add subgroups feature gpuweb/gpuweb#4963 (57 comments by @alan-baker, @dneto0, @jrprice, @kainino0x, @Kangz, @kdashg)
-
wgsl: @align(n) must divide required-align-of, for all structs gpuweb/gpuweb#4978 (12 comments by @alan-baker, @dneto0, @jrprice, @kdashg)
-
Compat mode: make f16 and f32 rendering optional. gpuweb/gpuweb#4983 (10 comments by @greggman, @kainino0x, @mwyrzykowski, @SenorBlanco, @teoxoy)
-
Compat: Limit the number of texture+sampler combinations gpuweb/gpuweb#4989 (9 comments by @greggman, @kainino0x, @teoxoy)
-
Texel buffer proposal gpuweb/gpuweb#4912 (9 comments by @dneto0, @jrprice, @kainino0x, @Kangz, @kdashg)
-
Proposal: expose GPUAdapterInfo from GPUDevice gpuweb/gpuweb#4809 (9 comments by @kainino0x, @mwyrzykowski, @toji)
-
Allow empty bindGroupLayouts in GPUPipelineLayoutDescriptor gpuweb/gpuweb#4946 (8 comments by @beaufortfrancois, @kainino0x)
-
Clarify: texture+sampling validation applies to pairs used together in a builtin gpuweb/gpuweb#4945 (5 comments by @greggman, @kainino0x)
-
Change GLSL loop examples to WGSL examples. gpuweb/gpuweb#4997 (4 comments by @dj2, @greggman, @mehmetoguzderin)
-
Adding clearer notes and examples for dynamic offset ordering gpuweb/gpuweb#4982 (4 comments by @kainino0x, @toji)
-
Plainly state alignment requirement for scalars and vectors gpuweb/gpuweb#4965 (4 comments by @alan-baker, @dneto0, @kdashg)
-
Add note about toneMapping mode in configure gpuweb/gpuweb#4922 (4 comments by @beaufortfrancois, @kainino0x, @Kangz)
-
Allow `featureLevel: "compatibility"`, which does nothing gpuweb/gpuweb#4897 (4 comments by @kainino0x, @Kangz)
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613 (4 comments by @alan-baker, @beaufortfrancois, @greggman, @kainino0x)
-
Check bindGroupLayout entries are empty gpuweb/gpuweb#4958 (4 comments by @beaufortfrancois, @kainino0x)
-
[push-constants] remove per-stage immediates open question gpuweb/gpuweb#4779 (2 comments by @fyellin, @teoxoy)
-
Remove the [[destroy started]] state, go straight to 'lost' gpuweb/gpuweb#4903 (1 comment by @kainino0x)
-
TR publications
-
7 documents published.
-
WebGPU published 4 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 3 times as a Working Draft. History of WebGPU Shading Language
-
Participants
-
2 participants joined.
-
Islam El-Ashi (Dfinity Stiftung) joined.
-
Albin Bernhardsson (ARM Limited) joined.
-
-
2 organizations joined.
-
Dfinity Stiftung joined.
-
ARM Limited joined.
-
GitHub
gpuweb/gpuweb
-
12 issues opened.
-
Provide identifier info in `GPUShaderModule.getCompilationInfo` gpuweb/gpuweb#4943, by @greggman.
-
Feature: Virtualized shader bindings gpuweb/gpuweb#4939, by @magcius.
-
Exposing limited static analysis of WGSL to the CPU (JS) re. shared data, types etc. gpuweb/gpuweb#4937, by @7ombie.
-
Build error: You used Status ED, but your Group (WEBGPU) is limited to the statuses ... gpuweb/gpuweb#4924, by @kainino0x.
-
Improve documentation for setBindGroup about order of dynamic offsets. gpuweb/gpuweb#4921, by @floooh.
-
wgsl: clarify pointer template arguments gpuweb/gpuweb#4920, by @k2d222.
-
Predictable image formation for HDR color spaces gpuweb/gpuweb#4919, by @donmccurdy.
-
Can `GPUOutOfMemoryError` only be generated by `createBuffer()`? gpuweb/gpuweb#4916, by @teoxoy.
-
Aliasing restrictions in WGSL spec are underspecified and potentially misleading gpuweb/gpuweb#4914, by @bavalpey.
-
Clarification needed on meaning of GPUImageDataLayout gpuweb/gpuweb#4908, by @trbabb.
-
Format validation missing from `getCurrentTexture` gpuweb/gpuweb#4906, by @kainino0x.
-
14 issues closed.
-
Aliasing restrictions in WGSL spec are underspecified and potentially misleading gpuweb/gpuweb#4914.
-
Can `GPUOutOfMemoryError` only be generated by `createBuffer()`? gpuweb/gpuweb#4916.
-
Clarify uses of `infinity` in the spec gpuweb/gpuweb#3135.
-
Mistake in recursive decent parser definition? gpuweb/gpuweb#4713.
-
wgsl: clarify pointer template arguments gpuweb/gpuweb#4920.
-
support blending 32 bit float textures gpuweb/gpuweb#3556.
-
Format validation missing from `getCurrentTexture` gpuweb/gpuweb#4906.
-
Resource copying/clearing/updating investigations gpuweb/gpuweb#28.
-
Investigation: GPURenderBundle state inheritance gpuweb/gpuweb#382.
-
Binary WGSL format in createShaderModule() WebAPI gpuweb/gpuweb#2571.
-
Clarification needed on meaning of GPUImageDataLayout gpuweb/gpuweb#4908.
-
1 issue reopened.
-
Feature detection for HDR canvas gpuweb/gpuweb#4828, by @kainino0x.
-
-
31 issues received 101 new comments.
-
Feature detection for HDR canvas gpuweb/gpuweb#4828 (14 comments by @beaufortfrancois, @ccameron-chromium, @greggman, @kainino0x, @mwyrzykowski)
-
Asynchronous shader reflection information gpuweb/gpuweb#624 (9 comments by @7ombie, @eliemichel, @greggman, @kainino0x)
-
Format validation missing from `getCurrentTexture` gpuweb/gpuweb#4906 (9 comments by @greggman, @kainino0x, @sagudev)
-
smoothstep: please remove error check when edge0 > edge1 for compile-time constants gpuweb/gpuweb#4900 (8 comments by @cwfitzgerald, @dneto0, @felixfaire, @kainino0x, @magcius, @reindernijhoff)
-
Predictable image formation for HDR color spaces gpuweb/gpuweb#4919 (6 comments by @beaufortfrancois, @ccameron-chromium, @donmccurdy, @magcius)
-
Aliasing restrictions in WGSL spec are underspecified and potentially misleading gpuweb/gpuweb#4914 (5 comments by @bavalpey, @dneto0)
-
Can `GPUOutOfMemoryError` only be generated by `createBuffer()`? gpuweb/gpuweb#4916 (5 comments by @kainino0x, @teoxoy, @toji)
-
Build error: You used Status ED, but your Group (WEBGPU) is limited to the statuses ... gpuweb/gpuweb#4924 (5 comments by @dneto0, @kainino0x, @tidoust)
-
Mistake in recursive decent parser definition? gpuweb/gpuweb#4713 (4 comments by @dj2, @dneto0, @timokoesters)
-
Dealing with holes in the pipeline layout. gpuweb/gpuweb#2043 (2 comments by @beaufortfrancois, @Kangz)
-
Exposing limited static analysis of WGSL to the CPU (JS) re. shared data, types etc. gpuweb/gpuweb#4937 (2 comments by @7ombie, @greggman)
-
16-bit normalized texture support gpuweb/gpuweb#3001 (2 comments by @beaufortfrancois, @jankrassnigg)
-
Compat: Should requesting compat mode on a device that supports core, return core or compat? gpuweb/gpuweb#4731 (2 comments by @kainino0x, @Kangz)
-
Investigation: GPURenderBundle state inheritance gpuweb/gpuweb#382 (2 comments by @kainino0x, @Kangz)
-
`textureSampleLevel` arguably doesn't work on Intel Mac in compute shaders gpuweb/gpuweb#4818 (1 comment by @c0de517e-rblx)
-
HDR ASTC texture feature "texture-compression-astc-hdr" gpuweb/gpuweb#3856 (1 comment by @kainino0x)
-
Support for gl_PrimitiveID / SV_PrimitiveID -like construct? gpuweb/gpuweb#1786 (1 comment by @JMS55)
-
Improve documentation for setBindGroup about order of dynamic offsets. gpuweb/gpuweb#4921 (1 comment by @kainino0x)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (1 comment by @SenorBlanco)
-
Position: WebGPU should specify both a bytecode format and a human-authorable format for WGSL gpuweb/gpuweb#3175 (1 comment by @kainino0x)
-
WebGL - WebGPU interoperability #18 "copy interaction" - what about an easy post process option gpuweb/gpuweb#3574 (1 comment by @kainino0x)
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841 (1 comment by @kainino0x)
-
-
27 pull requests opened.
-
Remove superfluous exposed on GPUDevice onuncapturederror gpuweb/gpuweb#4947, by @beaufortfrancois.
-
Allow empty bindGroupLayouts in GPUPipelineLayoutDescriptor gpuweb/gpuweb#4946, by @beaufortfrancois.
-
Clarify: texture+sampling validation applies to pairs used together in a builtin gpuweb/gpuweb#4945, by @dneto0.
-
Add xrCompatible adapter flag gpuweb/gpuweb#4942, by @toji.
-
Allow out-of-memory errors in createTexture/createQuerySet gpuweb/gpuweb#4941, by @kainino0x.
-
Add a proposal for sized-binding-arrays gpuweb/gpuweb#4940, by @Kangz.
-
Add 16-bit normalized texture format feature gpuweb/gpuweb#4938, by @beaufortfrancois.
-
Fix misspelling gpuweb/gpuweb#4936, by @fyellin.
-
wgsl: cleanup uses of "infinity" in "differences from IEEE" and fp conversion sections gpuweb/gpuweb#4935, by @dneto0.
-
wgsl: rebuild grammar module only if grammar contents has changed gpuweb/gpuweb#4934, by @dneto0.
-
wgsl: Remove [INF] macro and its uses gpuweb/gpuweb#4933, by @dneto0.
-
Similarity of variable and value declarations is a non-normative note gpuweb/gpuweb#4932, by @mehmetoguzderin.
-
Move specs to CG-DRAFT status gpuweb/gpuweb#4931, by @kainino0x.
-
wgsl: Define 'finite range' gpuweb/gpuweb#4930, by @dneto0.
-
avoid excessive rebuild and install of treesitter WGSL grammar gpuweb/gpuweb#4929, by @dneto0.
-
Fix WGSL grammar flow for Python virtual environments gpuweb/gpuweb#4928, by @dneto0.
-
Compatibility Mode: add vertex storage buffer and texture limits gpuweb/gpuweb#4927, by @SenorBlanco.
-
Move specs from webgpu (community group) to gpuwg (working group) gpuweb/gpuweb#4925, by @kainino0x.
-
Add an appendix for CTS status gpuweb/gpuweb#4923, by @alan-baker.
-
Add note about toneMapping mode in configure gpuweb/gpuweb#4922, by @beaufortfrancois.
-
Require delta in up/down shuffle to be dynamically uniform gpuweb/gpuweb#4917, by @alan-baker.
-
Fix typo "vec3(e4,e4,e6)" -> "vec3(e4,e5,e6)" gpuweb/gpuweb#4915, by @seven332.
-
Make image reads and writes non-private gpuweb/gpuweb#4913, by @alan-baker.
-
Texel buffer proposal gpuweb/gpuweb#4912, by @jrprice.
-
Check for supported context formats in `configure()` on content timeline gpuweb/gpuweb#4911, by @sagudev.
-
Update toneMapping mode in configure gpuweb/gpuweb#4909, by @beaufortfrancois.
-
Validate format in getCurrentTexture gpuweb/gpuweb#4907, by @kainino0x.
-
-
19 pull requests merged.
-
Clarify: texture+sampling validation applies to pairs used together in a builtin gpuweb/gpuweb#4945.
-
Remove superfluous exposed on GPUDevice onuncapturederror gpuweb/gpuweb#4947.
-
Allow out-of-memory errors in createTexture/createQuerySet gpuweb/gpuweb#4941.
-
Fix misspelling gpuweb/gpuweb#4936.
-
wgsl: rebuild grammar module only if grammar contents has changed gpuweb/gpuweb#4934.
-
wgsl: Remove [INF] macro and its uses gpuweb/gpuweb#4933.
-
Add an appendix for CTS status gpuweb/gpuweb#4923.
-
Require delta in up/down shuffle to be dynamically uniform gpuweb/gpuweb#4917.
-
Similarity of variable and value declarations is a non-normative note gpuweb/gpuweb#4932.
-
wgsl: Define 'finite range' gpuweb/gpuweb#4930.
-
Move specs to CG-DRAFT status gpuweb/gpuweb#4931.
-
avoid excessive rebuild and install of treesitter WGSL grammar gpuweb/gpuweb#4929.
-
Fix WGSL grammar flow for Python virtual environments gpuweb/gpuweb#4928.
-
Add "float32-blendable" feature gpuweb/gpuweb#4896.
-
Check for supported context formats in `configure()` on content timeline gpuweb/gpuweb#4911.
-
Fix typo "vec3(e4,e4,e6)" -> "vec3(e4,e5,e6)" gpuweb/gpuweb#4915.
-
Make image reads and writes non-private gpuweb/gpuweb#4913.
-
Update Tree-sitter version gpuweb/gpuweb#4827.
-
-
3 pull requests closed.
-
wgsl: describe sequenced execution gpuweb/gpuweb#2097.
-
Update toneMapping mode in configure gpuweb/gpuweb#4909.
-
Validate format in getCurrentTexture gpuweb/gpuweb#4907.
-
-
23 pull requests received 123 new comments.
-
Texel buffer proposal gpuweb/gpuweb#4912 (34 comments by @alan-baker, @dneto0, @JMS55, @jrprice, @kainino0x, @Kangz, @teoxoy)
-
Check for supported context formats in `configure()` on content timeline gpuweb/gpuweb#4911 (19 comments by @beaufortfrancois, @greggman, @jimblandy, @kainino0x, @mwyrzykowski, @sagudev, @tidoust)
-
Add note about toneMapping mode in configure gpuweb/gpuweb#4922 (11 comments by @beaufortfrancois, @kainino0x, @kdashg)
-
Compatibility Mode: add vertex storage buffer and texture limits gpuweb/gpuweb#4927 (9 comments by @greggman, @kainino0x, @SenorBlanco, @teoxoy)
-
wgsl: cleanup uses of "infinity" in "differences from IEEE" and fp conversion sections gpuweb/gpuweb#4935 (6 comments by @alan-baker, @dneto0)
-
Remove textureGather on integer textures gpuweb/gpuweb#4904 (4 comments by @greggman, @JMS55, @Kangz)
-
Add 16-bit normalized texture format feature gpuweb/gpuweb#4938 (4 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Fix WGSL grammar flow for Python virtual environments gpuweb/gpuweb#4928 (4 comments by @dneto0, @mehmetoguzderin)
-
avoid excessive rebuild and install of treesitter WGSL grammar gpuweb/gpuweb#4929 (3 comments by @dneto0)
-
Allow empty bindGroupLayouts in GPUPipelineLayoutDescriptor gpuweb/gpuweb#4946 (2 comments by @Kangz)
-
Update toneMapping mode in configure gpuweb/gpuweb#4909 (2 comments by @beaufortfrancois, @greggman)
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613 (2 comments by @beaufortfrancois, @kainino0x)
-
Allow `featureLevel: "compatibility"`, which does nothing gpuweb/gpuweb#4897 (2 comments by @kainino0x, @toji)
-
Clarify: texture+sampling validation applies to pairs used together in a builtin gpuweb/gpuweb#4945 (1 comment by @dneto0)
-
wgsl: rebuild grammar module only if grammar contents has changed gpuweb/gpuweb#4934 (1 comment by @dneto0)
-
gpuweb/cts
-
6 issues opened.
-
Move validation errors to top of log gpuweb/cts#4021, by @greggman.
-
timeout check triggers when expanding a set of tests in standalone gpuweb/cts#4019, by @greggman.
-
Fix `identifySamplePoints` for non-float formats gpuweb/cts#4017, by @greggman.
-
Add `historical` test coverage for removed features and renames gpuweb/cts#4005, by @ErichDonGubler.
-
reftest tests timeout on throws gpuweb/cts#3993, by @sagudev.
-
speed up texture builin tests gpuweb/cts#3975, by @greggman.
-
-
1 issue closed.
-
reftest tests timeout on throws gpuweb/cts#3993.
-
-
3 issues received 3 new comments.
-
timeout check triggers when expanding a set of tests in standalone gpuweb/cts#4019 (1 comment by @greggman)
-
Add `historical` test coverage for removed features and renames gpuweb/cts#4005 (1 comment by @kainino0x)
-
-
45 pull requests opened.
-
Fix readTextureToTexelViews to handle all types of formats and viewDimensions gpuweb/cts#4025, by @greggman.
-
Fix fragment shader subgroup builtin io test gpuweb/cts#4024, by @jiangzhaoming.
-
Add tests for GPUDevice.adapterInfo gpuweb/cts#4023, by @beaufortfrancois.
-
Stop trying to filter sint/uint formats. gpuweb/cts#4022, by @greggman.
-
Refactor to not render to rgba32float gpuweb/cts#4020, by @greggman.
-
Make textureGather/Compare not load weights gpuweb/cts#4018, by @greggman.
-
Test fixes for alignment validation landing gpuweb/cts#4016, by @petermcneeleychromium.
-
skip in_render_misc related tests for compat mode gpuweb/cts#4015, by @shrekshao.
-
Fix textureSample:sampled_2d_coords offset gpuweb/cts#4014, by @greggman.
-
Fix subgroupAdd and subgroupMul execution tests gpuweb/cts#4013, by @alan-baker.
-
index.html: Remove CSS attribute that did nothing and now breaks layout gpuweb/cts#4011, by @kainino0x.
-
Texture builtins: move addressMode? to case gpuweb/cts#4010, by @greggman.
-
Show more queries that are too long. gpuweb/cts#4009, by @greggman.
-
Remove non ASCII from texture diagrams gpuweb/cts#4008, by @greggman.
-
attemptGarbageCollection periodically (every N subcases) gpuweb/cts#4007, by @kainino0x.
-
Reduce default maxSubcasesInFlight instead of using batching gpuweb/cts#4006, by @kainino0x.
-
Texture Builtins: offset to case from subcase gpuweb/cts#4004, by @greggman.
-
Add float32-blendable feature validation tests gpuweb/cts#4003, by @beaufortfrancois.
-
Fix typescript error gpuweb/cts#4002, by @greggman.
-
Use batching for shader,validation,expression,call,builtin,*:values: gpuweb/cts#4001, by @kainino0x.
-
Revert "robust_access_vertex shader add float tolerance (#3968)" gpuweb/cts#4000, by @shrekshao.
-
texture builtin offset subcase to case gpuweb/cts#3999, by @greggman.
-
fix texture_formats:canvas_configuration test for supported context format gpuweb/cts#3998, by @sagudev.
-
Update tests for validating canvas texture formats in `configure()` gpuweb/cts#3997, by @beaufortfrancois.
-
Make identifySamplePoints handle compare gpuweb/cts#3996, by @greggman.
-
Include depth textures in more tests gpuweb/cts#3995, by @greggman.
-
clear reftest-wait if reftest throws gpuweb/cts#3994, by @sagudev.
-
Get per stage GPU mix weights on demand gpuweb/cts#3992, by @greggman.
-
Refactor texture-builins for compat no storage gpuweb/cts#3991, by @greggman.
-
Execution for quad swaps gpuweb/cts#3990, by @alan-baker.
-
Fix textureSampleBias gpuweb/cts#3989, by @greggman.
-
Fix textureSampleBias test gpuweb/cts#3988, by @greggman.
-
Quad broadcast execution tests gpuweb/cts#3987, by @alan-baker.
-
WGSL texture builtins all stages gpuweb/cts#3986, by @greggman.
-
Padding tests for vec3h in arrays gpuweb/cts#3985, by @alan-baker.
-
Add textureSampleCompare/Level WGSL tests gpuweb/cts#3984, by @greggman.
-
Fix texture-builtin speedup gpuweb/cts#3983, by @greggman.
-
Test indexing of a matrix using non-const index gpuweb/cts#3982, by @sagudev.
-
WGSL textureSample/Bias/Grad tests gpuweb/cts#3981, by @greggman.
-
Speed up Texture Builtin tests gpuweb/cts#3980, by @greggman.
-
in_render_common: fix compat texture view restriction gpuweb/cts#3979, by @shrekshao.
-
Subgroup bitwise tests gpuweb/cts#3978, by @alan-baker.
-
Switch to GPUCanvasConfiguration viewFormats array gpuweb/cts#3977, by @beaufortfrancois.
-
textureStore: fix compat texture view restriction gpuweb/cts#3976, by @shrekshao.
-
-
42 pull requests merged.
-
Stop trying to filter sint/uint formats. gpuweb/cts#4022.
-
Refactor to not render to rgba32float gpuweb/cts#4020.
-
Make textureGather/Compare not load weights gpuweb/cts#4018.
-
skip in_render_misc related tests for compat mode gpuweb/cts#4015.
-
Fix textureSample:sampled_2d_coords offset gpuweb/cts#4014.
-
Fix subgroupAdd and subgroupMul execution tests gpuweb/cts#4013.
-
Test indexing of a matrix using non-const index gpuweb/cts#3982.
-
Texture builtins: move addressMode? to case gpuweb/cts#4010.
-
Show more queries that are too long. gpuweb/cts#4009.
-
Execution for quad swaps gpuweb/cts#3990.
-
Quad broadcast execution tests gpuweb/cts#3987.
-
Subgroup bitwise tests gpuweb/cts#3978.
-
Remove non ASCII from texture diagrams gpuweb/cts#4008.
-
attemptGarbageCollection periodically (every N subcases) gpuweb/cts#4007.
-
Reduce default maxSubcasesInFlight instead of using batching gpuweb/cts#4006.
-
Texture Builtins: offset to case from subcase gpuweb/cts#4004.
-
Add float32-blendable feature validation tests gpuweb/cts#4003.
-
Use batching for shader,validation,expression,call,builtin,*:values: gpuweb/cts#4001.
-
Revert "robust_access_vertex shader add float tolerance (#3968)" gpuweb/cts#4000.
-
Fix typescript error gpuweb/cts#4002.
-
Make identifySamplePoints handle compare gpuweb/cts#3996.
-
Include depth textures in more tests gpuweb/cts#3995.
-
texture builtin offset subcase to case gpuweb/cts#3999.
-
Refactor texture-builins for compat no storage gpuweb/cts#3991.
-
Get per stage GPU mix weights on demand gpuweb/cts#3992.
-
Update tests for validating canvas texture formats in `configure()` gpuweb/cts#3997.
-
clear reftest-wait if reftest throws gpuweb/cts#3994.
-
Fix textureSampleBias gpuweb/cts#3989.
-
Fix textureSampleBias test gpuweb/cts#3988.
-
WGSL texture builtins all stages gpuweb/cts#3986.
-
Add textureSampleCompare/Level WGSL tests gpuweb/cts#3984.
-
Padding tests for vec3h in arrays gpuweb/cts#3985.
-
Fix texture-builtin speedup gpuweb/cts#3983.
-
WGSL textureSample/Bias/Grad tests gpuweb/cts#3981.
-
Speed up Texture Builtin tests gpuweb/cts#3980.
-
Switch to GPUCanvasConfiguration viewFormats array gpuweb/cts#3977.
-
in_render_common: fix compat texture view restriction gpuweb/cts#3979.
-
textureStore: fix compat texture view restriction gpuweb/cts#3976.
-
Add some tests for GPUCanvasContext getConfiguration gpuweb/cts#3973.
-
-
12 pull requests received 51 new comments.
-
Fix fragment shader subgroup builtin io test gpuweb/cts#4024 (11 comments by @alan-baker, @jiangzhaoming)
-
Add some tests for GPUCanvasContext getConfiguration gpuweb/cts#3973 (8 comments by @beaufortfrancois, @kainino0x)
-
fix texture_formats:canvas_configuration test for supported context format gpuweb/cts#3998 (7 comments by @beaufortfrancois, @Kangz, @sagudev)
-
Add validation tests for Multi-Draw Indirect gpuweb/cts#3962 (3 comments by @beaufortfrancois, @Sirtsu55)
-
Add float32-blendable feature validation tests gpuweb/cts#4003 (2 comments by @beaufortfrancois, @kainino0x)
-
Fix readTextureToTexelViews to handle all types of formats and viewDimensions gpuweb/cts#4025 (1 comment by @shrekshao)
-
TR publications
-
18 documents published.
-
WebGPU published 11 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 7 times as a Working Draft. History of WebGPU Shading Language
-
Participants
-
1 participant joined.
-
Shannon Woods (NVIDIA) joined.
-
-
2 participants left.
-
Ben Clayton (Google LLC) left.
-
Dean Jackson (Apple Inc.) left.
-
GitHub
gpuweb/gpuweb
-
17 issues opened.
-
bound variables in packages gpuweb/gpuweb#4905, by @mighdoll.
-
smoothstep: please remove error check when edge0 > edge1 for compile-time constants gpuweb/gpuweb#4900, by @reindernijhoff.
-
wgsl: support atomics on f32 (in workgroup and storage buffer) gpuweb/gpuweb#4894, by @dneto0.
-
Support pipeline-sized arrays in storage bindings gpuweb/gpuweb#4892, by @Runi-c.
-
Allow Early Fragment Tests to be forced gpuweb/gpuweb#4891, by @toji.
-
Reserve "noncoherent"? "non_coherent"? gpuweb/gpuweb#4880, by @kdashg.
-
empty array entries in createPipelineLayout and other place should/should not be ignored? gpuweb/gpuweb#4879, by @greggman.
-
Guarantees about early-z fragment discard gpuweb/gpuweb#4878, by @toji.
-
Make GPUAdapterInfo a dictionary instead of an interface? gpuweb/gpuweb#4876, by @saschanaz.
-
Alpha to Coverage spec and conformance on Qualcomm hardware gpuweb/gpuweb#4867, by @RafaelCintron.
-
Canvas contents after device destroy or device loss gpuweb/gpuweb#4859, by @kainino0x.
-
`textureGather` issues gpuweb/gpuweb#4857, by @greggman.
-
Feature request: "enum" support in WGSL gpuweb/gpuweb#4856, by @magcius.
-
Derivatives of `centroid`-interpolated variables gpuweb/gpuweb#4854, by @kainino0x.
-
WGSL spec normatively references itself gpuweb/gpuweb#4853, by @dneto0.
-
Error on viewFormat with no compatible usage? gpuweb/gpuweb#4852, by @toji.
-
Spec or browser or user issue: cannot call GPUBuffer.mapAsync() with size=undefined? gpuweb/gpuweb#4848, by @juj.
-
-
17 issues closed.
-
Canvas contents after device destroy or device loss gpuweb/gpuweb#4859.
-
Guarantees about early-z fragment discard gpuweb/gpuweb#4878.
-
Alpha to Coverage spec and conformance on Qualcomm hardware gpuweb/gpuweb#4867.
-
Zero value of position attribute in vertex shaders? gpuweb/gpuweb#4845.
-
Make GPUAdapterInfo a dictionary instead of an interface? gpuweb/gpuweb#4876.
-
Use enum tokens when they are the only possible grammar usage. gpuweb/gpuweb#4505.
-
Missing accuracies for derivatives gpuweb/gpuweb#4839.
-
Reserve "noncoherent"? "non_coherent"? gpuweb/gpuweb#4880.
-
WGSL spec normatively references itself gpuweb/gpuweb#4853.
-
Error on viewFormat with no compatible usage? gpuweb/gpuweb#4852.
-
Support for 3D Textures with ASTC gpuweb/gpuweb#4701.
-
What happens on the device timeline after `GPUDevice.destroy` is called? gpuweb/gpuweb#4294.
-
Support for Wasm memory64? gpuweb/gpuweb#2252.
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426.
-
More strongly push developers toward correct use of GPUAdapter gpuweb/gpuweb#4784.
-
-
45 issues received 175 new comments.
-
Feature detection for HDR canvas gpuweb/gpuweb#4828 (22 comments by @beaufortfrancois, @ccameron-chromium, @greggman, @kainino0x, @Kangz, @mdrejhon, @mwyrzykowski, @toji)
-
Guarantees about early-z fragment discard gpuweb/gpuweb#4878 (16 comments by @kainino0x, @Kangz, @kdashg, @litherum, @mwyrzykowski, @toji)
-
Canvas contents after device destroy or device loss gpuweb/gpuweb#4859 (11 comments by @kainino0x, @Kangz, @mwyrzykowski, @toji)
-
Investigation: Bindless resources gpuweb/gpuweb#380 (10 comments by @AndrewAday, @crystalthoughts, @devshgraphicsprogramming, @magcius, @Usnul)
-
wgsl: support atomics on f32 (in workgroup and storage buffer) gpuweb/gpuweb#4894 (9 comments by @dneto0, @jowens, @petermcneeleychromium)
-
smoothstep: is it an error when `low == high` and/or `low > high` (when constant/pipline-evaluated) gpuweb/gpuweb#4561 (8 comments by @greggman, @johanholwerda, @kainino0x, @reindernijhoff)
-
Alpha to Coverage spec and conformance on Qualcomm hardware gpuweb/gpuweb#4867 (6 comments by @kainino0x, @Kangz, @RafaelCintron, @toji)
-
Zero value of position attribute in vertex shaders? gpuweb/gpuweb#4845 (6 comments by @dneto0, @mwyrzykowski)
-
Hardware feature levels (and compat) gpuweb/gpuweb#4656 (6 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
smoothstep: please remove error check when edge0 > edge1 for compile-time constants gpuweb/gpuweb#4900 (5 comments by @greggman, @reindernijhoff)
-
Derivatives of `centroid`-interpolated variables gpuweb/gpuweb#4854 (5 comments by @alan-baker, @dneto0, @jimblandy, @kainino0x, @kdashg)
-
Error on viewFormat with no compatible usage? gpuweb/gpuweb#4852 (5 comments by @kainino0x, @Kangz, @toji)
-
empty array entries in createPipelineLayout and other place should/should not be ignored? gpuweb/gpuweb#4879 (4 comments by @greggman, @kainino0x, @Kangz)
-
copyExternalImageToTexture/importExternalTexture: Need a way to opt-out of color space conversion gpuweb/gpuweb#4826 (4 comments by @ccameron-chromium, @greggman, @kainino0x, @shaoboyan)
-
Spec or browser or user issue: cannot call GPUBuffer.mapAsync() with size=undefined? gpuweb/gpuweb#4848 (4 comments by @juj, @kainino0x)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (3 comments by @Kangz, @SenorBlanco, @teoxoy)
-
More strongly push developers toward correct use of GPUAdapter gpuweb/gpuweb#4784 (3 comments by @jimblandy, @Kangz, @mwyrzykowski)
-
Double precision floats (IEEE-754 binary64) gpuweb/gpuweb#2805 (2 comments by @alienself, @lslzl3000)
-
Make GPUAdapterInfo a dictionary instead of an interface? gpuweb/gpuweb#4876 (2 comments by @beaufortfrancois, @kainino0x)
-
array_index out of bounds behavior is not specified for texture builtins gpuweb/gpuweb#4782 (2 comments by @jrprice, @petermcneeleychromium)
-
What happens on the device timeline after `GPUDevice.destroy` is called? gpuweb/gpuweb#4294 (2 comments by @kainino0x, @Kangz)
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841 (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)
-
Clarify in spec that getMappedRange() ranges should be used to optimize map invalidations/copies gpuweb/gpuweb#4805 (1 comment by @kainino0x)
-
Optional feature request: filterable 32-bit float textures gpuweb/gpuweb#3601 (1 comment by @kainino0x)
-
Allow `GPUBufferUsage.QUERY_RESOLVE | GPUBufferUsage.MAP_READ` buffer usage gpuweb/gpuweb#4383 (1 comment by @haoxli)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (1 comment by @Kangz)
-
maxInterStageShaderComponents seems to overlap with maxInterStageShaderVariables gpuweb/gpuweb#4688 (1 comment by @Kangz)
-
Missing dynamic offset validation in "Validate encoder bind groups" gpuweb/gpuweb#4811 (1 comment by @Kangz)
-
-
40 pull requests opened.
-
Remove textureGather on integer textures gpuweb/gpuweb#4904, by @kainino0x.
-
Remove the [[destroy started]] state, go straight to 'lost' gpuweb/gpuweb#4903, by @kainino0x.
-
Add [[lastPresentedImage]] as fallback for canvas image gpuweb/gpuweb#4902, by @kainino0x.
-
Add GPUCanvasContext getConfiguration() method gpuweb/gpuweb#4899, by @beaufortfrancois.
-
Compat: describe as a featureLevel below core, upgradeable to core gpuweb/gpuweb#4898, by @kainino0x.
-
Allow `featureLevel: "compatibility"`, which does nothing gpuweb/gpuweb#4897, by @kainino0x.
-
Add "float32-blendable" feature gpuweb/gpuweb#4896, by @kainino0x.
-
WGSL: Replace incorrect use of semicolon in struct declaration gpuweb/gpuweb#4893, by @k2d222.
-
Loosen Alpha-to-coverage guarantees gpuweb/gpuweb#4890, by @toji.
-
subgroups: fix quadSwap* issues gpuweb/gpuweb#4889, by @dneto0.
-
wgsl: clarify inherited-from accuracy applies after fp rules and shortcuts gpuweb/gpuweb#4888, by @dneto0.
-
wgsl: vertex position.w = 0 is a dynamic error gpuweb/gpuweb#4887, by @dneto0.
-
wgsl: reserve 'noncoherent' and 'non_coherent' gpuweb/gpuweb#4886, by @dneto0.
-
Specify that writing to storage skips early z gpuweb/gpuweb#4885, by @toji.
-
Add derivatives to accuracy table gpuweb/gpuweb#4884, by @alan-baker.
-
[editorial] Grammatical fix for some comments gpuweb/gpuweb#4883, by @alan-baker.
-
Address the remaining TODO in the WGSL spec for sample footprint gpuweb/gpuweb#4882, by @mehmetoguzderin.
-
wgsl: Tweak wording of array constructor limit gpuweb/gpuweb#4881, by @jrprice.
-
[WGSL] Remove normative self-reference gpuweb/gpuweb#4877, by @alan-baker.
-
Add column `Extension` in `Built-in input and output values` gpuweb/gpuweb#4875, by @Jiawei-Shao.
-
Allow only strings in `featureLevel` gpuweb/gpuweb#4874, by @kainino0x.
-
Add `texture-compression-astc-sliced-3d` gpuweb/gpuweb#4873, by @mehmetoguzderin.
-
Add note to warn on unusable viewFormats gpuweb/gpuweb#4872, by @toji.
-
Fill out the Sample Frequency Shading section gpuweb/gpuweb#4871, by @toji.
-
[editorial] Restore validation error accidentally removed in #4860 gpuweb/gpuweb#4869, by @kainino0x.
-
Add API surface for feature levels gpuweb/gpuweb#4868, by @kainino0x.
-
[editorial] remove assertion from "dot" syntax gpuweb/gpuweb#4866, by @kainino0x.
-
[editorial] default GPUProgrammableStage.constants to {} gpuweb/gpuweb#4865, by @kainino0x.
-
[editorial] Make a definition for vertex format sizes gpuweb/gpuweb#4863, by @kainino0x.
-
[editorial] Use the word "return" instead of "stop" gpuweb/gpuweb#4862, by @kainino0x.
-
[editorial] Use `.[=list/size=]` instead of `.length` gpuweb/gpuweb#4861, by @kainino0x.
-
[editorial] Fix fencepost error in maxBindGroupsPlusVertexBuffers validation gpuweb/gpuweb#4860, by @kainino0x.
-
Two's complement for AbstractInt and wording improvement for i32 gpuweb/gpuweb#4858, by @mehmetoguzderin.
-
Filling out the output merging section gpuweb/gpuweb#4855, by @toji.
-
Add missing validation rule about |maxFragmentShaderInputVariables| gpuweb/gpuweb#4851, by @Jiawei-Shao.
-
E-pic-c/SECURITY.me gpuweb/gpuweb#4850, by @E-pic-C.
-
Describing depth/stencil processing in Detailed Operations gpuweb/gpuweb#4849, by @toji.
-
wgsl: dynamically infinite loops are dynamic errors gpuweb/gpuweb#4847, by @dneto0.
-
Fix typo expresion -> expres**s**ion gpuweb/gpuweb#4846, by @sagudev.
-
-
37 pull requests merged.
-
Add [[lastPresentedImage]] as fallback for canvas image gpuweb/gpuweb#4902.
-
Add GPUCanvasContext getConfiguration() method gpuweb/gpuweb#4899.
-
Allow only strings in `featureLevel` gpuweb/gpuweb#4874.
-
Address the remaining TODO in the WGSL spec for sample footprint gpuweb/gpuweb#4882.
-
WGSL: Replace incorrect use of semicolon in struct declaration gpuweb/gpuweb#4893.
-
Specify that writing to storage skips early z gpuweb/gpuweb#4885.
-
Loosen Alpha-to-coverage guarantees gpuweb/gpuweb#4890.
-
subgroups: fix quadSwap* issues gpuweb/gpuweb#4889.
-
wgsl: clarify inherited-from accuracy applies after fp rules and shortcuts gpuweb/gpuweb#4888.
-
wgsl: vertex position.w = 0 is a dynamic error gpuweb/gpuweb#4887.
-
Add derivatives to accuracy table gpuweb/gpuweb#4884.
-
wgsl: reserve 'noncoherent' and 'non_coherent' gpuweb/gpuweb#4886.
-
wgsl: Tweak wording of array constructor limit gpuweb/gpuweb#4881.
-
[editorial] Grammatical fix for some comments gpuweb/gpuweb#4883.
-
[push-constants] remove per-stage immediates open question gpuweb/gpuweb#4779.
-
Add column `Extension` in `Built-in input and output values` gpuweb/gpuweb#4875.
-
[WGSL] Remove normative self-reference gpuweb/gpuweb#4877.
-
Fill out the Sample Frequency Shading section gpuweb/gpuweb#4871.
-
Add note to warn on unusable viewFormats gpuweb/gpuweb#4872.
-
Add `texture-compression-astc-sliced-3d` gpuweb/gpuweb#4873.
-
Add API surface for feature levels gpuweb/gpuweb#4868.
-
[editorial] Restore validation error accidentally removed in #4860 gpuweb/gpuweb#4869.
-
[editorial] default GPUProgrammableStage.constants to {} gpuweb/gpuweb#4865.
-
[editorial] Use the word "return" instead of "stop" gpuweb/gpuweb#4862.
-
[editorial] remove assertion from "dot" syntax gpuweb/gpuweb#4866.
-
[editorial] Make a definition for vertex format sizes gpuweb/gpuweb#4863.
-
[editorial] Use `.[=list/size=]` instead of `.length` gpuweb/gpuweb#4861.
-
[editorial] Fix fencepost error in maxBindGroupsPlusVertexBuffers validation gpuweb/gpuweb#4860.
-
Two's complement for AbstractInt and wording improvement for i32 gpuweb/gpuweb#4858.
-
Filling out the output merging section gpuweb/gpuweb#4855.
-
Describing depth/stencil processing in Detailed Operations gpuweb/gpuweb#4849.
-
Uncapturederror and preventDefault gpuweb/gpuweb#4708.
-
Add usage to GPUTextureViewDescriptor gpuweb/gpuweb#4746.
-
Add missing validation rule about |maxFragmentShaderInputVariables| gpuweb/gpuweb#4851.
-
wgsl: dynamically infinite loops are dynamic errors gpuweb/gpuweb#4847.
-
Fix typo expresion -> expres**s**ion gpuweb/gpuweb#4846.
-
-
3 pull requests closed.
-
Feature Levels gpuweb/gpuweb#4761.
-
Annotate associativity for easier reading and downstream tooling gpuweb/gpuweb#3889.
-
E-pic-c/SECURITY.me gpuweb/gpuweb#4850.
-
-
29 pull requests received 134 new comments.
-
Describing depth/stencil processing in Detailed Operations gpuweb/gpuweb#4849 (26 comments by @kainino0x, @Kangz, @toji)
-
Clarify that atomic `and` and `or` are bitwise operations, slightly restructure, and fix missing prepositions gpuweb/gpuweb#4864 (13 comments by @alan-baker, @mehmetoguzderin)
-
Add GPUCanvasContext getConfiguration() method gpuweb/gpuweb#4899 (11 comments by @beaufortfrancois, @greggman, @kainino0x, @toji)
-
Add multi draw indirect feature gpuweb/gpuweb#2315 (7 comments by @beaufortfrancois, @kainino0x, @Sirtsu55)
-
Replace asynchronous .requestAdapterInfo() with synchronous .info gpuweb/gpuweb#4662 (4 comments by @beaufortfrancois, @mwyrzykowski)
-
Compat: describe as a featureLevel below core, upgradeable to core gpuweb/gpuweb#4898 (3 comments by @SenorBlanco)
-
wgsl: reserve 'noncoherent' and 'non_coherent' gpuweb/gpuweb#4886 (3 comments by @dneto0, @mehmetoguzderin)
-
[editorial] Fix fencepost error in maxBindGroupsPlusVertexBuffers validation gpuweb/gpuweb#4860 (3 comments by @kainino0x, @toji)
-
Add [[lastPresentedImage]] as fallback for canvas image gpuweb/gpuweb#4902 (2 comments by @kainino0x)
-
Allow `featureLevel: "compatibility"`, which does nothing gpuweb/gpuweb#4897 (1 comment by @kainino0x)
-
WGSL: Replace incorrect use of semicolon in struct declaration gpuweb/gpuweb#4893 (1 comment by @w3cbot)
-
Add column `Extension` in `Built-in input and output values` gpuweb/gpuweb#4875 (1 comment by @kainino0x)
-
[editorial] default GPUProgrammableStage.constants to {} gpuweb/gpuweb#4865 (1 comment by @kainino0x)
-
Annotate associativity for easier reading and downstream tooling gpuweb/gpuweb#3889 (1 comment by @mehmetoguzderin)
-
gpuweb/cts
-
11 issues opened.
