Commit Graph
11529 Commits
Author SHA1 Message Date
teoxoy 5dfcef2ade [naga] add name_reuse test case 2026-10-27 11:30:28 +00:00
teoxoy 2b206a0321 [msl] Hoist body temporaries used by continuing blocks
The MSL backend emits a loop's `continuing` block at the top of the
`while` loop, above the body. Values the body computed and `continuing`
referenced were then either re-evaluated in place, reading whatever
state the body had left behind, or, for statement results such as call
results, hit an `unreachable!` panic.

Now, when a loop has a `continuing` block or `break_if`, the
temporaries it needs from the body are declared ahead of the loop, and
the body assigns to them instead of declaring them.
2026-10-27 11:30:28 +00:00
teoxoy f788839ec1 [naga] add loop continuing block tests 2026-10-27 11:30:28 +00:00
teoxoy f0c49fbe8e [msl] use expression index as name for statement results in remaining
places
2026-10-27 11:30:28 +00:00
Sam 93cd1193c2 [remote-types] Add derives and reprTransparent on MapMode (#10752) 2026-10-27 09:21:50 +00:00
Dmitry Samoylenko 466dc0ba08 perf(metal): Use Arcs rather than copying VertexBufferMappings (#10630) 2026-10-27 00:53:37 +00:00
Andy Leiserson 112b3af5ed chore(ci): Increase timeout for wasm test job (#10744) 2026-10-27 00:04:06 +00:00
Jim Blandy ad5a70aea1 [core] Validate shader stages after applying pipeline overrides.
The WebGPU specification requires pipeline creation to validate
properties of shader stages, like workgroup sizes, that may depend on
the values of pipeline-overridable constants. Previously, wgpu-core
validated stages using an `Interface` built from the module before
overrides were applied, with results contrary to spec. Override
processing was then carried out by `wgpu_hal`, after all validation
had supposedly been taken care of.

This commit has `create_shader_module` save the validated Naga module
instead of creating a `wgpu_hal` shader module. At pipeline creation
time, `StageCompiler::compile` now applies the stage's override values
with `process_overrides`, validates the stage using an `Interface`
built from the result, and, if that validation succeeds, returns a
`CompiledStage` holding the `wgpu_hal` `ShaderModule` created from it.
Dropping the `CompiledStage` destroys that module once the pipeline
has been created.

Since the override values have already been applied, we pass
`wgpu_hal` empty sets of constants for these modules.

Since the `naga::back::pipeline_constants` module is now used by
`wgpu_core` rather than `wgpu_hal` backends, it can't be conditionally
available based on Naga's backend features. Include it
unconditionally.

This change removes `Interface::finalize_entry_point_name`, since we
no longer build an `Interface` at shader module creation. Instead, we
choose entry points from the `naga::Module` directly.

Add new tests: `tests/tests/wgpu-validation/api/overrides.rs`

Fixes #9774.
2026-10-26 23:42:33 +00:00
Jim Blandy e136c379cb [core] Validate pipeline stages with a shared StageCompiler helper.
Introduce `StageCompiler`, in the new `wgpu_core::device::stage_compiler`
module, which performs the per-stage work that
`create_compute_pipeline_or_error_inner` and
`create_render_pipeline_or_error_inner` previously repeated for each
stage: checking the shader module's validity and device, choosing
the entry point, checking the stage against the module's interface,
and tracking which stages were validated.

`StageCompiler::compile` returns a `CompiledStage`, which borrows the
module the stage should use, and records the entry point name. From
this we can build a `hal::ProgrammableStage` to use when creating the
hal pipeline.

Have `StageCompiler::compile` reject a passthrough stage immediately
if the pipeline's layout is being derived. `ImplicitLayoutError::Passthrough`
now names only that stage, rather than every passthrough stage in the
pipeline.

Remove `ImplicitLayoutError::ReflectionError`, along with the check
that reported it. The condition it was concerned with, deriving a
layout from a stage we couldn't reflect, is already rejected as
`ImplicitLayoutError::Passthrough`. The only way left to reach it was
a bug: for mesh pipelines without a fragment stage, it expected the
last stage to be a vertex stage, and so rejected them incorrectly.

