Commit Graph
3491 Commits
Author SHA1 Message Date
bors[bot]andGordon-F b7ba481a91 Merge #1282
1282: Update naga to gfx-18 r=kvark a=Gordon-F

Same as https://github.com/gfx-rs/gfx/pull/3695. Just waiting for new naga relese.

Co-authored-by: Gordon-F <ishaposhnik@icloud.com>
2021-03-25 14:01:12 +00:00
Gordon-F c1f8203f3c Update naga to gfx-18 2021-03-25 16:59:16 +03:00
bors[bot]andDzmitry Malyshau d8cbb8615a [rs] Merge #813
813: Fix dropping for command encoders r=grovesNL a=kvark

We weren't dropping them properly, and since the ID is just an obfuscated `Box<RealStuff>` that made the contents leak.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-03-25 03:29:02 +00:00
Dzmitry Malyshau 17e95abfe0 [rs] Fix dropping for command encoders 2021-03-24 22:34:00 -04:00
bors[bot]andJoshua Groves e5dfafb3cc Merge #1281
1281: Rename `level_count` to `mip_level_count` r=kvark a=grovesNL

**Description**
This PR renames `level_count` to `mip_level_count` on `TextureViewDescriptor`. This follows the naming in the specification, located at https://gpuweb.github.io/gpuweb/#texture-view-creation

**Testing**
`cargo build` and `cargo test` only, naming changes will be made downstream in wgpu-rs too.

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-24 14:56:00 +00:00
Joshua Groves 9885bdf320 Rename level_count to mip_level_count
This follows the naming in the specification, located at
https://gpuweb.github.io/gpuweb/#texture-view-creation
2021-03-24 09:14:30 -02:30
bors[bot]andJoshua Groves 95aa26dc60 [rs] Merge #811
811: Update to latest wgpu-{core,types} r=kvark a=grovesNL



Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-24 02:43:47 +00:00
Joshua Groves e65ef710a5 [rs] Update to latest wgpu-{core,types} 2021-03-24 00:00:16 -02:30
Pâris DOUADY 39e1835f72 [rs] Passthrough cross feature to wgpu-core (#810)
* Passthrough cross feature to wgpu-core

Hopefully it can even be removed when `naga` is more mature. :-)

* remove unconditional cross for wgc
2021-03-23 13:01:01 -04:00
bors[bot]andDzmitry Malyshau 9c98ebc48c Merge #1276
1276: Align create_swap_chain function signature r=cwfitzgerald a=kvark

**Connections**
Follow-up to #1034

**Description**
Swapchains are a bit special. Browsers don't use wgpu-core's swapchains, so we never bothered to convert them into the error model. But we still need this for Deno and webgpu-headers.

**Testing**
Untested, but also harmless

Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>
2021-03-23 14:39:57 +00:00
Dzmitry Malyshau f953266bbe Align create_swap_chain function signature 2021-03-23 10:39:13 -04:00
bors[bot]andDan Wilhelm 56465df92b [rs] Merge #809
809: Add overflow detection to hello-compute r=kvark a=danwilhelm

Description
The `hello-compute` example sometimes gives incorrect results if an intermediate Collatz value overflows. 

Proposed Solution
We can detect overflows before they occur, since the `3n+1` step is the only possible overflow. (WGSL only supports u32.)

Example
`cargo run --example hello-compute 77031 837799 8400511 63728127` 
> Incorrect Current Output: `[350, 524, 312, 346]`
> Correct Output: `[350, 524, 685, 949]`
> Proposed Solution: `[350, 524, OVERFLOW, OVERFLOW]`

Notes
- The `hello-compute` example appears to be [borrowed from naga](https://github.com/gfx-rs/naga/blob/master/tests/in/collatz.wgsl), so this would cause the implementations to differ.
- Hexadecimal literals are not yet implemented in naga for WGSL shaders, so the shader literals are not as pretty as they could be.
- Revised terminology `times` -> `steps` to be consistent with [Wikipedia](https://en.wikipedia.org/wiki/Collatz_conjecture).


Co-authored-by: Dan Wilhelm <dan@danwilhelm.com>
2021-03-23 14:32:47 +00:00
bors[bot]andJoshua Groves 2bc5526fbe Merge #1280
1280: Rename color/depth attachments to match spec r=kvark a=grovesNL

**Description**
- Rename `ColorAttachmentDescriptor` to `RenderPassColorAttachment` (https://gpuweb.github.io/gpuweb/#color-attachments)
- Rename `DepthStencilAttachmentDescriptor` to `RenderPassDepthStencilAttachment` (https://gpuweb.github.io/gpuweb/#depth-stencil-attachments)
- Rename `attachment` fields on both attachments to `view`