-
Add coverage for `GPUShaderModuleDescriptor.compilationHints` gpuweb/cts#3971, by @ErichDonGubler.
-
Test `texture-compression-astc-sliced-3d` gpuweb/cts#3967, by @kainino0x.
-
Params type for beforeAllSubcases is improperly specialized gpuweb/cts#3965, by @kainino0x.
-
Add tests for multi-draw-indirect gpuweb/cts#3961, by @kainino0x.
-
op: Fix alpha_to_coverage_mask test for Qualcomm gpuweb/cts#3960, by @kainino0x.
-
op: canvas read as an image source on device destroyed gpuweb/cts#3956, by @kainino0x.
-
wgsl: reserve 'noncoherent' and 'non_coherent' gpuweb/cts#3955, by @dneto0.
-
Add tests for the usage field of GPUTextureViewDescriptor gpuweb/cts#3953, by @vonture.
-
Fix all the tests about `maxInterStageShaderComponents` gpuweb/cts#3944, by @Jiawei-Shao.
-
Add tests for WGPU_ARRAY_SLICE_UNDEFINED and friends. gpuweb/cts#3926, by @Kangz.
-
5 issues closed.
-
Add tests for the usage field of GPUTextureViewDescriptor gpuweb/cts#3953.
-
wgsl: reserve 'noncoherent' and 'non_coherent' gpuweb/cts#3955.
-
Test clip-distances feature gpuweb/cts#3773.
-
Fix all the tests about `maxInterStageShaderComponents` gpuweb/cts#3944.
-
-
1 issue reopened.
-
9 issues received 17 new comments.
-
`npm run wpt` generates `*.worker.js` tests that accidentally get run by `wptrunner` gpuweb/cts#3834 (5 comments by @ErichDonGubler, @kainino0x)
-
Add tests for WGPU_ARRAY_SLICE_UNDEFINED and friends. gpuweb/cts#3926 (2 comments by @kainino0x, @Kangz)
-
web_platform: Check the result of transferToImageBitmap after presentation gpuweb/cts#2366 (1 comment by @kainino0x)
-
Add coverage for `GPUShaderModuleDescriptor.compilationHints` gpuweb/cts#3971 (1 comment by @ErichDonGubler)
-
-
38 pull requests opened.
-
Add some tests for GPUCanvasContext getConfiguration gpuweb/cts#3973, by @beaufortfrancois.
-
Clean output directories when building less than `all` gpuweb/cts#3972, by @kainino0x.
-
Add Rendering tests for Multi Draw Indirect gpuweb/cts#3970, by @Sirtsu55.
-
Fix reifyTextureViewDescriptor gpuweb/cts#3969, by @kainino0x.
-
robust_access_vertex shader add float tolerance gpuweb/cts#3968, by @shrekshao.
-
Roll `@webgpu/types` to 0.1.46 gpuweb/cts#3966, by @kainino0x.
-
Initial commit for subgroupInclusive add/mul cts gpuweb/cts#3964, by @petermcneeleychromium.
-
Fix textureNumLayers view_type subcase gpuweb/cts#3963, by @shrekshao.
-
Add validation tests for Multi-Draw Indirect gpuweb/cts#3962, by @beaufortfrancois.
-
Allow for trailing comma in var template declaration gpuweb/cts#3959, by @petermcneeleychromium.
-
Add validation tests on the limit of max vertex output location with … gpuweb/cts#3958, by @Jiawei-Shao.
-
wgsl: test reservation of 'noncoherent' and 'non_coherent' gpuweb/cts#3957, by @dneto0.
-
Test usage field of GPUTextureViewDescriptor gpuweb/cts#3954, by @vonture.
-
Add shader execution tests about `clip_distances` gpuweb/cts#3952, by @Jiawei-Shao.
-
Add validation tests on clip_distances and maxInterStageShaderVariables gpuweb/cts#3951, by @Jiawei-Shao.
-
Fix more compat textureBindingViewDimension validation gpuweb/cts#3950, by @shrekshao.
-
Add shader validation tests about built-in variable `clip_distances` gpuweb/cts#3949, by @Jiawei-Shao.
-
Fix TestCaseRecord proxy gpuweb/cts#3948, by @greggman.
-
Minor fix for extra comma when specifying access mode gpuweb/cts#3947, by @petermcneeleychromium.
-
Add shader validation tests about `clip_distances` and the extension gpuweb/cts#3946, by @Jiawei-Shao.
-
Remove all the tests about `maxInterStageShaderComponents` gpuweb/cts#3945, by @Jiawei-Shao.
-
Remove non-standard GPUAdapter requestAdapterInfo() method gpuweb/cts#3941, by @beaufortfrancois.
-
WGSL textureSample tests gpuweb/cts#3940, by @greggman.
-
Add Origin Trial token for Subgroups in localhost:8080 gpuweb/cts#3939, by @dneto0.
-
Skip identifying sample points unless debugging gpuweb/cts#3937, by @greggman.
-
Fragment tests for subgroupAny and subgroupAll gpuweb/cts#3936, by @alan-baker.
-
More WGSL flat issues gpuweb/cts#3935, by @greggman.
-
Fixes for no attachments being an error in webgpu/compat gpuweb/cts#3934, by @Kangz.
-
Compat: rg32xxx texture usage is not a WGSL error gpuweb/cts#3933, by @greggman.
-
Fix interpolate test for compat gpuweb/cts#3932, by @greggman.
-
Improve cube-array error logging gpuweb/cts#3931, by @greggman.
-
Same pipeline identify sample points gpuweb/cts#3930, by @greggman.
-
Get mip level mix weights just once. gpuweb/cts#3929, by @greggman.
-
Remove GPUShaderModuleDescriptor.sourceMap gpuweb/cts#3928, by @beaufortfrancois.
-
Fixes for no attachments being an error. gpuweb/cts#3927, by @Kangz.
-
-
35 pull requests merged.
-
Fix reifyTextureViewDescriptor gpuweb/cts#3969.
-
Test usage field of GPUTextureViewDescriptor gpuweb/cts#3954.
-
Clean output directories when building less than `all` gpuweb/cts#3972.
-
robust_access_vertex shader add float tolerance gpuweb/cts#3968.
-
Initial commit for subgroupInclusive add/mul cts gpuweb/cts#3964.
-
Roll `@webgpu/types` to 0.1.46 gpuweb/cts#3966.
-
Fix textureNumLayers view_type subcase gpuweb/cts#3963.
-
Allow for trailing comma in var template declaration gpuweb/cts#3959.
-
wgsl: test reservation of 'noncoherent' and 'non_coherent' gpuweb/cts#3957.
-
Add shader execution tests about `clip_distances` gpuweb/cts#3952.
-
Fix more compat textureBindingViewDimension validation gpuweb/cts#3950.
-
wgsl: Add short-circuiting validation tests gpuweb/cts#3567.
-
Add shader validation tests about built-in variable `clip_distances` gpuweb/cts#3949.
-
Remove non-standard GPUAdapter requestAdapterInfo() method gpuweb/cts#3941.
-
Fix TestCaseRecord proxy gpuweb/cts#3948.
-
Add shader validation tests about `clip_distances` and the extension gpuweb/cts#3946.
-
Remove all the tests about `maxInterStageShaderComponents` gpuweb/cts#3945.
-
Add Origin Trial token for Subgroups in localhost:8080 gpuweb/cts#3939.
-
Skip identifying sample points unless debugging gpuweb/cts#3937.
-
Fragment tests for subgroupAny and subgroupAll gpuweb/cts#3936.
-
More WGSL flat issues gpuweb/cts#3935.
-
Fixes for no attachments being an error in webgpu/compat gpuweb/cts#3934.
-
wgsl: Add another phony assignment test case gpuweb/cts#3925.
-
Compat: rg32xxx texture usage is not a WGSL error gpuweb/cts#3933.
-
Fix interpolate test for compat gpuweb/cts#3932.
-
Improve cube-array error logging gpuweb/cts#3931.
-
Same pipeline identify sample points gpuweb/cts#3930.
-
Get mip level mix weights just once. gpuweb/cts#3929.
-
Remove GPUShaderModuleDescriptor.sourceMap gpuweb/cts#3928.
-
Fixes for no attachments being an error. gpuweb/cts#3927.
-
-
2 pull requests closed.
-
WGSL textureSample tests gpuweb/cts#3940.
-
Minor fix for extra comma when specifying access mode gpuweb/cts#3947.
-
-
16 pull requests received 54 new comments.
-
Add the largest float representable as integer (signed/unsigned) to F32 testing (execution) set gpuweb/cts#3942 (8 comments by @petermcneeleychromium, @zoddicus)
-
Skip identifying sample points unless debugging gpuweb/cts#3937 (5 comments by @greggman, @kainino0x)
-
robust_access_vertex shader add float tolerance gpuweb/cts#3968 (4 comments by @greggman, @shrekshao)
-
Remove non-standard GPUAdapter requestAdapterInfo() method gpuweb/cts#3941 (3 comments by @ErichDonGubler, @kainino0x)
-
Avoid subnormals in expression validation for quantizing types (in normalize and refract) gpuweb/cts#3938 (3 comments by @dneto0, @petermcneeleychromium)
-
Add shader validation tests about built-in variable `clip_distances` gpuweb/cts#3949 (2 comments by @jiangzhaoming, @Jiawei-Shao)
-
Minor fix for extra comma when specifying access mode gpuweb/cts#3947 (2 comments by @dj2, @petermcneeleychromium)
-
Add Origin Trial token for Subgroups in localhost:8080 gpuweb/cts#3939 (2 comments by @dneto0, @kainino0x)
-
Add validation tests on clip_distances and maxInterStageShaderVariables gpuweb/cts#3951 (1 comment by @jiangzhaoming)
TR publications
-
11 documents published.
-
WebGPU published 6 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 5 times as a Working Draft. History of WebGPU Shading Language
-
Participants
-
1 participant joined.
-
Geoff Lang (Google LLC) joined.
-
GitHub
gpuweb/gpuweb
-
24 issues opened.
-
Zero value of position attribute in vertex shaders? gpuweb/gpuweb#4845, by @mwyrzykowski.
-
Reintroduce source maps gpuweb/gpuweb#4844, by @kainino0x.
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841, by @greggman.
-
Missing accuracies for derivatives gpuweb/gpuweb#4839, by @alan-baker.
-
Subgroup limits aren't really limits gpuweb/gpuweb#4834, by @alan-baker.
-
make it easier to see that you can't assign to a multi-element swizzle gpuweb/gpuweb#4833, by @dneto0.
-
WGSL doesn't support matrices with elements of integer type gpuweb/gpuweb#4829, by @aleino-nv.
-
Feature detection for HDR canvas gpuweb/gpuweb#4828, by @kainino0x.
-
Tracking issue for "must not be provided" rules that could be lifted gpuweb/gpuweb#4824, by @kainino0x.
-
`textureSampleLevel` arguably doesn't work on Intel Mac in compute shaders gpuweb/gpuweb#4818, by @greggman.
-
Support for `mat / scalar` gpuweb/gpuweb#4817, by @greggman.
-
Support for mixed type `>>=` operation gpuweb/gpuweb#4816, by @mehmetoguzderin.
-
Clamp overload for scalar min and max with vector value gpuweb/gpuweb#4815, by @mehmetoguzderin.
-
Dynamically choosing a render target array index slice from the vertex shader gpuweb/gpuweb#4812, by @magcius.
-
Missing dynamic offset validation in "Validate encoder bind groups" gpuweb/gpuweb#4811, by @teoxoy.
-
Allow getting adapter information from GPUDevice gpuweb/gpuweb#4810, by @jiangzhaoming.
-
remove `sourceMap` from `GPUShaderModuleDescriptor` gpuweb/gpuweb#4808, by @greggman.
-
Make the last 2 arguments to `copyBufferToBuffer` optional gpuweb/gpuweb#4807, by @greggman.
-
Multiple viewports are not supported by WebGPU gpuweb/gpuweb#4806, by @mwyrzykowski.
-
Compat: sample locations don't match the spec in OpenGL gpuweb/gpuweb#4804, by @greggman.
-
Inconsistency hyphenation in WGSL spec gpuweb/gpuweb#4803, by @alan-baker.
-
-
18 issues closed.
-
Missing dynamic offset validation in "Validate encoder bind groups" gpuweb/gpuweb#4811.
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558.
-
`textureSampleLevel` arguably doesn't work on Intel Mac in compute shaders gpuweb/gpuweb#4818.
-
Implementation Status: Firefox Nightly WebGPU is blocked by default on macOS gpuweb/gpuweb#4572.
-
remove `sourceMap` from `GPUShaderModuleDescriptor` gpuweb/gpuweb#4808.
-
mapAsync early-reject doesn't generate a validation error gpuweb/gpuweb#4690.
-
maxInterStageShaderComponents seems to overlap with maxInterStageShaderVariables gpuweb/gpuweb#4688.
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568.
-
make it easier to see that you can't assign to a multi-element swizzle gpuweb/gpuweb#4833.
-
WGSL built in functions are very under specified gpuweb/gpuweb#4527.
-
The value of the fragment builtin `@position` gpuweb/gpuweb#4777.
-
wgsl: clarify what happens for IEEE 754 "invalid operation" cases, e.g. acos(2) gpuweb/gpuweb#4219.
-
Define domain for float builtins and describe errors gpuweb/gpuweb#3220.
-
Precision when behavior of float built-in functions is implementation-specific gpuweb/gpuweb#2884.
-
Clarify undefined behavior of built in functions gpuweb/gpuweb#2765.
-
-
46 issues received 120 new comments.
-
`textureSampleLevel` arguably doesn't work on Intel Mac in compute shaders gpuweb/gpuweb#4818 (12 comments by @dneto0, @greggman, @Kangz, @kdashg, @mwyrzykowski)
-
Multiple viewports are not supported by WebGPU gpuweb/gpuweb#4806 (10 comments by @Kangz, @magcius, @mwyrzykowski, @toji)
-
Clarify in spec that getMappedRange() ranges should be used to optimize map invalidations/copies gpuweb/gpuweb#4805 (8 comments by @kainino0x, @Kangz, @vmilea)
-
Allow getting adapter information from GPUDevice gpuweb/gpuweb#4810 (7 comments by @gyagp, @jiangzhaoming, @Kangz, @mwyrzykowski, @qjia7)
-
What happens on the device timeline after `GPUDevice.destroy` is called? gpuweb/gpuweb#4294 (5 comments by @kainino0x, @teoxoy)
-
remove `sourceMap` from `GPUShaderModuleDescriptor` gpuweb/gpuweb#4808 (5 comments by @kainino0x, @kdashg, @litherum, @mwyrzykowski)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (5 comments by @Kangz, @SenorBlanco, @teoxoy)
-
Subgroup limits aren't really limits gpuweb/gpuweb#4834 (4 comments by @beaufortfrancois, @dneto0, @kainino0x)
-
bgra8unorm storage textures without bgra8unorm-storage, should generate WGSL error? gpuweb/gpuweb#4711 (4 comments by @dneto0, @Kangz, @kdashg)
-
List of issues to discuss with TC39, WASM, JS engine teams gpuweb/gpuweb#747 (4 comments by @kainino0x, @MikhailGorobets)
-
The value of the fragment builtin `@position` gpuweb/gpuweb#4777 (4 comments by @greggman, @Kangz, @kdashg, @mwyrzykowski)
-
copyExternalImageToTexture/importExternalTexture: Need a way to opt-out of color space conversion gpuweb/gpuweb#4826 (3 comments by @greggman, @kainino0x)
-
add subgroups, and make them portable if possible gpuweb/gpuweb#4306 (3 comments by @dneto0, @kenrussell)
-
Clamp overload for scalar min and max with vector value gpuweb/gpuweb#4815 (3 comments by @alan-baker, @jimblandy, @kdashg)
-
Missing dynamic offset validation in "Validate encoder bind groups" gpuweb/gpuweb#4811 (3 comments by @Kangz, @mwyrzykowski)
-
More strongly push developers toward correct use of GPUAdapter gpuweb/gpuweb#4784 (2 comments by @kainino0x, @toji)
-
Feature detection for HDR canvas gpuweb/gpuweb#4828 (2 comments by @beaufortfrancois, @ccameron-chromium)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (2 comments by @greggman, @magcius)
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558 (2 comments by @kainino0x, @Kangz)
-
Make the last 2 arguments to `copyBufferToBuffer` optional gpuweb/gpuweb#4807 (2 comments by @greggman, @kainino0x)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (2 comments by @kainino0x)
-
Dynamically choosing a render target array index slice from the vertex shader gpuweb/gpuweb#4812 (2 comments by @magcius, @mwyrzykowski)
-
Implementation Status: Firefox Nightly WebGPU is blocked by default on macOS gpuweb/gpuweb#4572 (1 comment by @rinsuki)
-
`textureGather` with integer texture formats doesn't work on AMD devices gpuweb/gpuweb#4841 (1 comment by @Kangz)
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568 (1 comment by @dneto0)
-
inter-stage interpolation validation is incorrectly described as triggering shader-creatin errors gpuweb/gpuweb#4836 (1 comment by @dneto0)
-
wgsl spec editorial: correctly reference the floating point standard IEEE-754-2019 gpuweb/gpuweb#1197 (1 comment by @dneto0)
-
wgsl: clarify what happens for IEEE 754 "invalid operation" cases, e.g. acos(2) gpuweb/gpuweb#4219 (1 comment by @dneto0)
-
Precision when behavior of float built-in functions is implementation-specific gpuweb/gpuweb#2884 (1 comment by @dneto0)
-
writing to storage buffers/textures in fragment shaders with multisample targets is not portable. gpuweb/gpuweb#4696 (1 comment by @kainino0x)
-
reuse of an GPUAdapter to request a GPUDevice does not reject. gpuweb/gpuweb#4352 (1 comment by @Kangz)
-
On device loss, should mapAsync throw AbortError instead of OperationError? gpuweb/gpuweb#4774 (1 comment by @Kangz)
-
-
15 pull requests opened.
-
Add dynamic offset to "Validate encoder bind groups" validation function gpuweb/gpuweb#4843, by @teoxoy.
-
Additional cases for non-exact texel copies gpuweb/gpuweb#4842, by @toji.
-
Remove GPUShaderModuleDescriptor.sourceMap gpuweb/gpuweb#4840, by @kainino0x.
-
Rename image copies -> texel copies gpuweb/gpuweb#4838, by @kainino0x.
-
interstage-interpolation checks generate pipeline-creation errors, no… gpuweb/gpuweb#4837, by @dneto0.
-
Better explain that you can't assign to a multi-letter swizzle gpuweb/gpuweb#4835, by @dneto0.
-
Clarify validation of storage texture texel-format and access modes gpuweb/gpuweb#4831, by @dneto0.
-
Update Tree-sitter version gpuweb/gpuweb#4827, by @mehmetoguzderin.
-
wsgl: clarify fp edge cases gpuweb/gpuweb#4825, by @dneto0.
-
wsgl: remove TODO about acos, asin accuracy for f16 gpuweb/gpuweb#4823, by @dneto0.
-
Add inclusive scan add and mul operations gpuweb/gpuweb#4822, by @alan-baker.
-
subgroups: subgroupBroadcast 'id' parameter is const-expression gpuweb/gpuweb#4820, by @dneto0.
-
wgsl: define 'extended real' numbers gpuweb/gpuweb#4819, by @dneto0.
-
Fixed two "it's" -> "its" typos. gpuweb/gpuweb#4813, by @Tomycj.
-
Proposal: expose GPUAdapterInfo from GPUDevice gpuweb/gpuweb#4809, by @jiangzhaoming.
-
-
22 pull requests merged.
-
Disable async operations after device destruction starts gpuweb/gpuweb#4788.
-
Add dynamic offset to "Validate encoder bind groups" validation function gpuweb/gpuweb#4843.
-
Additional cases for non-exact texel copies gpuweb/gpuweb#4842.
-
Change mapAsync to also early-reject if already mapped gpuweb/gpuweb#4787.
-
Remove GPUShaderModuleDescriptor.sourceMap gpuweb/gpuweb#4840.
-
Generate a validation error on mapAsync early-reject gpuweb/gpuweb#4786.
-
Remove maxInterStageShaderComponents gpuweb/gpuweb#4783.
-
Allow unknown limits to be requested with value `undefined` gpuweb/gpuweb#4781.
-
interstage-interpolation checks generate pipeline-creation errors, no… gpuweb/gpuweb#4837.
-
Add inclusive scan add and mul operations gpuweb/gpuweb#4822.
-
subgroups: subgroupBroadcast 'id' parameter is const-expression gpuweb/gpuweb#4820.
-
Better explain that you can't assign to a multi-letter swizzle gpuweb/gpuweb#4835.
-
Clarify validation of storage texture texel-format and access modes gpuweb/gpuweb#4831.
-
wsgl: clarify fp edge cases gpuweb/gpuweb#4825.
-
wsgl: remove TODO about acos, asin accuracy for f16 gpuweb/gpuweb#4823.
-
Fix rules about dual source blending in `validating GPUFragmentState` gpuweb/gpuweb#4791.
-
Add extra example for operations involving abstract numerics gpuweb/gpuweb#4795.
-
wgsl: define 'extended real' numbers gpuweb/gpuweb#4819.
-
[editorial] Add timelines to all attributes/slots, assorted editing gpuweb/gpuweb#4796.
-
Add note that enabling features can have performance impact gpuweb/gpuweb#4778.
-
Fixed two "it's" -> "its" typos. gpuweb/gpuweb#4813.
-
Add shader module creation validation for features gpuweb/gpuweb#4802.
-
-
2 pull requests closed.
-
wgsl: fold expression semantics text into definition of type rule gpuweb/gpuweb#2444.
-
Clarify typography of abstract int 0 and abstract float 0 gpuweb/gpuweb#4485.
-
-
24 pull requests received 92 new comments.
-
Disable async operations after device destruction starts gpuweb/gpuweb#4788 (7 comments by @kainino0x, @toji)
-
Generate a validation error on mapAsync early-reject gpuweb/gpuweb#4786 (5 comments by @kainino0x, @mwyrzykowski)
-
Add dynamic offset to "Validate encoder bind groups" validation function gpuweb/gpuweb#4843 (4 comments by @kainino0x, @teoxoy, @toji)
-
subgroups: subgroupBroadcast 'id' parameter is const-expression gpuweb/gpuweb#4820 (3 comments by @alan-baker)
-
Add extra example for operations involving abstract numerics gpuweb/gpuweb#4795 (3 comments by @dneto0, @matthew-wong1, @w3cbot)
-
Proposal: expose GPUAdapterInfo from GPUDevice gpuweb/gpuweb#4809 (2 comments by @Kangz, @mwyrzykowski)
-
Fix rules about dual source blending in `validating GPUFragmentState` gpuweb/gpuweb#4791 (2 comments by @Jiawei-Shao, @Kangz)
-
Allow unknown limits to be requested with value `undefined` gpuweb/gpuweb#4781 (1 comment by @kainino0x)
-
Better explain that you can't assign to a multi-letter swizzle gpuweb/gpuweb#4835 (1 comment by @dneto0)
-
wgsl: fold expression semantics text into definition of type rule gpuweb/gpuweb#2444 (1 comment by @dneto0)
-
[editorial] Add timelines to all attributes/slots, assorted editing gpuweb/gpuweb#4796 (1 comment by @toji)
gpuweb/cts
-
3 issues opened.
-
Support wildcards/glob in queries gpuweb/cts#3909, by @greggman.
-
Change compat shader module errors to pipeline errors gpuweb/cts#3893, by @greggman.
-
Running CTS on React Native gpuweb/cts#3892, by @wcandillon.
-
-
2 issues closed.
-
Refactor WGSL textureLoad execution tests to test more texels gpuweb/cts#3871.
-
Add tests against the extension `dual_source_blending` gpuweb/cts#3810.
-
-
1 issue reopened.
-
3 issues received 7 new comments.
-
`npm run wpt` generates `*.worker.js` tests that accidentally get run by `wptrunner` gpuweb/cts#3834 (2 comments by @ErichDonGubler, @kainino0x)
-
37 pull requests opened.
-
wgsl: Add another phony assignment test case gpuweb/cts#3925, by @jrprice.
-
Execution tests for subgroupAny and subgroupAll gpuweb/cts#3924, by @alan-baker.
-
Fix another edge case for textureGather. gpuweb/cts#3923, by @greggman.
-
Fix textureGather gpuweb/cts#3922, by @greggman.
-
Print more info for texture builtins gpuweb/cts#3921, by @greggman.
-
Remaining subgroup validation tests gpuweb/cts#3920, by @alan-baker.
-
Only report on first texture builtin error gpuweb/cts#3919, by @greggman.
-
Fix for textureGather gpuweb/cts#3918, by @greggman.
-
Skip tests failing when VideoFrame is not supported. gpuweb/cts#3917, by @Kangz.
-
Fix subgroup_size fragment tests to be valid WGSL. gpuweb/cts#3916, by @Kangz.
-
Tests for subgroup builtin values in fragment shaders gpuweb/cts#3915, by @alan-baker.
-
WGSL textureGatherCompare tests gpuweb/cts#3914, by @greggman.
-
Choose smaller texture sizes gpuweb/cts#3912, by @greggman.
-
Print more info in certain asserts gpuweb/cts#3911, by @greggman.
-
Reduce the number of shader modules used in texture tests gpuweb/cts#3910, by @greggman.
-
ldexp validation: conditionally include tested code in the shader gpuweb/cts#3908, by @dneto0.
-
Fix for texture_utils.ts gpuweb/cts#3907, by @greggman.
-
extractBits validation: conditionally include tested code in the shader gpuweb/cts#3906, by @dneto0.
-
insertBits validation: conditionally include tested code in the shader gpuweb/cts#3905, by @dneto0.
-
clamp validation: conditionally include test function gpuweb/cts#3904, by @dneto0.
-
Fix textureSampleLevel tests. gpuweb/cts#3903, by @greggman.
-
Compat: Fix textureLoad with depth texture test gpuweb/cts#3902, by @greggman.
-
fix smoothstep execution: only compute valid cases when const gpuweb/cts#3901, by @dneto0.
-
fix smoothstep validation gpuweb/cts#3900, by @dneto0.
-
validation: clamp low_high test must reference the test function gpuweb/cts#3899, by @dneto0.
-
WGSL textureGather tests gpuweb/cts#3898, by @greggman.
-
CTS for subgroupAdd and subgroupMul gpuweb/cts#3897, by @alan-baker.
-
WGSL execution tests for textureSampleBaseClampToEdge gpuweb/cts#3896, by @greggman.
-
Add Chrome OT token for "WebGPU Subgroups Feature" gpuweb/cts#3895, by @dneto0.
-
Switch compat WGSL errors to pipeline errors gpuweb/cts#3894, by @greggman.
-
WGSL textureLoad tests for external texture gpuweb/cts#3891, by @greggman.
-
Add WGSL textureLoad tests for storage formats gpuweb/cts#3890, by @greggman.
-
Stop Sooner gpuweb/cts#3889, by @greggman.
-
WGSL textureSampleLevel execution tests gpuweb/cts#3888, by @greggman.
-
Fix error in case gpu_external_texture_expiration:use_import_to_refre… gpuweb/cts#3887, by @shaoboyan.
-
Add validation tests on the access of the alpha channel of Src1 gpuweb/cts#3886, by @Jiawei-Shao.
-
-
38 pull requests merged.
-
Execution tests for subgroupAny and subgroupAll gpuweb/cts#3924.
-
Fix another edge case for textureGather. gpuweb/cts#3923.
-
Fix textureGather gpuweb/cts#3922.
-
Skip tests failing when VideoFrame is not supported. gpuweb/cts#3917.
-
Print more info for texture builtins gpuweb/cts#3921.
-
Remaining subgroup validation tests gpuweb/cts#3920.
-
CTS for subgroupAdd and subgroupMul gpuweb/cts#3897.
-
Only report on first texture builtin error gpuweb/cts#3919.
-
Fix for textureGather gpuweb/cts#3918.
-
Fix subgroup_size fragment tests to be valid WGSL. gpuweb/cts#3916.
-
Add Chrome OT token for "WebGPU Subgroups Feature" gpuweb/cts#3895.
-
Tests for subgroup builtin values in fragment shaders gpuweb/cts#3915.
-
WGSL textureGatherCompare tests gpuweb/cts#3914.
-
WGSL textureGather tests gpuweb/cts#3898.
-
Choose smaller texture sizes gpuweb/cts#3912.
-
Print more info in certain asserts gpuweb/cts#3911.
-
Reduce the number of shader modules used in texture tests gpuweb/cts#3910.
-
ldexp validation: conditionally include tested code in the shader gpuweb/cts#3908.
-
Stop Sooner gpuweb/cts#3889.
-
Fix for texture_utils.ts gpuweb/cts#3907.
-
Compat: Fix textureLoad with depth texture test gpuweb/cts#3902.
-
clamp validation: conditionally include test function gpuweb/cts#3904.
-
fix smoothstep execution: only compute valid cases when const gpuweb/cts#3901.
-
Fix textureSampleLevel tests. gpuweb/cts#3903.
-
fix smoothstep validation gpuweb/cts#3900.
-
validation: clamp low_high test must reference the test function gpuweb/cts#3899.
-
WGSL execution tests for textureSampleBaseClampToEdge gpuweb/cts#3896.
-
feat(standalone): add `title` to test node gpuweb/cts#3880.
-
WGSL textureLoad tests for external texture gpuweb/cts#3891.
-
Switch compat WGSL errors to pipeline errors gpuweb/cts#3894.
-
Add WGSL textureLoad tests for storage formats gpuweb/cts#3890.
-
WGSL textureSampleLevel execution tests gpuweb/cts#3888.
-
Add validation tests on the access of the alpha channel of Src1 gpuweb/cts#3886.
-
Use device subgroup limits in subgroup_size tests gpuweb/cts#3885.
-
-
11 pull requests received 82 new comments.
-
Execution tests for subgroupAny and subgroupAll gpuweb/cts#3924 (11 comments by @alan-baker, @dneto0)
-
insertBits validation: conditionally include tested code in the shader gpuweb/cts#3905 (1 comment by @dneto0)
-
Fix error in case gpu_external_texture_expiration:use_import_to_refre… gpuweb/cts#3887 (1 comment by @shaoboyan)
TR publications
-
18 documents published.
-
WebGPU Shading Language published 6 times as a Working Draft. History of WebGPU Shading Language
-
WebGPU published 12 times as a Working Draft. History of WebGPU
-
Participants
-
2 participants joined.
-
Xiaoshen X (W3C Invited Experts) joined.
-
Tarek Sherif (Datadog) joined.
-
-
1 organization joined.
-
Datadog joined.
-
GitHub
gpuweb/gpuweb
-
18 issues opened.
-
[editorial] Use infra spec's [=struct=] and [=tuple=] gpuweb/gpuweb#4800, by @kainino0x.
-
More strongly push developers toward correct use of GPUAdapter gpuweb/gpuweb#4784, by @kainino0x.
-
array_index out of bounds behavior is not specified for texture builtins gpuweb/gpuweb#4782, by @greggman.
-
The value of the fragment builtin `@position` gpuweb/gpuweb#4777, by @greggman.
-
What should limits related to features repo on the device when the feature is not enabled? gpuweb/gpuweb#4776, by @greggman.
-
Improve styling of colspans and checkmarks in tables gpuweb/gpuweb#4775, by @kainino0x.
-
On device loss, should mapAsync throw AbortError instead of OperationError? gpuweb/gpuweb#4774, by @kainino0x.
-
Extend `bgra8unorm-storage` feature to allow `read-only` access gpuweb/gpuweb#4759, by @kainino0x.
-
Confused on the definition and effect of toneMapping gpuweb/gpuweb#4756, by @domenic.
-
WebGPU and WGSL W3C Working Drafts on Chrome desktop consume way too much RAM: up to 542 MB gpuweb/gpuweb#4749, by @Tomycj.
-
Implementation Status Page Status Clarification gpuweb/gpuweb#4745, by @radiantone.
-
Clarify the timeline of execution of create*PipelineAsync gpuweb/gpuweb#4742, by @kainino0x.
-
Add `subgroups-f16` feature dependency validation to subgroups proposal? gpuweb/gpuweb#4734, by @jiangzhaoming.
-
Depth Bias does not affect line and point primitives. gpuweb/gpuweb#4729, by @toji.
-
Should invalid combinations of storage texture formats and access generate a WGSL error gpuweb/gpuweb#4728, by @greggman.
-
Should the definition of "isFiltering" be more explicit? gpuweb/gpuweb#4727, by @greggman.
-
-
65 issues closed.
-
array_index out of bounds behavior is not specified for texture builtins gpuweb/gpuweb#4782.
-
clarify when const-exprs subexpressions are evaluated gpuweb/gpuweb#4558.
-
Clarify whether type checking occurs for short-circuited RHS expressions gpuweb/gpuweb#4551.
-
Pointer chaining in WGSL gpuweb/gpuweb#1530.
-
Potential parsing conflict with const literals and postfix expressions. gpuweb/gpuweb#2772.
-
wgsl: describe how the name `vec4` in `vec4<f32>` is resolved gpuweb/gpuweb#3280.
-
handles aren't storable, but are the contents of variables gpuweb/gpuweb#1260.
-
Document that internal memories are zero-initialized: workgroup, output gpuweb/gpuweb#341.
-
Issue with type converting module scoped variables gpuweb/gpuweb#1104.
-
update description of shader-output mask gpuweb/gpuweb#1539.
-
wsgl:editorial: missing access mode in a few descriptions gpuweb/gpuweb#1992.
-
editorial: do we need a definition of "object"? gpuweb/gpuweb#1538.
-
wgsl: enable_directive grammar and example mismatch gpuweb/gpuweb#2173.
-
Clarify behavior of float division/remainder by 0 gpuweb/gpuweb#2798.
-
editorial: add example showing how to write an i32 of most negative value gpuweb/gpuweb#2872.
-
Behaviour of overflowing arithmetic operations gpuweb/gpuweb#2875.
-
wgsl: key use cases for Vector and Matrix seem out of place gpuweb/gpuweb#2976.
-
Clarification of dynamic nature of function calls gpuweb/gpuweb#3199.
-
Clarification around `sampled type` in the Texture Builtin section. gpuweb/gpuweb#2849.
-
Clarification of WGSL builtin function availability gpuweb/gpuweb#3192.
-
Recursive structs? gpuweb/gpuweb#4700.
-
Does an override need to be statically used by the entry point to be allowed? gpuweb/gpuweb#4624.
-
On device loss, should mapAsync throw AbortError instead of OperationError? gpuweb/gpuweb#4774.
-
reuse of an GPUAdapter to request a GPUDevice does not reject. gpuweb/gpuweb#4352.
-
Consider moving sliced 3D compressed BC texture to an extension. gpuweb/gpuweb#4705.
-
Function attribute duplication gpuweb/gpuweb#3203.
-
Editorial: Out-of-bounds accesses can occur without interacting with memory gpuweb/gpuweb#3273.
-
Accuracy of `mix` does not cover all common implementations of linear blend gpuweb/gpuweb#3260.
-
Should WGSL expose the concept of coherent buffers? gpuweb/gpuweb#1621.
-
Add to text: Global vars/consts declaration initializers must be const_expr gpuweb/gpuweb#1975.
-
wgsl editorial: "entry point IO types" are not types gpuweb/gpuweb#2442.
-
`@interpolate` values on reserved word list gpuweb/gpuweb#2913.
-
Clarify note about write-only storage textures in 4.4.6 Address Spaces. gpuweb/gpuweb#3047.
-
Remove the `Type-defining Keywords` section gpuweb/gpuweb#3277.
-
wgsl: add a note saying function prototypes are not required gpuweb/gpuweb#3779.
-
Update arithmetic expressions with results of over/underflow on abstract types gpuweb/gpuweb#3132.
-
integer bit shift by bit width or more gpuweb/gpuweb#289.
-
[wgsl] Limit number of function parameters to 255 gpuweb/gpuweb#1959.
-
Consider reserving keywords for a module system gpuweb/gpuweb#2607.
-
Remove remaining mentions of IO-shareable gpuweb/gpuweb#3735.
-
Clarification in limits section gpuweb/gpuweb#3217.
-
Change `creation-time constant` in Value Declarations to `const`? gpuweb/gpuweb#3061.
-
Make lost devices appear to function as much as possible gpuweb/gpuweb#1629.
-
Clarify the timeline of execution of create*PipelineAsync gpuweb/gpuweb#4742.
-
Clarify that create*PipelineAsync must not generate a validation error gpuweb/gpuweb#4714.
-
Distinction between content-timeline and device-timeline state gpuweb/gpuweb#2962.
-
Depth Bias does not affect line and point primitives. gpuweb/gpuweb#4729.
-
VideoFrame should use display size, not coded size gpuweb/gpuweb#4525.
-
HTMLCanvasElement copyExternalImageToTexture under specifed? gpuweb/gpuweb#4255.
-
Spec allows resolveTarget to be 3d but doesn't allow selecting a depth slice gpuweb/gpuweb#4600.
-
The output range of atan is wrong in the doc gpuweb/gpuweb#4376.
-
Define exclusivity of `Function Tag` and `Parameter Tag` values. gpuweb/gpuweb#3223.
-
describe the "sample footprint" gpuweb/gpuweb#2343.
-
Validation for setViewport should be `minDepth <= maxDepth` gpuweb/gpuweb#4461.
-
No way to destroy a shader? gpuweb/gpuweb#4116.
-
79 issues received 184 new comments.
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (18 comments by @greggman, @kainino0x, @Kangz, @SenorBlanco, @teoxoy)
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426 (12 comments by @kainino0x, @Kangz, @kdashg, @teoxoy)
-
Should invalid combinations of storage texture formats and access generate a WGSL error gpuweb/gpuweb#4728 (11 comments by @alan-baker, @greggman, @jimblandy, @kainino0x, @kdashg, @teoxoy)
-
array_index out of bounds behavior is not specified for texture builtins gpuweb/gpuweb#4782 (7 comments by @alan-baker, @greggman, @jimblandy, @kdashg, @magcius)
-
maxInterStageShaderComponents seems to overlap with maxInterStageShaderVariables gpuweb/gpuweb#4688 (7 comments by @kainino0x, @Kangz, @kdashg, @mwyrzykowski)
-
mapAsync early-reject doesn't generate a validation error gpuweb/gpuweb#4690 (6 comments by @kainino0x, @Kangz, @teoxoy)
-
What happens on the device timeline after `GPUDevice.destroy` is called? gpuweb/gpuweb#4294 (5 comments by @kainino0x, @teoxoy)
-
On device loss, should mapAsync throw AbortError instead of OperationError? gpuweb/gpuweb#4774 (5 comments by @jimblandy, @kdashg)
-
writing to storage buffers/textures in fragment shaders with multisample targets is not portable. gpuweb/gpuweb#4696 (4 comments by @greggman, @Kangz, @mwyrzykowski)
-
Allow all vertex input attributes to be considered OOB if at least one is (in the "process vertices" algorithm) gpuweb/gpuweb#4668 (4 comments by @Kangz, @mwyrzykowski, @teoxoy)
-
The `bgra8unorm-storage` feature should most likely only enable `write-only` access gpuweb/gpuweb#4377 (4 comments by @alan-baker, @kainino0x, @Kangz)
-
Confused on the definition and effect of toneMapping gpuweb/gpuweb#4756 (4 comments by @ccameron-chromium, @domenic, @kainino0x, @Kangz)
-
WebGPU and WGSL W3C Working Drafts on Chrome desktop consume way too much RAM: up to 542 MB gpuweb/gpuweb#4749 (4 comments by @greggman, @mehmetoguzderin, @Tomycj)
-
Add `subgroups-f16` feature dependency validation to subgroups proposal? gpuweb/gpuweb#4734 (3 comments by @alan-baker, @teoxoy)
-
What should limits related to features repo on the device when the feature is not enabled? gpuweb/gpuweb#4776 (3 comments by @greggman, @Kangz)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (3 comments by @kainino0x, @Kangz, @teoxoy)
-
The value of the fragment builtin `@position` gpuweb/gpuweb#4777 (3 comments by @dneto0, @greggman, @kainino0x)
-
Should copyExternalImageToTexture allow a slice of a 3d texture as a destination? gpuweb/gpuweb#4697 (3 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
Clarify that create*PipelineAsync must not generate a validation error gpuweb/gpuweb#4714 (3 comments by @jimblandy, @Kangz)
-
wgsl: clarify that type inference occurs independent of the numeric value of a constant-time expression gpuweb/gpuweb#2830 (2 comments by @alan-baker, @dneto0)
-
bgra8unorm storage textures without bgra8unorm-storage, should generate WGSL error? gpuweb/gpuweb#4711 (2 comments by @kainino0x, @Kangz)
-
reuse of an GPUAdapter to request a GPUDevice does not reject. gpuweb/gpuweb#4352 (2 comments by @kdashg, @teoxoy)
-
Criteria for pushing the WebGPU and WGSL specification to Candidate Recommendation gpuweb/gpuweb#3107 (2 comments by @kdashg)
-
Improve color and display-size readability of diagrams and code blocks and other elements in WGSL spec gpuweb/gpuweb#3649 (2 comments by @kdashg, @mehmetoguzderin)
-
Clarify the timeline of execution of create*PipelineAsync gpuweb/gpuweb#4742 (2 comments by @kainino0x)
-
Resurrect vertex formats dropped for macOS < 10.13 gpuweb/gpuweb#4549 (2 comments by @kainino0x, @Kangz)
-
Consider moving sliced 3D compressed BC texture to an extension. gpuweb/gpuweb#4705 (2 comments by @jimblandy, @Kangz)
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568 (1 comment by @alan-baker)
-
Potential parsing conflict with const literals and postfix expressions. gpuweb/gpuweb#2772 (1 comment by @dneto0)
-
wsgl:editorial: missing access mode in a few descriptions gpuweb/gpuweb#1992 (1 comment by @alan-baker)
-
editorial: add example showing how to write an i32 of most negative value gpuweb/gpuweb#2872 (1 comment by @alan-baker)
-
wgsl: key use cases for Vector and Matrix seem out of place gpuweb/gpuweb#2976 (1 comment by @alan-baker)
-
Clarification around `sampled type` in the Texture Builtin section. gpuweb/gpuweb#2849 (1 comment by @alan-baker)
-
Do pipeline-creation errors encompass the whole module or just the GPUProgrammableStage? gpuweb/gpuweb#4648 (1 comment by @kdashg)
-
Accuracy of `mix` does not cover all common implementations of linear blend gpuweb/gpuweb#3260 (1 comment by @kdashg)
-
Add to text: Global vars/consts declaration initializers must be const_expr gpuweb/gpuweb#1975 (1 comment by @alan-baker)
-
Clarify note about write-only storage textures in 4.4.6 Address Spaces. gpuweb/gpuweb#3047 (1 comment by @alan-baker)
-
Spec web page word highlight feature has low contrast in dark mode gpuweb/gpuweb#4598 (1 comment by @greggman)
-
Accuracy of `acosh` is does not include Vulkan CTS alternative gpuweb/gpuweb#3287 (1 comment by @kdashg)
-
treesitter custom scanner should treat blankspace, block comments, and line-ending comments consistently gpuweb/gpuweb#3876 (1 comment by @kdashg)
-
Make lost devices appear to function as much as possible gpuweb/gpuweb#1629 (1 comment by @kainino0x)
-
Distinction between content-timeline and device-timeline state gpuweb/gpuweb#2962 (1 comment by @kainino0x)
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558 (1 comment by @kainino0x)
-
Spec says “should warn” for unicode homographs. Should we allow control of that one via diagnostics? gpuweb/gpuweb#3688 (1 comment by @kdashg)
-
HTMLCanvasElement copyExternalImageToTexture under specifed? gpuweb/gpuweb#4255 (1 comment by @toji)
-
Spec allows resolveTarget to be 3d but doesn't allow selecting a depth slice gpuweb/gpuweb#4600 (1 comment by @toji)
-
Does an override need to be statically used by the entry point to be allowed? gpuweb/gpuweb#4624 (1 comment by @teoxoy)
-
Compat: Should requesting compat mode on a device that supports core, return core or compat? gpuweb/gpuweb#4731 (1 comment by @Kangz)
-
Define exclusivity of `Function Tag` and `Parameter Tag` values. gpuweb/gpuweb#3223 (1 comment by @dneto0)
-
Validation for setViewport should be `minDepth <= maxDepth` gpuweb/gpuweb#4461 (1 comment by @kainino0x)
-
Clamping behavior when comparison sampling `depth24plus` gpuweb/gpuweb#4653 (1 comment by @kainino0x)
-
-
50 pull requests opened.