In this commit, `StageCompiler` only does validation, but by the end
of this pull request, it will be compiling shader modules. Having its
name overstate its role for one commit seems clearer than introducing
everything in one commit and renaming it in the next. The comments,
however, are accurate in each commit.

This commit also changes compute pipelines to report errors in the
same way as render pipelines: an invalid shader module is now reported
as a `CreatePipelineError::Stage` error, and the shader module is
checked after the pipeline layout, not before.
2026-10-26 23:42:33 +00:00
Jim Blandy 78917c587b [core] Introduce validation::PipelineResources.
Gather the state that `Interface::check_stage` accumulates across all
of a pipeline's stages (the bind group layouts, minimum binding sizes,
and immediate data usage) into a new `PipelineResources` struct, which
`check_stage` takes by `&mut`.

Remove `immediates` from `StageIo`, so that `StageIo` holds only the
information one stage passes to the next.

Use `layout` to refer to the `BindingLayoutSource`, not `layouts`,
since that value describes a single pipeline layout. In the derived
case, it does hold an array of bind group layouts, but that is only
one variant; the value's name should make sense regardless which
variant is present.
2026-10-26 23:42:33 +00:00
Jim Blandy 6c6fb05421 [core] create_render_pipeline_or_error_inner: Tidy temporary vars.
In `wgpu_core/src/device/resource.rs`, in the function
`Device::create_render_pipeline_or_error_inner`, remove the leading
`_` from the following local variables' names, and simply leave them
uninitialized in their declaration:

- `_vertex_entry_point_name`
- `_task_entry_point_name`
- `_mesh_entry_point_name`

Rust is smart enough to see that these are always initialized before
they are used, and thus don't need to be initialized at their
declaration.
2026-10-26 23:42:33 +00:00
Jim Blandy 4f4581fdff [core] Combine ShaderModuleState with ShaderMetaData.
In `wgpu_core::pipeline`, delete `ShaderMetaData` and turn
`ShaderModuleState` into the equivalent enum, giving both the
`NagaModule` and `Passthrough` variants their own `raw` field.

Arrange for uses of `ShaderModuleState` to extract the raw `wgpu_hal`
module in the `match` expression that handles the two variants.

In isolation, this change looks like mere rearrangement, not a
cleanup, but hopefully either way it's easy to see that it has no
effect on behavior. Subsequent commits will remove `raw` from the
`NagaModule` variant; this refactor will allow that to affect only the
`NagaModule` `match` arms, leaving the passthrough code unaffected.
2026-10-26 23:42:33 +00:00
Jim Blandy 00583c146b [core] Move ShaderMetaData, PassthroughInterface to pipeline.rs.
Move the definitions of `ShaderMetaData` and `PassthroughInterface`
from `wgpu_core::validation`, which does not use them, to
`wgpu_core::pipeline`, where they are used in `ShaderModuleState`, and
created.
2026-10-26 23:42:33 +00:00
Jim Blandy 48b67d4826 [core] Refactor ShaderMetaData use sites to use match.
Slightly reorder the sites that consume `ShaderMetaData` values, so
that the code for the `NagaModule` and `Passthrough` cases can be
gathered into `match` statements.
2026-10-26 23:42:33 +00:00
Jim Blandy 69863648a8 [core] Turn ShaderMetaData into a Naga-vs.-passthrough enum.
Rename the `ShaderMetaData::Interface` variant to `NagaModule`,
and make it a struct-like variant.

Delete `ShaderMetaData::interface`, and update its call sites to use a
`match` expression directly on the enum instead. This makes the use
sites clearer, by making it apparent which style of module we're
processing.
2026-10-26 23:42:33 +00:00
renovate[bot] 7f5c0714ff Update cts digest to b009414 (#10719)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2026-10-26 22:43:21 +00:00
sagudev 0c7c12193a [deno] Minimize OpState borrow
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 319adfbcaa webgpu:api,validation,buffer,mapping:* now pass
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 401fc01b7b [deno] Fix substract overflow
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev f2f0acb748 webgpu:api,validation,buffer,mapping:mapAsync,* now pass
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 3208fe8c94 [deno] Ensure map state is still pending after defer
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 2eec736394 [deno] Do poll after Device::destroy to call
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev e04cd38d59 [deno] Call GPUBuffer::unmap procedure on GPUBuffer::destroy
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev cea73e5dac [deno] proper deferred mapAsync promise resolution
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 06dc3fe338 [deno] Check buffer state in content timeline
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev e12b2c27f1 [deno] Fixup DOMException builders
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev a8cfdef04e [core] Support passing MapMode as (potentially invalid) bitflags
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev ba46bd1452 [deno] Map BufferErrors to Abort/Operation as in spec
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 9e80fc9d1d [core] Do device validity check at the start of try_map_async
this is important as we need to return Abort error instead of Operation error which is raised on invalid buffer

Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
sagudev 4e98175c04 [core] Destruct BufferMappingOperation into mode and callback arguments
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-26 18:49:57 +00:00
John DelaneyandAndreas Reich 100c142937 Backend Caching of Heavy Requests (#10513)
Co-authored-by: Andreas Reich <r_andreas2@web.de>
2026-10-25 12:53:00 +00:00
Douglas DwyerandInner Daemons 322168c778 fix(wgpu): Make TextureBlitter respect sample_type when downsampling (#10657)
Co-authored-by: Inner Daemons <85136135+inner-daemons@users.noreply.github.com>
2026-10-25 12:33:50 +00:00
JMS55 2edafa5240 Record 1 barrier for N Queue::write_buffer/texture calls by deferring to submit time (#10319) 2026-10-25 11:57:55 +00:00
Andy Leiserson 63e505290e fix(core): Check that clear_texture targets all aspects (#10684) 2026-10-22 17:16:48 +00:00
renovate[bot] bbb9b06e90 Update Cargo.lock (#10528)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2026-10-22 15:50:16 +00:00
Sam 4fb3ccaa77 [core] Do not forget to fire user callback in configure (#10686)
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2026-10-22 06:42:02 +00:00
Andy Leiserson 09a0eba3aa fix(metal): Update sampler binding array limit from Metal docs (#10649) 2026-10-22 06:37:24 +00:00
Horu c0c938bb0b Discard unused StagingBelt's free chunks (#10383) 2026-10-22 06:35:37 +00:00
Andy Leiserson 4e3916783a Remove wgpu features for indicating Vulkan extension availability (#10635) 2026-10-22 06:20:43 +00:00
Andy Leiserson ade0590d81 Revert "Fix merge queue merging non-approved commits (#10552)" (#10691) 2026-10-22 05:01:14 +00:00
Inner DaemonsandAndy Leiserson 0f0262acdd Correctly zero init multiplanar textures (#9700)
Co-authored-by: Andy Leiserson <aleiserson@mozilla.com>
2026-10-22 00:11:46 +00:00
renovate[bot] 8e4718d5a1 Update zizmorcore/zizmor-action action to v0.7.1 (#10669)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2026-10-21 11:55:19 +00:00
Jim Blandy ccd6966eed [core] Clean up Interface::finalize_entry_point_name. (#10556) 2026-10-21 10:03:11 +00:00
Jim Blandy 28a16856de [core] Include the shader stage in MissingEntryPoint errors. (#10544) 2026-10-21 09:23:00 +00:00
Jim Blandy 3050396df2 [core] Combine Create{Compute,Render}PipelineError. (#10569) 2026-10-20 19:51:32 +00:00
renovate[bot] 4ebd7ca482 chore(deps): update cts digest to 9c85c65 (#10665)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2026-10-20 18:56:41 +00:00
Lexo Liu 6511b8f2da Revert "vulkan: work around Arm drivers ignoring negative viewport heights" (#10674) 2026-10-20 18:15:13 +00:00
Simon Hausmann 05ae83af5e fix(metal): permit acquiring drawables until the window has been visible (#10302) 2026-10-20 17:38:14 +00:00
Andy Leiserson 0342e2fba5 fix(core): Check same-device for external texture planes (#10636) 2026-10-20 16:56:33 +00:00
Abir Deol e084b66aab [naga spv-out] Declare LocalInvocationIndex once in mesh entry points (#10497) 2026-10-18 19:01:10 +00:00