**Testing**
None (just `cargo build`) because the renaming downstream should be trivial.

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-23 14:27:52 +00:00
Joshua Groves a5693ab2a3 Rename color/depth attachments to match spec
- Rename `ColorAttachmentDescriptor` to `RenderPassColorAttachment`
- Rename `DepthStencilAttachmentDescriptor` to `RenderPassDepthStencilAttachment`
- Rename `attachment` fields on both attachments to `view`
2021-03-23 09:31:59 -02:30
Dan Wilhelm 985c1feb4e [rs] Add overflow detection to hello-compute 2021-03-23 01:49:39 -07:00
bors[bot]andJoshua Groves fb0288a377 Merge #1279
1279: Fix some typos r=kvark a=grovesNL

**Description**
Fix a few small typos in doc comments

**Testing**
None

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-23 06:44:11 +00:00
bors[bot]andJoshua Groves b90199bec3 Merge #1278
1278: Rename copy views and layout to match spec r=kvark a=grovesNL

**Description**
Match the specification naming changes from https://github.com/gpuweb/gpuweb/pull/1375

**Testing**
None because the changes downstream to wgpu-rs and wgpu-native should be trivial (renaming only)

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-23 03:41:46 +00:00
Joshua Groves 306eaf3969 Fix some typos 2021-03-23 01:01:32 -02:30
Joshua Groves 9058f3ba62 Rename copy views and layout to match spec 2021-03-23 00:48:42 -02:30
bors[bot]andJoshua Groves d18cc67131 [rs] Merge #785
785: Update to latest web-sys bindings r=kvark a=grovesNL

We can wait for https://github.com/rustwasm/wasm-bindgen/pull/2482 first

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-03-23 02:56:58 +00:00
Joshua Groves 76f260c7c6 [rs] Update to latest web-sys bindings 2021-03-23 00:22:30 -02:30
bors[bot]andAndreas Reich 8dd9bd8af9 Merge #1275
1275: rows_per_image & rows_per_image are now optional r=kvark a=Wumpf

Fixes #988
Was surprisingly hard, hope it's correct; testing could be more extensive...
Ideas for more elegant handling very welcome.

**Testing**
Tested against wgpu-rs samples


Co-authored-by: Andreas Reich <r_andreas2@web.de>
2021-03-22 02:36:36 +00:00
Andreas Reich 720c73fa9c Review feedback / fix validate_linear_texture_data 2021-03-21 09:01:31 +01:00
bors[bot]andHiroaki Yutani f11765de4d [rs] Merge #806
806: Remove "how to compile the shaders" section on README r=kvark a=yutannihilation

I believe all the examples now use WGSL, so this section is no longer needed.

Co-authored-by: Hiroaki Yutani <yutani.ini@gmail.com>
2021-03-21 02:29:25 +00:00
Hiroaki Yutani 20a903726a [rs] Remove "how to compile the shaders" section on README
I believe all the examples now use WGSL, so this section is no longer needed.
2021-03-20 23:58:30 +09:00
Andreas Reich 50043875e5 rows_per_image & rows_per_image are now optional
Fixes #988
2021-03-20 12:33:28 +01:00
bors[bot]andDzmitry Malyshau 3add458e99 [rs] Merge #805
805: Update naga to gfx-17 r=kvark a=kvark

Gets Naga validation, MSL fixes, GLSL varying fixes, and better pipeline creation errors.

Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>
2021-03-19 20:37:42 +00:00
Dzmitry Malyshau 332de311e0 [rs] Update naga to gfx-17 2021-03-19 16:36:30 -04:00
bors[bot]andGordon-F c831c5512a Merge #1273
1273: Properly return native shader module errors to users r=kvark a=Gordon-F

**Testing**
Disable `cross` feature, break `naga` and see what happens 😄 

```log
wgpu error: Validation Error

Caused by:
    In Device::create_render_pipeline
    Internal error in stage VERTEX: Error compiling the shader "\"Compilation failed: \\n\\nprogram_source:1:1: error: unknown type name \\\'sadgasdgasdgasdg\\\'\\nsadgasdgasdgasdg aasdgadsgb>\\n^\\nprogram_source:1:28: error: expected \\\';\\\' after top level declarator\\nsadgasdgasdgasdg aasdgadsgb>\\n   
```



Co-authored-by: Gordon-F <ishaposhnik@icloud.com>
2021-03-19 18:13:28 +00:00
Gordon-F d20c0199d7 Properly return native shader module errors to users 2021-03-19 21:09:54 +03:00
bors[bot]andDzmitry Malyshau 92936954d8 Merge #1274
1274: Update naga to gfx-17 r=kvark a=kvark