-
Add shader module creation validation for features gpuweb/gpuweb#4802, by @alan-baker.
-
Clarify const-expression eval and type checking gpuweb/gpuweb#4801, by @alan-baker.
-
[editorial] Simplify attachment usage validation gpuweb/gpuweb#4799, by @kainino0x.
-
[editorial] Make usage scope validation algorithmic gpuweb/gpuweb#4797, by @kainino0x.
-
[editorial] Add timelines to all attributes/slots, assorted editing gpuweb/gpuweb#4796, by @kainino0x.
-
Add extra example for operations involving abstract numerics gpuweb/gpuweb#4795, by @matthew-wong1.
-
Remove mentions of multithreaded API gpuweb/gpuweb#4794, by @kainino0x.
-
[editorial] Link to a definition for [=implementation-defined=] gpuweb/gpuweb#4793, by @kainino0x.
-
Clarify array layer OOB for texture sampling gpuweb/gpuweb#4792, by @alan-baker.
-
Fix rules about dual source blending in `validating GPUFragmentState` gpuweb/gpuweb#4791, by @Jiawei-Shao.
-
If any vertex is OOB, the whole draw call may behave as OOB gpuweb/gpuweb#4790, by @kainino0x.
-
Reject requestDevice() if the adapter has already been used gpuweb/gpuweb#4789, by @kainino0x.
-
Disable async operations after device destruction starts gpuweb/gpuweb#4788, by @kainino0x.
-
Change mapAsync to also early-reject if already mapped gpuweb/gpuweb#4787, by @kainino0x.
-
Generate a validation error on mapAsync early-reject gpuweb/gpuweb#4786, by @kainino0x.
-
Change mapAsync to reject with AbortError on device loss gpuweb/gpuweb#4785, by @kainino0x.
-
Remove maxInterStageShaderComponents gpuweb/gpuweb#4783, by @kainino0x.
-
Allow unknown limits to be requested with value `undefined` gpuweb/gpuweb#4781, by @kainino0x.
-
[editorial] Improve more language relating to limit defaults gpuweb/gpuweb#4780, by @kainino0x.
-
[push-constants] remove per-stage immediates open question gpuweb/gpuweb#4779, by @teoxoy.
-
Add note that enabling features can have performance impact gpuweb/gpuweb#4778, by @kainino0x.
-
Note that leading zeros are forbidden because no octals. gpuweb/gpuweb#4773, by @kdashg.
-
`mix` inherits accuracy from either `x * (1.0 - z) + y * z` or `x + (y - x) * z`. gpuweb/gpuweb#4772, by @kdashg.
-
Clarify textureSampleBias bias param gpuweb/gpuweb#4771, by @alan-baker.
-
Mention any-order function definition in #functions, and note lack of forward decls. gpuweb/gpuweb#4770, by @kdashg.
-
Remove statement that uniformity is linear in number of tokens gpuweb/gpuweb#4769, by @alan-baker.
-
Add inverse channel transfer function gpuweb/gpuweb#4768, by @alan-baker.
-
Move overflow rules for concrete integers gpuweb/gpuweb#4767, by @alan-baker.
-
Clarify inherited-from allows flushing gpuweb/gpuweb#4766, by @alan-baker.
-
Clarify stage restrictions for IO attributes gpuweb/gpuweb#4765, by @alan-baker.
-
Add `texture-compression-bc-sliced-3d` gpuweb/gpuweb#4763, by @mehmetoguzderin.
-
Require `module` to be "valid to use with" `device` gpuweb/gpuweb#4762, by @teoxoy.
-
Feature Levels gpuweb/gpuweb#4761, by @kainino0x.
-
Fix some language which ignores that limits get clamped to defaults gpuweb/gpuweb#4760, by @kainino0x.
-
Clarify counting of case values for their maximum limit. gpuweb/gpuweb#4758, by @kdashg.
-
Add missing "valid to use with" checks to `GPURenderPassDescriptor` Valid Usage gpuweb/gpuweb#4757, by @teoxoy.
-
Resolve line rasterization editorial note gpuweb/gpuweb#4754, by @toji.
-
Resolve editorial note for rasterization points gpuweb/gpuweb#4753, by @toji.
-
Suppress error generation in create*PipelineAsync, clarify async steps gpuweb/gpuweb#4752, by @kainino0x.
-
Add alt text to CSS "content" gpuweb/gpuweb#4751, by @kainino0x.
-
Add usage to GPUTextureViewDescriptor gpuweb/gpuweb#4746, by @toji.
-
Point and line primitives must not have depth bias gpuweb/gpuweb#4743, by @toji.
-
bgra8unorm-storage only enables write-only gpuweb/gpuweb#4741, by @toji.
-
Update VideoFrames to use display size gpuweb/gpuweb#4740, by @toji.
-
Clarify blend factors must be validated against optional features gpuweb/gpuweb#4739, by @Jiawei-Shao.
-
Define canvas image contents when unconfigured gpuweb/gpuweb#4738, by @toji.
-
Resolve read-only depth/stencil inline issue gpuweb/gpuweb#4737, by @toji.
-
Fix GPUBlendFactor indentation gpuweb/gpuweb#4736, by @beaufortfrancois.
-
wgsl: Fix range of atan: starts at -pi/2, not pi/2 gpuweb/gpuweb#4732, by @dneto0.
-
Update viewport depth range validation gpuweb/gpuweb#4730, by @toji.
-
-
41 pull requests merged.
-
Explain the endianness of pack4xI8* and pack4xU8* builtins gpuweb/gpuweb#4425.
-
Clarify array layer OOB for texture sampling gpuweb/gpuweb#4792.
-
Clarify textureSampleBias bias param gpuweb/gpuweb#4771.
-
Clarify const-expression eval and type checking gpuweb/gpuweb#4801.
-
[editorial] Simplify attachment usage validation gpuweb/gpuweb#4799.
-
[editorial] Improve more language relating to limit defaults gpuweb/gpuweb#4780.
-
[editorial] Make usage scope validation algorithmic gpuweb/gpuweb#4797.
-
[editorial] Link to a definition for [=implementation-defined=] gpuweb/gpuweb#4793.
-
Change mapAsync to reject with AbortError on device loss gpuweb/gpuweb#4785.
-
Remove mentions of multithreaded API gpuweb/gpuweb#4794.
-
Reject requestDevice() if the adapter has already been used gpuweb/gpuweb#4789.
-
Add `texture-compression-bc-sliced-3d` gpuweb/gpuweb#4763.
-
If any vertex is OOB, the whole draw call may behave as OOB gpuweb/gpuweb#4790.
-
Add inverse channel transfer function gpuweb/gpuweb#4768.
-
Remove statement that uniformity is linear in number of tokens gpuweb/gpuweb#4769.
-
Clarify stage restrictions for IO attributes gpuweb/gpuweb#4765.
-
Clarify inherited-from allows flushing gpuweb/gpuweb#4766.
-
Move overflow rules for concrete integers gpuweb/gpuweb#4767.
-
Note that leading zeros are forbidden because no octals. gpuweb/gpuweb#4773.
-
Clarify counting of case values for their maximum limit. gpuweb/gpuweb#4758.
-
Require `module` to be "valid to use with" `device` gpuweb/gpuweb#4762.
-
Resolve editorial note for rasterization points gpuweb/gpuweb#4753.
-
Fix some language which ignores that limits get clamped to defaults gpuweb/gpuweb#4760.
-
Suppress error generation in create*PipelineAsync, clarify async steps gpuweb/gpuweb#4752.
-
Add missing "valid to use with" checks to `GPURenderPassDescriptor` Valid Usage gpuweb/gpuweb#4757.
-
bgra8unorm-storage only enables write-only gpuweb/gpuweb#4741.
-
Resolve line rasterization editorial note gpuweb/gpuweb#4754.
-
Add alt text to CSS "content" gpuweb/gpuweb#4751.
-
Push constant proposal gpuweb/gpuweb#4612.
-
Point and line primitives must not have depth bias gpuweb/gpuweb#4743.
-
Update VideoFrames to use display size gpuweb/gpuweb#4740.
-
Clarify blend factors must be validated against optional features gpuweb/gpuweb#4739.
-
Define canvas image contents when unconfigured gpuweb/gpuweb#4738.
-
Resolve read-only depth/stencil inline issue gpuweb/gpuweb#4737.
-
Fix GPUBlendFactor indentation gpuweb/gpuweb#4736.
-
Clarify scope of pipeline creation gpuweb/gpuweb#4724.
-
wgsl: Fix range of atan: starts at -pi/2, not pi/2 gpuweb/gpuweb#4732.
-
Add description of sample footprint gpuweb/gpuweb#4683.
-
Update viewport depth range validation gpuweb/gpuweb#4730.
-
Define level of detail gpuweb/gpuweb#4694.
-
-
3 pull requests closed.
-
Clarify the use of `@blend_src` in certain shader stages gpuweb/gpuweb#4723.
-
Move Detailed Operations section into separate doc gpuweb/gpuweb#4633.
-
-
2 pull requests reopened.
-
Fix rules about dual source blending in `validating GPUFragmentState` gpuweb/gpuweb#4791, by @Jiawei-Shao.
-
Clarify array layer OOB for texture sampling gpuweb/gpuweb#4792, by @alan-baker.
-
-
31 pull requests received 181 new comments.
-
Add `texture-compression-bc-sliced-3d` gpuweb/gpuweb#4763 (37 comments by @kainino0x, @Kangz, @kdashg, @mehmetoguzderin)
-
[editorial] Make usage scope validation algorithmic gpuweb/gpuweb#4797 (16 comments by @kainino0x, @Kangz)
-
Push constant proposal gpuweb/gpuweb#4612 (15 comments by @kainino0x, @Kangz, @kdashg, @magcius, @mwyrzykowski, @shaoboyan, @teoxoy, @toji)
-
Clarify array layer OOB for texture sampling gpuweb/gpuweb#4792 (12 comments by @alan-baker, @greggman)
-
Suppress error generation in create*PipelineAsync, clarify async steps gpuweb/gpuweb#4752 (11 comments by @kainino0x, @toji)
-
Fix rules about dual source blending in `validating GPUFragmentState` gpuweb/gpuweb#4791 (8 comments by @Jiawei-Shao, @kainino0x, @Kangz)
-
[editorial] Improve more language relating to limit defaults gpuweb/gpuweb#4780 (5 comments by @jimblandy, @kainino0x)
-
Add usage to GPUTextureViewDescriptor gpuweb/gpuweb#4746 (5 comments by @greggman, @kainino0x, @toji)
-
Clarify textureSampleBias bias param gpuweb/gpuweb#4771 (4 comments by @alan-baker, @dneto0, @kdashg)
-
Clarify const-expression eval and type checking gpuweb/gpuweb#4801 (4 comments by @alan-baker, @jrprice)
-
If any vertex is OOB, the whole draw call may behave as OOB gpuweb/gpuweb#4790 (4 comments by @kainino0x, @Kangz, @teoxoy)
-
Note that leading zeros are forbidden because no octals. gpuweb/gpuweb#4773 (3 comments by @kdashg, @mehmetoguzderin)
-
Mention any-order function definition in #functions, and note lack of forward decls. gpuweb/gpuweb#4770 (3 comments by @alan-baker, @Kangz, @kdashg)
-
Resolve editorial note for rasterization points gpuweb/gpuweb#4753 (3 comments by @kainino0x, @toji)
-
Clarify blend factors must be validated against optional features gpuweb/gpuweb#4739 (3 comments by @Jiawei-Shao, @kainino0x, @toji)
-
Guarantee that OOB writes to storage textures will be ignored gpuweb/gpuweb#4194 (1 comment by @kdashg)
-
`mix` inherits accuracy from either `x * (1.0 - z) + y * z` or `x + (y - x) * z`. gpuweb/gpuweb#4772 (1 comment by @alan-baker)
-
gpuweb/cts
-
13 issues opened.
-
Test that bias in textureSampleBias is clamped to [-16.0, 15.99] gpuweb/cts#3883, by @alan-baker.
-
Refactor WGSL textureLoad execution tests to test more texels gpuweb/cts#3871, by @greggman.
-
Create*PipelineAsync on lost device gpuweb/cts#3867, by @sagudev.
-
Rejection on createComputePipelineAsync causes ERROR on WPT test harness gpuweb/cts#3865, by @sagudev.
-
Consolidate texture format lists for tests gpuweb/cts#3854, by @greggman.
-
Texture format capability info does not take device into account gpuweb/cts#3848, by @kainino0x.
-
skipIfTextureFormatNotUsableAsStorageTexture does not handle bgra8unorm well gpuweb/cts#3847, by @alan-baker.
-
-
16 issues closed.
-
add f16 to kScalarTypeInfo in src/webgpu/shader/types.ts gpuweb/cts#3405.
-
Create*PipelineAsync on lost device gpuweb/cts#3867.
-
Rejection on createComputePipelineAsnyc causes timout in tests gpuweb/cts#3809.
-
wgsl: test scalar value constructors from abstracts gpuweb/cts#3254.
-
There are more createRenderPipeline without targets. gpuweb/cts#3806.
-
Enforce TS's `readonly` more strictly gpuweb/cts#3126.
-
wgsl: Bitwise tests gpuweb/cts#1631.
-
wgsl: if LHS is signed, the right shift is arithmetic gpuweb/cts#1853.
-
-
1 issue reopened.
-
wgsl: add validation and execution tests for all storage texture formats gpuweb/cts#2268, by @alan-baker.
-
-
17 issues received 27 new comments.
-
bitwise_shift tests appear to be non-conformant, fails on Chrome and Safari Technology Preview gpuweb/cts#3857 (4 comments by @kainino0x, @mwyrzykowski, @zoddicus)
-
Texture format capability info does not take device into account gpuweb/cts#3848 (3 comments by @greggman, @kainino0x)
-
Rejection on createComputePipelineAsnyc causes timout in tests gpuweb/cts#3809 (2 comments by @sagudev)
-
wgsl: add validation and execution tests for all storage texture formats gpuweb/cts#2268 (2 comments by @alan-baker)
-
wgsl: test scalar value constructors from abstracts gpuweb/cts#3254 (2 comments by @alan-baker, @zoddicus)
-
wgsl: matrix_scalar_compund tests fail because compoundAssignmentBuilder assumes LHS and RHS are the same type gpuweb/cts#3842 (2 comments by @dneto0)
-
`npm run wpt` generates `*.worker.js` tests that accidentally get run by `wptrunner` gpuweb/cts#3834 (2 comments by @ErichDonGubler)
-
Rejection on createComputePipelineAsync causes ERROR on WPT test harness gpuweb/cts#3865 (1 comment by @sagudev)
-
eslint (with plugins) takes a very long time and uses a lot of memory gpuweb/cts#3135 (1 comment by @kainino0x)
-
Expectation in dynamic_state.spec.ts does not align with the specification gpuweb/cts#3307 (1 comment by @kainino0x)
-
[wgsl] Validate errors are not generated for overrides which aren't present in the module. gpuweb/cts#3747 (1 comment by @alan-baker)
-
-
44 pull requests opened.
-
Use device subgroup limits in subgroup_size tests gpuweb/cts#3885, by @alan-baker.
-
Add dual source blending blend factors to `kBlendFactors` gpuweb/cts#3884, by @Jiawei-Shao.
-
Add operation tests on all blend factors using `Src1` gpuweb/cts#3882, by @Jiawei-Shao.
-
Add validation tests on blend factors with Src1 and @blend_src gpuweb/cts#3881, by @Jiawei-Shao.
-
feat(standalone): add `title` to test node gpuweb/cts#3880, by @ErichDonGubler.
-
Add validation tests on the blend factors using the alpha channel of src1 gpuweb/cts#3879, by @Jiawei-Shao.
-
Add validation tests about dual source blending and color targets gpuweb/cts#3878, by @Jiawei-Shao.
-
Refactor WGSL textureLoad tests to test more texels gpuweb/cts#3877, by @greggman.
-
phony assignment is not valid at module scope gpuweb/cts#3876, by @dneto0.
-
test phony assignment execution gpuweb/cts#3875, by @dneto0.
-
Cover GPUExternalTexture with WebCam Cases gpuweb/cts#3874, by @shaoboyan.
-
Fix div-by-0 partial eval tests gpuweb/cts#3873, by @sudonatalie.
-
WGSL textureLoad tests for multisampled and depth textures gpuweb/cts#3872, by @greggman.
-
generateTypes: Avoid generating test case with invalid alignment gpuweb/cts#3870, by @dneto0.
-
wgsl: Add coverage for swizzle on a pointer gpuweb/cts#3869, by @jrprice.
-
Correct expectation for shift equal to bit width gpuweb/cts#3868, by @sudonatalie.
-
Catch and fail GPUPipelineError gpuweb/cts#3866, by @sagudev.
-
refactor shader/types.ts: Add accessSuffixes to type info gpuweb/cts#3864, by @dneto0.
-
Add CTS test for new depthBias validation gpuweb/cts#3863, by @toji.
-
Subgroups basic cts gpuweb/cts#3862, by @alan-baker.
-
WGSL textureLoad exe tests 1d, 3d, and arrayed gpuweb/cts#3861, by @greggman.
-
[compat] Fix or skip context dependent res cases gpuweb/cts#3858, by @sudonatalie.
-
WGSL textureNumSamples execution tests gpuweb/cts#3856, by @greggman.
-
WGSL textureNumLevels execution tests gpuweb/cts#3855, by @greggman.
-
WGSL textureNumLayers execution test gpuweb/cts#3853, by @greggman.
-
WGSL builtin textureLoad for texture_2d tests gpuweb/cts#3852, by @greggman.
-
Skip some context_dependent_resolution cases gpuweb/cts#3851, by @sudonatalie.
-
wgsl: Add missing f32 from abstract conversion tests gpuweb/cts#3850, by @zoddicus.
-
Add storage texel format tests for textureLoad gpuweb/cts#3849, by @alan-baker.
-
Fix compound assignment storage_rw when resultType != storage(resultT… gpuweb/cts#3846, by @dneto0.
-
Fix type costruction of RHS in storage_rw compoundAssignmentBuilder gpuweb/cts#3843, by @dneto0.
-
Add format tests for textureStore gpuweb/cts#3841, by @alan-baker.
-
Add validation tests on the types of `blend_src(0)` and `blend_src(1)` gpuweb/cts#3840, by @Jiawei-Shao.
-
Add shader validation tests on the usage of @blend_src gpuweb/cts#3839, by @Jiawei-Shao.
-
Fix typo in test name: "disjoinRanges" should be "disjointRanges". gpuweb/cts#3838, by @jimblandy.
-
compat: fix more flat either issues gpuweb/cts#3837, by @greggman.
-
Add overflow cases for left shift of abstract int gpuweb/cts#3836, by @alan-baker.
-
Rename *.worker.js to *.as_worker.js gpuweb/cts#3835, by @kainino0x.
-
Add tests that confirm abstract shifted by concrete is abstract gpuweb/cts#3833, by @alan-baker.
-
Perform compound assignments directly on RW storage gpuweb/cts#3832, by @alan-baker.
-
Early evaluation expression validation tests gpuweb/cts#3831, by @alan-baker.
-
Add validation tests on the use of struct with `blend_src` in shader io gpuweb/cts#3830, by @Jiawei-Shao.
-
Add shader syntax validation tests against `@blend_src` gpuweb/cts#3829, by @Jiawei-Shao.
-
-
42 pull requests merged.
-
Subgroups basic cts gpuweb/cts#3862.
-
Add dual source blending blend factors to `kBlendFactors` gpuweb/cts#3884.
-
Add operation tests on all blend factors using `Src1` gpuweb/cts#3882.
-
Add validation tests on blend factors with Src1 and @blend_src gpuweb/cts#3881.
-
Cover GPUExternalTexture with WebCam Cases gpuweb/cts#3874.
-
Refactor WGSL textureLoad tests to test more texels gpuweb/cts#3877.
-
Add validation tests about dual source blending and color targets gpuweb/cts#3878.
-
test phony assignment execution gpuweb/cts#3875.
-
phony assignment is not valid at module scope gpuweb/cts#3876.
-
WGSL textureLoad tests for multisampled and depth textures gpuweb/cts#3872.
-
wgsl: Add coverage for swizzle on a pointer gpuweb/cts#3869.
-
Fix div-by-0 partial eval tests gpuweb/cts#3873.
-
generateTypes: Avoid generating test case with invalid alignment gpuweb/cts#3870.
-
refactor shader/types.ts: Add accessSuffixes to type info gpuweb/cts#3864.
-
WGSL textureLoad exe tests 1d, 3d, and arrayed gpuweb/cts#3861.
-
Correct expectation for shift equal to bit width gpuweb/cts#3868.
-
Catch and fail GPUPipelineError gpuweb/cts#3866.
-
Add CTS test for new depthBias validation gpuweb/cts#3863.
-
WGSL builtin textureLoad for texture_2d tests gpuweb/cts#3852.
-
[compat] Fix or skip context dependent res cases gpuweb/cts#3858.
-
WGSL textureNumLevels execution tests gpuweb/cts#3855.
-
WGSL textureNumSamples execution tests gpuweb/cts#3856.
-
Fix typo in test name: "disjoinRanges" should be "disjointRanges". gpuweb/cts#3838.
-
WGSL textureNumLayers execution test gpuweb/cts#3853.
-
wgsl: Add missing f32 from abstract conversion tests gpuweb/cts#3850.
-
Add storage texel format tests for textureLoad gpuweb/cts#3849.
-
Add format tests for textureStore gpuweb/cts#3841.
-
compat: fix more flat either issues gpuweb/cts#3837.
-
Fix type costruction of RHS in storage_rw compoundAssignmentBuilder gpuweb/cts#3843.
-
Add shader validation tests on the usage of @blend_src gpuweb/cts#3839.
-
Rename *.worker.js to *.as_worker.js gpuweb/cts#3835.
-
Add overflow cases for left shift of abstract int gpuweb/cts#3836.
-
Add tests that confirm abstract shifted by concrete is abstract gpuweb/cts#3833.
-
Perform compound assignments directly on RW storage gpuweb/cts#3832.
-
Expand context dependent name test to include flat sampling gpuweb/cts#3824.
-
Early evaluation expression validation tests gpuweb/cts#3831.
-
Add shader syntax validation tests against `@blend_src` gpuweb/cts#3829.
-
-
1 pull request closed.
-
Skip some context_dependent_resolution cases gpuweb/cts#3851.
-
-
22 pull requests received 128 new comments.
-
Run batchFinishedCallback on processBatch to catch unhandled rejections by test functions gpuweb/cts#3844 (6 comments by @kainino0x, @sagudev)
-
WGSL textureLoad tests for multisampled and depth textures gpuweb/cts#3872 (4 comments by @greggman, @kainino0x)
-
WGSL textureLoad exe tests 1d, 3d, and arrayed gpuweb/cts#3861 (3 comments by @greggman, @kainino0x)
-
Fix typo in test name: "disjoinRanges" should be "disjointRanges". gpuweb/cts#3838 (1 comment by @kainino0x)
-
Fix compound assignment storage_rw when resultType != storage(resultT… gpuweb/cts#3846 (1 comment by @dneto0)
TR publications
-
17 documents published.
-
WebGPU published 12 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 5 times as a Working Draft. History of WebGPU Shading Language
-
GitHub
gpuweb/gpuweb
-
15 issues opened.
-
Clarify that create*PipelineAsync must not generate a validation error gpuweb/gpuweb#4714, by @sagudev.
-
Mistake in recursive decent parser definition? gpuweb/gpuweb#4713, by @timokoesters.
-
bgra8unorm storage textures without bgra8unorm-storage, should generate WGSL error? gpuweb/gpuweb#4711, by @greggman.
-
Define conversion from GPUColor to texture format gpuweb/gpuweb#4707, by @toji.
-
Consider moving sliced 3D compressed BC texture to an extension. gpuweb/gpuweb#4705, by @Kangz.
-
WGSL API proposal: shaderLog gpuweb/gpuweb#4704, by @mwyrzykowski.
-
Support for 3D Textures with ASTC gpuweb/gpuweb#4701, by @mehmetoguzderin.
-
Recursive structs? gpuweb/gpuweb#4700, by @Alphapage.
-
Should copyExternalImageToTexture allow a slice of a 3d texture as a destination? gpuweb/gpuweb#4697, by @greggman.
-
GPUDeviceLostReason should contain more values gpuweb/gpuweb#4695, by @mwyrzykowski.
-
Optional<T> type, like in Rust gpuweb/gpuweb#4692, by @Alphapage.
-
mapAsync early-reject doesn't generate a validation error gpuweb/gpuweb#4690, by @kainino0x.
-
maxInterStageShaderComponents seems to overlap with maxInterStageShaderVariables gpuweb/gpuweb#4688, by @mwyrzykowski.
-
10 issues closed.
-
Extended brightness range rendering gpuweb/gpuweb#4239.
-
Dual Source Blending Investigation + Feature Proposal gpuweb/gpuweb#4283.
-
Add support for 3D compressed textures. gpuweb/gpuweb#3183.
-
GPUDeviceLostReason should contain more values gpuweb/gpuweb#4695.
-
Standard Sample Pattern gpuweb/gpuweb#4005.
-
Type inconsistency between drawIndexed and drawIndexedIndirect calls gpuweb/gpuweb#4674.
-
The format of "64-bit unsigned integers" is never specified gpuweb/gpuweb#4289.
-
Optional<T> type, like in Rust gpuweb/gpuweb#4692.
-
Clarification regarding double negation (--) operator? gpuweb/gpuweb#4669.
-
-
40 issues received 121 new comments.
-
Consider moving sliced 3D compressed BC texture to an extension. gpuweb/gpuweb#4705 (14 comments by @kainino0x, @Kangz, @magcius, @mehmetoguzderin, @teoxoy)
-
WebGPU Compatibility Mode gpuweb/gpuweb#4266 (9 comments by @greggman, @kainino0x, @Kangz, @sebbbi, @SenorBlanco)
-
Static Samplers Proposal gpuweb/gpuweb#4687 (8 comments by @austinEng, @kainino0x, @Kangz, @magcius, @Sirtsu55)
-
GPUDeviceLostReason should contain more values gpuweb/gpuweb#4695 (7 comments by @kainino0x, @mwyrzykowski)
-
Clarification regarding double negation (--) operator? gpuweb/gpuweb#4669 (7 comments by @dneto0, @jimblandy, @kdashg, @mwyrzykowski)
-
WGSL API proposal: shaderLog gpuweb/gpuweb#4704 (6 comments by @alan-baker, @Kangz, @magcius, @mwyrzykowski)
-
Allow all vertex input attributes to be considered OOB if at least one is (in the "process vertices" algorithm) gpuweb/gpuweb#4668 (5 comments by @kainino0x, @teoxoy)
-
Support for arrays of textures gpuweb/gpuweb#822 (4 comments by @gitbetter, @Kangz, @TheMostDiligent)
-
Compatibility Mode: Support OpenGLES devices with zero SSBOs and image uniforms in vertex, fragment stages gpuweb/gpuweb#4721 (4 comments by @greggman, @kainino0x, @teoxoy)
-
Do pipeline-creation errors encompass the whole module or just the GPUProgrammableStage? gpuweb/gpuweb#4648 (3 comments by @alan-baker, @jimblandy, @kdashg)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (3 comments by @kainino0x, @Kangz)
-
Add support for 3D compressed textures. gpuweb/gpuweb#3183 (3 comments by @kainino0x, @mehmetoguzderin, @teoxoy)
-
Collaborate with the immersive web groups for WebXR interop gpuweb/gpuweb#2778 (3 comments by @alienself, @Kangz, @mikemainguy)
-
Clarify that create*PipelineAsync must not generate a validation error gpuweb/gpuweb#4714 (2 comments by @kainino0x)
-
Case for optional GPUBindGroupLayoutEntry entries gpuweb/gpuweb#851 (2 comments by @kainino0x, @MikhailGorobets)
-
`requestAdapterInfo()` should be required to prompt (in at least some circumstances) gpuweb/gpuweb#3962 (2 comments by @kainino0x, @mwyrzykowski)
-
VideoFrame should use display size, not coded size gpuweb/gpuweb#4525 (2 comments by @hlevring, @shaoboyan)
-
writing to storage buffers/textures in fragment shaders with multisample targets is not portable. gpuweb/gpuweb#4696 (2 comments by @greggman, @kainino0x)
-
add subgroups, and make them portable if possible gpuweb/gpuweb#4306 (2 comments by @dneto0, @raphlinus)
-
Incorrect validation error with dispatchWorkgroupsIndirect gpuweb/gpuweb#4632 (2 comments by @Kangz, @magcius)
-
maxInterStageShaderComponents seems to overlap with maxInterStageShaderVariables gpuweb/gpuweb#4688 (1 comment by @kainino0x)
-
mapAsync early-reject doesn't generate a validation error gpuweb/gpuweb#4690 (1 comment by @kainino0x)
-
Should there be a limit on the number of open GPUCommandEncoder instances? gpuweb/gpuweb#4622 (1 comment by @Kangz)
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558 (1 comment by @toji)
-
Should copyExternalImageToTexture allow a slice of a 3d texture as a destination? gpuweb/gpuweb#4697 (1 comment by @kainino0x)
-
Type inconsistency between drawIndexed and drawIndexedIndirect calls gpuweb/gpuweb#4674 (1 comment by @toji)
-
-
22 pull requests opened.
-
Add algorithm for linear interpolation gpuweb/gpuweb#4725, by @toji.
-
Clarify scope of pipeline creation gpuweb/gpuweb#4724, by @alan-baker.
-
Clarify the use of `@blend_src` in certain shader stages gpuweb/gpuweb#4723, by @Jiawei-Shao.
-
Clarify workgroupSize calculation gpuweb/gpuweb#4722, by @toji.
-
Extensions editorial: linking and syncing gpuweb/gpuweb#4720, by @mehmetoguzderin.
-
Add compute algorithm to detailed operations gpuweb/gpuweb#4719, by @toji.
-
Rename `subgroups-f16` to `subgroups_f16` for WGSL gpuweb/gpuweb#4718, by @jrprice.
-
Remove or resolve some editorial notes gpuweb/gpuweb#4717, by @toji.
-
Fix \ escaping in Python strings gpuweb/gpuweb#4716, by @dneto0.
-
[subgroups proposal] correct that the `id` must be a const-expression, not `e` gpuweb/gpuweb#4715, by @teoxoy.
-
[Editorial] Fix a missing parenthesis gpuweb/gpuweb#4712, by @jiangzhaoming.
-
Add missing `dual-source-blending` in `GPUFeatureName` gpuweb/gpuweb#4710, by @Jiawei-Shao.
-
Uncapturederror and preventDefault gpuweb/gpuweb#4708, by @greggman.
-
Algorithm steps for attachment load/store gpuweb/gpuweb#4706, by @toji.
-
Remove outdated inline issue gpuweb/gpuweb#4703, by @toji.
-
Improve render bundle steps gpuweb/gpuweb#4702, by @toji.
-
More rigorously define the relationship between device/GPUDevice gpuweb/gpuweb#4699, by @toji.
-
Add queue steps for copyExternalImageToTexture gpuweb/gpuweb#4698, by @toji.
-
Define level of detail gpuweb/gpuweb#4694, by @toji.
-
Clarify that using the standard sample patterns is mandatory gpuweb/gpuweb#4693, by @toji.
-
Fixed baseVertex type for drawIndexedIndirect gpuweb/gpuweb#4691, by @toji.
-
Rename image->texel image, block row->texel block row gpuweb/gpuweb#4689, by @toji.
-
-
24 pull requests merged.
-
Add algorithm for linear interpolation gpuweb/gpuweb#4725.
-
Clarify workgroupSize calculation gpuweb/gpuweb#4722.
-
Add extended range gpuweb/gpuweb#4500.
-
Remove or resolve some editorial notes gpuweb/gpuweb#4717.
-
Add compute algorithm to detailed operations gpuweb/gpuweb#4719.
-
Extensions editorial: linking and syncing gpuweb/gpuweb#4720.
-
Rename `subgroups-f16` to `subgroups_f16` for WGSL gpuweb/gpuweb#4718.
-
[subgroups proposal] correct that the `id` must be a const-expression, not `e` gpuweb/gpuweb#4715.
-
Guarantee that OOB writes to storage textures will be ignored gpuweb/gpuweb#4194.
-
Fix \ escaping in Python strings gpuweb/gpuweb#4716.
-
[Editorial] Fix a missing parenthesis gpuweb/gpuweb#4712.
-
Add missing `dual-source-blending` in `GPUFeatureName` gpuweb/gpuweb#4710.
-
Add optional feature dual_source_blending gpuweb/gpuweb#4621.
-
Algorithm steps for attachment load/store gpuweb/gpuweb#4706.
-
Improve render bundle steps gpuweb/gpuweb#4702.
-
Add queue steps for copyExternalImageToTexture gpuweb/gpuweb#4698.
-
Remove outdated inline issue gpuweb/gpuweb#4703.
-
More rigorously define the relationship between device/GPUDevice gpuweb/gpuweb#4699.
-
Clarify that using the standard sample patterns is mandatory gpuweb/gpuweb#4693.
-
Fixed baseVertex type for drawIndexedIndirect gpuweb/gpuweb#4691.
-
Add a definition for 64-bit unsigned ints gpuweb/gpuweb#4666.
-
Adjusts subgroup built-in name verbs to match atomics operations gpuweb/gpuweb#4627.
-
Make `unmap()` device timeline validation explicit gpuweb/gpuweb#4679.
-
Rename image->texel image, block row->texel block row gpuweb/gpuweb#4689.
-
-
1 pull request closed.
-
Adjusts subgroup built-in function names to match GLSL gpuweb/gpuweb#4528.
-
-
25 pull requests received 131 new comments.
-
Push constant proposal gpuweb/gpuweb#4612 (18 comments by @kainino0x, @Kangz, @litherum, @mwyrzykowski, @shaoboyan, @toji)
-
Clarify scope of pipeline creation gpuweb/gpuweb#4724 (13 comments by @alan-baker, @dj2, @dneto0, @mehmetoguzderin)
-
Add optional feature dual_source_blending gpuweb/gpuweb#4621 (10 comments by @dneto0, @Jiawei-Shao, @kainino0x, @Kangz, @kdashg)
-
Guarantee that OOB writes to storage textures will be ignored gpuweb/gpuweb#4194 (7 comments by @alan-baker, @dneto0, @jimblandy, @kainino0x, @kdashg, @mwyrzykowski)
-
Move Detailed Operations section into separate doc gpuweb/gpuweb#4633 (7 comments by @kainino0x, @teoxoy, @toji)
-
Add algorithm for linear interpolation gpuweb/gpuweb#4725 (5 comments by @greggman, @kainino0x, @toji)
-
Make sure the device is valid when creating a texture view gpuweb/gpuweb#4678 (2 comments by @kainino0x, @toji)
-
Replace asynchronous .requestAdapterInfo() with synchronous .info gpuweb/gpuweb#4662 (2 comments by @beaufortfrancois)
-
Adjusts subgroup built-in name verbs to match atomics operations gpuweb/gpuweb#4627 (1 comment by @kdashg)
-
gpuweb/cts
-
12 issues opened.
-
Add test for `@interpolate(flat, either)` gpuweb/cts#3815, by @greggman.
-
Add tests against the extension `dual_source_blending` gpuweb/cts#3810, by @Jiawei-Shao.
-
Rejection on createComputePipelineAsnyc causes timout in tests gpuweb/cts#3809, by @sagudev.
-
There are more createRenderPipeline without targets. gpuweb/cts#3806, by @sagudev.
-
determinant tests are too strict for f16 gpuweb/cts#3795, by @dneto0.
-
rgba16float canvases with weird values in them gpuweb/cts#3790, by @kainino0x.
-
webgpu cts not run when ubunt + Discrete graphics card. gpuweb/cts#3783, by @82157402.
-
Test `[SameObject] attribute`s return the same object each time gpuweb/cts#3779, by @kainino0x.
-
10 issues closed.
-
Add test for `@interpolate(flat, either)` gpuweb/cts#3815.
-
webgpu cts not run when ubunt + Discrete graphics card. gpuweb/cts#3783.
-
Track all the buffers and textures with TrackForCleanup() gpuweb/cts#1752.
-
wgsl: address space names are not keywords gpuweb/cts#1902.
-
Add coverage for validation of >=1 target in render passes gpuweb/cts#3755.
-
determinant tests are too strict for f16 gpuweb/cts#3795.
-
Some expectations in web_platform,copyToTexture, are unexpected gpuweb/cts#3767.
-
-
1 issue reopened.
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761, by @mehmetoguzderin.
-
-
11 issues received 24 new comments.
-
webgpu cts not run when ubunt + Discrete graphics card. gpuweb/cts#3783 (5 comments by @82157402, @greggman, @kainino0x)
-
Rejection on createComputePipelineAsnyc causes timout in tests gpuweb/cts#3809 (3 comments by @kainino0x, @sagudev)
-
Some expectations in web_platform,copyToTexture, are unexpected gpuweb/cts#3767 (3 comments by @kainino0x, @mwyrzykowski)
-
wgsl: Add execution test for @interpolate(flat,first), and validation tests for `either` and `first` gpuweb/cts#3818 (2 comments by @greggman)
-
wgsl: support abstract arrays, and constructor that infers result type. gpuweb/cts#1705 (1 comment by @alan-baker)
-
bgra8unorm storage textures without bgra8unorm-storage, should generate WGSL error? gpuweb/cts#3804 (1 comment by @greggman)
-
Test `[SameObject] attribute`s return the same object each time gpuweb/cts#3779 (1 comment by @kainino0x)
-
-
40 pull requests opened.
-
Add lang="en" to standalone index.html gpuweb/cts#3828, by @kainino0x.
-
print_environment: compat, fallback, features, and limits gpuweb/cts#3827, by @kainino0x.
-
trackForCleanup(requestDevice()) -> requestDeviceTracked() gpuweb/cts#3826, by @kainino0x.
-
Expand context dependent name test to include flat sampling gpuweb/cts#3824, by @alan-baker.
-
trackForCleanup all devices, and await device.lost while cleaning up gpuweb/cts#3823, by @kainino0x.
-
Add execution test for @interpolate(flat, either) gpuweb/cts#3822, by @greggman.
-
trackForCleanup all buffers and textures gpuweb/cts#3821, by @kainino0x.
-
Partial evaluation validation for constructors gpuweb/cts#3820, by @alan-baker.
-
Set document.title to query in standalone CTS gpuweb/cts#3819, by @greggman.
-
Rework insertBits count_offset tests to include runtime values gpuweb/cts#3817, by @alan-baker.
-
Require query sets to be tracked for cleanup gpuweb/cts#3816, by @kainino0x.
-
Compat: Refactor tests to use @interpolate(flat, either) gpuweb/cts#3814, by @greggman.
-
Compat: Add a test for requestDevice gpuweb/cts#3813, by @greggman.
-
Add shader validation tests about using `blend_src` and enabling extension gpuweb/cts#3812, by @Jiawei-Shao.
-
Test createRenderPipeline fail without any attachment gpuweb/cts#3808, by @sagudev.
-
Add dummy targets to render pipelines gpuweb/cts#3807, by @sagudev.
-
Compat: fix rgXXX format as storage texture issues gpuweb/cts#3805, by @greggman.
-
Compat: Fix textureLoad WGSL validation tests for compat spec changes. gpuweb/cts#3803, by @greggman.
-
Compat: Fix interpolation tests for compat spec changes gpuweb/cts#3802, by @greggman.
-
Rework extractBits count_offset tests to include runtime values gpuweb/cts#3801, by @alan-baker.
-
Add validation tests for smoothstep partial eval errors gpuweb/cts#3800, by @alan-baker.
-
Update api tests for 3D textures with BC compression gpuweb/cts#3799, by @mehmetoguzderin.
-
Add partial evaluation validation tests for ldexp gpuweb/cts#3798, by @alan-baker.
-
Rework clamp low_high tests to support partial evaluation gpuweb/cts#3797, by @alan-baker.
-
Partial evaluation error tests for bit shifts gpuweb/cts#3796, by @alan-baker.
