The purpose of the PR is to support Naga modules everywhere.
As a requirement, it updates the gfx-rs version used.
Most of the logic is dedicated towards building a shader interface,
where previously we just used naga's IR. Now we have our own mini-IR.
659: Temporarily skip debug labels for web backend r=kvark a=grovesNL
Debug labels aren't currently supported by gecko. We can re-enable these later.
Co-authored-by: Joshua Groves <josh@joshgroves.com>
655: Update to gfx from master r=kvark a=kvark
Depends on https://github.com/gfx-rs/wgpu/pull/1057
Expect the unexpected wrt the gpu-descriptor, which hasn't been tested much!
Also refactors the CI a bit to reduce the number of jobs by moving out `wasm` into a separate job.
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
1057: Move to gfx master r=cwfitzgerald a=kvark
**Connections**
Fixes#1027
**Description**
Switches us to gfx dependency on github.
Note: there is problem with libc dependency - https://github.com/timothee-haudebourg/khronos-egl/issues/7 - that would need to be resolved before we get this version of the code into either Gecko or Servo.
**Testing**
Ran wgpu-rs examples
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
654: Use web-sys release instead of patch r=kvark a=grovesNL
`patch` isn't necessary anymore because there's a new release of web-sys and wasm-bindgen
Fixes#637
Co-authored-by: Joshua Groves <josh@joshgroves.com>
1056: Replace gfx-descriptor by gpu-descriptor r=cwfitzgerald a=kvark
**Connections**
Unblocks #1027
cc @zakarumych
**Description**
Removes the last link we had between us and gfx-rs.
**Testing**
Tested on wgpu-rs
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
1054: Flush write_texture staging memory r=kvark a=kvark
**Connections**
Fixes part of https://github.com/gfx-rs/wgpu-rs/issues/648
**Description**
Adds the missing flush to the staging buffer in `write_texture`
**Testing**
Untested
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
Used thiserror for the error enum
Moved the contents of the error module to the main module
Made the error an actual enum and not just a string
Splitted the write function on the feature manager into a function for
checking for feature availability
Splitted the backend into two more modules (features, errors)
Added a shorthand type for the backend Result
Reorganized function ordering
Moved from using a struct to write image dimension to a function
Started documenting the backend
1051: Register init-destination buffer in the pending writes r=kvark a=kvark
**Connections**
Fixes#1049
**Description**
At some recent point we introduced the map of destination buffers for copies, and I missed the case where buffers are mapped at creation.
**Testing**
Tested on the test-case in #1049
Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>
1050: player: use the wgpu-cores's gfx-select macro r=kvark a=kvark
**Connections**
Looks like I broke the player in #1034, surprised we didn't notice, and that `cargo test` still passed. The problem was having the `Empty` backend for the device.
**Description**
Fixes the `device` backend in the player, switches the backend set to be `PRIMARY`, and re-uses the `wgc::gfx-select`.
**Testing**
Replaying a trace from #1049
Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>