* Initial commit
* Other initial changes
* Updated shader snapshots
* Added new HLSL limitation
* Moved error to global variable error
* Added docs to per_primitive
* Added a little bit more docs here and there in IR
* Adding validation to ensure that task shaders have a task payload
* Updated spec to reflect the change to payload variables
* Updated the mesh shading spec because it was goofy
* some doc tweaks
* Tried to clarify docs a little
* Tried to update spec
* Removed a warning
* Addressed comment about docs mistake
* Review in progress
- Extensive revisions to `docs/api-specs/mesh_shading.md`.
- Doc comments.
- Ensure `Module` stays at the bottom of `ir/mod.rs`.
- Avoid a clone.
- Rename some arguments to be more specific.
- Minor readability tweaks.
* mesh_shading.md: more tweaks
* Ran cargo fmt
* Small tweaks
* [naga] Move definition of `ShaderStage::compute_like` to `proc`.
Move the definition of `naga::ShaderStage::compute_like` from
`naga::ir` into `naga::proc`. We generally want ot keep methods out of
`naga::ir`, since the IR itself is complicated enough already.
* Replace TODO comment with followup issue.
* Update analyzer.rs
Co-authored-by: Jim Blandy <jimb@red-bean.com>
* Removed stuff in accordance with Jim's recommendation
* minor changes for readability
* Pull mesh shader output type validation out into its own function.
* doc fixes
* remove duplicated task payload validation
* Quick little changes
* Another quick fix
* Quick fix
* Removed unnecessary TODO statement
* A
* Tried to be more expressive
* Made functions only work in mesh shader entry points
* Various validation fix attempts
* Undid capabilities resize
* WGSL PR is up :)
* Update naga/src/ir/mod.rs
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
* Update naga/src/front/wgsl/error.rs
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
* Update naga/src/ir/mod.rs
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
* Other Erich suggestion
* Updated docs & validation for some builtins
* Added some docs & removed contentious "// TODO"
* Fixed bad validation, formatted mesh shader wgsl
* Rewrote the IR and parser significantly
* Improved validation slightly, remvoed obselete crap, fixed bug in compaction, made clippy happy
* Added changelog entry
* Made parser respect enable extension
* Updated mesh shader spec
* Cleaned up the mesh shader analyzer function
* Updated changelog
* Updated validation from spv-write
* Slight tweaks
* Added capabilitiy thing blah blah blah
* Fixed compaction & clean stuff up
* Fixed compaction and such pt 2
* Paved the way for primitive-index stuff
* Little bit of cleanup
* Added new features/capabilities for point primitives and added snapshots for stuff
* Updated spec to rename something
* Updated some limits stuff
* Made some improvements
* Fixed doctests & webgpu stuff
* Fixed experimental features stuff
* A little more scaffolding
* Added some initial validation
* Added per primitive validation
* Improved per primitive validation
* Added some more validation
* Fixed primitive index validation
* Updated the spec
* Updated comments on some limits
* Made some fixes
* Fixed new validation rule for task payload size
* Update wgpu-core/src/validation.rs
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
* Update wgpu-core/src/validation.rs
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
* A small tweak to mesh shader recommended minimums
* Addressed 2 comments
* A few more comments addressed with limits
* Addressed more comments
* Update wgpu-core/src/validation.rs
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
* Split up error
* Addressed another comment by changing workgroup size validation
* Final comment addressed
* Update wgpu-info/src/human.rs
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update wgpu-core/src/validation.rs
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update wgpu-info/src/human.rs
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Fixed compilation and added some new test functions
---------
Co-authored-by: SupaMaggie70Incorporated <85136135+SupaMaggie70Incorporated@users.noreply.github.com>
Co-authored-by: Jim Blandy <jimb@red-bean.com>
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
The plumbing to get it out of `wgpu_core` is awkward but I didn’t see
a clearly better option. I think it would make sense to have something
like `Global::get_label<T>(id: Id<T>) -> String` but that looks like a
lot more work.
This allows us to save some work when a bind group is replaced,
although WebGPU still requires us to fail a submission if any
resource in any bind group that was ever referenced during
encoding is destroyed.
Fixes#8399
* Don't remove resources from the usage scope, it is not necessary, and
not useful, since we don't verify the scope is empty afterwards.
* Avoid some unnecessary `unsafe`.
In the future `transfer` will want to use the `Arc` versions of these
types. I have exported them from `wgpu_core::command::ffi`, to document
that these are used for FFI purposes, and from `wgpu_core::command`, for
backwards compatibility, although I also change the in-tree uses to
use them from `wgpu_types` instead of from `wgpu_core::command`.
It is emptied by `reset_queries` at the end of every render pass, so
it's just keeping an allocation alive, not holding any state. It seems
unlikely that there is sufficient performance gain from reusing the
memory allocation to justify the complexity of additional state at
higher layers.
Previously, the check was skipped if the copy was a single row, which is
not correct. The check should be made whenever bytes_per_row is
specified. It is permissible not to specify bytes_per_row if the copy is
a single row, but if it is specified, it must be aligned.
Also removes a redundant check of the `offset` alignment.
Since the offset and bytesPerRow alignment checks are not part of
"validating linear texture data", I chose to remove that instance of
them. These checks are now in `validate_texture_buffer_copy`, which
does not correspond 1:1 with the spec, but has a comment explaining how
it does correspond.
TransferError now has separate variants for texture copy formats that
are only forbidden in combination with specific aspects
(CopyFrom/ToForbiddenTextureFormatAspect), and texture copy formats that
are always forbidden, irrespective of the aspect
(CopyFrom/ToForbiddenTextureFormat).
This produces a less confusing error message by not mentioning the
aspect it is not relevant.
P010 is a 4:2:0 chroma subsampled planar format, similar to NV12. Each
component uses 16 bits of storage, of which only the high 10 bits are
used. On DX12 this maps to DXGI_FORMAT_P010, and on Vulkan this maps to
G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16.
The existing "nv12" gpu test module has been renamed to
"planar_texture", and a new test P010_TEXTURE_CREATION_SAMPLING has
been added similar to the existing NV12_TEXTURE_CREATION_SAMPLING. The
remaining tests in this module have been converted to validation tests,
and now test both NV12 and P010 formats.