-
Cleanup for virtualMipSize usage gpuweb/cts#3794, by @greggman.
-
Enable linting for .js files gpuweb/cts#3793, by @kainino0x.
-
Make `virtualMipSize` take a `GPUExtent3D` gpuweb/cts#3792, by @greggman.
-
Remove /* eslint-disable prettier/prettier */ where possible. gpuweb/cts#3791, by @greggman.
-
Rework case generation for determinant test gpuweb/cts#3789, by @dneto0.
-
Add PRNG::uniformInt gpuweb/cts#3788, by @dneto0.
-
Fix div/rem early eval error cases gpuweb/cts#3787, by @alan-baker.
-
Revert "Update validation tests for 3D textures with BC compression" gpuweb/cts#3786, by @kainino0x.
-
Use standard html instead of quirks html gpuweb/cts#3785, by @greggman.
-
Add uncapturederror test gpuweb/cts#3784, by @greggman.
-
WGSL textureSample tests for 3D and cubemaps gpuweb/cts#3782, by @greggman.
-
Out-of-bounds textureStore tests gpuweb/cts#3781, by @alan-baker.
-
Add more vertex shader res restrictions on tests gpuweb/cts#3780, by @sudonatalie.
-
Update validation tests for 3D textures with BC compression gpuweb/cts#3777, by @mehmetoguzderin.
-
-
39 pull requests merged.
-
print_environment: compat, fallback, features, and limits gpuweb/cts#3827.
-
Add lang="en" to standalone index.html gpuweb/cts#3828.
-
trackForCleanup(requestDevice()) -> requestDeviceTracked() gpuweb/cts#3826.
-
trackForCleanup all devices, and await device.lost while cleaning up gpuweb/cts#3823.
-
Add execution test for @interpolate(flat, either) gpuweb/cts#3822.
-
trackForCleanup all buffers and textures gpuweb/cts#3821.
-
Partial evaluation validation for constructors gpuweb/cts#3820.
-
Compat: Refactor tests to use @interpolate(flat, either) gpuweb/cts#3814.
-
Set document.title to query in standalone CTS gpuweb/cts#3819.
-
Compat: Add a test for requestDevice gpuweb/cts#3813.
-
Require query sets to be tracked for cleanup gpuweb/cts#3816.
-
Out-of-bounds textureStore tests gpuweb/cts#3781.
-
Rework insertBits count_offset tests to include runtime values gpuweb/cts#3817.
-
Add dummy targets to render pipelines gpuweb/cts#3807.
-
Test createRenderPipeline fail without any attachment gpuweb/cts#3808.
-
Fix div/rem early eval error cases gpuweb/cts#3787.
-
Partial evaluation error tests for bit shifts gpuweb/cts#3796.
-
Add validation tests for smoothstep partial eval errors gpuweb/cts#3800.
-
Rework extractBits count_offset tests to include runtime values gpuweb/cts#3801.
-
Compat: fix rgXXX format as storage texture issues gpuweb/cts#3805.
-
Compat: Fix interpolation tests for compat spec changes gpuweb/cts#3802.
-
Add partial evaluation validation tests for ldexp gpuweb/cts#3798.
-
Rework clamp low_high tests to support partial evaluation gpuweb/cts#3797.
-
Rework case generation for determinant test gpuweb/cts#3789.
-
Cleanup for virtualMipSize usage gpuweb/cts#3794.
-
Add PRNG::uniformInt gpuweb/cts#3788.
-
Make `virtualMipSize` take a `GPUExtent3D` gpuweb/cts#3792.
-
Add uncapturederror test gpuweb/cts#3784.
-
Enable linting for .js files gpuweb/cts#3793.
-
Remove /* eslint-disable prettier/prettier */ where possible. gpuweb/cts#3791.
-
WGSL textureSample tests for 3D and cubemaps gpuweb/cts#3782.
-
Use standard html instead of quirks html gpuweb/cts#3785.
-
Revert "Update validation tests for 3D textures with BC compression" gpuweb/cts#3786.
-
Add more vertex shader res restrictions on tests gpuweb/cts#3780.
-
Update validation tests for 3D textures with BC compression gpuweb/cts#3777.
-
Compat: Skip more tests for sample_* validation gpuweb/cts#3776.
-
-
18 pull requests received 100 new comments.
-
Out-of-bounds textureStore tests gpuweb/cts#3781 (21 comments by @alan-baker, @dneto0, @greggman, @kainino0x)
-
trackForCleanup all devices, and await device.lost while cleaning up gpuweb/cts#3823 (8 comments by @greggman, @kainino0x)
-
Update validation tests for 3D textures with BC compression gpuweb/cts#3777 (6 comments by @kainino0x, @mehmetoguzderin)
-
Compat: Refactor tests to use @interpolate(flat, either) gpuweb/cts#3814 (4 comments by @greggman, @jrprice)
-
Add shader validation tests about using `blend_src` and enabling extension gpuweb/cts#3812 (3 comments by @jiangzhaoming, @Jiawei-Shao)
-
Add execution test for @interpolate(flat, either) gpuweb/cts#3822 (2 comments by @greggman, @jrprice)
-
Remove /* eslint-disable prettier/prettier */ where possible. gpuweb/cts#3791 (1 comment by @kainino0x)
-
TR publications
-
20 documents published.
-
WebGPU published 14 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 6 times as a Working Draft. History of WebGPU Shading Language
-
Participants
-
4 participants left.
-
Todd Kosloff (OPPO) left.
-
he feng (OPPO) left.
-
Xunran Ding (OPPO) left.
-
Weixin Dong (OPPO) left.
-
-
1 organization left.
-
OPPO left.
-
GitHub
gpuweb/gpuweb
-
19 issues opened.
-
Static Samplers Proposal gpuweb/gpuweb#4687, by @Sirtsu55.
-
Type inconsistency between drawIndexed and drawIndexedIndirect calls gpuweb/gpuweb#4674, by @mwyrzykowski.
-
Clarification regarding double negation (--) operator? gpuweb/gpuweb#4669, by @mwyrzykowski.
-
Unclear who "consumer" is. gpuweb/gpuweb#4660, by @fyellin.
-
GL_EXT_fragment_shader_barycentric gpuweb/gpuweb#4658, by @alexwennstrom.
-
Hardware feature levels (and compat) gpuweb/gpuweb#4656, by @kainino0x.
-
Clamping behavior when comparison sampling `depth24plus` gpuweb/gpuweb#4653, by @austinEng.
-
Syntax problem in English gpuweb/gpuweb#4649, by @12ff54e.
-
Do pipeline-creation errors encompass the whole module or just the GPUProgrammableStage? gpuweb/gpuweb#4648, by @alan-baker.
-
Clarify supported context formats definition gpuweb/gpuweb#4644, by @beaufortfrancois.
-
Incorrect validation error with dispatchWorkgroupsIndirect gpuweb/gpuweb#4632, by @brendan-duncan.
-
Does an override need to be statically used by the entry point to be allowed? gpuweb/gpuweb#4624, by @kainino0x.
-
Should bitcast tests allow FTZ on inputs/outputs? gpuweb/gpuweb#4623, by @zoddicus.
-
Should there be a limit on the number of open GPUCommandEncoder instances? gpuweb/gpuweb#4622, by @mwyrzykowski.
-
-
29 issues closed.
-
Task: Define the copies with 1d and 3d textures gpuweb/gpuweb#2690.
-
Bytes per texel of `depth24plus-stencil8` could be 4 - 5 gpuweb/gpuweb#3771.
-
Should there be a limit on the number of open GPUCommandEncoder instances? gpuweb/gpuweb#4622.
-
HW Clip Planes support gpuweb/gpuweb#390.
-
Synchronously query GPUAdapterInfo gpuweb/gpuweb#4536.
-
Clarify supported context formats definition gpuweb/gpuweb#4644.
-
Unclear who "consumer" is. gpuweb/gpuweb#4660.
-
Syntax problem in English gpuweb/gpuweb#4649.
-
Consider deduplicating "must be a single aspect" validation gpuweb/gpuweb#1192.
-
bitcast behavior regarding NaNs is unexpected? gpuweb/gpuweb#4450.
-
GL_EXT_fragment_shader_barycentric gpuweb/gpuweb#4658.
-
Y orientation of viewport coordinates vs. NDC gpuweb/gpuweb#4492.
-
Add missing limits to correspondence doc gpuweb/gpuweb#4273.
-
Should command buffers be invalidated by an invalid submit? gpuweb/gpuweb#4599.
-
Should bitcast tests allow FTZ on inputs/outputs? gpuweb/gpuweb#4623.
-
Allow 2d-array texture view as renderable texture view gpuweb/gpuweb#4607.
-
Reduce limits for composite types gpuweb/gpuweb#4547.
-
Reduce limits for the number of case selector values in a switch statement gpuweb/gpuweb#4548.
-
Reduce limit for maximum number of elements in const-expression of array type gpuweb/gpuweb#4546.
-
Reduce byte-size limit for function and private address spaces gpuweb/gpuweb#4545.
-
publish-TR-webgpu: Attribute “xlink” not allowed on element “svg” at this point gpuweb/gpuweb#3222.
-
Initialization of `GPUObjectBase` internal slot is left unspecified gpuweb/gpuweb#2871.
-
editorial: set label on all created objects gpuweb/gpuweb#3332.
-
-
3 issues reopened.
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558, by @kainino0x.
-
Add support for 3D compressed textures. gpuweb/gpuweb#3183, by @kainino0x.
-
Numeric conversion rules wording is inconsistent gpuweb/gpuweb#4538, by @dneto0.
-
-
54 issues received 175 new comments.
-
Should there be a limit on the number of open GPUCommandEncoder instances? gpuweb/gpuweb#4622 (35 comments by @aleino-nv, @grovesNL, @kainino0x, @Kangz, @litherum, @mwyrzykowski, @teoxoy)
-
Support "uniform texel buffer" and "storage texel buffer" ? gpuweb/gpuweb#162 (12 comments by @jimblandy, @Kangz, @kdashg, @litherum, @mwyrzykowski, @sebbbi)
-
GPURenderCommandsMixin.draw / drawIndexed allows for the execution of far too many vertices gpuweb/gpuweb#4659 (10 comments by @greggman, @kainino0x, @mwyrzykowski, @toji)
-
bitcast behavior regarding NaNs is unexpected? gpuweb/gpuweb#4450 (8 comments by @alan-baker, @kainino0x, @mwyrzykowski, @sciecode)
-
Does an override need to be statically used by the entry point to be allowed? gpuweb/gpuweb#4624 (8 comments by @alan-baker, @dneto0, @kainino0x, @Kangz, @kdashg)
-
It should be possible to define an 'override' that evaluates depending on other 'override' constants gpuweb/gpuweb#4680 (7 comments by @ben-clayton, @dj2, @JD31, @jrprice)
-
Add support for 3D compressed textures. gpuweb/gpuweb#3183 (6 comments by @kainino0x, @Kangz, @mehmetoguzderin, @mwyrzykowski, @teoxoy)
-
Hardware feature levels (and compat) gpuweb/gpuweb#4656 (6 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
Clamping behavior when comparison sampling `depth24plus` gpuweb/gpuweb#4653 (6 comments by @austinEng, @kainino0x, @Kangz)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (4 comments by @kainino0x, @Kangz, @kdashg)
-
Should bitcast tests allow FTZ on inputs/outputs? gpuweb/gpuweb#4623 (4 comments by @alan-baker, @ben-clayton, @jimblandy, @kdashg)
-
Numeric conversion rules wording is inconsistent gpuweb/gpuweb#4538 (4 comments by @dneto0, @kdashg)
-
Clarification regarding double negation (--) operator? gpuweb/gpuweb#4669 (3 comments by @dneto0, @mwyrzykowski)
-
Add subgroup barrier to subgroups proposal gpuweb/gpuweb#4437 (3 comments by @alienself, @munrocket, @raphlinus)
-
Do pipeline-creation errors encompass the whole module or just the GPUProgrammableStage? gpuweb/gpuweb#4648 (3 comments by @alan-baker, @greggman, @mighdoll)
-
Should command buffers be invalidated by an invalid submit? gpuweb/gpuweb#4599 (3 comments by @Kangz, @litherum, @mwyrzykowski)
-
Allow 2d-array texture view as renderable texture view gpuweb/gpuweb#4607 (3 comments by @Kangz, @magcius)
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568 (3 comments by @dj2, @greggman, @kdashg)
-
PNG ImageBitmap (WebAPI) incompatibility with `copyExternalImageToTexture` for physically-based workflow. gpuweb/gpuweb#4609 (3 comments by @Kangz, @sciecode)
-
Add a note that copyT2T might not preserve the exact bit values gpuweb/gpuweb#2558 (2 comments by @kainino0x)
-
Allow all vertex input attributes to be considered OOB if at least one is (in the "process vertices" algorithm) gpuweb/gpuweb#4668 (2 comments by @kainino0x, @teoxoy)
-
"get a copy of the image contents of a context" does not handle case where canvas is unconfigured gpuweb/gpuweb#4606 (2 comments by @kainino0x, @Kangz)
-
Consider how to make the spec a better developer reference gpuweb/gpuweb#2957 (2 comments by @greggman, @kainino0x)
-
Compat: Disallow `textureSampleLevel`/`textureSampleCompareLevel` with `texture_depth_2d_array` gpuweb/gpuweb#4519 (2 comments by @Kangz, @kdashg)
-
Spec allows resolveTarget to be 3d but doesn't allow selecting a depth slice gpuweb/gpuweb#4600 (2 comments by @Kangz, @magcius)
-
publish-TR-webgpu: Attribute “xlink” not allowed on element “svg” at this point gpuweb/gpuweb#3222 (2 comments by @kainino0x, @toji)
-
reuse of an GPUAdapter to request a GPUDevice does not reject. gpuweb/gpuweb#4352 (1 comment by @teoxoy)
-
Type inconsistency between drawIndexed and drawIndexedIndirect calls gpuweb/gpuweb#4674 (1 comment by @kainino0x)
-
Support for gl_PrimitiveID / SV_PrimitiveID -like construct? gpuweb/gpuweb#1786 (1 comment by @mwyrzykowski)
-
Incorrect validation error with dispatchWorkgroupsIndirect gpuweb/gpuweb#4632 (1 comment by @kainino0x)
-
Texture Format RGBA32Float Does Not Support Storage Texture Access StorageTextureAccess::ReadWrite gpuweb/gpuweb#4650 (1 comment by @Jiawei-Shao)
-
Difference between alignment of struct with size 8 vs `vec2<f32>` in uniform buffers. gpuweb/gpuweb#4497 (1 comment by @kdashg)
-
Compat: Disallow textureLoad() of depth textures in WGSL via validation. gpuweb/gpuweb#4512 (1 comment by @kdashg)
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513 (1 comment by @kdashg)
-
smoothstep: is it an error when `low == high` and/or `low > high` (when constant/pipline-evaluated) gpuweb/gpuweb#4561 (1 comment by @kdashg)
-
u32(abstractint) errors on out of bounds value, u32(abstractfloat) and u32(f32) clamps out of bounds value gpuweb/gpuweb#4569 (1 comment by @kdashg)
-
Remove the need to hard-code texel format for `texture_storage_2d` in shaders gpuweb/gpuweb#4298 (1 comment by @brendan-duncan)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (1 comment by @Kangz)
-
[editorial] Explain the relationship between color management and `-srgb` formats gpuweb/gpuweb#4587 (1 comment by @greggman)
-
Initialization of `GPUObjectBase` internal slot is left unspecified gpuweb/gpuweb#2871 (1 comment by @toji)
-
Use enum tokens when they are the only possible grammar usage. gpuweb/gpuweb#4505 (1 comment by @alan-baker)
-
-
48 pull requests opened.
-
Complete remaining texture copy algorithms gpuweb/gpuweb#4686, by @toji.
-
Restrict compressed 3D to BCn gpuweb/gpuweb#4685, by @mehmetoguzderin.
-
[editorial] Fix var and value table footnote gpuweb/gpuweb#4684, by @alan-baker.
-
Add description of sample footprint gpuweb/gpuweb#4683, by @toji.
-
Detail binding size check at draw/dispatch gpuweb/gpuweb#4682, by @toji.
-
Add missing `multisample.count` validation gpuweb/gpuweb#4681, by @teoxoy.
-
Make `unmap()` device timeline validation explicit gpuweb/gpuweb#4679, by @teoxoy.
-
Make sure the device is valid when creating a texture view gpuweb/gpuweb#4678, by @teoxoy.
-
Allow compressed formats for 3D textures gpuweb/gpuweb#4677, by @mehmetoguzderin.
-
Spec orientation of each cube face, using a diagram gpuweb/gpuweb#4675, by @kainino0x.
-
Fix outdated [[internals]].[[state]] in diagrams gpuweb/gpuweb#4673, by @kainino0x.
-
Complete algorithm for copyTextureToTexture gpuweb/gpuweb#4672, by @toji.
-
Fixes for invoke-mermaid.sh and invoke-bikeshed.sh gpuweb/gpuweb#4671, by @kainino0x.
-
Point to TC39 draft of source map spec instead of sourcemaps.info gpuweb/gpuweb#4670, by @kainino0x.
-
Remove createShaderModule and createCommandEncoder issues gpuweb/gpuweb#4667, by @toji.
-
Add a definition for 64-bit unsigned ints gpuweb/gpuweb#4666, by @toji.
-
Clarify list of context compatible formats gpuweb/gpuweb#4665, by @toji.
-
Minor grammatical fix gpuweb/gpuweb#4664, by @toji.
-
Clarify the consumer of a GPUProgrammableStage gpuweb/gpuweb#4663, by @toji.
-
Replace asynchronous .requestAdapterInfo() with synchronous .info gpuweb/gpuweb#4662, by @kainino0x.
-
Better definition for 'images' gpuweb/gpuweb#4661, by @toji.
-
Update issue templates for new label, new Dawn URLs gpuweb/gpuweb#4655, by @kainino0x.
-
`indirectBuffer` should be added as `input` to the `usage scope` gpuweb/gpuweb#4654, by @teoxoy.
-
Subgroups proposal: Consider subgroup_size used in compute stage uniform gpuweb/gpuweb#4652, by @jiangzhaoming.
-
Compatibility: one aspect per texture view. gpuweb/gpuweb#4647, by @SenorBlanco.
-
Describe how comparison samplers interact with filtering gpuweb/gpuweb#4646, by @toji.
-
Add a better definition for 'texture' gpuweb/gpuweb#4645, by @toji.
-
Clarify that pipeline-overrides don't have to be statically used to be set gpuweb/gpuweb#4642, by @dneto0.
-
Add coordinate system diagrams and fix NDC to framebuffer equation gpuweb/gpuweb#4641, by @toji.
-
Allow bitcast to flush to zero gpuweb/gpuweb#4640, by @alan-baker.
-
add first and either to interpolate gpuweb/gpuweb#4639, by @greggman.
-
Compat: sample_mask and sample_index shader module error gpuweb/gpuweb#4638, by @greggman.
-
Compat: disallow @interpolate(flat, first) gpuweb/gpuweb#4637, by @greggman.
-
Compat: rgXXX formats generate error at shader module creation gpuweb/gpuweb#4636, by @greggman.
-
Compat: linear and sample error at shader module creation gpuweb/gpuweb#4635, by @greggman.
-
Compat: depth textureLoad error at shader module creation gpuweb/gpuweb#4634, by @greggman.
-
Move Detailed Operations section into separate doc gpuweb/gpuweb#4633, by @toji.
-
Fix 2d-array texture view as a renderable texture view gpuweb/gpuweb#4631, by @haoxli.
-
Deduped most image/buffer copy validation gpuweb/gpuweb#4629, by @toji.
-
Add missing limits to the correspondence doc gpuweb/gpuweb#4628, by @toji.
-
Adjusts subgroup built-in name verbs to match atomics operations gpuweb/gpuweb#4627, by @munrocket.
-
[non-normative] Drawing buffer can also be lazily cleared gpuweb/gpuweb#4626, by @kainino0x.
-
Making valid, invalid, and expired regular properties gpuweb/gpuweb#4625, by @toji.
-
Add optional feature dual_source_blending gpuweb/gpuweb#4621, by @Jiawei-Shao.
-
Indicate timeline for majority of algorithms gpuweb/gpuweb#4620, by @toji.
-
Extract privacy and security considerations into a separate BS source file gpuweb/gpuweb#4618, by @toji.
-
Remove [[internal]] concept gpuweb/gpuweb#4617, by @toji.
-
Add early evaluation errors for smoothstep gpuweb/gpuweb#4616, by @alan-baker.
-
-
46 pull requests merged.
-
Complete remaining texture copy algorithms gpuweb/gpuweb#4686.
-
Restrict compressed 3D to BCn gpuweb/gpuweb#4685.
-
[editorial] Fix var and value table footnote gpuweb/gpuweb#4684.
-
Complete algorithm for copyTextureToTexture gpuweb/gpuweb#4672.
-
Add optional feature `clip-distances` gpuweb/gpuweb#4588.
-
Detail binding size check at draw/dispatch gpuweb/gpuweb#4682.
-
Add missing `multisample.count` validation gpuweb/gpuweb#4681.
-
Make sure the device is valid when creating a texture view gpuweb/gpuweb#4678.
-
Allow compressed formats for 3D textures gpuweb/gpuweb#4677.
-
Replace asynchronous .requestAdapterInfo() with synchronous .info gpuweb/gpuweb#4662.
-
Spec orientation of each cube face, using a diagram gpuweb/gpuweb#4675.
-
Subgroups proposal: Consider subgroup_size used in compute stage uniform gpuweb/gpuweb#4652.
-
Fixes for invoke-mermaid.sh and invoke-bikeshed.sh gpuweb/gpuweb#4671.
-
Fix outdated [[internals]].[[state]] in diagrams gpuweb/gpuweb#4673.
-
Point to TC39 draft of source map spec instead of sourcemaps.info gpuweb/gpuweb#4670.
-
Clarify list of context compatible formats gpuweb/gpuweb#4665.
-
Better definition for 'images' gpuweb/gpuweb#4661.
-
Clarify the consumer of a GPUProgrammableStage gpuweb/gpuweb#4663.
-
Remove createShaderModule and createCommandEncoder issues gpuweb/gpuweb#4667.
-
Minor grammatical fix gpuweb/gpuweb#4664.
-
add first and either to interpolate gpuweb/gpuweb#4639.
-
Making valid, invalid, and expired regular properties gpuweb/gpuweb#4625.
-
Update issue templates for new label, new Dawn URLs gpuweb/gpuweb#4655.
-
Deduped most image/buffer copy validation gpuweb/gpuweb#4629.
-
`indirectBuffer` should be added as `input` to the `usage scope` gpuweb/gpuweb#4654.
-
Describe how comparison samplers interact with filtering gpuweb/gpuweb#4646.
-
Add a better definition for 'texture' gpuweb/gpuweb#4645.
-
Compat: depth textureLoad error at shader module creation gpuweb/gpuweb#4634.
-
Compat: linear and sample error at shader module creation gpuweb/gpuweb#4635.
-
Compat: rgXXX formats generate error at shader module creation gpuweb/gpuweb#4636.
-
Compat: disallow @interpolate(flat, first) gpuweb/gpuweb#4637.
-
Compat: sample_mask and sample_index shader module error gpuweb/gpuweb#4638.
-
Add coordinate system diagrams and fix NDC to framebuffer equation gpuweb/gpuweb#4641.
-
Add missing limits to the correspondence doc gpuweb/gpuweb#4628.
-
Compatibility: one aspect per texture view. gpuweb/gpuweb#4647.
-
Clarify that pipeline-overrides don't have to be statically used to be set gpuweb/gpuweb#4642.
-
Allow bitcast to flush to zero gpuweb/gpuweb#4640.
-
Fix 2d-array texture view as a renderable texture view gpuweb/gpuweb#4631.
-
[non-normative] Drawing buffer can also be lazily cleared gpuweb/gpuweb#4626.
-
Indicate timeline for majority of algorithms gpuweb/gpuweb#4620.
-
wgsl: Reduce shader limits gpuweb/gpuweb#4578.
-
Extract privacy and security considerations into a separate BS source file gpuweb/gpuweb#4618.
-
Remove [[internal]] concept gpuweb/gpuweb#4617.
-
Use 'create a new WebGPU Object' for all objects gpuweb/gpuweb#4615.
-
Add early evaluation errors for smoothstep gpuweb/gpuweb#4616.
-
-
3 pull requests closed.
-
[editorial] wgsl builtin function notes gpuweb/gpuweb#3670.
-
Add synchronous GPUAdapter.info gpuweb/gpuweb#4550.
-
Generalize attribute grammar gpuweb/gpuweb#3911.
-
-
38 pull requests received 333 new comments.
-
Add optional feature dual_source_blending gpuweb/gpuweb#4621 (52 comments by @alan-baker, @dj2, @Jiawei-Shao, @kainino0x, @Kangz, @kdashg, @teoxoy, @toji)
-
Push constant proposal gpuweb/gpuweb#4612 (50 comments by @alan-baker, @JMS55, @kainino0x, @Kangz, @litherum, @magcius, @mwyrzykowski, @shaoboyan)
-
Add coordinate system diagrams and fix NDC to framebuffer equation gpuweb/gpuweb#4641 (25 comments by @kainino0x, @toji)
-
Add optional feature `clip-distances` gpuweb/gpuweb#4588 (15 comments by @alan-baker, @Jiawei-Shao, @kainino0x, @Kangz)
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613 (13 comments by @alan-baker, @kainino0x, @Kangz, @toji)
-
Move Detailed Operations section into separate doc gpuweb/gpuweb#4633 (11 comments by @kainino0x, @toji)
-
Replace asynchronous .requestAdapterInfo() with synchronous .info gpuweb/gpuweb#4662 (8 comments by @beaufortfrancois, @kainino0x, @mwyrzykowski, @toji)
-
Making valid, invalid, and expired regular properties gpuweb/gpuweb#4625 (8 comments by @kainino0x, @toji)
-
Describe how comparison samplers interact with filtering gpuweb/gpuweb#4646 (8 comments by @kainino0x, @toji)
-
Add synchronous GPUAdapter.info gpuweb/gpuweb#4550 (6 comments by @jimblandy, @kainino0x, @Kangz, @toji)
-
Add a definition for 64-bit unsigned ints gpuweb/gpuweb#4666 (5 comments by @alan-baker, @kainino0x, @toji)
-
Make `unmap()` device timeline validation explicit gpuweb/gpuweb#4679 (5 comments by @kainino0x, @teoxoy)
-
Make sure the device is valid when creating a texture view gpuweb/gpuweb#4678 (3 comments by @kainino0x, @Kangz, @teoxoy)
-
Subgroups proposal: Consider subgroup_size used in compute stage uniform gpuweb/gpuweb#4652 (1 comment by @jiangzhaoming)
-
Point to TC39 draft of source map spec instead of sourcemaps.info gpuweb/gpuweb#4670 (1 comment by @toji)
-
Remove createShaderModule and createCommandEncoder issues gpuweb/gpuweb#4667 (1 comment by @kainino0x)
-
`indirectBuffer` should be added as `input` to the `usage scope` gpuweb/gpuweb#4654 (1 comment by @kainino0x)
-
Invalidate all command buffers passed to submit, whether the submit is valid or not. gpuweb/gpuweb#4608 (1 comment by @toji)
-
Adjusts subgroup built-in name verbs to match atomics operations gpuweb/gpuweb#4627 (1 comment by @w3cbot)
-
gpuweb/cts
-
16 issues opened.
-
Test clip-distances feature gpuweb/cts#3773, by @dneto0.
-
Some expectations in web_platform,copyToTexture, are unexpected gpuweb/cts#3767, by @mwyrzykowski.
-
Failing matrix multiplication tests gpuweb/cts#3765, by @sudonatalie.
-
Allow specifying subcase parameters in a query gpuweb/cts#3763, by @greggman.
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761, by @mehmetoguzderin.
-
textureDimensions(texture_external) needs testing gpuweb/cts#3759, by @Kangz.
-
Add coverage for validation of >=1 target in render passes gpuweb/cts#3755, by @ErichDonGubler.
-
Assert there are no uncaptured errors. gpuweb/cts#3753, by @sagudev.
-
[Refactor] wgsl: Consolidate validation tests gpuweb/cts#3741, by @alan-baker.
-
Add memory layout tests for arrays of matrices gpuweb/cts#3735, by @jrprice.
-
add abstract-float matrix types to kMatTypes in src/webgpu/util/conversion.ts gpuweb/cts#3723, by @dneto0.
-
wgsl: test smoothstep early evaluation error when low >= high gpuweb/cts#3722, by @dneto0.
-
-
146 issues closed.
-
Add return type tests for texture builtins gpuweb/cts#3586.
-
Failing matrix multiplication tests gpuweb/cts#3765.
-
val: Test depthSlice from GPURenderPassColorAttachment gpuweb/cts#3144.
-
wgsl: & and * are lower precedence than .member and [], even in LHS gpuweb/cts#1981.
-
wgsl: test smoothstep early evaluation error when low >= high gpuweb/cts#3722.
-
Move builtin must_use tests into per-builtin validation suites gpuweb/cts#3502.
-
wgsl: Execution tests for atomic initialization gpuweb/cts#1487.
-
wgsl: Validate various structs with size/align validate correctly. gpuweb/cts#1456.
-
wgsl: validate struct parsing gpuweb/cts#1501.
-
Add memory layout tests for arrays of matrices gpuweb/cts#3735.
-
address space restrictions vs. shader stage gpuweb/cts#2398.
-
wgsl: brace-enclosed statement nesting depth limited to 127 gpuweb/cts#1956.
-
wgsl: Validate limits gpuweb/cts#1710.
-
add vec2(), vec3(), vec4() gpuweb/cts#2481.
-
wgsl: array element count is an unrestricted expression gpuweb/cts#2308.
-
test case: invalid to have bare type constructor or type conversion gpuweb/cts#1411.
-
wgsl: validate atomics disallow initializer expressions gpuweb/cts#1486.
-
Reduce or defer creation of strings for diagnostics gpuweb/cts#3166.
-
wgsl: Implement 'heavy' and 'lite' modes for running numeric tests gpuweb/cts#3093.
-
wgsl: Speed up AbstractFloat (especially Matrix) tests gpuweb/cts#3039.
-
wgsl: conversion expressions gpuweb/cts#2302.
-
Refactor floating point Case generation to follow integer patterns gpuweb/cts#2276.
-
wgsl: Validate array type comparisions gpuweb/cts#1491.
-
wgsl: Validate `array` parsing gpuweb/cts#1492.
-
wgsl: validate array creation gpuweb/cts#1490.
-
wgsl: validate `atomic` parsing gpuweb/cts#1493.
-
wgsl: validate `mat` parsing gpuweb/cts#1494.
-
wgsl: validate creation-time constants gpuweb/cts#1559.
-
wgsl: Validate `override` declarations gpuweb/cts#1557.
-
Add a case for `atomicSub(ptr, -2147483648i)` (int-min) gpuweb/cts#3100.
-
[wgsl] Add discard inside continuing block execution test. gpuweb/cts#3317.
-
wgsl: validate assignment gpuweb/cts#1644.
-
wgsl: interpolation execution tests gpuweb/cts#1696.
-
wgsl : validate function call parameter evaluation ordering gpuweb/cts#1688.
-
wgsl: Arithmetic Expression tests gpuweb/cts#1626.
-
wgsl: Address of test gpuweb/cts#1632.
-
wgsl: Validate increment and decrement gpuweb/cts#1646.
-
wgsl: indirection expression gpuweb/cts#1633.
-
wgsl: validate `continue` statement gpuweb/cts#1681.
-
wgsl: validate switch statements gpuweb/cts#1648.
-
wgsl: validate if statements gpuweb/cts#1647.
-
wgsl: validate `while` statement gpuweb/cts#1670.
-
wgsl: validate `for` statement gpuweb/cts#1669.
-
wgsl: validate `loop` statement gpuweb/cts#1666.
-
wgsl: validate creation-time functions gpuweb/cts#1689.
-
wgsl: validate `return` statement gpuweb/cts#1682.
-
wgsl: Execution tests for `discard` statement gpuweb/cts#1683.
-
wgsl: validate `function_call` statement gpuweb/cts#1684.
-
Support reading structs as outputs from builtin tests gpuweb/cts#2028.
-
wgsl: add short predeclared type aliases for matrix types. gpuweb/cts#2074.
-
wgsl: validate user-defined functions gpuweb/cts#1687.
-
wgsl: indexing X by non-const-expr index requires X to be concrete gpuweb/cts#1893.
-
wgsl: short-named aliases for concrete vector types gpuweb/cts#2036.
-
wgsl: validate non-finite const eval for smoothstep gpuweb/cts#2007.
-
wgsl: validate non-finite const eval for binary arithmetic gpuweb/cts#2008.
-
wgsl: validate non-finite const eval for normalize gpuweb/cts#2004.
-
wgsl: validate non-finite const eval for pack* gpuweb/cts#2005.
-
wgsl: validate non-finite const eval for refract gpuweb/cts#2006.
-
wgsl: validate non-finite const eval for fma gpuweb/cts#2001.
-
wgsl: validate non-finite const eval for length gpuweb/cts#2002.
-
wgsl: validate non-finite const eval for mix gpuweb/cts#2003.
-
wgsl: validate non-finite const eval for clamp gpuweb/cts#1998.
-
wgsl: validate non-finite const eval for cross gpuweb/cts#1999.
-
wgsl: validate non-finite const eval for dot gpuweb/cts#2000.
-
wgsl: f16 validation tests for `reflect` gpuweb/cts#2731.
-
wgsl: f16 validation tests for `pow` gpuweb/cts#2726.
-
wgsl: f16 validation tests for `normalize` gpuweb/cts#2723.
-
wgsl: f16 validation tests for `mix` gpuweb/cts#2720.
-
wgsl: f16 validation tests for `ldexp` gpuweb/cts#2714.
-
wgsl: f16 validation tests for `frexp` gpuweb/cts#2711.
-
wgsl: f16 validation tests for `fma` gpuweb/cts#2708.
-
wgsl: f16 validation tests for `faceForward` gpuweb/cts#2705.
-
wgsl: f16 validation tests for `dot` gpuweb/cts#2696.
-
wgsl: f16 validation tests for `distance` gpuweb/cts#2693.
-
wgsl: f16 validation tests for `cross` gpuweb/cts#2687.
-
wgsl: f16 validation tests for `atan2` gpuweb/cts#2675.
-
wgsl: f16 validation tests for remainder (`%`) gpuweb/cts#2640.
-
wgsl: f16 validation tests for division (`/`) gpuweb/cts#2638.
-
wgsl: f16 validation tests for multiplication (`*`) gpuweb/cts#2635.
-
wgsl: f16 validation tests for subtraction (`-`) gpuweb/cts#2634.
-
wgsl: f16 validation tests for addition `(+)` gpuweb/cts#2632.
-
wgsl: AbstractInt execution tests for `sign` builtin gpuweb/cts#2584.
-
wgsl: f32 validation tests for `mix` gpuweb/cts#2719.
-
wgsl: AbstractFloat validation tests for `normalize` gpuweb/cts#2721.
-
wgsl: f32 validation tests for `normalize` gpuweb/cts#2722.
-
wgsl: AbstractFloat validation tests for `pow` gpuweb/cts#2724.
-
wgsl: f32 validation tests for `pow` gpuweb/cts#2725.
-
wgsl: Implement `fract` validation tests gpuweb/cts#3286.
-
wgsl: validate non-finite const eval for abs gpuweb/cts#1996.
-
wgsl: f32 validation tests for addition (`+`) gpuweb/cts#2631.
-
wgsl: f32 validation tests for subtraction (`-`) gpuweb/cts#2633.
-
wgsl: f32 validation tests for multiplication (`*`) gpuweb/cts#2636.
-
wgsl: f32 validation tests for division (`/`) gpuweb/cts#2637.
-
wgsl: f32 validation tests for remainder (`%`) gpuweb/cts#2639.
-
wgsl: AbstractFloat validation tests for subtraction (`-`) gpuweb/cts#2649.
-
wgsl: AbstractFloat validation tests for addition (`+`) gpuweb/cts#2650.
-
wgsl: AbstractFloat validation tests for multiplication (`*`) gpuweb/cts#2651.
-
wgsl: AbstractFloat validation tests for division (`/`) gpuweb/cts#2652.
-
wgsl: AbstractFloat validation tests for remainder (`%`) gpuweb/cts#2653.
-
wgsl: AbstractFloat validation tests for `atan2` gpuweb/cts#2673.
-
wgsl: f32 validation tests for `atan2` gpuweb/cts#2674.
-
wgsl: AbstractFloat validation tests for `cross` gpuweb/cts#2685.
-
wgsl: f32 validation tests for `cross` gpuweb/cts#2686.
-
wgsl: AbstractFloat validation tests for `distance` gpuweb/cts#2691.
-
wgsl: f32 validation tests for `distance` gpuweb/cts#2692.
-
wgsl: AbstractFloat validation tests for `dot` gpuweb/cts#2694.
-
wgsl: f32 validation tests for `dot` gpuweb/cts#2695.
-
wgsl: AbstractFloat validation tests for `faceForward` gpuweb/cts#2703.
-
wgsl: f32 validation tests for `faceForward` gpuweb/cts#2704.
-
wgsl: AbstractFloat validation tests for `fma` gpuweb/cts#2706.
-
wgsl: f32 validation tests for `fma` gpuweb/cts#2707.
-
wgsl: AbstractFloat validation tests for `frexp` gpuweb/cts#2709.
-
wgsl: f32 validation tests for `frexp` gpuweb/cts#2710.
-
wgsl: AbstractFloat validation tests for `ldexp` gpuweb/cts#2712.
-
wgsl: f32 validation tests for `ldexp` gpuweb/cts#2713.
-
wgsl: AbstractFloat validation tests for `mix` gpuweb/cts#2718.
-
wgsl: f32 validation tests for `quantizeToF16` gpuweb/cts#2727.
-
wgsl: f32 validation tests for `pack2x16float` gpuweb/cts#2728.
-
wgsl: AbstractFloat validation tests for `reflect` gpuweb/cts#2729.
-
wgsl: f32 validation tests for `reflect` gpuweb/cts#2730.
-
wgsl: Execution tests for vector swizzles gpuweb/cts#3169.
-
wgsl: Execution tests for vector accessors gpuweb/cts#3170.
-
wgsl: Execution tests for array and vector accessors gpuweb/cts#3171.
-
wgsl: Execution tests for matrix accessors gpuweb/cts#3172.
-
wgsl: Implement `all` builtin validation tests gpuweb/cts#3272.
-
wgsl: Implement `any` builtin validation tests gpuweb/cts#3273.
-
wgsl: Implement `select` builtin validation tests gpuweb/cts#3274.
-
wgsl: Add AbstractInt entries to `kAllIntegerScalarsAndVectors` gpuweb/cts#3275.
-
wgsl: Implement `refract` validation tests gpuweb/cts#3288.
-
wgsl: Implement `smoothstep` validation tests gpuweb/cts#3289.
-
wgsl: Implement `step` validation tests gpuweb/cts#3290.
-
wgsl: Test swizzling of convertible types gpuweb/cts#3709.
-
-
2 issues reopened.
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761, by @kainino0x.
-
Enforce TS's `readonly` more strictly gpuweb/cts#3126, by @kainino0x.
-
-
116 issues received 129 new comments.
-
`webgpu:api,operation,labels:wrappers_do_not_share_labels:*` calls `createRenderPipeline` with no target gpuweb/cts#3754 (5 comments by @ErichDonGubler, @kainino0x, @Kangz)
-
Test `texture-compression-bc-sliced-3d` gpuweb/cts#3761 (4 comments by @greggman, @kainino0x, @mehmetoguzderin)
-
limit the overrides in the shader interface, to those that are "statically accessed" gpuweb/cts#1991 (3 comments by @dj2, @kainino0x)
-
Invalid CTS shader,execution,expression test attempts to assign concrete vec2<i32> to vec2<f32>? gpuweb/cts#3743 (2 comments by @mwyrzykowski)
-
Add texture type wgsl validation tests for all texture builtin functions gpuweb/cts#3594 (1 comment by @greggman)
-
Some expectations in web_platform,copyToTexture, are unexpected gpuweb/cts#3767 (1 comment by @mwyrzykowski)
-
webgpu:shader,execution,expression,call,builtin,bitcast:vec2f_to_vec4h should allow preservation of denormals gpuweb/cts#3758 (1 comment by @zoddicus)
-
[wgsl] Validate errors are not generated for overrides which aren't present in the module. gpuweb/cts#3747 (1 comment by @dj2)
-
wgsl: shader-creation and pipelin-creation errors with left shift and concrete unsigned LHS gpuweb/cts#1925 (1 comment by @alan-baker)
-
wgsl: Validate various structs with size/align validate correctly. gpuweb/cts#1456 (1 comment by @alan-baker)
-
test case: invalid to have bare type constructor or type conversion gpuweb/cts#1411 (1 comment by @alan-baker)
-
Move builtin must_use tests into per-builtin validation suites gpuweb/cts#3502 (1 comment by @alan-baker)
-
wgsl: Implement 'heavy' and 'lite' modes for running numeric tests gpuweb/cts#3093 (1 comment by @zoddicus)
-
`F32Vector` and `F32Matrix` could store if they are contain only finite values gpuweb/cts#2253 (1 comment by @zoddicus)
-
Should CTS assert instead of quantizing when generating u32/i32 cases? gpuweb/cts#2189 (1 comment by @zoddicus)
-
wgsl: Refactor various generateFooCases and makeFooCase utilities to use Generics gpuweb/cts#3185 (1 comment by @zoddicus)
-
wgsl: unsupported exponent value for ldexp can generate creation-time and pipeline-time errors gpuweb/cts#1927 (1 comment by @dj2)
-
wgsl: corner cases for scoping of formal parameters to user-declared functions gpuweb/cts#2157 (1 comment by @dj2)
-
wgsl: shader-creation and pipeline-creation errors for invalid expressions gpuweb/cts#1814 (1 comment by @dj2)
-
wgsl: indexing X by non-const-expr index requires X to be concrete gpuweb/cts#1893 (1 comment by @dj2)
-
wgsl: & and * are lower precedence than .member and [], even in LHS gpuweb/cts#1981 (1 comment by @dj2)
-
wgsl: enforce that declared array element count is element_count_expression gpuweb/cts#1731 (1 comment by @dj2)
-
wgsl: AbstractFloat validation tests for multiplication (`*`) gpuweb/cts#2651 (1 comment by @zoddicus)
-
wgsl: test smoothstep early evaluation error when low >= high gpuweb/cts#3722 (1 comment by @dneto0)
-
wgsl: Add AbstractInt entries to `kAllIntegerScalarsAndVectors` gpuweb/cts#3275 (1 comment by @zoddicus)
-
wgsl: Need to distinguish between going OOB and an indeterminate value in FP framework gpuweb/cts#2771 (1 comment by @zoddicus)
-
-
39 pull requests opened.
