1609: [Vulkan] Initialize wgpu objects from raw handles r=kvark a=zarik5
**Connections**
This PR is a successor of https://github.com/gfx-rs/gfx/pull/3762
**Description**
The `handle_is_external` flag mechanism was not enough to ensure safety, when for example the Instance is wrapped in `Arc<>` and a library cannot keep track of all its clones. This is the case with the bevy render rewrite. The solution was to let wgpu be in charge of destroying the handles, but it also has to keep a reference to a "drop guard" which is always dropped after the handle is destroyed. For the OpenXR integration, this drop guard will be `xr::Instance`.
For now this is just a proof of concept. Only instance creation is handled, and there is even type error in `wgc::Instance::from_hal()`.
I have a few concerns:
* Is it ok to expose `hal::Instance` from the wgpu crate? Or should the user pass all the parameters so `hal::Instance` can be constructed internally? This second options is more disruptive, since the wgpu-types crate would need to import all platform-specific crates to define the structure/enum to hold the raw handles.
* Without counting the call to create the `hal::Instance`, there are 4 indirection calls to save the raw instance. Can this be optimized in any way?
Do you think it is possible to merge wgpu-hal into wgpu-core? This could help with reducing the distance from the surface level API to the platform specific APIs even more.
**Testing**
Vulkan/Android (Oculus Quest) using [this sample](https://github.com/zarik5/openxrs/blob/wgpu-test/openxr/examples/vulkan.rs).
Co-authored-by: Riccardo Zaglia <riccardo.zaglia5@gmail.com>
1607: Fix Downlevel Vertex Stage Storage Buffer Check r=kvark a=cwfitzgerald
**Connections**
Fixes a bug in #1599. Also follows up on #1583 marking all of hello-compute as a failure.
**Testing**
Tests now pass on rpi4.
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
1599: Fix Limits for lava/llvmpipe and re-enable and re-work CI r=kvark,groves,wumpf a=cwfitzgerald
**Connections**
No longer fixes#1551.
**Description**
This PR has a couple things going on at once.
- Fixes limits for llvmpipe, lavapipe, and rpi4.
- Added a downlevel limit to express that RPI4 does not allow storage buffers in vertex shaders on GL.
- Added a `Limits::downlevel_default()` that takes minimum limits from GLES and adds some more documentation on how to choose limits.
- Moved all examples to the new downlevel limits
- Reworks CI to re-enable software testing and unify instructions.
**Testing**
It is
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
DYNAMIC_INDEXING is checked by default, since WGPU doesn't allow
for any kind of specialization constants and constant indexing is
nigh useless.
STORAGE_RESOURCE_BINDING_ARRAY is enabled iff the device supports:
- both or neither of uniform and buffer arrays with dynamic indexing;
- both or neither of sampled and storage images with dynamic indexing.
NONUNIFORM_INDEXING is enabled iff for ALL types of descriptor arrays
*that are supported for dynamic indexing*, nonuniform indexing is
also allowed.
These flags have a limitation in that some platforms
(eg.
https://vulkan.gpuinfo.org/displayreport.php?id=11692#features_core_12)
may support a strange subset of those features (example device
supporting non-uniform indexing for textures and storage buffers,
but NOT for storage textures or uniform buffers). In that case feature
will be reported as missing, even though it is partially present.
In the author's opinion, this flag set is convenient for the
users to query; however, due to aforementioned limitations, it would be
desirable to obtain statistics about "edge-case" platforms and what
percentage of reports they comprise.
1562: Fix check for QUERY_SET_MAX_QUERIES was using >= instead of > r=cwfitzgerald a=HalfVoxel
Fixes#1560
Co-authored-by: Aron Granberg <aron.granberg@gmail.com>
1525: Add Naga bypass to allow feeding raw SPIR-V shader data to the backend. r=kvark a=ElectronicRU
**Connections**
Fixes#1520 .
**Description**
While Naga checking is undoubtedly very useful, it currently lags behind
what is possible in SPIR-V and even what is promised by WGPU (ie binding
arrays). This adds an unsafe method to wgpu::Device to allow feeding
raw SPIR-V data into the backend, and adds a feature flag to request a
backend supporting this operation.
**Testing**
`texture_arrays` example is runnable now, which uncovers an additional bug in Vulkan backend - binding arrays are instead treated like sole bindings, and indices shift, too. Lots of errors from validation layer ensue.
Co-authored-by: Alex S <alex0player@gmail.com>
- Remove Cow from wgpu-hal API surface. You graze on in our hearts.
- Rename feature flag to SPIRV_SHADER_PASSTHROUGH in anticipation
of crate feature `spirv` to reduce confusion.
- Add some more documentation about behaviour and intended avenues of
use to the feature.
1528: Support TEXTURE_SPECIFIC_FORMAT_CAPABILITIES for rendertargets. r=kvark a=ElectronicRU
**Description**
With TEXTURE_SPECIFIC_FORMAT_FEATURES enabled, one would assume that render pipeline creation would use extended features
for color/depth-stencil render targets. However, it does not. This aims to fix that discrepancy.
Co-authored-by: Alex S <alex0player@gmail.com>
- rename ShaderInput variants
- rename feature flag
- hard error on Metal backend trying to compile SPIR-V
- fix test failing because of feature flag bits changing