**Connections**
Includes https://github.com/gfx-rs/gfx/pull/3688

**Description**
Here comes the "gfx-17" naga train!

**Testing**
Tested on wgpu-rs examples

Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>
2021-03-19 17:44:28 +00:00
Dzmitry Malyshau c52dad9d47 Update naga to gfx-17 2021-03-19 13:43:11 -04:00
bors[bot]andDzmitry Malyshau a2d8e10c7d [rs] Merge #792
792: Use implicit layout for hello-compute r=kvark a=kvark



Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-03-19 14:34:13 +00:00
bors[bot]andDzmitry Malyshau b3bc3d02e0 [rs] Merge #803
803: Shader fixes with binary expressions r=kvark a=kvark

Goes along with https://github.com/gfx-rs/naga/pull/599

Co-authored-by: Dzmitry Malyshau <kvark@fastmail.com>
2021-03-19 05:36:54 +00:00
Dzmitry Malyshau b9c60811f8 [rs] Shader fixes with binary expressions 2021-03-19 01:34:19 -04:00
bors[bot]andDzmitry Malyshau c45ea6f5b1 [rs] Merge #800
800: Schedule event waits after redraw events are cleared r=kvark a=kvark

As @adamnemecek pointed out, our examples were busy waiting.
I investigated a bit and figured that our `WaitScheduled` was always interrupted by `RedrawEventsCleared`. So scheduling the wait in it works. Tested on Linux and macOS. CPU time of the cube example drops from 100% to single-digit.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-03-18 03:00:21 +00:00
Dzmitry Malyshau 070688a84d [rs] Schedule event waits after redraw events are cleared 2021-03-17 22:58:35 -04:00
bors[bot]andDzmitry Malyshau ba6015e2c7 Merge #1272
1272: Add sample count to framebuffer key r=kvark a=kvark

**Connections**
Fixes #1271

**Description**
The framebuffer is created for a pass, so naturally the hashmap key for a framebuffer has to include the pass key. Otherwise, we'd be overwriting the framebuffer entry, or using a framebuffer from a different pass. The sample count was missing from our FB key.

**Testing**
Tested on a trace from #1271 that I hand-ported to latest master.

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2021-03-17 16:15:30 +00:00
Dzmitry Malyshau 9673635125 Add sample count to framebuffer key 2021-03-17 12:13:11 -04:00
bors[bot]andAron Granberg 0542ad585a [rs] Merge #798
798: Fix all the clippy lints r=kvark a=HalfVoxel

Clippy is now happy

Co-authored-by: Aron Granberg <aron.granberg@gmail.com>
2021-03-17 14:02:03 +00:00
Aron Granberg 162fe18e7d [rs] Fix all the clippy lints and add clippy/fmt CI 2021-03-17 14:40:25 +01:00
bors[bot]andDzmitry Malyshau af38afd58d [rs] Merge #799
799: Update naga to gfx-16 r=kvark a=kvark

Makes WGSL constants much more ergonomic to write, and fixes the issues we had with the water example.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-03-17 13:37:45 +00:00
Dzmitry Malyshau 2fae02d8b2 [rs] Update naga to gfx-16 2021-03-17 09:35:54 -04:00
bors[bot]andConnor Fitzgerald f89244e3c3 [rs] Merge #797
797: Move from `tracing` to `log` + `profiling` r=kvark a=cwfitzgerald

Continuation of https://github.com/gfx-rs/wgpu/pull/1268 into wgpu-rs

This also includes the naga gfx-15 train.

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2021-03-17 13:17:03 +00:00
Aron Granberg aa45e7668b [rs] Fix excessive buffer padding (#795)
* Fix excessive buffer padding

* Prettify

* cargo fmt

* Simplify code

* Style
2021-03-17 09:09:58 -04:00
bors[bot]andDzmitry Malyshau bb9a5a85d2 Merge #1270
1270: Update naga to gfx-16 r=kvark a=kvark

**Connections**
Picks up https://github.com/gfx-rs/gfx/pull/3684

**Description**
Just an update

**Testing**
Tested on wgpu-rs examples

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-03-17 13:08:07 +00:00
Dzmitry Malyshau b97f5a5b24 Update naga to gfx-16 2021-03-17 09:07:08 -04:00
Connor Fitzgerald 3a131f5f60 [rs] Move from tracing to log + profiling 2021-03-17 01:34:38 -04:00
bors[bot]andConnor Fitzgerald 2f3b398e38 Merge #1268
1268: Move from `tracing` to `log` + `profiling` r=kvark a=cwfitzgerald

**Connections**

Fixes #1211. Fixes #1189.

**Description**

Moves from `tracing` to `log` + `profiling`. I also removed some of the offending spans that were only for pushing commands onto a vector of commands.

**Testing**

Tested on wgpu-rs

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2021-03-17 05:30:36 +00:00
Connor Fitzgerald cf6f9c3ca8 Move from tracing to log + profiling 2021-03-17 01:22:30 -04:00