-
Compat: Skip more tests for sample_* validation gpuweb/cts#3776, by @jrprice.
-
Compat: Make compat tests run on core and expect to pass gpuweb/cts#3775, by @greggman.
-
Fixup storage texture var restrictions in test gpuweb/cts#3774, by @sudonatalie.
-
Add vertex variable resource restrictions to test gpuweb/cts#3772, by @sudonatalie.
-
Revert "Update validation tests for 3D textures with compression" gpuweb/cts#3771, by @kainino0x.
-
Roll @webgpu/types gpuweb/cts#3770, by @kainino0x.
-
Compat: Convert pipeline creation tests to shader module creation tests. gpuweb/cts#3769, by @greggman.
-
Provide valid data for astc textures gpuweb/cts#3768, by @greggman.
-
Fix matrix overflow test case gpuweb/cts#3766, by @sudonatalie.
-
Add a test which does override evaluation. gpuweb/cts#3764, by @dj2.
-
Update validation tests for 3D textures with compression gpuweb/cts#3762, by @mehmetoguzderin.
-
WGSL builtin textureSample compressed formats gpuweb/cts#3760, by @greggman.
-
WGSL builtin textureSample w/derivatives test gpuweb/cts#3757, by @greggman.
-
Refactor const parse tests gpuweb/cts#3752, by @alan-baker.
-
Test * and [] precendence on LHS gpuweb/cts#3751, by @alan-baker.
-
Test that let-decls of pointers are allowed as parameters gpuweb/cts#3750, by @alan-baker.
-
Add early evaluation error tests for smoothstep gpuweb/cts#3749, by @alan-baker.
-
Refactor more parse tests gpuweb/cts#3748, by @alan-baker.
-
Refactor expression parse tests gpuweb/cts#3746, by @alan-baker.
-
Move builtin must_use tests to specific builtin tests gpuweb/cts#3745, by @alan-baker.
-
shader/exeuction: Fix abstract-numeric constructor tests gpuweb/cts#3744, by @ben-clayton.
-
Refactor statement validation tests gpuweb/cts#3742, by @alan-baker.
-
Add more structure validation tests gpuweb/cts#3740, by @alan-baker.
-
Add layout tests for matrices in arrays gpuweb/cts#3739, by @alan-baker.
-
shader/execution: Add array of matrix index tests gpuweb/cts#3738, by @ben-clayton.
-
call unregisterAllServiceWorkers only if serviceWorker are actually available gpuweb/cts#3737, by @sagudev.
-
Test variable address space restrictions for shader stages gpuweb/cts#3736, by @alan-baker.
-
Add vec2(), vec3(), and vec4() execution tests gpuweb/cts#3734, by @alan-baker.
-
More array validation tests gpuweb/cts#3733, by @alan-baker.
-
More atomic validation tests gpuweb/cts#3732, by @alan-baker.
-
More matrix validation gpuweb/cts#3731, by @alan-baker.
-
More const declatation validation tests gpuweb/cts#3730, by @alan-baker.
-
More override validation tests gpuweb/cts#3729, by @alan-baker.
-
Add test for atomicSub with i32 min value gpuweb/cts#3728, by @alan-baker.
-
Add discard in continuing block execution test gpuweb/cts#3727, by @alan-baker.
-
wgsl: Implement abstract shift tests (<< and >>) gpuweb/cts#3726, by @zoddicus.
-
Vector and matrix constructor validation tests gpuweb/cts#3725, by @alan-baker.
-
Fix concreteTypeOf(Type.f16, [Type.f16]) gpuweb/cts#3724, by @dneto0.
-
-
46 pull requests merged.
-
Compat: Make compat tests run on core and expect to pass gpuweb/cts#3775.
-
Replace GPUAdapter requestAdapterInfo() method with info attribute gpuweb/cts#3679.
-
Roll @webgpu/types gpuweb/cts#3770.
-
Fixup storage texture var restrictions in test gpuweb/cts#3774.
-
Add vertex variable resource restrictions to test gpuweb/cts#3772.
-
Revert "Update validation tests for 3D textures with compression" gpuweb/cts#3771.
-
Provide valid data for astc textures gpuweb/cts#3768.
-
Add a test which does override evaluation. gpuweb/cts#3764.
-
Update validation tests for 3D textures with compression gpuweb/cts#3762.
-
Fix matrix overflow test case gpuweb/cts#3766.
-
WGSL builtin textureSample compressed formats gpuweb/cts#3760.
-
Validate invalidation of submitted command buffers gpuweb/cts#3716.
-
WGSL builtin textureSample w/derivatives test gpuweb/cts#3757.
-
Refactor const parse tests gpuweb/cts#3752.
-
Test * and [] precendence on LHS gpuweb/cts#3751.
-
Test that let-decls of pointers are allowed as parameters gpuweb/cts#3750.
-
Add early evaluation error tests for smoothstep gpuweb/cts#3749.
-
Refactor more parse tests gpuweb/cts#3748.
-
Refactor expression parse tests gpuweb/cts#3746.
-
Fix concreteTypeOf(Type.f16, [Type.f16]) gpuweb/cts#3724.
-
Move builtin must_use tests to specific builtin tests gpuweb/cts#3745.
-
shader/exeuction: Fix abstract-numeric constructor tests gpuweb/cts#3744.
-
Refactor statement validation tests gpuweb/cts#3742.
-
Add more structure validation tests gpuweb/cts#3740.
-
Add layout tests for matrices in arrays gpuweb/cts#3739.
-
shader/execution: Add array of matrix index tests gpuweb/cts#3738.
-
Improve validation tests for unpack builtin functions gpuweb/cts#3630.
-
Test variable address space restrictions for shader stages gpuweb/cts#3736.
-
wgsl: Add CTS tests for WGSL limits gpuweb/cts#3389.
-
Add vec2(), vec3(), and vec4() execution tests gpuweb/cts#3734.
-
wgsl: Implement abstract shift tests (<< and >>) gpuweb/cts#3726.
-
More array validation tests gpuweb/cts#3733.
-
More atomic validation tests gpuweb/cts#3732.
-
More matrix validation gpuweb/cts#3731.
-
More const declatation validation tests gpuweb/cts#3730.
-
More override validation tests gpuweb/cts#3729.
-
Add test for atomicSub with i32 min value gpuweb/cts#3728.
-
spelling: "take affect" -> "take effect" gpuweb/cts#3720.
-
Add discard in continuing block execution test gpuweb/cts#3727.
-
Vector and matrix constructor validation tests gpuweb/cts#3725.
-
Validation tests for vector constructors from elements gpuweb/cts#3721.
-
wgsl: Add conversions to abstract vector swizzle tests gpuweb/cts#3711.
-
wgsl: Refactor existing bitwise shift execution tests gpuweb/cts#3718.
-
-
1 pull request closed.
-
WGSL validation test for textureDimensions gpuweb/cts#3556.
-
-
17 pull requests received 99 new comments.
-
WGSL builtin textureSample w/derivatives test gpuweb/cts#3757 (23 comments by @greggman, @kainino0x)
-
WGSL builtin textureSample compressed formats gpuweb/cts#3760 (20 comments by @greggman, @kainino0x)
-
fix: specify `rgba8unorm` render target for `webgpu:api,operation,labels:wrappers_do_not_share_labels:*` gpuweb/cts#3756 (7 comments by @ErichDonGubler, @kainino0x)
-
wgsl: Implement abstract shift tests (<< and >>) gpuweb/cts#3726 (7 comments by @amaiorano, @zoddicus)
-
Validate invalidation of submitted command buffers gpuweb/cts#3716 (6 comments by @kainino0x, @toji)
-
Add more cases to canvas reftest with device loss gpuweb/cts#3715 (6 comments by @kainino0x, @toji, @w07i)
-
Compat: Convert pipeline creation tests to shader module creation tests. gpuweb/cts#3769 (5 comments by @greggman, @jrprice)
-
Add vertex variable resource restrictions to test gpuweb/cts#3772 (5 comments by @alan-baker, @sudonatalie)
-
Replace GPUAdapter requestAdapterInfo() method with info attribute gpuweb/cts#3679 (5 comments by @beaufortfrancois, @greggman, @kainino0x)
-
Test variable address space restrictions for shader stages gpuweb/cts#3736 (4 comments by @alan-baker, @dj2)
-
Improve validation tests for unpack builtin functions gpuweb/cts#3630 (4 comments by @ben-clayton, @haoxli)
-
Revert "Update validation tests for 3D textures with compression" gpuweb/cts#3771 (1 comment by @mehmetoguzderin)
-
Add cts to cover copyEI2T copying from fully transparent image gpuweb/cts#3621 (1 comment by @kainino0x)
-
TR publications
-
11 documents published.
-
WebGPU Shading Language published 3 times as a Working Draft. History of WebGPU Shading Language
-
WebGPU published 8 times as a Working Draft. History of WebGPU
-
Participants
-
22 participants joined.
-
Erich Gubler (Mozilla Foundation) joined.
-
Stephen White (Google LLC) joined.
-
lin li (China Mobile Communications Corporation) joined.
-
Yunchao He (ByteDance) joined.
-
Shaobo Yan (Intel Corporation) joined.
-
Connor Fitzgerald (W3C Invited Experts) joined.
-
Lokbondo Kung (Google LLC) joined.
-
Austin Eng (Google LLC) joined.
-
Shuai (Shrek) Shao (Google LLC) joined.
-
Antonio Maiorano (Google LLC) joined.
-
Ryan Harrison (Google LLC) joined.
-
dan sinclair (Google LLC) joined.
-
Ben Clayton (Google LLC) joined.
-
James Price (Google LLC) joined.
-
Teodor Tanasoaia (Mozilla Foundation) joined.
-
Kenji Baheux (Google LLC) joined.
-
Gregg Tavares (Google LLC) joined.
-
Hao Li (Intel Corporation) joined.
-
Mike Wyrzykowski (Apple Inc.) joined.
-
Jie Chen (Intel Corporation) joined.
-
Zhaoming Jiang (Intel Corporation) joined.
-
Krists Auders (Microsoft Corporation) joined.
-
GitHub
gpuweb/gpuweb
-
23 issues opened.
-
Allow 2d-array texture view as renderable texture view gpuweb/gpuweb#4607, by @haoxli.
-
Limit on number of active occlusion queries? gpuweb/gpuweb#4603, by @kainino0x.
-
Has isInf() been removed? gpuweb/gpuweb#4602, by @coramat.
-
Spec allows resolveTarget to be 3d but doesn't allow selecting a depth slice gpuweb/gpuweb#4600, by @kainino0x.
-
Should command buffers be invalidated by an invalid submit? gpuweb/gpuweb#4599, by @kainino0x.
-
Spec web page word highlight feature has low contrast in dark mode gpuweb/gpuweb#4598, by @greggman.
-
Should @diagnostic be allowed in the middle of lines? gpuweb/gpuweb#4595, by @mwyrzykowski.
-
[editorial] Explain the relationship between color management and `-srgb` formats gpuweb/gpuweb#4587, by @kainino0x.
-
List duplicates in spec writing style guide (or elsewhere) gpuweb/gpuweb#4579, by @mehmetoguzderin.
-
Unused bindings throw error gpuweb/gpuweb#4577, by @Rus-Abd.
-
Standardize #include for language server tooling gpuweb/gpuweb#4576, by @stefnotch.
-
[Feature Request] Allow readonly usage of atomics gpuweb/gpuweb#4574, by @stefnotch.
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573, by @kainino0x.
-
Implementation Status: Firefox Nightly WebGPU is blocked by default on macOS gpuweb/gpuweb#4572, by @rinsuki.
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568, by @greggman.
-
writeTexture is missing check for [[destroyed]] gpuweb/gpuweb#4565, by @kainino0x.
-
Do we need a `high-precision` vs `speed` option? gpuweb/gpuweb#4562, by @greggman.
-
clarify when const-exprs subexpressions are evaluated gpuweb/gpuweb#4558, by @dneto0.
-
const/override binary expression clarification. gpuweb/gpuweb#4557, by @Kangz.
-
31 issues closed.
-
Specify swizzles as text rather than grammar gpuweb/gpuweb#4307.
-
Increase distinguishing factor between token types gpuweb/gpuweb#3707.
-
Attribute syntax should not specialize to built-in language attributes gpuweb/gpuweb#3520.
-
Should @diagnostic be allowed in the middle of lines? gpuweb/gpuweb#4595.
-
Has isInf() been removed? gpuweb/gpuweb#4602.
-
Clarify the set of attachments in render pass's colorAttachments gpuweb/gpuweb#4390.
-
Provide a description of every descriptor member gpuweb/gpuweb#2815.
-
Simplify resolveTarget validation language gpuweb/gpuweb#3767.
-
Content timeline step (6) for copyExternalImageToTexture() is inaccurate gpuweb/gpuweb#4456.
-
writeBuffer contents are written on device timeline gpuweb/gpuweb#4252.
-
Slight type problem with `GPURenderPipeline.[[writesDepth]]` gpuweb/gpuweb#4416.
-
Spec is unclear about 3D texture mipmap support gpuweb/gpuweb#4491.
-
Should `GPUQueue.writeTexture` check `[[destroyed]]`? gpuweb/gpuweb#4253.
-
HTMLImageElement width, height needs spec in GPUImageCopyExternalImage gpuweb/gpuweb#4256.
-
Clarification of behavior around GPUStoreOp.discard? gpuweb/gpuweb#4398.
-
Clarify whether `canvas.width = canvas.width` invalidates a WebGPU canvas gpuweb/gpuweb#4452.
-
Should alphaToCoverageEnabled require a blendable format? gpuweb/gpuweb#4506.
-
Does `entryPoint` defaulting apply to `compilationHints`? gpuweb/gpuweb#4510.
-
Unused bindings throw error gpuweb/gpuweb#4577.
-
Accuracy for AbstractFloat `fract` give unintuitive/unexpected results gpuweb/gpuweb#4523.
-
Standardize #include for language server tooling gpuweb/gpuweb#4576.
-
API should be exposed to ServiceWorker gpuweb/gpuweb#4197.
-
Compat: validating out "2d" of more than 1 layer gpuweb/gpuweb#4378.
-
writeTexture is missing check for [[destroyed]] gpuweb/gpuweb#4565.
-
Do we need a `high-precision` vs `speed` option? gpuweb/gpuweb#4562.
-
const/override binary expression clarification. gpuweb/gpuweb#4557.
-
Compat: Disallow bgra8unorm-srgb textures. gpuweb/gpuweb#4514.
-
Broken Chromium bug tracker link on Implementation Status wiki page gpuweb/gpuweb#4553.
-
texture_depth_2d needs more information gpuweb/gpuweb#4555.
-
-
53 issues received 127 new comments.
-
VideoFrame should use display size, not coded size gpuweb/gpuweb#4525 (16 comments by @greggman, @kainino0x, @Kangz, @kdashg, @shaoboyan)
-
Standardize #include for language server tooling gpuweb/gpuweb#4576 (10 comments by @dj2, @greggman, @kdashg, @mighdoll, @stefnotch)
-
Clarification of behavior around GPUStoreOp.discard? gpuweb/gpuweb#4398 (7 comments by @kainino0x, @mwyrzykowski)
-
[editorial] Explain the relationship between color management and `-srgb` formats gpuweb/gpuweb#4587 (6 comments by @kainino0x, @kdashg)
-
[editorial] Consider if there's a much better name for "image copies" gpuweb/gpuweb#4573 (4 comments by @kainino0x, @Kangz)
-
Spec web page word highlight feature has low contrast in dark mode gpuweb/gpuweb#4598 (4 comments by @greggman, @kainino0x, @litherum)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (4 comments by @kdashg, @SenorBlanco, @teoxoy)
-
WGSL built in functions are very under specified gpuweb/gpuweb#4527 (4 comments by @dneto0, @jimblandy, @kdashg, @starmole)
-
const/override binary expression clarification. gpuweb/gpuweb#4557 (4 comments by @dj2, @Kangz, @teoxoy)
-
Should @diagnostic be allowed in the middle of lines? gpuweb/gpuweb#4595 (3 comments by @ben-clayton, @jimblandy, @mwyrzykowski)
-
PNG ImageBitmap (WebAPI) incompatibility with `copyExternalImageToTexture` for physically-based workflow. gpuweb/gpuweb#4609 (3 comments by @sciecode, @shaoboyan)
-
Investigation: Generate mipmaps gpuweb/gpuweb#386 (3 comments by @devshgraphicsprogramming, @JolifantoBambla, @toji)
-
[high-level] Should predeclared enumerant values (e.g. `position`) should be shadowable by user variables? (e.g. `let position =`) gpuweb/gpuweb#4539 (3 comments by @kdashg)
-
clarify when const-exprs subexpressions are evaluated gpuweb/gpuweb#4558 (3 comments by @dneto0, @kdashg)
-
Attribute syntax should not specialize to built-in language attributes gpuweb/gpuweb#3520 (2 comments by @kdashg, @mehmetoguzderin)
-
u32(abstractint) errors on out of bounds value, u32(abstractfloat) and u32(f32) clamps out of bounds value gpuweb/gpuweb#4569 (2 comments by @alan-baker, @jimblandy)
-
Should command buffers be invalidated by an invalid submit? gpuweb/gpuweb#4599 (2 comments by @jimblandy, @toji)
-
Clarify whether type checking occurs for short-circuited RHS expressions gpuweb/gpuweb#4551 (2 comments by @dneto0, @kdashg)
-
Accuracy for AbstractFloat `fract` give unintuitive/unexpected results gpuweb/gpuweb#4523 (2 comments by @kdashg, @zoddicus)
-
Compat: WGSL validation at createShaderModule or createPipeline time? gpuweb/gpuweb#4568 (2 comments by @greggman)
-
Do we need a `high-precision` vs `speed` option? gpuweb/gpuweb#4562 (2 comments by @greggman, @rconde01)
-
Increase distinguishing factor between token types gpuweb/gpuweb#3707 (1 comment by @mehmetoguzderin)
-
HTMLImageElement width, height needs spec in GPUImageCopyExternalImage gpuweb/gpuweb#4256 (1 comment by @toji)
-
Reduce limits for the number of case selector values in a switch statement gpuweb/gpuweb#4548 (1 comment by @kdashg)
-
Reduce limit for maximum number of elements in const-expression of array type gpuweb/gpuweb#4546 (1 comment by @kdashg)
-
Reduce byte-size limit for function and private address spaces gpuweb/gpuweb#4545 (1 comment by @kdashg)
-
Implementation Status: Firefox Nightly WebGPU is blocked by default on macOS gpuweb/gpuweb#4572 (1 comment by @kainino0x)
-
Accuracy of `mix` does not cover all common implementations of linear blend gpuweb/gpuweb#3260 (1 comment by @zoddicus)
-
smoothstep: is it an error when `low == high` and/or `low > high` (when constant/pipline-evaluated) gpuweb/gpuweb#4561 (1 comment by @alan-baker)
-
wgsl: clarify what happens for IEEE 754 "invalid operation" cases, e.g. acos(2) gpuweb/gpuweb#4219 (1 comment by @dneto0)
-
Broken Chromium bug tracker link on Implementation Status wiki page gpuweb/gpuweb#4553 (1 comment by @kainino0x)
-
-
32 pull requests opened.
-
Use 'create a new WebGPU Object' for all objects gpuweb/gpuweb#4615, by @toji.
-
[editorial] Reorder attribute tables to put descriptions first gpuweb/gpuweb#4614, by @dneto0.
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613, by @kainino0x.
-
Push constant proposal gpuweb/gpuweb#4612, by @shaoboyan.
-
Invalidate all command buffers passed to submit, whether the submit is valid or not. gpuweb/gpuweb#4608, by @toji.
-
Descriptions for remaining descriptor members gpuweb/gpuweb#4605, by @toji.
-
Rephrase resolveTarget extent validation gpuweb/gpuweb#4604, by @toji.
-
Add logo to the spec and favicon gpuweb/gpuweb#4601, by @toji.
-
[non-normative] Explain -srgb formats are not color managed/tagged gpuweb/gpuweb#4597, by @kainino0x.
-
Validate that copyExternalImageToTexture dest is not destroyed gpuweb/gpuweb#4596, by @toji.
-
Move write portion of writeBuffer onto queue timeline gpuweb/gpuweb#4594, by @toji.
-
Require that every submitted command buffer must be unique gpuweb/gpuweb#4593, by @toji.
-
Attachment resolveTargets must also be pairwise disjoint gpuweb/gpuweb#4592, by @toji.
-
Only use depthWriteEnabled if provided gpuweb/gpuweb#4591, by @toji.
-
Clarify valid uses for each texture dimension gpuweb/gpuweb#4590, by @toji.
-
Fix invalid copyExternalImageToTexture origin references gpuweb/gpuweb#4589, by @toji.
-
Add optional feature `clip-distances` gpuweb/gpuweb#4588, by @Jiawei-Shao.
-
Clarify when canvas resize algorithm applies gpuweb/gpuweb#4586, by @toji.
-
Specify that textures used with alphaToCoverage must be blendable gpuweb/gpuweb#4585, by @toji.
-
Specify that attachment discard may be lazy gpuweb/gpuweb#4584, by @toji.
-
Better define anisotropic filtering gpuweb/gpuweb#4583, by @toji.
-
Note: shader module hints do not consider default entry points gpuweb/gpuweb#4581, by @toji.
-
Remove resolved TODOs gpuweb/gpuweb#4580, by @toji.
-
wgsl: Reduce shader limits gpuweb/gpuweb#4578, by @jrprice.
-
Issue templates: fix links, remove blank option, improve descriptions/sorting gpuweb/gpuweb#4575, by @kainino0x.
-
[editorial] Update old incorrectly styled timeline boxes, clarify timeline examples gpuweb/gpuweb#4571, by @kainino0x.
-
[editorial] Tweak "requirements" style for clarity, in a few instances gpuweb/gpuweb#4570, by @kainino0x.
-
Specify that writeTexture must fail if the texture is destroyed gpuweb/gpuweb#4567, by @toji.
-
Drop ASTC requirement in compatibility mode. gpuweb/gpuweb#4566, by @SenorBlanco.
-
Compat: disallow textureLoad with depth textures gpuweb/gpuweb#4564, by @greggman.
-
[editorial] Clarify ldexp overloads gpuweb/gpuweb#4563, by @alan-baker.
-
Move Attribute Values out of Enumerants gpuweb/gpuweb#4559, by @mehmetoguzderin.
-
-
28 pull requests merged.
-
[editorial] Reorder attribute tables to put descriptions first gpuweb/gpuweb#4614.
-
[editorial] Clarify ldexp overloads gpuweb/gpuweb#4563.
-
Better Specialize Syntax, Fix Typos, and Improve Grammar Validation Script gpuweb/gpuweb#4554.
-
Attachment resolveTargets must also be pairwise disjoint gpuweb/gpuweb#4592.
-
Descriptions for remaining descriptor members gpuweb/gpuweb#4605.
-
Rephrase resolveTarget extent validation gpuweb/gpuweb#4604.
-
Add logo to the spec and favicon gpuweb/gpuweb#4601.
-
Better define anisotropic filtering gpuweb/gpuweb#4583.
-
Fix invalid copyExternalImageToTexture origin references gpuweb/gpuweb#4589.
-
Move write portion of writeBuffer onto queue timeline gpuweb/gpuweb#4594.
-
Only use depthWriteEnabled if provided gpuweb/gpuweb#4591.
-
Clarify valid uses for each texture dimension gpuweb/gpuweb#4590.
-
Validate that copyExternalImageToTexture dest is not destroyed gpuweb/gpuweb#4596.
-
Require that every submitted command buffer must be unique gpuweb/gpuweb#4593.
-
Specify that attachment discard may be lazy gpuweb/gpuweb#4584.
-
Clarify when canvas resize algorithm applies gpuweb/gpuweb#4586.
-
Specify that textures used with alphaToCoverage must be blendable gpuweb/gpuweb#4585.
-
Note: shader module hints do not consider default entry points gpuweb/gpuweb#4581.
-
Remove resolved TODOs gpuweb/gpuweb#4580.
-
Issue templates: fix links, remove blank option, improve descriptions/sorting gpuweb/gpuweb#4575.
-
[editorial] Tweak "requirements" style for clarity, in a few instances gpuweb/gpuweb#4570.
-
wgsl: Add a carve out for AbstractFloat `fract` accuracy gpuweb/gpuweb#4541.
-
Compat: disallow textureLoad with depth textures gpuweb/gpuweb#4564.
-
Clarify behavior of out-of-gamut canvas gpuweb/gpuweb#4476.
-
Specify that writeTexture must fail if the texture is destroyed gpuweb/gpuweb#4567.
-
Move Attribute Values out of Enumerants gpuweb/gpuweb#4559.
-
Specify copy rules for different canvas contexts gpuweb/gpuweb#4366.
-
-
29 pull requests received 168 new comments.
-
Better Specialize Syntax, Fix Typos, and Improve Grammar Validation Script gpuweb/gpuweb#4554 (50 comments by @alan-baker, @dneto0, @mehmetoguzderin)
-
Add optional feature `clip-distances` gpuweb/gpuweb#4588 (19 comments by @beaufortfrancois, @Jiawei-Shao, @kainino0x, @teoxoy)
-
Push constant proposal gpuweb/gpuweb#4612 (15 comments by @alan-baker, @dneto0, @magcius, @shaoboyan)
-
Attachment resolveTargets must also be pairwise disjoint gpuweb/gpuweb#4592 (11 comments by @kainino0x, @toji)
-
[editorial] Tweak "requirements" style for clarity, in a few instances gpuweb/gpuweb#4570 (10 comments by @kainino0x, @toji)
-
Add synchronous GPUAdapter.info gpuweb/gpuweb#4550 (7 comments by @beaufortfrancois, @kainino0x, @toji)
-
Invalidate all command buffers passed to submit, whether the submit is valid or not. gpuweb/gpuweb#4608 (3 comments by @kainino0x, @toji)
-
Compat: disallow textureLoad with depth textures gpuweb/gpuweb#4564 (3 comments by @greggman, @kainino0x)
-
Drop ASTC requirement in compatibility mode. gpuweb/gpuweb#4566 (3 comments by @kainino0x, @SenorBlanco, @teoxoy)
-
Issue templates: fix links, remove blank option, improve descriptions/sorting gpuweb/gpuweb#4575 (2 comments by @kainino0x, @toji)
-
Allow out-of-standard-gamut colors with float canvases gpuweb/gpuweb#4526 (2 comments by @ccameron-chromium)
-
Define conventions/rules for features that add new limits gpuweb/gpuweb#4613 (1 comment by @kainino0x)
-
Specify that textures used with alphaToCoverage must be blendable gpuweb/gpuweb#4585 (1 comment by @kainino0x)
-
Note: shader module hints do not consider default entry points gpuweb/gpuweb#4581 (1 comment by @kainino0x)
-
gpuweb/cts
-
30 issues opened.
-
wgsl: Test swizzling of convertible types gpuweb/cts#3709, by @zoddicus.
-
getDeviceMismatchedBindingResource returns mismatched buffers gpuweb/cts#3707, by @sagudev.
-
At module-scope is not allowed gpuweb/cts#3705, by @dj2.
-
wgsl: fract tests unexpectedly ballooning runtime gpuweb/cts#3701, by @zoddicus.
-
Test more types in `webgpu:shader,validation,statement,*` tests gpuweb/cts#3685, by @ben-clayton.
-
Test that abstract matrices OOB conditions cause validation errors. gpuweb/cts#3682, by @dj2.
-
wgsl: Update CTS for AF fract gpuweb/cts#3672, by @zoddicus.
-
copyToTexture needs to cover partially-transparent content source. gpuweb/cts#3668, by @shaoboyan.
-
wgsl: fract AF execution tests are out of spec gpuweb/cts#3663, by @zoddicus.
-
errors in webgpu:shader,execution,expression,constructor,non_zero:* gpuweb/cts#3657, by @kainino0x.
-
atan2 WGSL validation test fail - f16 not enabled gpuweb/cts#3653, by @greggman.
-
wgls: Implement `unpack2x16`[snorm|unorm|float] validation tests gpuweb/cts#3620, by @haoxli.
-
wgls: Implement `unpack4x8`[snorm|snorm] validation tests gpuweb/cts#3618, by @haoxli.
-
Fix web_platform tests to run on dedicated, shared, and service workers gpuweb/cts#3616, by @kainino0x.
-
Improvements to validation builtin tests gpuweb/cts#3613, by @amaiorano.
-
wgsl: Implement `pack2x16float` validation tests gpuweb/cts#3610, by @Jiawei-Shao.
-
wgsl: Implement `pack2x16unorm` validation tests gpuweb/cts#3609, by @Jiawei-Shao.
-
wgsl: Implement `pack2x16snorm` validation tests gpuweb/cts#3608, by @Jiawei-Shao.
-
wgsl: Implement `pack4x8unorm` validation tests gpuweb/cts#3606, by @Jiawei-Shao.
-
clarify when const-exprs are evaluated gpuweb/cts#3596, by @dneto0.
-
Add texture type wgsl validation tests for all texture builtin functions gpuweb/cts#3594, by @greggman.
-
Skip or fail tests if the worker= mode is not supported gpuweb/cts#3591, by @kainino0x.
-
Add return type tests for texture builtins gpuweb/cts#3586, by @greggman.
-
Service worker mode not working in standalone in static deployment gpuweb/cts#3583, by @kainino0x.
-
-
74 issues closed.
-
getDeviceMismatchedBindingResource returns mismatched buffers gpuweb/cts#3707.
-
wgsl: execution test for fwidthFine builtin gpuweb/cts#1257.
-
wgsl: execution test for fwidth builtin gpuweb/cts#1255.
-
wgsl: execution test for fwidthCoarse builtin gpuweb/cts#1256.
-
wgsl: Validate logical expressions gpuweb/cts#1623.
-
wgsl: validate struct accessor gpuweb/cts#1622.
-
wgsl: validate array access expressions gpuweb/cts#1621.
-
wgsl: validate matrix access expressions gpuweb/cts#1611.
-
wgsl: Execution tests, constructors gpuweb/cts#1606.
-
wgsl: fract tests unexpectedly ballooning runtime gpuweb/cts#3701.
-
wgsl: validation function scoped variables gpuweb/cts#1603.
-
wgsl: Execution tests for var initialization gpuweb/cts#1565.
-
wgsl: execution test for override declaration gpuweb/cts#1558.
-
wgsl: Validate let declarations gpuweb/cts#1551.
-
wgsl: Validate texture data conversions gpuweb/cts#1544.
-
wgsl: validate extra components when reading / writing texture data gpuweb/cts#1543.
-
wgsl: Pointer execution tests gpuweb/cts#1540.
-
Ensure shader creation-time evaluation gpuweb/cts#1519.
-
Add abstract-numeric `ScalarType`s gpuweb/cts#1517.
-
wgsl: execution tests for ordering of atomic writes gpuweb/cts#1489.
-
wgsl: Validate atomic type creation gpuweb/cts#1485.
-
wgsl: Validate matrix creation with non-floating point types gpuweb/cts#1484.
-
wgsl: Testing for large abstract values gpuweb/cts#1482.
-
wgsl: execution tests for scoping rules gpuweb/cts#1468.
-
At module-scope is not allowed gpuweb/cts#3705.
-
wgsl: validate `global_directive` gpuweb/cts#1465.
-
wgsl: execution tests for `@location` gpuweb/cts#1454.
-
wgsl: execution tests for `@interpolate` gpuweb/cts#1451.
-
wgsl: Add execution tests for `@binding` and `@group` values gpuweb/cts#1440.
-
wgsl: Add `@builtin` execution tests gpuweb/cts#1436.
-
wgsl: `@align` execution tests gpuweb/cts#1433.
-
wgsl: Non UTF-8 text is a valid encoding. gpuweb/cts#1429.
-
wgsl: execution test for workgroupBarrier builtin gpuweb/cts#1295.
-
wgsl: execution test for storageBarrier builtin gpuweb/cts#1294.
-
wgsl: More Abstract number tests gpuweb/cts#1323.
-
wgsl: execution test for dpdyCoarse builtin gpuweb/cts#1253.
-
wgsl: execution test for dpdyFine builtin gpuweb/cts#1254.
-
wgsl: execution test for dpdy builtin gpuweb/cts#1252.
-
wgsl: execution test for dpdxFine builtin gpuweb/cts#1251.
-
wgsl: execution test for dpdxCoarse builtin gpuweb/cts#1250.
-
wgsl: Add execution test for dpdx builtin gpuweb/cts#1249.
-
wgsl: literal parsing execution tests gpuweb/cts#1192.
-
Test that abstract matrices OOB conditions cause validation errors. gpuweb/cts#3682.
-
Test more types in `webgpu:shader,validation,statement,*` tests gpuweb/cts#3685.
-
errors in webgpu:shader,execution,expression,constructor,non_zero:* gpuweb/cts#3657.
-
wgsl: Update CTS for AF fract gpuweb/cts#3672.
-
wgsl: fract AF execution tests are out of spec gpuweb/cts#3663.
-
Test that fine derivatives are actually fine gpuweb/cts#3065.
-
wgsl: test unrestricted_pointer_parameters feature gpuweb/cts#3056.
-
atan2 WGSL validation test fail - f16 not enabled gpuweb/cts#3653.
-
wgsl: Implement `pack2x16unorm` validation tests gpuweb/cts#3609.
-
wgsl: Implement `pack2x16float` validation tests gpuweb/cts#3610.
-
wgsl: Implement `pack2x16snorm` validation tests gpuweb/cts#3608.
-
wgsl: Implement `pack4x8unorm` validation tests gpuweb/cts#3606.
-
wgls: Implement `unpack2x16`[snorm|unorm|float] validation tests gpuweb/cts#3620.
-
wgls: Implement `unpack4x8`[snorm|snorm] validation tests gpuweb/cts#3618.
-
Skip or fail tests if the worker= mode is not supported gpuweb/cts#3591.
-
Service worker mode not working in standalone in static deployment gpuweb/cts#3583.
-
clarify when const-exprs are evaluated gpuweb/cts#3596.
-
Add test for GPUCompilationMessage using UTF-16 gpuweb/cts#3489.
-
-
75 issues received 106 new comments.
-
`webgpu:shader,validation,expression,call,builtin,distance:values:type="f16";*` test failures gpuweb/cts#3623 (10 comments by @ben-clayton, @toji, @zoddicus)
-
errors in webgpu:shader,execution,expression,constructor,non_zero:* gpuweb/cts#3657 (7 comments by @ben-clayton, @kainino0x)
-
`npm test` is failing with `Generated test variant would produce too-long -actual.txt filename` gpuweb/cts#3676 (5 comments by @ben-clayton, @greggman, @kainino0x)
-
error in webgpu:shader,execution,expression,access,structure,index:const_nested:member_types gpuweb/cts#3654 (4 comments by @ben-clayton, @kainino0x)
-
Service worker mode not working in standalone in static deployment gpuweb/cts#3583 (4 comments by @beaufortfrancois, @kainino0x)
-
error in webgpu:shader,validation,expression,call,builtin,normalize:values:stage="override";type="vec4%3Cf16%3E gpuweb/cts#3655 (3 comments by @kainino0x, @toji)
-
wgsl: fract tests unexpectedly ballooning runtime gpuweb/cts#3701 (2 comments by @kainino0x, @zoddicus)
-
should test use of implicit derivatives in point rasterization mode gpuweb/cts#1182 (2 comments by @dj2, @jrprice)
-
Constant evaluation shader test expectations do not allow for rounding as specified in the specification? gpuweb/cts#3261 (2 comments by @mwyrzykowski, @zoddicus)
-
wgsl: test unrestricted_pointer_parameters feature gpuweb/cts#3056 (2 comments by @ben-clayton, @dneto0)
-
Add option to discard device after a certain number of test cases have used it gpuweb/cts#2898 (1 comment by @ashiaabobs)
-
web_platform: copyExternalImageToTexture with more source types (PNG, ImageData) with color spaces gpuweb/cts#1476 (1 comment by @kainino0x)
-
Add canvas tests for RGBA16Float values with regard to SDR luminance gpuweb/cts#1093 (1 comment by @kainino0x)
-
wgsl: validate extra components when reading / writing texture data gpuweb/cts#1543 (1 comment by @dj2)
-
wgsl: validate all arithmetic operations fail if one of the types is an atomic. gpuweb/cts#1488 (1 comment by @dj2)
-
wgsl: Validate vector creation with non scalar and non-abstract types fails gpuweb/cts#1483 (1 comment by @dj2)
-
wgsl: For each builtin that is `@const` update tests to run at both compile time and runtime. gpuweb/cts#1481 (1 comment by @dj2)
-
wgsl: Formal parameter of pointer type must be instantiable as a variable gpuweb/cts#1395 (1 comment by @dj2)
-
Test more types in `webgpu:shader,validation,statement,*` tests gpuweb/cts#3685 (1 comment by @ben-clayton)
-
How to differentiate between backends due to internal details/rounding gpuweb/cts#930 (1 comment by @zoddicus)
-
Fix web_platform tests to run on dedicated, shared, and service workers gpuweb/cts#3616 (1 comment by @kainino0x)
-
smoothstep: is it an error when `low == high` and/or `low > high` (when constant/pipline-evaluated) gpuweb/cts#3601 (1 comment by @amaiorano)
-
-
114 pull requests opened.
-
Validation tests for vector constructors from elements gpuweb/cts#3721, by @alan-baker.
-
spelling: "take affect" -> "take effect" gpuweb/cts#3720, by @dneto0.
-
getDeviceMismatchedBindingResource return correct buffer gpuweb/cts#3719, by @sagudev.
-
wgsl: Refactor existing bitwise shift execution tests gpuweb/cts#3718, by @zoddicus.
-
Zero and value constructor validation tests gpuweb/cts#3717, by @alan-baker.
-
Validate invalidation of submitted command buffers gpuweb/cts#3716, by @toji.
-
Add more cases to canvas reftest with device loss gpuweb/cts#3715, by @kainino0x.
-
[wgsl] Move decl,ptr_spelling to types,pointer. gpuweb/cts#3714, by @dj2.
-
[wgsl] Combine var/let tests gpuweb/cts#3713, by @dj2.
-
[wgsl] Move the align validation tests. gpuweb/cts#3712, by @dj2.
-
wgsl: Add conversions to abstract vector swizzle tests gpuweb/cts#3711, by @zoddicus.
-
Structure type validation tests gpuweb/cts#3710, by @alan-baker.
-
wgsl: Add validation tests for vector access of abstract types gpuweb/cts#3708, by @zoddicus.
-
Add eslint n/no-restricted-import, n/no-restricted-require gpuweb/cts#3706, by @kainino0x.
-
Don't automatically include all types from dependencies gpuweb/cts#3704, by @kainino0x.
-
[wgsl] Add `@size` const eval test gpuweb/cts#3703, by @dj2.
-
Buildfix: use correct include for 'assert' gpuweb/cts#3702, by @kainino0x.
-
Add matrix overflow test cases. gpuweb/cts#3700, by @dj2.
-
Test that submit validates command buffers uniqueness gpuweb/cts#3699, by @toji.
-
Fix all the tests about resource limits on inter-stage shader variables gpuweb/cts#3697, by @Jiawei-Shao.
-
Validation tests for trailing commas in type specifiers gpuweb/cts#3696, by @alan-baker.
-
Validation for address space and access mode in var decls gpuweb/cts#3695, by @alan-baker.
-
wgsl: Implement `transpose` validation tests gpuweb/cts#3694, by @zoddicus.
-
[wgsl] Add var initialization tests. gpuweb/cts#3693, by @dj2.
-
Validation tests for duplicate attributes gpuweb/cts#3692, by @alan-baker.
-
wgsl: Migrate AF multiplication tests to use lookup table gpuweb/cts#3691, by @zoddicus.
-
shader/validation: Expand tested statement condition types gpuweb/cts#3690, by @ben-clayton.
-
Validation tests for texture/Dimensions/Num/Layers/Levels/Samples gpuweb/cts#3689, by @greggman.
-
wgsl: Add validation tests for continue statement gpuweb/cts#3688, by @jrprice.
-
wgsl: Add more validation tests for length gpuweb/cts#3687, by @jrprice.
-
wgsl: Add execution tests for fwidth builtins gpuweb/cts#3686, by @jrprice.
-
wgsl: Migrate AF 'fract' to use look up table gpuweb/cts#3684, by @zoddicus.
-
Structure access expression validation gpuweb/cts#3683, by @alan-baker.
-
add missing listings and shorten test names gpuweb/cts#3681, by @greggman.
-
Matrix access expression validation tests gpuweb/cts#3680, by @alan-baker.
-
Replace GPUAdapter requestAdapterInfo() method with info attribute gpuweb/cts#3679, by @beaufortfrancois.
-
shader/execution: Fix `f16` structure member access tests gpuweb/cts#3678, by @ben-clayton.
-
shader/execution: Fix abstract non_zero constructor tests gpuweb/cts#3677, by @ben-clayton.
-
Fix texel data test gpuweb/cts#3675, by @Jiawei-Shao.
-
[style] Always specify loadOp before storeOp gpuweb/cts#3674, by @kainino0x.
-
Add various matrix validation test. gpuweb/cts#3673, by @dj2.
-
shader/validation: Move `break if` expression type validation out of `parse` gpuweb/cts#3671, by @ben-clayton.
-
Array access expression validation gpuweb/cts#3670, by @alan-baker.
-
Add reftests: canvas display after device lost gpuweb/cts#3669, by @w07i.
-
wgsl: Migrate AF Subtraction to using lookup table gpuweb/cts#3667, by @zoddicus.
-
Add WGSL validation return, value tests for textureSampleXXX builtins gpuweb/cts#3666, by @greggman.
-
Add return and value validation tests for textureLoad/textureStore gpuweb/cts#3665, by @greggman.
