This PR removes optional backend features in favor of run-time
selection at the `request_adapter` entry point. Adapters from all
supported gpu backends on the platform are reported.
It also makes gfx-rs dependency to be public (for now).
This change merges `Id` with `GenericId` and abstracts away its contents, exposing itself as just u64. This will allow us to encode the backend type into it, in addition to index and epoch.
It also refactors `TypedId` to have a single function extracting all the fields, since this is how it's used all the times anyway.
307: Add cubemap hint to multiple of 6 layers textures r=kvark a=LaylConway
This is an attempt at addressing #306. Vulkan has specific requirements on what can and can't be marked with the CUBE hint.
Co-authored-by: Layl <2385329-layl@users.noreply.gitlab.com>
302: Fixed pipeline barriers that are not transitions r=grovesNL a=kvark
The actual fix is a one-liner: `u.start != u.end` bit in `PendingTransition::record`. The case is relatively new - as of #281, which I haven't tested extensively.
The PR also improves our logging for further assistance with similar issues... but the most annoying piece is that I would find this much much earlier if I didn't ignore the result here: `let _ = state.change(...)`. Let it be the lesson to all of us :)
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
301: Raw window handle support instead of winit r=grovesNL a=kvark
This removes `winit` and paves the way for https://github.com/gfx-rs/wgpu-rs/issues/64
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
299: Rename buffer to encoder in wgpu_command_buffer_copy_* functions r=kvark a=yanchith
This also regenerates `wgpu.h` and updates uses of `wgpu_command_buffer_copy_buffer_to_buffer` in `compute/main.c`.
EDIT: all good...
~Note that I couldn't regenerate `wgpu-remote.h` because of the following cbindgen error:~
```
ERROR: Cannot find a mangling for generic path GenericPath { path: Path { name: "Adapter" }, export_name: "Adapter", generics: [Path(GenericPath { path: Path { name: "Backend" }, export_name: "Backend", generics: [], ctype: None })], ctype: None }. This usually means that a type referenced by this generic was incompatible or not found.
```
Should we investigate before merging? Maybe I am just using an unlucky nightly?
Closes: #250
Co-authored-by: yanchith <yanchi.toth@gmail.com>
297: Update rendy-memory and rendy-descriptor to 0.4 from crates.io r=kvark a=yanchith
I also locally tested for regressions in `wgpu-rs` on the metal backend (only have macOS today) and the examples seemed ok.
Co-authored-by: yanchith <yanchi.toth@gmail.com>
291: Fix sample_mask field r=kvark a=rukai
Looks like it was just missed in https://github.com/gfx-rs/wgpu/pull/285
Co-authored-by: Rukai <rubickent@gmail.com>
285: Update API according to the upstream spec r=grovesNL a=kvark
Includes changes in texture view creation, enumeration mapping,
binding types, and more.
This would be a nice target for our 0.3 release.
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
284: Add start_slot parameter r=kvark a=danaugrs
A first attempt to fix https://github.com/gfx-rs/wgpu/issues/215.
Co-authored-by: Daniel Salvadori <danaugrs@gmail.com>
280: Remove pod add read_spirv r=kvark a=rukai
Looks like Pod is no longer used by wgpu-rs so I removed it.
read_spirv is needed in wgpu-rs now so I added it.
Co-authored-by: Rukai <rubickent@gmail.com>
267: Use extend_from_slice instead of copy_from_slice r=kvark a=kocsis1david
`copy_from_slice` gives an error because it expects the lengths to be the same.
Co-authored-by: David <kocsis1david@windowslive.com>