-
wgsl: Move AF `fract` execution tests using f32 intervals gpuweb/cts#3664, by @zoddicus.
-
Enumerant type validation tests gpuweb/cts#3662, by @alan-baker.
-
wgsl: Test atomic builtins with non-atomic types gpuweb/cts#3661, by @jrprice.
-
Add enable f16 for atan2 tests gpuweb/cts#3660, by @dneto0.
-
shader/execution: Fix composite abstract runtime indexing (again) gpuweb/cts#3659, by @ben-clayton.
-
Test validation of increment and decrement statements. gpuweb/cts#3658, by @dneto0.
-
Test that lhs reference of increment decrement evaluated once gpuweb/cts#3656, by @dneto0.
-
Add texture type and return type tests to gpuweb/cts#3652, by @greggman.
-
Add texture type and return type tests to gpuweb/cts#3651, by @greggman.
-
Add validation tests for phony assignment gpuweb/cts#3650, by @dneto0.
-
Pointer type validation tests gpuweb/cts#3649, by @alan-baker.
-
shader/validation: Add compound assignment validation gpuweb/cts#3648, by @ben-clayton.
-
[wgsl] Modernizes `modf`, `round`, and `saturate` to cover remaining types gpuweb/cts#3647, by @lokokung.
-
[wgsl] User function validation gpuweb/cts#3646, by @dj2.
-
shader/validation: Begin parsing and type validation for statements gpuweb/cts#3645, by @ben-clayton.
-
wgsl: Test that you cannot type `ref` gpuweb/cts#3644, by @jrprice.
-
wgsl: Add utility to help check for ambiguous overflows gpuweb/cts#3643, by @toji.
-
Remove stale TODO gpuweb/cts#3642, by @dneto0.
-
shader/execution: Add structure member access tests gpuweb/cts#3641, by @ben-clayton.
-
Add return type and texture type WGSL textureSample tests gpuweb/cts#3640, by @greggman.
-
wgsl: Improve the quality of AI -> u32/i32 cases gpuweb/cts#3639, by @zoddicus.
-
Validation tests do not belong in the execution test branch gpuweb/cts#3638, by @greggman.
-
Add parameter validation for `atan2` gpuweb/cts#3637, by @dj2.
-
Update petamoriken/float16 to v3.8.6 gpuweb/cts#3636, by @toji.
-
wgsl: Updated inverseSqrt builtin args validation gpuweb/cts#3635, by @toji.
-
Add missing f16 enables to my recent tests gpuweb/cts#3634, by @amaiorano.
-
wgsl: Update exp/exp2 builtins args validation gpuweb/cts#3633, by @toji.
-
wgsl: Update trig builtins arg validation gpuweb/cts#3632, by @toji.
-
shader/execution: Add more constructor tests gpuweb/cts#3631, by @ben-clayton.
-
Improve validation tests for unpack builtin functions gpuweb/cts#3630, by @haoxli.
-
Add more validation tests for unpack4xI8 and unpack4xU8 gpuweb/cts#3629, by @haoxli.
-
Modernizes ceil/clamp tests to cover remaining types gpuweb/cts#3628, by @lokokung.
-
Validation tests for builtin `extractBits` gpuweb/cts#3627, by @lokokung.
-
wgsl: Validation for refract builtin function gpuweb/cts#3626, by @toji.
-
For .unimplemented() tests, skip before iterating subcases gpuweb/cts#3625, by @kainino0x.
-
Add validation tests for type aliases gpuweb/cts#3624, by @alan-baker.
-
Validate unpack2x16snorm, unpack2x16unorm and unpack2x16float gpuweb/cts#3622, by @haoxli.
-
Add cts to cover copyEI2T copying from fully transparent image gpuweb/cts#3621, by @shaoboyan.
-
Validate unpack4x8snorm and unpack4x8unorm gpuweb/cts#3619, by @haoxli.
-
Fix tests that use rAF when run in shared/service worker gpuweb/cts#3617, by @kainino0x.
-
wgsl: Validation of ldexp builtin function gpuweb/cts#3615, by @toji.
-
Add tests for implicit conversion and overload resolution gpuweb/cts#3614, by @alan-baker.
-
Validate pack2x16float gpuweb/cts#3612, by @Jiawei-Shao.
-
Validate pack2x16snorm and pack2x16unorm gpuweb/cts#3611, by @Jiawei-Shao.
-
Validate pack4x8unorm gpuweb/cts#3607, by @Jiawei-Shao.
-
Fix wrong metadata in video resource meta lists gpuweb/cts#3605, by @shaoboyan.
-
Fail tests in worker=* mode, when worker can't be created gpuweb/cts#3604, by @kainino0x.
-
wgsl: Add tests for operator precedence gpuweb/cts#3603, by @jrprice.
-
Add smoothstep validation tests gpuweb/cts#3602, by @amaiorano.
-
Tests that context dependent names do not use name resolution gpuweb/cts#3600, by @alan-baker.
-
Compute service worker paths using import.meta gpuweb/cts#3599, by @kainino0x.
-
Add validation tests for any, all, and select gpuweb/cts#3598, by @amaiorano.
-
Fix a few web_platform tests to run or skip on workers gpuweb/cts#3597, by @kainino0x.
-
Add div/rem binary operation validation tests gpuweb/cts#3595, by @Kangz.
-
Make GitHub hide diff on hashes.json by default gpuweb/cts#3593, by @kainino0x.
-
Validation tests for insertBits. gpuweb/cts#3592, by @lokokung.
-
wgsl: Validation of frexp builtin function gpuweb/cts#3590, by @toji.
-
Validate pack4x8snorm gpuweb/cts#3589, by @shrekshao.
-
wgsl: Validation for faceForward builtin function gpuweb/cts#3588, by @toji.
-
wgsl: Validation of fract builtin function gpuweb/cts#3587, by @toji.
-
wgsl: Validation for distance builtin function gpuweb/cts#3585, by @toji.
-
Add out of range add/sub/mul operator tests. gpuweb/cts#3584, by @Kangz.
-
Finishing types:textures: external, depth, sampler gpuweb/cts#3582, by @shrekshao.
-
Compat: Fix WGSL textureDimensions tests gpuweb/cts#3581, by @greggman.
-
Print video error message if video fails play gpuweb/cts#3580, by @shaoboyan.
-
wgsl: Validation for dot builtin function gpuweb/cts#3579, by @toji.
-
WGSL execution TextureSample tests gpuweb/cts#3578, by @greggman.
-
-
122 pull requests merged.
-
getDeviceMismatchedBindingResource return correct buffer gpuweb/cts#3719.
-
wgsl: Add execution tests for fwidth builtins gpuweb/cts#3686.
-
Zero and value constructor validation tests gpuweb/cts#3717.
-
Add cts to cover copyEI2T copying from fully transparent image gpuweb/cts#3621.
-
Structure type validation tests gpuweb/cts#3710.
-
Add reftests: canvas display after device lost gpuweb/cts#3669.
-
[wgsl] Move decl,ptr_spelling to types,pointer. gpuweb/cts#3714.
-
[wgsl] Combine var/let tests gpuweb/cts#3713.
-
[wgsl] Move the align validation tests. gpuweb/cts#3712.
-
wgsl: Implement `transpose` validation tests gpuweb/cts#3694.
-
wgsl: Add validation tests for vector access of abstract types gpuweb/cts#3708.
-
Add eslint n/no-restricted-import, n/no-restricted-require gpuweb/cts#3706.
-
Test that submit validates command buffers uniqueness gpuweb/cts#3699.
-
Don't automatically include all types from dependencies gpuweb/cts#3704.
-
[wgsl] Add `@size` const eval test gpuweb/cts#3703.
-
Buildfix: use correct include for 'assert' gpuweb/cts#3702.
-
Add matrix overflow test cases. gpuweb/cts#3700.
-
wgsl: Validation for refract builtin function gpuweb/cts#3626.
-
[wgsl] Add var initialization tests. gpuweb/cts#3693.
-
Validation tests for trailing commas in type specifiers gpuweb/cts#3696.
-
Validation for address space and access mode in var decls gpuweb/cts#3695.
-
Validation tests for duplicate attributes gpuweb/cts#3692.
-
Fix texel data test gpuweb/cts#3675.
-
wgsl: Migrate AF multiplication tests to use lookup table gpuweb/cts#3691.
-
wgsl: Validation of fma builtin function gpuweb/cts#3577.
-
shader/validation: Expand tested statement condition types gpuweb/cts#3690.
-
wgsl: Validation for dot builtin function gpuweb/cts#3579.
-
Validation tests for texture/Dimensions/Num/Layers/Levels/Samples gpuweb/cts#3689.
-
wgsl: Add more validation tests for length gpuweb/cts#3687.
-
wgsl: Add validation tests for continue statement gpuweb/cts#3688.
-
Structure access expression validation gpuweb/cts#3683.
-
Matrix access expression validation tests gpuweb/cts#3680.
-
shader/execution: Fix abstract non_zero constructor tests gpuweb/cts#3677.
-
shader/execution: Fix `f16` structure member access tests gpuweb/cts#3678.
-
wgsl: Migrate AF 'fract' to use look up table gpuweb/cts#3684.
-
wgsl: Migrate AF Subtraction to using lookup table gpuweb/cts#3667.
-
add missing listings and shorten test names gpuweb/cts#3681.
-
Add various matrix validation test. gpuweb/cts#3673.
-
[style] Always specify loadOp before storeOp gpuweb/cts#3674.
-
Array access expression validation gpuweb/cts#3670.
-
shader/validation: Begin parsing and type validation for statements gpuweb/cts#3645.
-
wgsl: Validation for pow builtin function gpuweb/cts#3549.
-
Add return and value validation tests for textureLoad/textureStore gpuweb/cts#3665.
-
wgsl: Move AF `fract` execution tests using f32 intervals gpuweb/cts#3664.
-
Enumerant type validation tests gpuweb/cts#3662.
-
wgsl: Test atomic builtins with non-atomic types gpuweb/cts#3661.
-
Test validation of increment and decrement statements. gpuweb/cts#3658.
-
wgsl: Validation of ldexp builtin function gpuweb/cts#3615.
-
Test that lhs reference of increment decrement evaluated once gpuweb/cts#3656.
-
Add texture type and return type tests to gpuweb/cts#3652.
-
[wgsl] User function validation gpuweb/cts#3646.
-
Add validation tests for phony assignment gpuweb/cts#3650.
-
wgsl: Validation for mix builtin function gpuweb/cts#3573.
-
Pointer type validation tests gpuweb/cts#3649.
-
Add enable f16 for atan2 tests gpuweb/cts#3660.
-
shader/execution: Fix composite abstract runtime indexing (again) gpuweb/cts#3659.
-
shader/validation: Add compound assignment validation gpuweb/cts#3648.
-
Add more validation tests for unpack4xI8 and unpack4xU8 gpuweb/cts#3629.
-
Validate pack2x16float gpuweb/cts#3612.
-
wgsl: Add utility to help check for ambiguous overflows gpuweb/cts#3643.
-
Modernizes ceil/clamp tests to cover remaining types gpuweb/cts#3628.
-
wgsl: Test that you cannot type `ref` gpuweb/cts#3644.
-
Add return type and texture type WGSL textureSample tests gpuweb/cts#3640.
-
Remove stale TODO gpuweb/cts#3642.
-
shader/execution: Add structure member access tests gpuweb/cts#3641.
-
wgsl: Improve the quality of AI -> u32/i32 cases gpuweb/cts#3639.
-
Validate pack2x16snorm and pack2x16unorm gpuweb/cts#3611.
-
Validate pack4x8unorm gpuweb/cts#3607.
-
Validation tests do not belong in the execution test branch gpuweb/cts#3638.
-
Add parameter validation for `atan2` gpuweb/cts#3637.
-
Update petamoriken/float16 to v3.8.6 gpuweb/cts#3636.
-
shader/execution: Add more constructor tests gpuweb/cts#3631.
-
wgsl: Updated inverseSqrt builtin args validation gpuweb/cts#3635.
-
wgsl: Update trig builtins arg validation gpuweb/cts#3632.
-
Validation tests for builtin `extractBits` gpuweb/cts#3627.
-
Add missing f16 enables to my recent tests gpuweb/cts#3634.
-
wgsl: Update exp/exp2 builtins args validation gpuweb/cts#3633.
-
Add tests that builtin methods correctly detect being shadowed. gpuweb/cts#3366.
-
Validate unpack2x16snorm, unpack2x16unorm and unpack2x16float gpuweb/cts#3622.
-
Validate unpack4x8snorm and unpack4x8unorm gpuweb/cts#3619.
-
Validation tests for insertBits. gpuweb/cts#3592.
-
WGSL execution TextureSample tests gpuweb/cts#3578.
-
For .unimplemented() tests, skip before iterating subcases gpuweb/cts#3625.
-
Add validation tests for type aliases gpuweb/cts#3624.
-
Fix tests that use rAF when run in shared/service worker gpuweb/cts#3617.
-
Add validation tests for any, all, and select gpuweb/cts#3598.
-
wgsl: Add execution tests for derivative builtins gpuweb/cts#3459.
-
Add tests for implicit conversion and overload resolution gpuweb/cts#3614.
-
Fix wrong metadata in video resource meta lists gpuweb/cts#3605.
-
Fail tests in worker=* mode, when worker can't be created gpuweb/cts#3604.
-
Add smoothstep validation tests gpuweb/cts#3602.
-
wgsl: Add tests for operator precedence gpuweb/cts#3603.
-
Add div/rem binary operation validation tests gpuweb/cts#3595.
-
Compute service worker paths using import.meta gpuweb/cts#3599.
-
wgsl: Validation for distance builtin function gpuweb/cts#3585.
-
wgsl: Add i32 conversion execution tests for abstract inputs gpuweb/cts#3555.
-
WGSL validation test for textureGather/compare gpuweb/cts#3550.
-
wgsl: Add u32 conversion execution tests for abstract inputs gpuweb/cts#3553.
-
wgsl: Add f16 conversion execution tests for abstract inputs gpuweb/cts#3560.
-
Tests that context dependent names do not use name resolution gpuweb/cts#3600.
-
Better-preserve log messages sent from workers gpuweb/cts#3572.
-
Fix a few web_platform tests to run or skip on workers gpuweb/cts#3597.
-
wgsl: Validation for faceForward builtin function gpuweb/cts#3588.
-
wgsl: Validation of frexp builtin function gpuweb/cts#3590.
-
wgsl: Validation of fract builtin function gpuweb/cts#3587.
-
Add shader execution tests for call arg order gpuweb/cts#3566.
-
Add a test which prints what environment it runs in gpuweb/cts#3571.
-
Make GitHub hide diff on hashes.json by default gpuweb/cts#3593.
-
Add validation tests for *Bit builtin functions gpuweb/cts#3565.
-
Adds validation tests for count* builtin functions. gpuweb/cts#3498.
-
Validate pack4x8snorm gpuweb/cts#3589.
-
Add out of range add/sub/mul operator tests. gpuweb/cts#3584.
-
Finishing types:textures: external, depth, sampler gpuweb/cts#3582.
-
wgsl: Validation for reflect builtin function gpuweb/cts#3569.
-
Compat: Fix WGSL textureDimensions tests gpuweb/cts#3581.
-
Print video error message if video fails play gpuweb/cts#3580.
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated (take 2) gpuweb/cts#3570.
-
-
1 pull request closed.
-
Add texture type and return type tests to gpuweb/cts#3651.
-
-
76 pull requests received 465 new comments.
-
shader/validation: Begin parsing and type validation for statements gpuweb/cts#3645 (43 comments by @alan-baker, @ben-clayton, @dj2, @dneto0)
-
WGSL execution TextureSample tests gpuweb/cts#3578 (35 comments by @ben-clayton, @greggman, @kainino0x)
-
Add smoothstep validation tests gpuweb/cts#3602 (32 comments by @amaiorano, @ben-clayton, @kainino0x)
-
wgsl: Replace FPTraits with static data for AF addition execution tests gpuweb/cts#3568 (19 comments by @kainino0x, @zoddicus)
-
wgsl: Validation for refract builtin function gpuweb/cts#3626 (18 comments by @dneto0, @toji, @zoddicus)
-
WGSL validation test for textureGather/compare gpuweb/cts#3550 (17 comments by @alan-baker, @greggman)
-
Add validation tests for any, all, and select gpuweb/cts#3598 (16 comments by @amaiorano, @ben-clayton, @kainino0x)
-
wgsl: Add utility to help check for ambiguous overflows gpuweb/cts#3643 (15 comments by @dneto0, @toji, @zoddicus)
-
[wgsl] Add var initialization tests. gpuweb/cts#3693 (14 comments by @ben-clayton, @dj2, @dneto0, @zoddicus)
-
wgsl: Add validation tests for vector access of abstract types gpuweb/cts#3708 (13 comments by @ben-clayton, @zoddicus)
-
Fix texel data test gpuweb/cts#3675 (13 comments by @jiangzhaoming, @Jiawei-Shao, @kainino0x, @Kangz)
-
Add cts to cover copyEI2T copying from fully transparent image gpuweb/cts#3621 (9 comments by @kainino0x, @shaoboyan)
-
Test validation of increment and decrement statements. gpuweb/cts#3658 (9 comments by @dj2, @dneto0)
-
wgsl: Implement `transpose` validation tests gpuweb/cts#3694 (8 comments by @dneto0, @toji, @zoddicus)
-
Modernizes ceil/clamp tests to cover remaining types gpuweb/cts#3628 (7 comments by @dj2, @lokokung)
-
Add a test which prints what environment it runs in gpuweb/cts#3571 (7 comments by @greggman, @kainino0x)
-
wgsl: Refactor existing bitwise shift execution tests gpuweb/cts#3718 (6 comments by @ben-clayton, @zoddicus)
-
wgsl: Add conversions to abstract vector swizzle tests gpuweb/cts#3711 (6 comments by @ben-clayton, @zoddicus)
-
Improve validation tests for unpack builtin functions gpuweb/cts#3630 (5 comments by @ben-clayton, @haoxli)
-
Test that submit validates command buffers uniqueness gpuweb/cts#3699 (5 comments by @kainino0x, @lokokung, @toji)
-
shader/validation: Move `break if` expression type validation out of `parse` gpuweb/cts#3671 (5 comments by @ben-clayton, @dj2)
-
shader/execution: Fix abstract non_zero constructor tests gpuweb/cts#3677 (4 comments by @ben-clayton, @kainino0x)
-
wgsl: Migrate AF Subtraction to using lookup table gpuweb/cts#3667 (4 comments by @ben-clayton, @zoddicus)
-
Add return type and texture type WGSL textureSample tests gpuweb/cts#3640 (4 comments by @dj2, @greggman)
-
wgsl: Add execution tests for derivative builtins gpuweb/cts#3459 (4 comments by @ben-clayton, @jrprice)
-
Validation for address space and access mode in var decls gpuweb/cts#3695 (3 comments by @alan-baker, @jrprice)
-
Validation tests for duplicate attributes gpuweb/cts#3692 (3 comments by @alan-baker, @dj2, @jrprice)
-
shader/validation: Add compound assignment validation gpuweb/cts#3648 (3 comments by @ben-clayton, @Kangz)
-
Tests that context dependent names do not use name resolution gpuweb/cts#3600 (3 comments by @dneto0)
-
wgsl: Add execution tests for fwidth builtins gpuweb/cts#3686 (2 comments by @ben-clayton, @jrprice)
-
Validation tests for vector constructors from elements gpuweb/cts#3721 (2 comments by @alan-baker, @dj2)
-
Validate unpack2x16snorm, unpack2x16unorm and unpack2x16float gpuweb/cts#3622 (2 comments by @ben-clayton, @haoxli)
-
Add tests for implicit conversion and overload resolution gpuweb/cts#3614 (2 comments by @alan-baker, @ben-clayton)
-
wgsl: Validation for faceForward builtin function gpuweb/cts#3588 (2 comments by @ben-clayton, @toji)
-
Fix all the tests about resource limits on inter-stage shader variables gpuweb/cts#3697 (1 comment by @Kangz)
-
TR publications
-
9 documents published.
-
WebGPU Shading Language published 5 times as a Working Draft. History of WebGPU Shading Language
-
WebGPU published 4 times as a Working Draft. History of WebGPU
-
GitHub
gpuweb/gpuweb
-
29 issues opened.
-
texture_depth_2d needs more information gpuweb/gpuweb#4555, by @Alphapage.
-
Broken Chromium bug tracker link on Implementation Status wiki page gpuweb/gpuweb#4553, by @jjudd.
-
Clarify whether type checking occurs for short-circuited RHS expressions gpuweb/gpuweb#4551, by @jrprice.
-
Resurrect vertex formats dropped for macOS < 10.13 gpuweb/gpuweb#4549, by @kainino0x.
-
Reduce limits for the number of case selector values in a switch statement gpuweb/gpuweb#4548, by @jrprice.
-
Reduce limits for composite types gpuweb/gpuweb#4547, by @jrprice.
-
Reduce limit for maximum number of elements in const-expression of array type gpuweb/gpuweb#4546, by @jrprice.
-
Reduce byte-size limit for function and private address spaces gpuweb/gpuweb#4545, by @jrprice.
-
Numeric conversion rules wording is inconsistent gpuweb/gpuweb#4538, by @zoddicus.
-
Allow 3D Textures with 2D Compressed Formats gpuweb/gpuweb#4537, by @mehmetoguzderin.
-
Synchronously query GPUAdapterInfo gpuweb/gpuweb#4536, by @kainino0x.
-
Why no texel format for texture_2d, but f32, i32 or u32 gpuweb/gpuweb#4534, by @Alphapage.
-
GPUAdapter.limits seems unable to detect the true capabilities of the GPU gpuweb/gpuweb#4532, by @microaaron.
-
Function arguments cannot be modified? gpuweb/gpuweb#4531, by @microaaron.
-
WGSL built in functions are very under specified gpuweb/gpuweb#4527, by @starmole.
-
VideoFrame should use display size, not coded size gpuweb/gpuweb#4525, by @kainino0x.
-
Does writeTimestamp() belong to GPUCommandEncoder or GPURenderPassEncoder? gpuweb/gpuweb#4524, by @microaaron.
-
Accuracy for AbstractFloat `fract` give unintuitive/unexpected results gpuweb/gpuweb#4523, by @zoddicus.
-
Compat: Disallow vertex stride of 0 gpuweb/gpuweb#4515, by @Kangz.
-
Compat: Disallow bgra8unorm-srgb textures. gpuweb/gpuweb#4514, by @greggman.
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513, by @greggman.
-
Compat: Disallow textureLoad() of depth textures in WGSL via validation. gpuweb/gpuweb#4512, by @greggman.
-
Does `entryPoint` defaulting apply to `compilationHints`? gpuweb/gpuweb#4510, by @kainino0x.
-
Issues in "validating GPUDepthStencilState" algorithm gpuweb/gpuweb#4508, by @Kangz.
-
Will function overloading be supported in the future? gpuweb/gpuweb#4507, by @microaaron.
-
Should alphaToCoverageEnabled require a blendable format? gpuweb/gpuweb#4506, by @mwyrzykowski.
-
-
14 issues closed.
-
Allow 3D Textures with 2D Compressed Formats gpuweb/gpuweb#4537.
-
Function arguments cannot be modified? gpuweb/gpuweb#4531.
-
Investigation: multi-draw-indirect gpuweb/gpuweb#4349.
-
Why no texel format for texture_2d, but f32, i32 or u32 gpuweb/gpuweb#4534.
-
GPUAdapter.limits seems unable to detect the true capabilities of the GPU gpuweb/gpuweb#4532.
-
Does writeTimestamp() belong to GPUCommandEncoder or GPURenderPassEncoder? gpuweb/gpuweb#4524.
-
Compat: Validating out viewFormats / view reinterpretation gpuweb/gpuweb#4404.
-
Compat: Disallow vertex stride of 0 gpuweb/gpuweb#4515.
-
Luminance-clamping definition in the spec is not how it actually works gpuweb/gpuweb#4263.
-
Will function overloading be supported in the future? gpuweb/gpuweb#4507.
-
Compat: Validate out storage texture formats that are not supported in GLES gpuweb/gpuweb#4413.
-
Compat: Validating out `@interpolate(linear)` and `@interpolate(..., sample)` gpuweb/gpuweb#4419.
-
Should we constrain the location of user input-output stage variables WGSL gpuweb/gpuweb#1962.
-
Compat: can not copy multisample textures gpuweb/gpuweb#4468.
-
-
1 issue reopened.
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426, by @teoxoy.
-
-
49 issues received 148 new comments.
-
DrawIndirectCount gpuweb/gpuweb#1354 (21 comments by @alecazam, @cdiggins, @chingham, @Dampfwalze, @JesseRMeyer, @kainino0x, @lcodes, @logankaser, @mrshannon, @ringoz, @saydric)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (12 comments by @greggman, @kainino0x, @Kangz, @kdashg, @SenorBlanco)
-
Should alphaToCoverageEnabled require a blendable format? gpuweb/gpuweb#4506 (10 comments by @kainino0x, @kdashg, @mwyrzykowski, @toji)
-
Add support for 3D compressed textures. gpuweb/gpuweb#3183 (7 comments by @kainino0x, @mehmetoguzderin)
-
Use enum tokens when they are the only possible grammar usage. gpuweb/gpuweb#4505 (7 comments by @dneto0, @kdashg, @mehmetoguzderin)
-
Investigation : Push Constants gpuweb/gpuweb#75 (7 comments by @devshgraphicsprogramming, @greggman, @lcodes, @magcius, @shaoboyan)
-
Clarify whether type checking occurs for short-circuited RHS expressions gpuweb/gpuweb#4551 (6 comments by @alan-baker, @dj2, @dneto0, @jrprice)
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426 (6 comments by @austinEng, @kainino0x, @Kangz, @teoxoy)
-
Allow using a (float32-filterable 2d single level) texture view for an externalTexture binding. gpuweb/gpuweb#4504 (5 comments by @jimblandy, @Kangz, @kdashg)
-
Compat: Disallow `texture*()` of a `texture_depth_2d_array` with an offset gpuweb/gpuweb#4513 (4 comments by @kainino0x, @Kangz, @teoxoy)
-
Accuracy for AbstractFloat `fract` give unintuitive/unexpected results gpuweb/gpuweb#4523 (4 comments by @kdashg, @zoddicus)
-
mapSync on Workers - and possibly on the main thread gpuweb/gpuweb#2217 (3 comments by @juj, @Kangz)
-
Compat: Disallow bgra8unorm-srgb textures. gpuweb/gpuweb#4514 (3 comments by @greggman, @kainino0x, @Kangz)
-
Shader module compilation hint API is messy gpuweb/gpuweb#4338 (3 comments by @juj, @kainino0x, @Kangz)
-
VideoFrame should use display size, not coded size gpuweb/gpuweb#4525 (2 comments by @greggman, @kdashg)
-
Compat: validating out "2d" of more than 1 layer gpuweb/gpuweb#4378 (2 comments by @kainino0x, @Kangz)
-
Criteria for pushing the WebGPU and WGSL specification to Candidate Recommendation gpuweb/gpuweb#3107 (2 comments by @ibesora, @Kangz)
-
Does writeTimestamp() belong to GPUCommandEncoder or GPURenderPassEncoder? gpuweb/gpuweb#4524 (2 comments by @kainino0x, @microaaron)
-
Should we constrain the location of user input-output stage variables WGSL gpuweb/gpuweb#1962 (2 comments by @kainino0x, @kdashg)
-
Issues in "validating GPUDepthStencilState" algorithm gpuweb/gpuweb#4508 (2 comments by @kainino0x, @Kangz)
-
Compat: Disallow textureLoad() of depth textures in WGSL via validation. gpuweb/gpuweb#4512 (1 comment by @Kangz)
-
relax absolute error bound for asin f32: too tight for Intel f32 with DXC gpuweb/gpuweb#4477 (1 comment by @kdashg)
-
Does `entryPoint` defaulting apply to `compilationHints`? gpuweb/gpuweb#4510 (1 comment by @kainino0x)
-
Compat: Validate out storage texture formats that are not supported in GLES gpuweb/gpuweb#4413 (1 comment by @greggman)
-
Compat: Validating out `@interpolate(linear)` and `@interpolate(..., sample)` gpuweb/gpuweb#4419 (1 comment by @greggman)
-
Is it really feasible to covert between WGSL and SPIR-V seemlessly? gpuweb/gpuweb#620 (1 comment by @nevakrien)
-
-
18 pull requests opened.
-
Better Specialize Syntax, Fix Typos, and Improve Grammar Validation Script gpuweb/gpuweb#4554, by @mehmetoguzderin.
-
Make shader-creation errors possible for array sizes gpuweb/gpuweb#4552, by @alan-baker.
-
Add synchronous GPUAdapter.info gpuweb/gpuweb#4550, by @kainino0x.
-
Minor typos gpuweb/gpuweb#4544, by @greggman.
-
Fix some typos gpuweb/gpuweb#4543, by @greggman.
-
Compat: Disallow bgra8unorm-srgb textures gpuweb/gpuweb#4542, by @greggman.
-
wgsl: Add a carve out for AbstractFloat `fract` accuracy gpuweb/gpuweb#4541, by @zoddicus.
-
[editorial] Clarify that texture copy ranges are 'physical' gpuweb/gpuweb#4530, by @kainino0x.
-
[editorial] Clarify that override declarations must be at module scope gpuweb/gpuweb#4529, by @alan-baker.
-
Adjusts subgroup built-in function names to match GLSL gpuweb/gpuweb#4528, by @Lichtso.
-
Allow out-of-standard-gamut colors with float canvases gpuweb/gpuweb#4526, by @kainino0x.
-
Use y component in vec4 constructor gpuweb/gpuweb#4522, by @Gobbel2000.
-
Compat: Validate depthOrArrayLayers matches textureBindingViewDimension gpuweb/gpuweb#4520, by @greggman.
-
Subtract position from Vulkan inter stage limits gpuweb/gpuweb#4518, by @teoxoy.
-
Increase `maxInterStageShaderComponents` to 64 gpuweb/gpuweb#4517, by @teoxoy.
-
Add const_assert to statement behaviors gpuweb/gpuweb#4516, by @alan-baker.
-
Compat: Disallow texture format reinterpretation gpuweb/gpuweb#4511, by @greggman.
-
Lower maxInterStageShaderVariables to 15 gpuweb/gpuweb#4509, by @Jiawei-Shao.
-
-
13 pull requests merged.
-
Make shader-creation errors possible for array sizes gpuweb/gpuweb#4552.
-
Increase `maxInterStageShaderComponents` to 64 gpuweb/gpuweb#4517.
-
Compat: Disallow bgra8unorm-srgb textures gpuweb/gpuweb#4542.
-
Fix some typos gpuweb/gpuweb#4543.
-
Minor typos gpuweb/gpuweb#4544.
-
[editorial] Clarify that texture copy ranges are 'physical' gpuweb/gpuweb#4530.
-
[editorial] Clarify that override declarations must be at module scope gpuweb/gpuweb#4529.
-
Use y component in vec4 constructor gpuweb/gpuweb#4522.
-
Compat: Validate depthOrArrayLayers matches textureBindingViewDimension gpuweb/gpuweb#4520.
-
Add const_assert to statement behaviors gpuweb/gpuweb#4516.
-
Compat: Disallow texture format reinterpretation gpuweb/gpuweb#4511.
-
Clarify handling of out of range color values gpuweb/gpuweb#4499.
-
Fix the computation of inter-stage shader components gpuweb/gpuweb#4503.
-
-
1 pull request closed.
-
Lower maxInterStageShaderVariables to 15 gpuweb/gpuweb#4509.
-
-
1 pull request reopened.
-
Clarify behavior of out-of-gamut canvas gpuweb/gpuweb#4476, by @kainino0x.
-
-
16 pull requests received 39 new comments.
-
Clarify behavior of out-of-gamut canvas gpuweb/gpuweb#4476 (7 comments by @ccameron-chromium, @kainino0x)
-
Subtract position from Vulkan inter stage limits gpuweb/gpuweb#4518 (6 comments by @kainino0x, @teoxoy)
-
Allow out-of-standard-gamut colors with float canvases gpuweb/gpuweb#4526 (4 comments by @kainino0x)
-
Adjusts subgroup built-in function names to match GLSL gpuweb/gpuweb#4528 (3 comments by @alan-baker, @Lichtso)
-
Clarify handling of out of range color values gpuweb/gpuweb#4499 (3 comments by @kainino0x, @kdashg)
-
Require depthWriteEnabled and depthCompare only for formats with depth gpuweb/gpuweb#4318 (2 comments by @juj, @kainino0x)
-
Better Specialize Syntax, Fix Typos, and Improve Grammar Validation Script gpuweb/gpuweb#4554 (1 comment by @mehmetoguzderin)
-
gpuweb/cts
-
12 issues opened.
-
framework: add bgra8unorm to src/webgpu/shader/types.ts TexelFormats gpuweb/cts#3541, by @dneto0.
-
bitcast.bin gets updated when it seems like it shouldn't gpuweb/cts#3538, by @kainino0x.
-
Test using video from a webcam gpuweb/cts#3531, by @kainino0x.
-
`"const"` `f16` test regression gpuweb/cts#3526, by @ben-clayton.
-
Clarification on `src/webgpu/listing_meta.json` gpuweb/cts#3522, by @zoddicus.
-
Put optional_cts_websocket_logger behind an option gpuweb/cts#3504, by @kainino0x.
-
Move builtin must_use tests into per-builtin validation suites gpuweb/cts#3502, by @jrprice.
-
Add test for GPUCompilationMessage using UTF-16 gpuweb/cts#3489, by @stefnotch.
-
Restrict the number of AF matrix-matrix multiplication tests further gpuweb/cts#3483, by @zoddicus.
-
Investigate replacing CaseList with Case[] gpuweb/cts#3458, by @zoddicus.
-
wgsl: Apply case reduction, `selectNCases` to various AF Matrix tests gpuweb/cts#3457, by @zoddicus.
-
-
35 issues closed.
-
Clarification on `src/webgpu/listing_meta.json` gpuweb/cts#3522.
-
wgsl: AbstractFloat execution test for `step` builtin gpuweb/cts#2530.
-
wgsl: AbstractFloat execution tests for `frexp` gpuweb/cts#2588.
-
wgsl: Add abstract float/int tests to existing specs gpuweb/cts#1297.
-
Put optional_cts_websocket_logger behind an option gpuweb/cts#3504.
-
`"const"` `f16` test regression gpuweb/cts#3526.
-
wgsl: Add AbstractFloat entries to `kAllFloatsScalarsAndVectors` gpuweb/cts#3276.
-
wgsl: AbstractFloat execution tests for `remainder (%)` gpuweb/cts#2585.
-
wgsl: AbstractFloat execution test for `tan` builtin gpuweb/cts#2506.
-
wgsl : AbstractFloat execution test for `tanh` builtin. gpuweb/cts#2528.
-
wgsl: AbstractFloat execution test for `sqrt` builtin gpuweb/cts#2531.
-
wgsl: AbstractFloat execution test for `smoothstep` builtin gpuweb/cts#2534.
-
wgsl: AbstractFloat execution test for `sinh` builtin gpuweb/cts#2536.
-
wgsl: AbstractFloat execution tests for `cosh` gpuweb/cts#2539.
-
wgsl: AbstractFloat execution tests for `exp` builtin gpuweb/cts#2542.
-
wgsl: AbstractFloat execution tests for `exp2` builtin gpuweb/cts#2545.
-
wgsl: AbstractFloat execution tests for `faceForward` builtin gpuweb/cts#2546.
-
wgsl: AbstractFloat execution tests for `acos` builtin gpuweb/cts#2550.
-
wgsl: AbstractFloat execution tests for `asin` builtin gpuweb/cts#2552.
-
wgsl: AbstractFloat execution tests for `length` builtin gpuweb/cts#2554.
-
wgsl: AbstractFloat execution tests for `distance` builtin gpuweb/cts#2556.
-
wgsl: AbstractFloat execution tests for `acosh` builtin gpuweb/cts#2558.
-
wgsl: AbstractFloat execution tests for `asinh` builtin gpuweb/cts#2560.
-
wgsl: AbstractFloat execution tests for `atanh` builtin gpuweb/cts#2562.
-
wgsl: AbstractFloat execution tests for `normalize` builtin gpuweb/cts#2570.
-
wgsl: AbstractFloat execution tests for `pow` builtin gpuweb/cts#2572.
-
wgsl: AbstractFloat execution tests for `reflect` builtin gpuweb/cts#2578.
-
wgsl: AbstractFloat execution tests for `refract` builtin gpuweb/cts#2581.
-
Restrict the number of AF matrix-matrix multiplication tests further gpuweb/cts#3483.
-
wgsl: AbstractFloat execution test for `determinant` builtin gpuweb/cts#2523.
-
Investigate replacing CaseList with Case[] gpuweb/cts#3458.
-
wgsl: validate out of range const-eval for `asin` gpuweb/cts#1965.
-
wgsl: validate out of range const eval for `atanh` gpuweb/cts#1966.
-
wgsl: validate out of range const-eval for acos gpuweb/cts#1964.
-
-
15 issues received 32 new comments.
-
Clarification on `src/webgpu/listing_meta.json` gpuweb/cts#3522 (7 comments by @dneto0, @kainino0x, @mwyrzykowski, @zoddicus)
-
Add test for GPUCompilationMessage using UTF-16 gpuweb/cts#3489 (6 comments by @kainino0x, @stefnotch)
-
Restrict the number of AF matrix-matrix multiplication tests further gpuweb/cts#3483 (2 comments by @zoddicus)
-
Investigate handling of very small (near zero) AbstractFloats and ECMAScript `numbers` gpuweb/cts#2993 (1 comment by @zoddicus)
-
`webgpu:shader,execution,expression,unary,address_of_and_indirection:deref` for `f16` are failing gpuweb/cts#3514 (1 comment by @dneto0)
-
Per-aspect capability data for multi-aspect texture formats gpuweb/cts#2495 (1 comment by @kainino0x)
-
-
115 pull requests opened.
-
wgsl: Validation of fma builtin function gpuweb/cts#3577, by @toji.
-
wgsl: Validation for step builtin function gpuweb/cts#3576, by @toji.
-
wgsl: Validation for trunc builtin function gpuweb/cts#3575, by @toji.
-
wgsl: Update args validation for sign builtin gpuweb/cts#3574, by @toji.
-
wgsl: Validation for mix builtin function gpuweb/cts#3573, by @toji.
-
Better-preserve log messages sent from workers gpuweb/cts#3572, by @kainino0x.
-
Add a test which prints what environment it runs in gpuweb/cts#3571, by @kainino0x.
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated (take 2) gpuweb/cts#3570, by @kainino0x.
-
wgsl: Validation for reflect builtin function gpuweb/cts#3569, by @toji.
-
wgsl: Replace FPTraits with static data for AF addition execution tests gpuweb/cts#3568, by @zoddicus.
-
wgsl: Add short-circuiting validation tests gpuweb/cts#3567, by @jrprice.
-
Add shader execution tests for call arg order gpuweb/cts#3566, by @alan-baker.
-
Add validation tests for *Bit builtin functions gpuweb/cts#3565, by @lokokung.
-
Add some logging/asserts to mapping tests gpuweb/cts#3564, by @kainino0x.
-
Add parameter validation for `atanh` (#3559) gpuweb/cts#3563, by @dj2.
-
Add parameter validation for `atan` gpuweb/cts#3562, by @dj2.
-
wgsl: Implement `bitcast` with AbstractInt inputs gpuweb/cts#3561, by @zoddicus.
-
wgsl: Add f16 conversion execution tests for abstract inputs gpuweb/cts#3560, by @zoddicus.
-
Add parameter validation for `asinh` gpuweb/cts#3559, by @dj2.
-
Add validation test for asin invalid parameters gpuweb/cts#3558, by @dj2.
-
Add invalid function parameter validation. gpuweb/cts#3557, by @dj2.
-
WGSL validation test for textureDimensions gpuweb/cts#3556, by @greggman.
-
wgsl: Add i32 conversion execution tests for abstract inputs gpuweb/cts#3555, by @zoddicus.
-
Add `acosh` validation tests. gpuweb/cts#3554, by @dj2.
-
wgsl: Add u32 conversion execution tests for abstract inputs gpuweb/cts#3553, by @zoddicus.
-
Add validation testing for `acos` parameters. gpuweb/cts#3552, by @dj2.
-
wgsl: Cleanup up .unimplemented()s for numeric execution tests gpuweb/cts#3551, by @zoddicus.
-
WGSL validation test for textureGather/compare gpuweb/cts#3550, by @greggman.
-
wgsl: Validation for pow builtin function gpuweb/cts#3549, by @toji.
-
Don't validate or warn about listing_meta.json contents gpuweb/cts#3548, by @kainino0x.
-
Let-declaration validation gpuweb/cts#3547, by @alan-baker.
-
Remaining validation tests for const declarations gpuweb/cts#3546, by @alan-baker.
-
shader/execution: New access and constructor tests gpuweb/cts#3545, by @ben-clayton.
-
Add external_texture test. gpuweb/cts#3544, by @dj2.
-
wgsl: Validation for quantizeToF16 builting function gpuweb/cts#3543, by @toji.
-
wgsl: Validation tests for cross builtin function gpuweb/cts#3542, by @toji.
-
Remaining validation tests for overrides gpuweb/cts#3540, by @alan-baker.
-
Move more tests into subcases gpuweb/cts#3539, by @ben-clayton.
-
Move 'debug' option to globalTestConfig gpuweb/cts#3537, by @kainino0x.
-
Disable logToWebSocket by default gpuweb/cts#3536, by @kainino0x.
-
Array validation tests gpuweb/cts#3535, by @alan-baker.
-
wgsl: Implement AbstractFloat `ldexp` execution tests gpuweb/cts#3534, by @zoddicus.
-
wgsl: Implement `determinant` validation tests gpuweb/cts#3533, by @zoddicus.
-
Validation tests for matrix types gpuweb/cts#3532, by @alan-baker.
-
wgsl: Implement AbstractFloat `fract` execution tests gpuweb/cts#3530, by @zoddicus.
-
Revert "Restore ?worker and ?worker=1 to mean ?worker=dedicated (#3485)" gpuweb/cts#3529, by @ben-clayton.
-
WGSL validation tests for textureLoad and textureStore gpuweb/cts#3528, by @greggman.
-
wgsl: Implement AbstractFloat `frexp` execution tests gpuweb/cts#3527, by @zoddicus.
-
wgsl: Implement AbstractFloat `step` execution tests gpuweb/cts#3525, by @zoddicus.
-
Fix depth_clamp_and_clip test gpuweb/cts#3524, by @kainino0x.
-
Revert #3469 and #3466 gpuweb/cts#3523, by @ben-clayton.
-
Fix intra-invocation coherence 3d tests gpuweb/cts#3521, by @alan-baker.
-
Fix error dst color space in copyToTexture,video,* gpuweb/cts#3520, by @shaoboyan.
-
WGSL validation test for textureSampleBaseClampToEdge gpuweb/cts#3519, by @greggman.
-
Include offset variants for textureSample and textureSampleCompare gpuweb/cts#3518, by @greggman.
-
WGSL Validation test for textureSampleLevel gpuweb/cts#3517, by @greggman.
-
Test validation of arg numbers and types, must_use for degrees, radia… gpuweb/cts#3516, by @dneto0.
-
Fix `f16()` `F16Value.value` quantization gpuweb/cts#3515, by @ben-clayton.
-
Add a test for the error line position gpuweb/cts#3513, by @stefnotch.
-
WGSL validation tests for textureSampleGrad gpuweb/cts#3512, by @greggman.
-
Cleanup texture builtin WGSL validation tests gpuweb/cts#3511, by @greggman.
-
WGSL Validation test for textureSampleCompareLevel gpuweb/cts#3510, by @greggman.
-
wgsl: Add normalize builtin validation tests gpuweb/cts#3509, by @toji.
-
wgsl: Add execution tests for remaining AF builtins that should defer to f32 gpuweb/cts#3508, by @zoddicus.
-
Update types,textures and bgra8unorm_storage tests gpuweb/cts#3507, by @shrekshao.
-
Add validation tests for statement behavior analysis gpuweb/cts#3506, by @alan-baker.
-
Texture sample compare fix gpuweb/cts#3505, by @greggman.
-
Improve 'args' and 'must_use' validation tests for several builtins gpuweb/cts#3503, by @dneto0.
-
op: Add depthSlice tests in rendering/3d_texture_slices.spec.ts gpuweb/cts#3501, by @haoxli.
-
WGSL Validation tests for textureSampleCompare gpuweb/cts#3500, by @greggman.
-
WGSL validation tests for textureSampleBias gpuweb/cts#3499, by @greggman.
-
Adds validation tests for count* builtin functions. gpuweb/cts#3498, by @lokokung.
-
wgsl: Add validation test for min/max builtin functions gpuweb/cts#3497, by @toji.
-
wgsl: validate bad arg counts, arg types, must_use for sin, cos gpuweb/cts#3496, by @dneto0.
-
wgsl: validate bad arg counts, arg types, must_use for log, log2 gpuweb/cts#3495, by @dneto0.
-
wgsl: Add more validation tests for bitwise shift gpuweb/cts#3494, by @jrprice.
-
wgsl: validate floor negative test cases gpuweb/cts#3492, by @dneto0.
-
wgsl: valdate dot4I8Packed, dot4U8Packed on bool and bool vec gpuweb/cts#3491, by @dneto0.
-
shader/execution: Fix struct padding for `f16`s gpuweb/cts#3488, by @ben-clayton.
-
Remove ULP & Absolute Error support for AF gpuweb/cts#3487, by @zoddicus.
-
Further restrict AF matrix-matrix multiplication cases gpuweb/cts#3486, by @zoddicus.
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated gpuweb/cts#3485, by @kainino0x.
-
shader/execution: Fix composite abstract runtime indexing gpuweb/cts#3484, by @ben-clayton.
-
wgsl: Update value creation and type comparisons gpuweb/cts#3482, by @jrprice.
-
WGSL: Add binary + / - / * validation tests but for matrices gpuweb/cts#3481, by @Kangz.
-
WGSL: Add unary arithmetic negation validation tests. gpuweb/cts#3480, by @Kangz.
-
Re-add `gen_listings.ts` gpuweb/cts#3479, by @ben-clayton.
-
Test `textureSample` WGSL validation gpuweb/cts#3478, by @greggman.
-
wgsl: Add bitwise complement validation tests gpuweb/cts#3477, by @jrprice.
-
wgsl: Add and/or/xor validation tests gpuweb/cts#3476, by @jrprice.
-
shader/execution: Implement vector component selection tests gpuweb/cts#3475, by @ben-clayton.
-
wgsl: Implement AbstractFloat `determinant` execution tests gpuweb/cts#3474, by @zoddicus.
-
wgsl: Add logical negation validation tests gpuweb/cts#3473, by @jrprice.
-
Type texture gpuweb/cts#3472, by @shrekshao.
-
wgsl: Add comparison operator validation tests gpuweb/cts#3471, by @jrprice.
-
Rename kAllScalarsAndVectors to kAllNumericScalarsAndVectors gpuweb/cts#3470, by @jrprice.
-
Fix device_pool refactor gpuweb/cts#3469, by @greggman.
-
Begin array indexing tests gpuweb/cts#3468, by @ben-clayton.
-
Discard shader execution tests gpuweb/cts#3467, by @alan-baker.
-
Add AdapterLimitsGPUTest that sets all limits to the adapter's gpuweb/cts#3466, by @greggman.
-
wgsl: Use selectNCases to restrict the number of AF Matrix cases gpuweb/cts#3465, by @zoddicus.
-
Update All and Concrete type names gpuweb/cts#3464, by @dj2.
-
wgsl: Replace CaseList with Case[] gpuweb/cts#3463, by @zoddicus.
-
Expression framework refactoring gpuweb/cts#3462, by @ben-clayton.
-
Simplify some type constants gpuweb/cts#3461, by @dj2.
-
wgsl: Enable Absolute Error intervals for AbstractFloat gpuweb/cts#3460, by @zoddicus.
-
wgsl: Add execution tests for derivative builtins gpuweb/cts#3459, by @jrprice.
-
Add loop control flow tests with logical operators gpuweb/cts#3456, by @alan-baker.
-
Add `abs` parameter validation. gpuweb/cts#3455, by @dj2.
-
User IO passthrough shader tests gpuweb/cts#3454, by @alan-baker.
-
WGSL: Add validation tests for readonly_and_readwrite_storage_textures. gpuweb/cts#3453, by @Kangz.
-
Shader validation tests for atomics gpuweb/cts#3452, by @alan-baker.
-
GPUExternalTexture: Add cts to cover non-YUV format video frame gpuweb/cts#3451, by @shaoboyan.
-
-
103 pull requests merged.
-
wgsl: Validation for step builtin function gpuweb/cts#3576.
-
wgsl: Validation for trunc builtin function gpuweb/cts#3575.
-
wgsl: Update args validation for sign builtin gpuweb/cts#3574.
-
wgsl: Validation for quantizeToF16 builting function gpuweb/cts#3543.
-
wgsl: Cleanup up .unimplemented()s for numeric execution tests gpuweb/cts#3551.
-
wgsl: Implement `bitcast` with AbstractInt inputs gpuweb/cts#3561.
-
wgsl: Validation tests for cross builtin function gpuweb/cts#3542.
-
Add some logging/asserts to mapping tests gpuweb/cts#3564.
-
Add parameter validation for `atanh` (#3559) gpuweb/cts#3563.
-
Add parameter validation for `atan` gpuweb/cts#3562.
-
shader/execution: New access and constructor tests gpuweb/cts#3545.
-
Add a test for the error line position gpuweb/cts#3513.
-
Add parameter validation for `asinh` gpuweb/cts#3559.
-
Move 'debug' option to globalTestConfig gpuweb/cts#3537.
-
Add validation test for asin invalid parameters gpuweb/cts#3558.
-
Add invalid function parameter validation. gpuweb/cts#3557.
-
Add `acosh` validation tests. gpuweb/cts#3554.
-
Add validation testing for `acos` parameters. gpuweb/cts#3552.
-
Don't validate or warn about listing_meta.json contents gpuweb/cts#3548.
-
Let-declaration validation gpuweb/cts#3547.
-
Remaining validation tests for const declarations gpuweb/cts#3546.
-
op: Add depthSlice tests in rendering/3d_texture_slices.spec.ts gpuweb/cts#3501.
-
Add external_texture test. gpuweb/cts#3544.
-
add f16 to webgpu/shader/types.ts utilities gpuweb/cts#3407.
-
wgsl: Implement AbstractFloat `step` execution tests gpuweb/cts#3525.
-
wgsl: Implement AbstractFloat `frexp` execution tests gpuweb/cts#3527.
-
wgsl: Implement AbstractFloat `fract` execution tests gpuweb/cts#3530.
-
wgsl: Implement AbstractFloat `ldexp` execution tests gpuweb/cts#3534.
-
wgsl: Implement `determinant` validation tests gpuweb/cts#3533.
-
Move more tests into subcases gpuweb/cts#3539.
-
Remaining validation tests for overrides gpuweb/cts#3540.
-
Add loop control flow tests with logical operators gpuweb/cts#3456.
-
Disable logToWebSocket by default gpuweb/cts#3536.
-
Array validation tests gpuweb/cts#3535.
-
Validation tests for matrix types gpuweb/cts#3532.
-
WGSL validation tests for textureLoad and textureStore gpuweb/cts#3528.
-
Fix error dst color space in copyToTexture,video,* gpuweb/cts#3520.
-
wgsl: Add normalize builtin validation tests gpuweb/cts#3509.
-
Fix depth_clamp_and_clip test gpuweb/cts#3524.
-
Revert #3469 and #3466 gpuweb/cts#3523.
-
Fix intra-invocation coherence 3d tests gpuweb/cts#3521.
-
WGSL validation test for textureSampleBaseClampToEdge gpuweb/cts#3519.
-
WGSL Validation test for textureSampleLevel gpuweb/cts#3517.
-
wgsl: Add validation test for min/max builtin functions gpuweb/cts#3497.
-
WGSL validation tests for textureSampleBias gpuweb/cts#3499.
-
Include offset variants for textureSample and textureSampleCompare gpuweb/cts#3518.
-
Fix `f16()` `F16Value.value` quantization gpuweb/cts#3515.
-
WGSL: Add binary + / - / * validation tests but for matrices gpuweb/cts#3481.
-
WGSL: Add unary arithmetic negation validation tests. gpuweb/cts#3480.
-
WGSL Validation test for textureSampleCompareLevel gpuweb/cts#3510.
-
Add validation tests for statement behavior analysis gpuweb/cts#3506.
-
Improve 'args' and 'must_use' validation tests for several builtins gpuweb/cts#3503.
-
WGSL validation tests for textureSampleGrad gpuweb/cts#3512.
-
Update types,textures and bgra8unorm_storage tests gpuweb/cts#3507.
-
wgsl: validate bad arg counts, arg types, must_use for log, log2 gpuweb/cts#3495.
-
Cleanup texture builtin WGSL validation tests gpuweb/cts#3511.
-
wgsl: validate bad arg counts, arg types, must_use for sin, cos gpuweb/cts#3496.
-
Texture sample compare fix gpuweb/cts#3505.
-
WGSL Validation tests for textureSampleCompare gpuweb/cts#3500.
-
Discard shader execution tests gpuweb/cts#3467.
-
wgsl: Add more validation tests for bitwise shift gpuweb/cts#3494.
-
wgsl: validate floor negative test cases gpuweb/cts#3492.
-
Type texture gpuweb/cts#3472.
-
wgsl: valdate dot4I8Packed, dot4U8Packed on bool and bool vec gpuweb/cts#3491.
-
Remove ULP & Absolute Error support for AF gpuweb/cts#3487.
-
shader/execution: Fix struct padding for `f16`s gpuweb/cts#3488.
-
Test `textureSample` WGSL validation gpuweb/cts#3478.
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated gpuweb/cts#3485.
-
shader/execution: Fix composite abstract runtime indexing gpuweb/cts#3484.
-
Further restrict AF matrix-matrix multiplication cases gpuweb/cts#3486.
-
wgsl: Implement AbstractFloat `determinant` execution tests gpuweb/cts#3474.
-
wgsl: Update value creation and type comparisons gpuweb/cts#3482.
-
wgsl: Add bitwise complement validation tests gpuweb/cts#3477.
-
Re-add `gen_listings.ts` gpuweb/cts#3479.
-
wgsl: Add and/or/xor validation tests gpuweb/cts#3476.
-
wgsl: Add logical negation validation tests gpuweb/cts#3473.
-
shader/execution: Implement vector component selection tests gpuweb/cts#3475.
-
Add service worker support gpuweb/cts#3419.
-
Begin array indexing tests gpuweb/cts#3468.
-
wgsl: Add comparison operator validation tests gpuweb/cts#3471.
-
wgsl: Enable Absolute Error intervals for AbstractFloat gpuweb/cts#3460.
-
wgsl: Implement AbstractFloat `ceil` execution tests gpuweb/cts#3449.
-
wgsl: Implement AbstractInt bitwise-and execution tests gpuweb/cts#3436.
-
wgsl: Use selectNCases to restrict the number of AF Matrix cases gpuweb/cts#3465.
-
Fix device_pool refactor gpuweb/cts#3469.
-
Rename kAllScalarsAndVectors to kAllNumericScalarsAndVectors gpuweb/cts#3470.
-
Shader validation tests for atomics gpuweb/cts#3452.
-
Add AdapterLimitsGPUTest that sets all limits to the adapter's gpuweb/cts#3466.
-
Expression framework refactoring gpuweb/cts#3462.
-
Update All and Concrete type names gpuweb/cts#3464.
-
Simplify some type constants gpuweb/cts#3461.
-
wgsl: Replace CaseList with Case[] gpuweb/cts#3463.
-
User IO passthrough shader tests gpuweb/cts#3454.
-
wgsl: Implement AbstractFloat matrix multiplcation tests gpuweb/cts#3446.
-
Add `abs` parameter validation. gpuweb/cts#3455.
-
Tests for scoping diagnostics gpuweb/cts#3450.
-
GPUExternalTexture: Add cts to cover non-YUV format video frame gpuweb/cts#3451.
-
-
2 pull requests closed.
-
Generate Web Worker script for every test file gpuweb/cts#3415.
-
Compat: relax precision of rgba8unorm/snorm storage texture test gpuweb/cts#3448.
-
-
64 pull requests received 328 new comments.
-
op: Add depthSlice tests in rendering/3d_texture_slices.spec.ts gpuweb/cts#3501 (18 comments by @haoxli, @kainino0x)
-
shader/execution: New access and constructor tests gpuweb/cts#3545 (15 comments by @ben-clayton, @dneto0, @zoddicus)
-
WGSL validation tests for textureSampleBias gpuweb/cts#3499 (10 comments by @ben-clayton, @greggman)
-
Add validation tests for *Bit builtin functions gpuweb/cts#3565 (9 comments by @dj2, @dneto0, @lokokung)
-
wgsl: Validation for quantizeToF16 builting function gpuweb/cts#3543 (9 comments by @dneto0, @toji, @zoddicus)
-
wgsl: Implement AbstractFloat matrix multiplcation tests gpuweb/cts#3446 (8 comments by @ben-clayton, @zoddicus)
-
wgsl: Add execution tests for remaining AF builtins that should defer to f32 gpuweb/cts#3508 (7 comments by @ben-clayton, @zoddicus)
-
wgsl: Replace FPTraits with static data for AF addition execution tests gpuweb/cts#3568 (6 comments by @dneto0)
-
WGSL: Add binary + / - / * validation tests but for matrices gpuweb/cts#3481 (6 comments by @jrprice, @Kangz)
-
wgsl: Add short-circuiting validation tests gpuweb/cts#3567 (5 comments by @alan-baker, @dneto0, @jrprice)
-
wgsl: Cleanup up .unimplemented()s for numeric execution tests gpuweb/cts#3551 (5 comments by @dj2, @zoddicus)
-
WGSL Validation tests for textureSampleCompare gpuweb/cts#3500 (5 comments by @ben-clayton, @greggman)
-
Adds validation tests for count* builtin functions. gpuweb/cts#3498 (4 comments by @dneto0, @lokokung)
-
wgsl: Add validation test for min/max builtin functions gpuweb/cts#3497 (4 comments by @dneto0, @toji)
-
bytesPerBlock refactors; generalize validation,image_copy,buffer_related over aspects gpuweb/cts#3490 (4 comments by @kainino0x, @shrekshao)
-
Don't validate or warn about listing_meta.json contents gpuweb/cts#3548 (3 comments by @dneto0, @kainino0x)
-
Revert "Restore ?worker and ?worker=1 to mean ?worker=dedicated (#3485)" gpuweb/cts#3529 (3 comments by @kainino0x)
-
Test validation of arg numbers and types, must_use for degrees, radia… gpuweb/cts#3516 (3 comments by @ben-clayton, @dneto0)
-
wgsl: validate bad arg counts, arg types, must_use for sin, cos gpuweb/cts#3496 (3 comments by @ben-clayton, @dneto0)
-
wgsl: Implement AbstractFloat `step` execution tests gpuweb/cts#3525 (2 comments by @dneto0, @zoddicus)
-
wgsl: Convert all AbstractFloat tests with non-simple errors to using f32 intervals gpuweb/cts#3493 (2 comments by @dneto0, @zoddicus)
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated gpuweb/cts#3485 (2 comments by @beaufortfrancois)
-
wgsl: Add bitwise complement validation tests gpuweb/cts#3477 (2 comments by @ben-clayton, @jrprice)
-
Add AdapterLimitsGPUTest that sets all limits to the adapter's gpuweb/cts#3466 (2 comments by @ben-clayton, @greggman)
-
Restore ?worker and ?worker=1 to mean ?worker=dedicated (take 2) gpuweb/cts#3570 (1 comment by @kainino0x)
-
Compat: relax precision of rgba8unorm/snorm storage texture test gpuweb/cts#3448 (1 comment by @greggman)
-
wgsl: Add i32 conversion execution tests for abstract inputs gpuweb/cts#3555 (1 comment by @zoddicus)
-
wgsl: Add u32 conversion execution tests for abstract inputs gpuweb/cts#3553 (1 comment by @zoddicus)
-
Clean up deprecated .copyDst/.renderable/.renderTarget* from kTextureFormatInfo gpuweb/cts#3412 (1 comment by @kainino0x)
-
GPUExternalTexture: Add cts to cover non-YUV format video frame gpuweb/cts#3451 (1 comment by @shaoboyan)
TR publications
-
8 documents published.
-
WebGPU Shading Language published 5 times as a Working Draft. History of WebGPU Shading Language
-
WebGPU published 3 times as a Working Draft. History of WebGPU
-
Participants
-
1 participant left.
-
Van Nguyen (Vidbase, Inc) left.
-
-
1 organization left.
-
Vidbase, Inc left.
-
GitHub
gpuweb/cts
-
8 issues opened.
-
wgsl: relax asin(f32) absolute error to 6.81* 10^(-5) gpuweb/cts#3437, by @dneto0.
-
Add coverage for optional `GPUProgrammableStage.entryPoint` gpuweb/cts#3432, by @ErichDonGubler.
-
add f16 to kScalarTypeInfo in src/webgpu/shader/types.ts gpuweb/cts#3405, by @dneto0.
-
maxBindGroupsPlusVertexBuffers.spec.ts has invalid tests? gpuweb/cts#3376, by @mwyrzykowski.
-
Add component count of each texture format in format_info.js gpuweb/cts#3375, by @Jiawei-Shao.
-
Make much of src/webgpu/api/operation use max limits gpuweb/cts#3363, by @greggman.
-
-
12 issues closed.
-
wgsl: Comparison expression tests gpuweb/cts#1627.
-
wgsl: AbstractFloat execution test for `dot` builtin gpuweb/cts#2512.
-
wgsl: u32 execution test for `dot` builtin gpuweb/cts#2515.
-
wgsl: i32 execution test for `dot` builtin gpuweb/cts#2516.
-
wgsl: AbstractInt execution test for `dot` builtin gpuweb/cts#2514.
-
wgsl: relax asin(f32) absolute error to 6.81* 10^(-5) gpuweb/cts#3437.
-
support read-only, read-write storage textures gpuweb/cts#3078.
-
maxBindGroupsPlusVertexBuffers.spec.ts has invalid tests? gpuweb/cts#3376.
-
wgsl: Add AbstractInt `select` execution tests gpuweb/cts#3173.
-
-
8 issues received 20 new comments.
-
api,validation,capability_checks,limits,maxBindGroups:createPipelineLayout test has some invalid expectations gpuweb/cts#3346 (6 comments by @greggman, @kainino0x, @mwyrzykowski)
-
Make much of src/webgpu/api/operation use max limits gpuweb/cts#3363 (3 comments by @greggman, @kainino0x, @mwyrzykowski)
-
Two api,validation,gpu_external_texture_expiration tests fail due to issue PR 3312 gpuweb/cts#3364 (3 comments by @mwyrzykowski, @shaoboyan)
-
maxBindGroupsPlusVertexBuffers.spec.ts has invalid tests? gpuweb/cts#3376 (2 comments by @greggman, @mwyrzykowski)
-
Reduce unnecessary logging for validation tests gpuweb/cts#3165 (2 comments by @kainino0x, @zoddicus)
-
val: Test that two 'auto' BGLs at different group indices on the same pipeline are compatible gpuweb/cts#3388 (1 comment by @greggman)
-
Running webgpu:web_platform,external_texture,video:* multiple times hangs browser gpuweb/cts#3301 (1 comment by @mwyrzykowski)
-
-
86 pull requests opened.
-
Tests for scoping diagnostics gpuweb/cts#3450, by @alan-baker.
-
wgsl: Implement AbstractFloat `ceil` execution tests gpuweb/cts#3449, by @zoddicus.
-
Compat: relax precision of rgba8unorm/snorm storage texture test gpuweb/cts#3448, by @greggman.
-
Compat: Texture Intra Invocation Coherence gpuweb/cts#3447, by @greggman.
-
wgsl: Implement AbstractFloat matrix multiplcation tests gpuweb/cts#3446, by @zoddicus.
-
Compat: Skip workgroupUniformLoad tests > max invocations gpuweb/cts#3445, by @greggman.
-
Remove unneeded language feature skip gpuweb/cts#3444, by @jrprice.
-
wgsl: Implement AbstractFloat `dot` execution tests gpuweb/cts#3443, by @zoddicus.
-
Re-add temporarily TestWorker which aliases to TestDedicatedWorker gpuweb/cts#3442, by @beaufortfrancois.
-
Validation tests for resource compatibility of render pipelines gpuweb/cts#3441, by @alan-baker.
-
wgsl: Implement AbstractInt complement (~) execution tests gpuweb/cts#3440, by @zoddicus.
-
wgsl: Implement AbstractInt bitwise-xor execution tests gpuweb/cts#3439, by @zoddicus.
-
Relax asin(f32) absolute error to 6.81x10^(-5) gpuweb/cts#3438, by @dneto0.
-
wgsl: Implement AbstractInt bitwise-and execution tests gpuweb/cts#3436, by @zoddicus.
-
Reapply "Add shared worker support (#3345)" (#3428) gpuweb/cts#3435, by @dj2.
-
[wgsl] Validate continue inside switch control flow. gpuweb/cts#3434, by @dj2.
-
wgsl: Add validation tests for derivative builtins gpuweb/cts#3433, by @jrprice.
-
Resource compatibility validation gpuweb/cts#3431, by @alan-baker.
-
WIP: Implement AbstractInt bitwise execution tests gpuweb/cts#3430, by @zoddicus.
-
wgsl: Add workgroupUniformLoad validation tests gpuweb/cts#3429, by @jrprice.
-
Revert "Add shared worker support (#3345)" gpuweb/cts#3428, by @dj2.
-
wgsl: Add workgroupUniformLoad execution tests gpuweb/cts#3427, by @jrprice.
-
Validate address space layout constraints gpuweb/cts#3426, by @alan-baker.
-
Add compat mode testing to the PR checklist gpuweb/cts#3425, by @greggman.
-
[shader] Add tests for ptr parameters, used by builtins. gpuweb/cts#3424, by @ben-clayton.
-
Add timing information to server CTS runner gpuweb/cts#3423, by @ben-clayton.
-
op: 1D and 3D read-only storage textures gpuweb/cts#3422, by @Jiawei-Shao.
-
Compat: Skip new storage texture rg32 tests gpuweb/cts#3421, by @greggman.
-
op: read-only storage textures in vertex shader gpuweb/cts#3420, by @Jiawei-Shao.
-
Add service worker support gpuweb/cts#3419, by @beaufortfrancois.
-
op: read-only storage textures in fragment shader gpuweb/cts#3418, by @Jiawei-Shao.
-
op: bgra8unorm as read-only storage texture format gpuweb/cts#3417, by @Jiawei-Shao.
-
Add textureBarrier tests gpuweb/cts#3416, by @alan-baker.
-
Generate Web Worker script for every test file gpuweb/cts#3415, by @kainino0x.
-
Add recommended extensions to code-workspace gpuweb/cts#3414, by @amaiorano.
-
[shader/execution] Fixes for `textureDimensions()` gpuweb/cts#3413, by @ben-clayton.
-
wgsl: Implement `i32` & `u32` dot execution tests gpuweb/cts#3411, by @zoddicus.
-
Validation tests for unary address-of and indirection gpuweb/cts#3410, by @amaiorano.
-
fix comment in fragment_builtins.spec.ts gpuweb/cts#3409, by @greggman.
-
Just cleaning up a comment in fragment_builtins.spec.ts gpuweb/cts#3408, by @greggman.
-
add f16 to webgpu/shader/types.ts utilities gpuweb/cts#3407, by @dneto0.
-
wgsl: Implement AbstractInt `dot` execution tests gpuweb/cts#3406, by @zoddicus.
-
Test fragment_input @builtin(sample_mask) gpuweb/cts#3404, by @greggman.
-
Add force fallback adapter option gpuweb/cts#3403, by @greggman.
-
Tests for intra-invocation coherence of storage textures gpuweb/cts#3402, by @alan-baker.
-
wgsl: Implement AbstractInt comparison execution tests gpuweb/cts#3401, by @zoddicus.
-
wgsl: Implement AbstractInt remainder execution tests gpuweb/cts#3400, by @zoddicus.
-
wgsl: Implement AbstractInt division execution tests gpuweb/cts#3399, by @zoddicus.
-
wgsl: Implement AbstractInt multiplication execution tests gpuweb/cts#3398, by @zoddicus.
-
wgsl: Implement AbstractInt subtraction execution tests gpuweb/cts#3397, by @zoddicus.
-
Fix nits gpuweb/cts#3396, by @beaufortfrancois.
-
[shader/execution] Implement textureDimensions() gpuweb/cts#3395, by @ben-clayton.
-
Clean up deprecated .sampleType/.copySrc from kTextureFormatInfo gpuweb/cts#3394, by @kainino0x.
-
Separate tests for target format supporting usage and being a color format gpuweb/cts#3393, by @kainino0x.
-
wgsl: Implement AbstractFloat addition execution tests gpuweb/cts#3392, by @zoddicus.
-
Switch memory model tests to use PRNG gpuweb/cts#3391, by @alan-baker.
-
wgsl: Implement AbstractInt negation execution tests gpuweb/cts#3390, by @zoddicus.
-
wgsl: Add CTS tests for WGSL limits gpuweb/cts#3389, by @jrprice.
-
wgsl: Implement AbstractInt `sign` execution tests gpuweb/cts#3387, by @zoddicus.
-
Rename test to address_of_and_indirection gpuweb/cts#3386, by @amaiorano.
-
wgsl: Implement AbstractInt `max` execution tests gpuweb/cts#3385, by @zoddicus.
-
Add missing types to indirection expression tests gpuweb/cts#3384, by @amaiorano.
-
wgsl: Implement AbstractInt `min` execution tests gpuweb/cts#3383, by @zoddicus.
-
Test default bindgrouplayouts match in the same pipeline gpuweb/cts#3382, by @greggman.
-
Add validation tests for out of order location declarations gpuweb/cts#3381, by @alan-baker.
-
wgsl: Implement AbstractInt `clamp` execution tests gpuweb/cts#3380, by @zoddicus.
-
Fix assignment of AInt to u32 in the expression generator gpuweb/cts#3379, by @amaiorano.
-
wgsl: Add AbstractInt `select` execution tests gpuweb/cts#3378, by @zoddicus.
-
Fix maxBindGroupsPlusVertexBuffers test gpuweb/cts#3377, by @greggman.
-
Test default layout bindgroups gpuweb/cts#3374, by @greggman.
-
Fix bitcast of AFloat to F32 tests to allow FTZ gpuweb/cts#3373, by @amaiorano.
-
wgsl: Implement AbstractInt `abs` execution tests gpuweb/cts#3372, by @zoddicus.
-
Reduce permutations of most expensive tests gpuweb/cts#3371, by @ben-clayton.
-
op: read-only storage textures - Part I gpuweb/cts#3370, by @Jiawei-Shao.
-
Compat: skip more multisample copyT2T tests gpuweb/cts#3369, by @greggman.
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3368, by @greggman.
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3367, by @greggman.
-
Add tests that builtin methods correctly detect being shadowed. gpuweb/cts#3366, by @dj2.
-
Use 'trackForCleanUp' to destroy HTMLVideoElement gpuweb/cts#3365, by @shaoboyan.
-
Compat: Test that multisample copyT2T is validated out gpuweb/cts#3362, by @greggman.
-
Compat: Skip multisample copyT2T tests gpuweb/cts#3361, by @greggman.
-
Fix maxVertexBuffers test gpuweb/cts#3360, by @greggman.
-
wgsl: Validation tests for built-ins using AbstractInts gpuweb/cts#3359, by @zoddicus.
-
Remove usage of the Theora codec gpuweb/cts#3358, by @padenot.
-
Fix maxComputeWorkgroupStorageSize test gpuweb/cts#3357, by @greggman.
-
-
77 pull requests merged.
-
wgsl: Implement AbstractInt comparison execution tests gpuweb/cts#3401.
-
wgsl: Implement AbstractFloat `dot` execution tests gpuweb/cts#3443.
-
wgsl: Implement `i32` & `u32` dot execution tests gpuweb/cts#3411.
-
wgsl: Implement AbstractInt `dot` execution tests gpuweb/cts#3406.
-
Compat: Texture Intra Invocation Coherence gpuweb/cts#3447.
-
Validation tests for resource compatibility of render pipelines gpuweb/cts#3441.
-
Compat: Skip workgroupUniformLoad tests > max invocations gpuweb/cts#3445.
-
Remove unneeded language feature skip gpuweb/cts#3444.
-
wgsl: Implement AbstractInt complement (~) execution tests gpuweb/cts#3440.
-
Re-add temporarily TestWorker which aliases to TestDedicatedWorker gpuweb/cts#3442.
-
Relax asin(f32) absolute error to 6.81x10^(-5) gpuweb/cts#3438.
-
wgsl: Implement AbstractInt bitwise-xor execution tests gpuweb/cts#3439.
-
Resource compatibility validation gpuweb/cts#3431.
-
Add textureBarrier tests gpuweb/cts#3416.
-
WIP: Implement AbstractInt bitwise execution tests gpuweb/cts#3430.
-
Reapply "Add shared worker support (#3345)" (#3428) gpuweb/cts#3435.
-
wgsl: Add validation tests for derivative builtins gpuweb/cts#3433.
-
wgsl: Add workgroupUniformLoad validation tests gpuweb/cts#3429.
-
[wgsl] Validate continue inside switch control flow. gpuweb/cts#3434.
-
Add compat mode testing to the PR checklist gpuweb/cts#3425.
-
Revert "Add shared worker support (#3345)" gpuweb/cts#3428.
-
wgsl: Add workgroupUniformLoad execution tests gpuweb/cts#3427.
-
Validate address space layout constraints gpuweb/cts#3426.
-
op: 1D and 3D read-only storage textures gpuweb/cts#3422.
-
[shader] Add tests for ptr parameters, used by builtins. gpuweb/cts#3424.
-
Add timing information to server CTS runner gpuweb/cts#3423.
-
Add shared worker support gpuweb/cts#3345.
-
Compat: Skip new storage texture rg32 tests gpuweb/cts#3421.
-
op: read-only storage textures in vertex shader gpuweb/cts#3420.
-
op: read-only storage textures in fragment shader gpuweb/cts#3418.
-
op: bgra8unorm as read-only storage texture format gpuweb/cts#3417.
-
Add recommended extensions to code-workspace gpuweb/cts#3414.
-
Validation tests for unary address-of and indirection gpuweb/cts#3410.
-
Clean up deprecated .sampleType/.copySrc from kTextureFormatInfo gpuweb/cts#3394.
-
[shader/execution] Fixes for `textureDimensions()` gpuweb/cts#3413.
-
Add force fallback adapter option gpuweb/cts#3403.
-
fix comment in fragment_builtins.spec.ts gpuweb/cts#3409.
-
Test fragment_input @builtin(sample_mask) gpuweb/cts#3404.
-
[shader/execution] Implement textureDimensions() gpuweb/cts#3395.
-
Tests for intra-invocation coherence of storage textures gpuweb/cts#3402.
-
Fix nits gpuweb/cts#3396.
-
wgsl: Implement AbstractInt remainder execution tests gpuweb/cts#3400.
-
wgsl: Implement AbstractInt multiplication execution tests gpuweb/cts#3398.
-
wgsl: Implement AbstractInt division execution tests gpuweb/cts#3399.
-
Dedup kTextureFormats, fix format table docs gpuweb/cts#3351.
-
Switch memory model tests to use PRNG gpuweb/cts#3391.
-
wgsl: Implement AbstractInt subtraction execution tests gpuweb/cts#3397.
-
Test default bindgrouplayouts match in the same pipeline gpuweb/cts#3382.
-
wgsl: Implement AbstractFloat addition execution tests gpuweb/cts#3392.
-
wgsl: Implement AbstractInt negation execution tests gpuweb/cts#3390.
-
wgsl: Implement AbstractInt `sign` execution tests gpuweb/cts#3387.
-
wgsl: Implement AbstractInt `max` execution tests gpuweb/cts#3385.
-
wgsl: Implement AbstractInt `min` execution tests gpuweb/cts#3383.
-
wgsl: Implement AbstractInt `clamp` execution tests gpuweb/cts#3380.
-
Rename test to address_of_and_indirection gpuweb/cts#3386.
-
Add missing types to indirection expression tests gpuweb/cts#3384.
-
Add validation tests for out of order location declarations gpuweb/cts#3381.
-
op: read-only storage textures - Part I gpuweb/cts#3370.
-
Test default layout bindgroups gpuweb/cts#3374.
-
Fix maxBindGroupsPlusVertexBuffers test gpuweb/cts#3377.
-
wgsl: Add AbstractInt `select` execution tests gpuweb/cts#3378.
-
Fix assignment of AInt to u32 in the expression generator gpuweb/cts#3379.
-
Fix bitcast of AFloat to F32 tests to allow FTZ gpuweb/cts#3373.
-
wgsl: Implement AbstractInt `abs` execution tests gpuweb/cts#3372.
-
wgsl: Validation tests for built-ins using AbstractInts gpuweb/cts#3359.
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3368.
-
Reduce permutations of most expensive tests gpuweb/cts#3371.
-
Compat: skip more multisample copyT2T tests gpuweb/cts#3369.
-
Use 'trackForCleanUp' to destroy HTMLVideoElement gpuweb/cts#3365.
-
Fix maxVertexBuffers test gpuweb/cts#3360.
-
Compat: Test that multisample copyT2T is validated out gpuweb/cts#3362.
-
Compat: Skip multisample copyT2T tests gpuweb/cts#3361.
-
Remove usage of the Theora codec gpuweb/cts#3358.
-
Fix maxComputeWorkgroupStorageSize test gpuweb/cts#3357.
-
Fix maxXXXPerShaderStage tests gpuweb/cts#3356.
-
-
2 pull requests closed.
-
Just cleaning up a comment in fragment_builtins.spec.ts gpuweb/cts#3408.
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3367.
-
-
40 pull requests received 229 new comments.
-
[shader/execution] Implement textureDimensions() gpuweb/cts#3395 (37 comments by @ben-clayton, @dneto0, @kainino0x)
-
Add tests that builtin methods correctly detect being shadowed. gpuweb/cts#3366 (19 comments by @ben-clayton, @dj2)
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3368 (13 comments by @greggman, @kainino0x, @mwyrzykowski)
-
wgsl: Validation tests for built-ins using AbstractInts gpuweb/cts#3359 (11 comments by @ben-clayton, @zoddicus)
-
Switch memory model tests to use PRNG gpuweb/cts#3391 (8 comments by @alan-baker, @reeselevine, @zoddicus)
-
Test fragment_input @builtin(sample_mask) gpuweb/cts#3404 (7 comments by @dneto0, @greggman, @shrekshao)
-
Test maxBindGroupsPlusVertexBuffers limits gpuweb/cts#3367 (6 comments by @greggman, @kainino0x, @mwyrzykowski)
-
wgsl: Implement AbstractInt comparison execution tests gpuweb/cts#3401 (5 comments by @ben-clayton, @zoddicus)
-
wgsl: Implement AbstractInt complement (~) execution tests gpuweb/cts#3440 (4 comments by @ben-clayton, @zoddicus)
-
WIP: Implement AbstractInt bitwise execution tests gpuweb/cts#3430 (4 comments by @ben-clayton, @zoddicus)
-
wgsl: Add AbstractInt `select` execution tests gpuweb/cts#3378 (4 comments by @ben-clayton, @zoddicus)
-
Fix assignment of AInt to u32 in the expression generator gpuweb/cts#3379 (4 comments by @amaiorano, @ben-clayton)
-
wgsl: Implement AbstractFloat matrix multiplcation tests gpuweb/cts#3446 (3 comments by @ben-clayton, @zoddicus)
-
wgsl: Implement AbstractFloat `ceil` execution tests gpuweb/cts#3449 (2 comments by @ben-clayton, @zoddicus)
-
op: read-only storage textures in vertex shader gpuweb/cts#3420 (2 comments by @jiangzhaoming, @Jiawei-Shao)
-
op: read-only storage textures in fragment shader gpuweb/cts#3418 (2 comments by @jiangzhaoming, @Kangz)
-
Validation tests for unary address-of and indirection gpuweb/cts#3410 (2 comments by @amaiorano, @ben-clayton)
-
Clean up deprecated .sampleType/.copySrc from kTextureFormatInfo gpuweb/cts#3394 (2 comments by @kainino0x)
-
Separate tests for target format supporting usage and being a color format gpuweb/cts#3393 (2 comments by @kainino0x, @lokokung)
-
fragment builtins: document missing test, or other places to find them gpuweb/cts#3339 (1 comment by @dneto0)
-
gpuweb/gpuweb
-
13 issues opened.
-
Use enum tokens when they are the only possible grammar usage. gpuweb/gpuweb#4505, by @dj2.
-
Provide a kind of memory layout mapping in BufferDescriptor/BufferInitDescriptor gpuweb/gpuweb#4498, by @Alphapage.
-
Difference between alignment of struct with size 8 vs `vec2<f32>` in uniform buffers. gpuweb/gpuweb#4497, by @Swoorup.
-
Add in a way to manually commit the canvas texture gpuweb/gpuweb#4496, by @greggman.
-
Y orientation of viewport coordinates vs. NDC gpuweb/gpuweb#4492, by @kpreid.
-
Spec is unclear about 3D texture mipmap support gpuweb/gpuweb#4491, by @ben-clayton.
-
Compat: Dealing with overflow for `drawIndexedIndirect` gpuweb/gpuweb#4490, by @greggman.
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489, by @SenorBlanco.
-
[editorial] Consistency of built-in function overloads gpuweb/gpuweb#4488, by @alan-baker.
-
Texture types with f16, e.g. `texture_2d<f16>` gpuweb/gpuweb#4483, by @kainino0x.
-
Remove "update the rendering" hook for service/shared worker gpuweb/gpuweb#4478, by @kainino0x.
-
relax absolute error bound for asin f32: too tight for Intel f32 with DXC gpuweb/gpuweb#4477, by @dneto0.
-
-
5 issues closed.
-
relax absolute error bound for asin f32: too tight for Intel f32 with DXC gpuweb/gpuweb#4477.
-
Provide a kind of memory layout mapping in BufferDescriptor/BufferInitDescriptor gpuweb/gpuweb#4498.
-
Shader extensions are redundant gpuweb/gpuweb#3961.
-
Compat: Dealing with overflow for `drawIndexedIndirect` gpuweb/gpuweb#4490.
-
Remove "update the rendering" hook for service/shared worker gpuweb/gpuweb#4478.
-
-
2 issues reopened.
-
Should we constrain the location of user input-output stage variables WGSL gpuweb/gpuweb#1962, by @Kangz.
-
Should we constrain the location of user input-output stage variables WGSL gpuweb/gpuweb#1962, by @Kangz.
-
-
28 issues received 112 new comments.
-
mapSync on Workers - and possibly on the main thread gpuweb/gpuweb#2217 (16 comments by @greggman, @kainino0x, @Kangz, @litherum, @lvyitian, @torokati44, @valadaptive)
-
Create a proposal for multi-worker gpuweb/gpuweb#354 (15 comments by @greggman, @kainino0x, @Kangz, @kenrussell, @louderspace, @syg, @tlively)
-
Use enum tokens when they are the only possible grammar usage. gpuweb/gpuweb#4505 (9 comments by @ben-clayton, @dj2, @mehmetoguzderin)
-
Compatibility mode: provoking vertex differs between OpenGL and WebGPU gpuweb/gpuweb#4489 (8 comments by @greggman, @jimblandy, @kainino0x, @kdashg, @SenorBlanco)
-
Should we constrain the location of user input-output stage variables WGSL gpuweb/gpuweb#1962 (7 comments by @Jiawei-Shao, @kainino0x, @Kangz)
-
Compat: can not copy multisample textures gpuweb/gpuweb#4468 (6 comments by @greggman, @kainino0x, @kdashg)
-
Dealing with holes in the pipeline layout. gpuweb/gpuweb#2043 (6 comments by @greggman, @kainino0x, @Kangz)
-
Extended brightness range rendering gpuweb/gpuweb#4239 (3 comments by @adixon-adobe, @louderspace, @ounterecker)
-
API should be exposed to ServiceWorker gpuweb/gpuweb#4197 (3 comments by @beaufortfrancois, @rickzx, @tqchen)
-
[editorial] Consistency of built-in function overloads gpuweb/gpuweb#4488 (3 comments by @ben-clayton, @mehmetoguzderin)
-
Difference between alignment of struct with size 8 vs `vec2<f32>` in uniform buffers. gpuweb/gpuweb#4497 (2 comments by @dneto0)
-
Provide a kind of memory layout mapping in BufferDescriptor/BufferInitDescriptor gpuweb/gpuweb#4498 (2 comments by @Alphapage, @Kangz)
-
Compat: Dealing with overflow for `drawIndexedIndirect` gpuweb/gpuweb#4490 (2 comments by @greggman)
-
Slight type problem with `GPURenderPipeline.[[writesDepth]]` gpuweb/gpuweb#4416 (2 comments by @kainino0x)
-
Remove "update the rendering" hook for service/shared worker gpuweb/gpuweb#4478 (2 comments by @beaufortfrancois, @kainino0x)
-
relax absolute error bound for asin f32: too tight for Intel f32 with DXC gpuweb/gpuweb#4477 (1 comment by @kdashg)
-
Evaluation order of assignment and compound assignment are flipped gpuweb/gpuweb#3928 (1 comment by @quantumagi)
-
Avoid indeterminate value result for `tanh` with large input (f32 88.7 or f16 11.09) gpuweb/gpuweb#4458 (1 comment by @kdashg)
-
-
12 pull requests opened.
-
Fix the computation of inter-stage shader components gpuweb/gpuweb#4503, by @Jiawei-Shao.
-
typos: s/paramters/parameters gpuweb/gpuweb#4502, by @ErichDonGubler.
-
Relax asin(f32) absolute error bound to 6.81x10^(-5) gpuweb/gpuweb#4501, by @dneto0.
-
Add extended range gpuweb/gpuweb#4500, by @ccameron-chromium.
-
Clarify handling of out of range color values gpuweb/gpuweb#4499, by @ccameron-chromium.
-
[wgsl] Assume `bool` is 4 bytes for array limits gpuweb/gpuweb#4487, by @jrprice.
-
[wgsl] Clarify the vec and matrix abstract overloads. gpuweb/gpuweb#4486, by @dj2.
-
Clarify typography of abstract int 0 and abstract float 0 gpuweb/gpuweb#4485, by @dneto0.
-
Remove the greater-or-equal constraint on maxBindingsPerBindGroup gpuweb/gpuweb#4484, by @nical.
-
[wgsl] Fixup bitcast return types. gpuweb/gpuweb#4482, by @dj2.
-
Add Compat multisample texture copy validation gpuweb/gpuweb#4480, by @greggman.
-
Remove 'update the rendering' hook for service/shared worker gpuweb/gpuweb#4479, by @beaufortfrancois.
-
-
11 pull requests merged.
-
typos: s/paramters/parameters gpuweb/gpuweb#4502.
-
Relax asin(f32) absolute error bound to 6.81x10^(-5) gpuweb/gpuweb#4501.
-
[wgsl] Assume `bool` is 4 bytes for array limits gpuweb/gpuweb#4487.
-
Add Compat multisample texture copy validation gpuweb/gpuweb#4480.
-
Remove 'update the rendering' hook for service/shared worker gpuweb/gpuweb#4479.
-
[wgsl] Fixup bitcast return types. gpuweb/gpuweb#4482.
-
Add Compat interpolation validation gpuweb/gpuweb#4424.
-
Mention that sample_index causes per-sample fragment shader execution gpuweb/gpuweb#4442.
-
Use "get the entry point" algorithm where necessary. gpuweb/gpuweb#4470.
-
Fix truncated line in "default pipeline layout" algorithm. gpuweb/gpuweb#4471.
-
Add SharedWorker support gpuweb/gpuweb#4465.
-
-
3 pull requests closed.
-
[wgsl] Clarify the vec and matrix abstract overloads. gpuweb/gpuweb#4486.
-
Add Compat rg32xxx storage texture restrictions gpuweb/gpuweb#4423.
-
Make setBindGroup non-throwing/non-allocating gpuweb/gpuweb#4475.
-
-
13 pull requests received 39 new comments.
-
Add extended range gpuweb/gpuweb#4500 (17 comments by @ccameron-chromium, @kainino0x, @mwyrzykowski)
-
Use "get the entry point" algorithm where necessary. gpuweb/gpuweb#4470 (5 comments by @jimblandy, @kainino0x)
-
Remove the greater-or-equal constraint on maxBindingsPerBindGroup gpuweb/gpuweb#4484 (4 comments by @kainino0x, @nical)
-
Define built-in constants and add two circle constants gpuweb/gpuweb#4473 (3 comments by @ben-clayton, @mehmetoguzderin)
-
Fix truncated line in "default pipeline layout" algorithm. gpuweb/gpuweb#4471 (2 comments by @jimblandy, @kainino0x)
-
Clarify typography of abstract int 0 and abstract float 0 gpuweb/gpuweb#4485 (1 comment by @mehmetoguzderin)
-
Remove 'update the rendering' hook for service/shared worker gpuweb/gpuweb#4479 (1 comment by @kainino0x)
-
TR publications
-
12 documents published.
-
WebGPU published 6 times as a Working Draft. History of WebGPU
-
WebGPU Shading Language published 6 times as a Working Draft. History of WebGPU Shading Language
-
Participants
-
1 participant joined.
-
Shuangyan Yue (Baidu, Inc.) joined.
-
-
1 participant left.
-
Antoine Dumas (Preset Computing) left.
-
-
1 organization left.
-
Preset Computing left.
-
GitHub
gpuweb/cts
-
35 issues opened.
-
Test CopyEI2T with full transparent source gpuweb/cts#3350, by @shaoboyan.
-
Cleanup: find and deduplicate fullscreen quad shader code gpuweb/cts#3344, by @kainino0x.
-
Add tests for @must_use attribute gpuweb/cts#3334, by @amaiorano.
-
wgsl: Take a look at how subnormals and FTZ are handled in assignment tests gpuweb/cts#3325, by @zoddicus.
-
[wgsl] Add discard inside continuing block execution test. gpuweb/cts#3317, by @dj2.
-
Expectation in dynamic_state.spec.ts does not align with the specification gpuweb/cts#3307, by @mwyrzykowski.
-
Make AbstractInts Go. gpuweb/cts#3302, by @dj2.
-
reftests: Delay calling context getCurrentTexture after configure gpuweb/cts#3292, by @w07i.
-
wgsl: Implement `step` validation tests gpuweb/cts#3290, by @zoddicus.
-
wgsl: Implement `smoothstep` validation tests gpuweb/cts#3289, by @zoddicus.
-
wgsl: Implement `refract` validation tests gpuweb/cts#3288, by @zoddicus.
-
wgsl: Implement `fract` validation tests gpuweb/cts#3286, by @zoddicus.
-
wgsl: Implement `floor` validation tests gpuweb/cts#3283, by @zoddicus.
-
Test videos with both flip and rotation gpuweb/cts#3281, by @kainino0x.
-
reftests: Test canvas that the context is only configured. gpuweb/cts#3279, by @w07i.
-
wgsl: Add AbstractFloat entries to `kAllFloatsScalarsAndVectors` gpuweb/cts#3276, by @zoddicus.
-
wgsl: Add AbstractInt entries to `kAllIntegerScalarsAndVectors` gpuweb/cts#3275, by @zoddicus.
-
wgsl: Implement `select` builtin validation tests gpuweb/cts#3274, by @zoddicus.
-
wgsl: Implement `any` builtin validation tests gpuweb/cts#3273, by @zoddicus.
-
wgsl: Implement `all` builtin validation tests gpuweb/cts#3272, by @zoddicus.
-
Implement operation tests for separate read-only depth-stencil gpuweb/cts#3270, by @Kangz.
-
`popErrorScope failed; did beginTestScope get missed?` gpuweb/cts#3268, by @ben-clayton.
-
Improve uniformity analysis coverage for nested LValue expressions gpuweb/cts#3266, by @jrprice.
-
wgsl: test scalar value constructors from abstracts gpuweb/cts#3254, by @dneto0.
-
web_platform: Canvas readback after device lost/destroyed gpuweb/cts#3252, by @kainino0x.
-
reftests: Canvas display after device lost/destroyed gpuweb/cts#3251, by @kainino0x.
-
wgsl: Implement `dot4I8Packed` execution tests gpuweb/cts#3248, by @zoddicus.
-
wgsl: Implement `dot4U8Packed` execution tests gpuweb/cts#3247, by @zoddicus.
-
Test writeBuffer/writeTexture with all possible buffer sources gpuweb/cts#3241, by @kainino0x.
-
How to create cts.html with variants for every test case? gpuweb/cts#3233, by @sagudev.
-
26 issues closed.
-
wgsl: Validation module scoped variables gpuweb/cts#1570.
-
Make AbstractInts Go. gpuweb/cts#3302.
-
wgsl: compound statement tests gpuweb/cts#1634.
-
Add tests for syntax sugar for dereferencing composites gpuweb/cts#3225.
-
wgsl: validate `break-if` statement gpuweb/cts#1672.
-
wgsl: test dp4a ( packed_4x8_integer_dot_product ) gpuweb/cts#3180.
-
wgsl: execution tests for `@workgroup_size` gpuweb/cts#1463.
-
test builtins and expressions in all shader stages gpuweb/cts#841.
-
wgsl: Validate that `@location` can duplicate between input/output gpuweb/cts#1706.
-
Fix CTS roll into Chromium gpuweb/cts#3028.
-
wgsl: Execution tests for `@id` gpuweb/cts#1445.
-
wgsl: Implement `floor` validation tests gpuweb/cts#3283.
-
wgsl: Implement `dot4I8Packed` execution tests gpuweb/cts#3248.
-
wgsl: Implement `dot4U8Packed` execution tests gpuweb/cts#3247.
-
Improve uniformity analysis coverage for nested LValue expressions gpuweb/cts#3266.
-
`popErrorScope failed; did beginTestScope get missed?` gpuweb/cts#3268.
-
wpt lint errors gpuweb/cts#2628.
-
wgsl: test unrestricted_pointer_parameters WGSL language extension gpuweb/cts#3181.
-
wgsl: Change usage of `abstract` to `abstract_float` gpuweb/cts#3184.
-
wgsl: Add FPInterval to the types for expected in FP unittests gpuweb/cts#3210.
-
-
26 issues received 42 new comments.
-
How to create cts.html with variants for every test case? gpuweb/cts#3233 (5 comments by @kainino0x, @sagudev)
-
reftests: Canvas display after device lost/destroyed gpuweb/cts#3251 (4 comments by @kainino0x, @w07i)
-
api,validation,capability_checks,limits,maxBindGroups:createPipelineLayout test has some invalid expectations gpuweb/cts#3346 (3 comments by @greggman, @mwyrzykowski)
-
Running webgpu:web_platform,external_texture,video:* multiple times hangs browser gpuweb/cts#3301 (3 comments by @kainino0x, @shaoboyan)
-
Constant evaluation shader test expectations do not allow for rounding as specified in the specification? gpuweb/cts#3261 (3 comments by @dneto0, @mwyrzykowski)
-
shader,execution,expression,binary,af_addition:scalar:* has incorrect expectations gpuweb/cts#3253 (2 comments by @mwyrzykowski)
-
fixture.ts is checking for GPUPipelineError.stack to exist but it is not required on DOMException gpuweb/cts#3222 (2 comments by @kainino0x, @mwyrzykowski)
-
Expectation in dynamic_state.spec.ts does not align with the specification gpuweb/cts#3307 (1 comment by @Kangz)
-
wgsl: 'fallthrough' removed, and 'default' can be one of the case values gpuweb/cts#1750 (1 comment by @jrprice)
-
wgsl: Validate that `@location` can duplicate between input/output gpuweb/cts#1706 (1 comment by @jrprice)
-
reftests: Test canvas that the context is only configured. gpuweb/cts#3279 (1 comment by @kainino0x)
-
wgsl: Add AbstractFloat entries to `kAllFloatsScalarsAndVectors` gpuweb/cts#3276 (1 comment by @zoddicus)
-
wgsl: test unrestricted_pointer_parameters WGSL language extension gpuweb/cts#3181 (1 comment by @ben-clayton)
-
Rotated videos in webgpu:web_platform,external_texture,video:* is ccw instead of cw gpuweb/cts#3232 (1 comment by @shaoboyan)
-
wgsl: Add FPInterval to the types for expected in FP unittests gpuweb/cts#3210 (1 comment by @zoddicus)
-
-
89 pull requests opened.
-
Fix maxXXXPerShaderStage tests gpuweb/cts#3356, by @greggman.
-
Use beginSubcases() more aggressively gpuweb/cts#3355, by @ben-clayton.
-
Fix maxBindGroups test so it can attempt more resources gpuweb/cts#3354, by @greggman.
-
shader/validation: Validate var initializers gpuweb/cts#3353, by @jrprice.
-
Dedup kTextureFormats, fix format table docs gpuweb/cts#3351, by @kainino0x.
-
Pass adapter limit maxTextureArrayLayers for OOM tests. gpuweb/cts#3349, by @lokokung.
-
shader/validation: Validate binding points on variables gpuweb/cts#3348, by @jrprice.
-
shader/validation: Test function addrspace at module-scope gpuweb/cts#3347, by @jrprice.
-
Add shared worker support gpuweb/cts#3345, by @beaufortfrancois.
-
wgsl: Add trivial execution test for each shader stage gpuweb/cts#3343, by @dneto0.
-
Fix resize test for offscreen canvas gpuweb/cts#3342, by @greggman.
-
wgsl: Test mixing stage attributes on the same function gpuweb/cts#3341, by @dneto0.
-
Extend compute pass multisampled copy texture helper to copy multiple pixels gpuweb/cts#3340, by @shrekshao.
-
fragment builtins: document missing test, or other places to find them gpuweb/cts#3339, by @dneto0.
-
shader/validation: Test store types for var decls gpuweb/cts#3338, by @jrprice.
-
wgsl: compound statement tests gpuweb/cts#3337, by @dneto0.
-
Add barrier sync tests using workgroupUniformLoad gpuweb/cts#3336, by @alan-baker.
-
Add more validation tests for @must_use function attribute gpuweb/cts#3335, by @amaiorano.
-
webgpu,api,operation,texture_view,write:format:* gpuweb/cts#3333, by @shrekshao.
-
Fix color space conversion test gpuweb/cts#3332, by @greggman.
-
wgsl: Implement handling AbstractInts as output from tests gpuweb/cts#3331, by @zoddicus.
-
Compat: Fix readonly depth-stencil sampling gpuweb/cts#3330, by @greggman.
-
Compat: Fix depth bias test gpuweb/cts#3329, by @greggman.
-
Compat: Fix storage texture read_write test gpuweb/cts#3328, by @greggman.
-
Add storage texture loads to uniformity tests gpuweb/cts#3327, by @jrprice.
-
Add textureBarrier() to uniformity tests gpuweb/cts#3326, by @jrprice.
-
shader/validation: Add tests for barrier builtins gpuweb/cts#3324, by @jrprice.
-
Use correctly rounded in bitcast test. gpuweb/cts#3323, by @dj2.
-
Shader execution tests for memory layout gpuweb/cts#3322, by @alan-baker.
-
Put common parts of texture_zero.spec.ts into a separate file gpuweb/cts#3321, by @greggman.
-
Move the query manipulation to parseQuery gpuweb/cts#3320, by @greggman.
-
Re-review initMethod = render in image_copy tests gpuweb/cts#3319, by @shrekshao.
-
Add inter-stage superset test gpuweb/cts#3318, by @greggman.
-
Shorten suite name in query gpuweb/cts#3316, by @greggman.
-
wgsl: Add tests for requires directives gpuweb/cts#3315, by @jrprice.
-
Add `continuing` validation tests. gpuweb/cts#3314, by @dj2.
-
Unload HTMLVideoElement actively after cases finished gpuweb/cts#3312, by @shaoboyan.
-
Update getCurrentTexture test to match spec. gpuweb/cts#3311, by @greggman.
-
Compat: Fix canvas configure test for compat gpuweb/cts#3310, by @greggman.
-
Fix drawToCanvas2D test gpuweb/cts#3309, by @greggman.
-
Remove workaround for Tint unary array access gpuweb/cts#3308, by @shrekshao.
-
Add support for AbstractInts as inputs gpuweb/cts#3306, by @zoddicus.
-
Add break-if validation tests gpuweb/cts#3305, by @dj2.
-
Add executions tests for unary indirection gpuweb/cts#3304, by @amaiorano.
-
Add arrayLength validation tests. gpuweb/cts#3303, by @dj2.
-
wgsl: Test all 4x8 integer pack/unpack builtin functions gpuweb/cts#3300, by @Jiawei-Shao.
-
Add AbstractFloat bitcast execution tests. gpuweb/cts#3299, by @dj2.
-
Add validation tests for pointer_composite_access gpuweb/cts#3298, by @amaiorano.
-
Test destroyed state of storage textures in bind groups gpuweb/cts#3297, by @austinEng.
-
wgsl: @size only allowed on creation-fixed types gpuweb/cts#3296, by @jrprice.
-
Add tests for workgroup_size execution gpuweb/cts#3295, by @alan-baker.
-
Add reftests: delay calling context getCurrentTexture after configure gpuweb/cts#3294, by @w07i.
-
wgsl: Add `dot4I8Packed` and `dot4U8Packed` execution tests gpuweb/cts#3293, by @Jiawei-Shao.
-
Relax the fragment builtin tests gpuweb/cts#3291, by @greggman.
-
wgsl: Add in some missing timing data for pack* & unpack* tests gpuweb/cts#3285, by @zoddicus.
-
wgsl: Add `floor` validation tests gpuweb/cts#3284, by @zoddicus.
-
op: Basic operation tests on 1D and 3D read-write storage textures gpuweb/cts#3282, by @Jiawei-Shao.
-
Use TexelView for readbackFromWebGPUCanvas tests gpuweb/cts#3280, by @greggman.
-
Add flip video files for GPUExternalTexture related cts gpuweb/cts#3278, by @shaoboyan.
-
Compat: Refactor compute_builtins test gpuweb/cts#3277, by @greggman.
-
Add separate readonly depth-stencil tests gpuweb/cts#3271, by @Kangz.
-
Add more uniformity tests gpuweb/cts#3269, by @alan-baker.
-
op: Basic operations with read_write storage texture gpuweb/cts#3267, by @Jiawei-Shao.
-
Expand image_copy test params only for compat + r8snorm/rg8snorm gpuweb/cts#3264, by @shrekshao.
-
Revert "Slightly expand image_copy tests to cover more snorm cases for compat" gpuweb/cts#3263, by @shrekshao.
-
Slightly expand image_copy tests to cover more snorm cases for compat gpuweb/cts#3260, by @shrekshao.
-
Compat: Test that depthBiasClamp must be 0 gpuweb/cts#3259, by @greggman.
-
Compat: Skip tests with depth_bias clamp != 0 gpuweb/cts#3258, by @greggman.
-
Compat: Refactor tests that use sample_index for compat gpuweb/cts#3257, by @greggman.
-
Compat: Test that sample_index generates validation error gpuweb/cts#3256, by @greggman.
-
Allow Path as query gpuweb/cts#3255, by @greggman.
-
Test @interpolate(..., centroid) gpuweb/cts#3250, by @greggman.
-
Remove grunt-ts gpuweb/cts#3249, by @greggman.
-
fullyExpandSubtrees & ability to disable no long path assert gpuweb/cts#3246, by @sagudev.
-
Update video files to improve external_texture* related cts coverage gpuweb/cts#3245, by @shaoboyan.
-
Consistently clean up VideoFrame objects gpuweb/cts#3244, by @kainino0x.
-
Test builtin(front_facing) gpuweb/cts#3243, by @greggman.
-
Compat: Check unsupported storage texture formats gpuweb/cts#3242, by @greggman.
-
wgsl: Improve vector and matrix OOB propegation gpuweb/cts#3240, by @zoddicus.
-
Val: bindable resource type compatibility between pipeline layout and shader gpuweb/cts#3239, by @Jiawei-Shao.
-
val: Add depthSlice tests to render_pass_descriptor:color_attachments gpuweb/cts#3238, by @haoxli.
-
Test fragment @builtin(sample_index) gpuweb/cts#3237, by @greggman.
-
Fix scaling issue in canvas image rendering test gpuweb/cts#3236, by @greggman.
-
Compat: make atomicCompareExchangeWeak tests handle limits gpuweb/cts#3235, by @greggman.
-
wgsl: Add discussion about how OOB-ness is inherited gpuweb/cts#3234, by @zoddicus.
-
-
84 pull requests merged.
-
Fix maxBindGroups test so it can attempt more resources gpuweb/cts#3354.
-
Use beginSubcases() more aggressively gpuweb/cts#3355.
-
shader/validation: Validate var initializers gpuweb/cts#3353.
-
wgsl: Implement handling AbstractInts as output from tests gpuweb/cts#3331.
-
shader/validation: Validate binding points on variables gpuweb/cts#3348.
-
Pass adapter limit maxTextureArrayLayers for OOM tests. gpuweb/cts#3349.
-
webgpu,api,operation,texture_view,write:format:* gpuweb/cts#3333.
-
wgsl: compound statement tests gpuweb/cts#3337.
-
wgsl: Add trivial execution test for each shader stage gpuweb/cts#3343.
-
wgsl: Test mixing stage attributes on the same function gpuweb/cts#3341.
-
shader/validation: Test function addrspace at module-scope gpuweb/cts#3347.
-
shader/validation: Test store types for var decls gpuweb/cts#3338.
-
Shader execution tests for memory layout gpuweb/cts#3322.
-
Add barrier sync tests using workgroupUniformLoad gpuweb/cts#3336.
-
Fix resize test for offscreen canvas gpuweb/cts#3342.
-
Add more validation tests for @must_use function attribute gpuweb/cts#3335.
-
Fix color space conversion test gpuweb/cts#3332.
-
Update getCurrentTexture test to match spec. gpuweb/cts#3311.
-
shader/validation: Add tests for barrier builtins gpuweb/cts#3324.
-
Compat: Fix readonly depth-stencil sampling gpuweb/cts#3330.
-
Compat: Fix storage texture read_write test gpuweb/cts#3328.
-
Compat: Fix depth bias test gpuweb/cts#3329.
-
Use correctly rounded in bitcast test. gpuweb/cts#3323.
-
Add executions tests for unary indirection gpuweb/cts#3304.
-
Add storage texture loads to uniformity tests gpuweb/cts#3327.
-
Add textureBarrier() to uniformity tests gpuweb/cts#3326.
-
Re-review initMethod = render in image_copy tests gpuweb/cts#3319.
-
Add support for AbstractInts as inputs gpuweb/cts#3306.
-
wgsl: Add tests for requires directives gpuweb/cts#3315.
-
Move the query manipulation to parseQuery gpuweb/cts#3320.
-
Put common parts of texture_zero.spec.ts into a separate file gpuweb/cts#3321.
-
Shorten suite name in query gpuweb/cts#3316.
-
Compat: Fix canvas configure test for compat gpuweb/cts#3310.
-
Add inter-stage superset test gpuweb/cts#3318.
-
Unload HTMLVideoElement actively after cases finished gpuweb/cts#3312.
-
Add `continuing` validation tests. gpuweb/cts#3314.
-
Add validation tests for pointer_composite_access gpuweb/cts#3298.
-
Fix drawToCanvas2D test gpuweb/cts#3309.
-
wgsl: @size only allowed on creation-fixed types gpuweb/cts#3296.
-
Add break-if validation tests gpuweb/cts#3305.
-
Remove workaround for Tint unary array access gpuweb/cts#3308.
-
wgsl: Test all 4x8 integer pack/unpack builtin functions gpuweb/cts#3300.
-
Add arrayLength validation tests. gpuweb/cts#3303.
-
Add AbstractFloat bitcast execution tests. gpuweb/cts#3299.
-
Test destroyed state of storage textures in bind groups gpuweb/cts#3297.
-
Add tests for workgroup_size execution gpuweb/cts#3295.
-
wgsl: Add `floor` validation tests gpuweb/cts#3284.
-
Add separate readonly depth-stencil tests gpuweb/cts#3271.
-
wgsl: Add `dot4I8Packed` and `dot4U8Packed` execution tests gpuweb/cts#3293.
-
Relax the fragment builtin tests gpuweb/cts#3291.
-
Use TexelView for readbackFromWebGPUCanvas tests gpuweb/cts#3280.
-
wgsl: Add in some missing timing data for pack* & unpack* tests gpuweb/cts#3285.
-
op: Basic operation tests on 1D and 3D read-write storage textures gpuweb/cts#3282.
-
Allow Path as query gpuweb/cts#3255.
-
Add flip video files for GPUExternalTexture related cts gpuweb/cts#3278.
-
Compat: Refactor compute_builtins test gpuweb/cts#3277.
-
fullyExpandSubtrees & ability to disable no long path assert gpuweb/cts#3246.
-
op: Basic operations with read_write storage texture gpuweb/cts#3267.
-
Add more uniformity tests gpuweb/cts#3269.
-
Update video files to improve external_texture* related cts coverage gpuweb/cts#3245.
-
Expand image_copy test params only for compat + r8snorm/rg8snorm gpuweb/cts#3264.
-
Compat: Skip tests with depth_bias clamp != 0 gpuweb/cts#3258.
-
Compat: Test that sample_index generates validation error gpuweb/cts#3256.
-
Compat: Refactor tests that use sample_index for compat gpuweb/cts#3257.
-
Compat: Test that depthBiasClamp must be 0 gpuweb/cts#3259.
-
Test @interpolate(..., centroid) gpuweb/cts#3250.
-
Remove grunt-ts gpuweb/cts#3249.
-
wgsl: Improve vector and matrix OOB propegation gpuweb/cts#3240.
-
Consistently clean up VideoFrame objects gpuweb/cts#3244.
-
Test builtin(front_facing) gpuweb/cts#3243.
-
Compat: Check unsupported storage texture formats gpuweb/cts#3242.
-
Compat: make atomicCompareExchangeWeak tests handle limits gpuweb/cts#3235.
-
wgsl: Add discussion about how OOB-ness is inherited gpuweb/cts#3234.
-
Test fragment @builtin(sample_index) gpuweb/cts#3237.
-
Fix scaling issue in canvas image rendering test gpuweb/cts#3236.
-
-
2 pull requests closed.
-
wgsl: Change 'abstract' -> 'abstract_float' gpuweb/cts#3216.
-
57 pull requests received 232 new comments.
-
Add support for AbstractInts as inputs gpuweb/cts#3306 (34 comments by @ben-clayton, @dj2, @zoddicus)
-
Update video files to improve external_texture* related cts coverage gpuweb/cts#3245 (13 comments by @kainino0x, @Kangz, @shaoboyan)
-
Add shader pointer parameter tests, including for `unrestricted_pointer_parameter` support gpuweb/cts#3265 (12 comments by @ben-clayton, @dneto0)
-
Add executions tests for unary indirection gpuweb/cts#3304 (10 comments by @amaiorano, @ben-clayton)
-
Unload HTMLVideoElement actively after cases finished gpuweb/cts#3312 (10 comments by @beaufortfrancois, @shaoboyan)
-
op: Basic operations with read_write storage texture gpuweb/cts#3267 (9 comments by @jiangzhaoming, @Kangz)
-
val: Add depthSlice tests to render_pass_descriptor:color_attachments gpuweb/cts#3238 (7 comments by @haoxli, @Kangz)
-
Val: bindable resource type compatibility between pipeline layout and shader gpuweb/cts#3239 (7 comments by @jiangzhaoming, @Jiawei-Shao, @Kangz)
-
Slightly expand image_copy tests to cover more snorm cases for compat gpuweb/cts#3260 (6 comments by @greggman, @shrekshao)
-
wgsl: Implement handling AbstractInts as output from tests gpuweb/cts#3331 (5 comments by @dj2, @zoddicus)
-
Extend compute pass multisampled copy texture helper to copy multiple pixels gpuweb/cts#3340 (5 comments by @kainino0x, @shrekshao)
-
fullyExpandSubtrees & ability to disable no long path assert gpuweb/cts#3246 (5 comments by @kainino0x, @sagudev)
-
Add flip video files for GPUExternalTexture related cts gpuweb/cts#3278 (4 comments by @greggman, @kainino0x, @shaoboyan)
-
wgsl: Change 'abstract' -> 'abstract_float' gpuweb/cts#3216 (4 comments by @ben-clayton, @jiangzhaoming, @zoddicus)
-
Compat: make atomicCompareExchangeWeak tests handle limits gpuweb/cts#3235 (3 comments by @greggman, @shrekshao)
-
Add more validation tests for @must_use function attribute gpuweb/cts#3335 (2 comments by @alan-baker, @amaiorano)
-
Re-review initMethod = render in image_copy tests gpuweb/cts#3319 (2 comments by @kainino0x, @shrekshao)
-
wgsl: @size only allowed on creation-fixed types gpuweb/cts#3296 (2 comments by @jrprice, @zoddicus)
-
Slightly expand more copyWidth params for image_copy tests to add compat coverage gpuweb/cts#3262 (2 comments by @shrekshao)
-
Revert "Slightly expand image_copy tests to cover more snorm cases for compat" gpuweb/cts#3263 (2 comments by @shrekshao)
-
fragment builtins: document missing test, or other places to find them gpuweb/cts#3339 (1 comment by @dneto0)
-
wgsl: Test all 4x8 integer pack/unpack builtin functions gpuweb/cts#3300 (1 comment by @Jiawei-Shao)
-
Add reftests: delay calling context getCurrentTexture after configure gpuweb/cts#3294 (1 comment by @kainino0x)
-
wgsl: Add `dot4I8Packed` and `dot4U8Packed` execution tests gpuweb/cts#3293 (1 comment by @ben-clayton)
-
shader/validation: Test ptr aliasing validation for `unrestricted_pointer_parameters` gpuweb/cts#3287 (1 comment by @dneto0)
-
wgsl: Add in some missing timing data for pack* & unpack* tests gpuweb/cts#3285 (1 comment by @zoddicus)
-
op: Basic operation tests on 1D and 3D read-write storage textures gpuweb/cts#3282 (1 comment by @jiangzhaoming)
-
Expand image_copy test params only for compat + r8snorm/rg8snorm gpuweb/cts#3264 (1 comment by @greggman)
-
gpuweb/gpuweb
-
19 issues opened.
-
[wgsl] Functions section defines WGSL builtins as always being available. gpuweb/gpuweb#4474, by @dj2.
-
Compat: can not copy multisample textures gpuweb/gpuweb#4468, by @greggman.
-
for dispatchWorkgroups, the workgroupCountY and workgroupCountZ default to 1 gpuweb/gpuweb#4467, by @dneto0.
-
Validation for setViewport should be `minDepth <= maxDepth` gpuweb/gpuweb#4461, by @mwyrzykowski.
-
Why is dispatchWorkgroupsIndirect() running much slower than dispatchWorkgroups()? gpuweb/gpuweb#4460, by @Loner789.
-
Per-pipeline binding points gpuweb/gpuweb#4459, by @martinfouilleul.
-
Avoid indeterminate value result for `tanh` with large input (f32 88.7 or f16 11.09) gpuweb/gpuweb#4458, by @jiangzhaoming.
-
Link tooltips overlaps link in table gpuweb/gpuweb#4457, by @guusw.
-
Content timeline step (6) for copyExternalImageToTexture() is inaccurate gpuweb/gpuweb#4456, by @mwyrzykowski.
-
Timestamp Write Dictionaries are Identical for Render Passes and Compute Passes gpuweb/gpuweb#4455, by @conorpo.
-
Clarify whether `canvas.width = canvas.width` invalidates a WebGPU canvas gpuweb/gpuweb#4452, by @kainino0x.
-
[editorial] Improve consistency of how values are represented in the spec gpuweb/gpuweb#4451, by @alan-baker.
-
bitcast behavior regarding NaNs is unexpected? gpuweb/gpuweb#4450, by @mwyrzykowski.
-
sampleCounts of 2, 8, and 16 gpuweb/gpuweb#4446, by @greggman.
-
Add anchors for WGSL language features. gpuweb/gpuweb#4440, by @Kangz.
-
Add subgroup barrier to subgroups proposal gpuweb/gpuweb#4437, by @raphlinus.
-
Is there an easier way to cast away an atomic specifier for subsequent shaders/dispatches? gpuweb/gpuweb#4436, by @kjpgit.
-
Some `Data Packing Built-in Functions` need more info gpuweb/gpuweb#4435, by @ben-clayton.
-
How to put barriers between dispatch() calls? gpuweb/gpuweb#4433, by @kjpgit.
-
-
17 issues closed.
-
for dispatchWorkgroups, the workgroupCountY and workgroupCountZ default to 1 gpuweb/gpuweb#4467.
-
Make Setting canvas.width and canvas.height lighter weight gpuweb/gpuweb#4388.
-
Link tooltips overlaps link in table gpuweb/gpuweb#4457.
-
Transfer a GPUTexture or GPUBuffer from one Worker to another. gpuweb/gpuweb#4244.
-
Must debug groups be balanced within a single command buffer? gpuweb/gpuweb#4172.
-
Per-pipeline binding points gpuweb/gpuweb#4459.
-
Timestamp Write Dictionaries are Identical for Render Passes and Compute Passes gpuweb/gpuweb#4455.
-
Conversion from AbstractInteger to i32 appears to be underspecified gpuweb/gpuweb#4415.
-
`AbstractInt` conversion ranking forces unfortunate compromise gpuweb/gpuweb#4291.
-
Weird behavior while using switch code block gpuweb/gpuweb#4430.
-
Limitation on the precision of hex floats? gpuweb/gpuweb#4350.
-
Occlusion Query - result type not specified gpuweb/gpuweb#4427.
-
Some `Data Packing Built-in Functions` need more info gpuweb/gpuweb#4435.
-
How to put barriers between dispatch() calls? gpuweb/gpuweb#4433.
-
-
47 issues received 94 new comments.
-
Make Setting canvas.width and canvas.height lighter weight gpuweb/gpuweb#4388 (7 comments by @greggman, @kdashg)
-
API should be exposed to ServiceWorker gpuweb/gpuweb#4197 (6 comments by @beaufortfrancois, @jimblandy, @kainino0x, @tqchen)
-
HW Clip Planes support gpuweb/gpuweb#390 (6 comments by @kdashg, @mwyrzykowski, @teoxoy, @unvestigate)
-
Validation for setViewport should be `minDepth <= maxDepth` gpuweb/gpuweb#4461 (5 comments by @kainino0x, @Kangz, @mwyrzykowski)
-
Why is dispatchWorkgroupsIndirect() running much slower than dispatchWorkgroups()? gpuweb/gpuweb#4460 (3 comments by @conorpo, @toji)
-
Compat: renderable float16, float32 textures gpuweb/gpuweb#4417 (3 comments by @greggman, @kdashg, @teoxoy)
-
Conversion from AbstractInteger to i32 appears to be underspecified gpuweb/gpuweb#4415 (3 comments by @dneto0, @kdashg, @mwyrzykowski)
-
add swizzles for scalar->vectors conversion too gpuweb/gpuweb#4428 (3 comments by @dj2, @greggman, @kdashg)
-
Assignment of swizzled values should be permitted gpuweb/gpuweb#737 (3 comments by @dneto0, @kainino0x, @myth0genesis)
-
Compat: Validate out storage texture formats that are not supported in GLES gpuweb/gpuweb#4413 (3 comments by @greggman, @SenorBlanco)
-
for dispatchWorkgroups, the workgroupCountY and workgroupCountZ default to 1 gpuweb/gpuweb#4467 (2 comments by @dneto0, @jrprice)
-
wgsl: can hex floats explicitly represent infinity and NaN as float literals? gpuweb/gpuweb#1769 (2 comments by @ben-clayton, @cybersoulK)
-
Investigation: Bindless resources gpuweb/gpuweb#380 (2 comments by @devshgraphicsprogramming, @feynon)
-
createTexture does not validate viewFormats against usage gpuweb/gpuweb#4426 (2 comments by @kdashg, @teoxoy)
-
Are "Fine" derivatives guaranteed to actually be fine? gpuweb/gpuweb#4325 (2 comments by @kainino0x, @kdashg)
-
[FEATURE REQUEST] Pretty-print utility for GPUBufferUsage and other bitflags gpuweb/gpuweb#4372 (2 comments by @greggman, @magcius)
-
Clarify whether `canvas.width = canvas.width` invalidates a WebGPU canvas gpuweb/gpuweb#4452 (1 comment by @greggman)
-
Transfer a GPUTexture or GPUBuffer from one Worker to another. gpuweb/gpuweb#4244 (1 comment by @kainino0x)
-
Must debug groups be balanced within a single command buffer? gpuweb/gpuweb#4172 (1 comment by @kainino0x)
-
API spec should highlight vec3f 16 byte alignment vs 12 byte size gpuweb/gpuweb#4192 (1 comment by @kainino0x)
-
Add a floored or euclidean modulo implementation to wgsl gpuweb/gpuweb#3987 (1 comment by @ch3coohlink)
-
Content timeline step (6) for copyExternalImageToTexture() is inaccurate gpuweb/gpuweb#4456 (1 comment by @kainino0x)
-
Timestamp Write Dictionaries are Identical for Render Passes and Compute Passes gpuweb/gpuweb#4455 (1 comment by @haoxli)
-
`AbstractInt` conversion ranking forces unfortunate compromise gpuweb/gpuweb#4291 (1 comment by @kdashg)
-
document ambiguity in tie-breaks for cube map sampling face selection gpuweb/gpuweb#4396 (1 comment by @kdashg)
-
Some `Data Packing Built-in Functions` need more info gpuweb/gpuweb#4435 (1 comment by @ben-clayton)
-
Surely it shouldn't take 5-15ms to get a trivial result back from GPU with mapAsync? gpuweb/gpuweb#4432 (1 comment by @gyagp)
-
-
21 pull requests opened.
-
Clarify behavior of out-of-gamut canvas gpuweb/gpuweb#4476, by @kainino0x.
-
Make setBindGroup non-throwing/non-allocating gpuweb/gpuweb#4475, by @kainino0x.
-
Define built-in constants and add two circle constants gpuweb/gpuweb#4473, by @mehmetoguzderin.
-
Infra related with outdated preview prune and preview extension gpuweb/gpuweb#4472, by @mehmetoguzderin.
-
Fix truncated line in "default pipeline layout" algorithm. gpuweb/gpuweb#4471, by @jimblandy.
-
Use "get the entry point" algorithm where necessary. gpuweb/gpuweb#4470, by @jimblandy.
-
Include GPUProgrammableStage validation in derived type validation. gpuweb/gpuweb#4469, by @jimblandy.
-
Fix typos in #4449. gpuweb/gpuweb#4466, by @jimblandy.
-
Add SharedWorker support gpuweb/gpuweb#4465, by @beaufortfrancois.
-
Add ServiceWorker support gpuweb/gpuweb#4464, by @beaufortfrancois.
-
[editorial] Fix "validating shader binding" argument name. gpuweb/gpuweb#4462, by @jimblandy.
-
Add LHS expr rules for * and & gpuweb/gpuweb#4454, by @alan-baker.
-
Remove ResizeObserver example gpuweb/gpuweb#4453, by @greggman.
-
Add scalar value constructors from abstracts gpuweb/gpuweb#4449, by @alan-baker.
-
Fix result type in occlusion query gpuweb/gpuweb#4447, by @haoxli.
-
Compatibility mode: depth bias clamp must be zero. gpuweb/gpuweb#4445, by @SenorBlanco.
-
Disallow sample_index builtin in Compatibility. gpuweb/gpuweb#4443, by @SenorBlanco.
-
Mention that sample_index causes per-sample fragment shader execution gpuweb/gpuweb#4442, by @greggman.
-
[editorial] fix character refs gpuweb/gpuweb#4441, by @alan-baker.
-
Update packed_4x8_integer_dot_product description gpuweb/gpuweb#4439, by @beaufortfrancois.
-
-
15 pull requests merged.
-
Include GPUProgrammableStage validation in derived type validation. gpuweb/gpuweb#4469.
-
Infra related with outdated preview prune and preview extension gpuweb/gpuweb#4472.
-
Fix typos in #4449. gpuweb/gpuweb#4466.
-
Update packed_4x8_integer_dot_product description gpuweb/gpuweb#4439.
-
Add Compat Limits to Proposal gpuweb/gpuweb#4397.
-
Add ServiceWorker support gpuweb/gpuweb#4464.
-
[editorial] Fix "validating shader binding" argument name. gpuweb/gpuweb#4462.
-
Remove ResizeObserver example gpuweb/gpuweb#4453.
-
Add LHS expr rules for * and & gpuweb/gpuweb#4454.
-
Compatibility mode: depth bias clamp must be zero. gpuweb/gpuweb#4445.
-
Add scalar value constructors from abstracts gpuweb/gpuweb#4449.
-
Disallow sample_index builtin in Compatibility. gpuweb/gpuweb#4443.
-
Fix result type in occlusion query gpuweb/gpuweb#4447.
-
[editorial] fix character refs gpuweb/gpuweb#4441.
-
-
13 pull requests received 42 new comments.
-
Add scalar value constructors from abstracts gpuweb/gpuweb#4449 (8 comments by @alan-baker, @ben-clayton, @dneto0, @mwyrzykowski)
-
Compatibility Mode: disallow two-component (RG) texture formats in storage texture bindings. gpuweb/gpuweb#4444 (5 comments by @greggman, @kainino0x, @SenorBlanco)
-
Add Compat interpolation validation gpuweb/gpuweb#4424 (5 comments by @greggman, @SenorBlanco, @teoxoy)
-
Define built-in constants and add two circle constants gpuweb/gpuweb#4473 (2 comments by @Kangz, @mehmetoguzderin)
-
Update packed_4x8_integer_dot_product description gpuweb/gpuweb#4439 (2 comments by @beaufortfrancois)
-
[editorial] Fix "validating shader binding" argument name. gpuweb/gpuweb#4462 (2 comments by @kainino0x)
-
Mention that sample_index causes per-sample fragment shader execution gpuweb/gpuweb#4442 (2 comments by @greggman, @Kangz)
-
Explain the endianness of pack4xI8* and pack4xU8* builtins gpuweb/gpuweb#4425 (1 comment by @ben-clayton)