Commit Graph
9013 Commits
Author SHA1 Message Date
Frizi 07376d11e8 Handle winit exception in web to avoid breaking async executor 2021-06-03 13:02:56 +02:00
Dzmitry Malyshau e4b0de2526 Links updates 2021-06-02 23:54:26 -04:00
Dzmitry Malyshau 900a141392 Work around CI failures on hello-compute 2021-06-02 22:48:33 -04:00
Dzmitry Malyshau a6ff3ea086 Merge changelogs and readmes 2021-06-02 21:50:16 -04:00
Dzmitry Malyshau 4fc478513a Merge ../wgpu-rs into reunion 2021-06-02 21:31:02 -04:00
Igor Shaposhnik 1a60ba01d9 Add note about MSRV in README 2021-06-02 21:19:09 -04:00
Frizi 2295508f0e Insert attribute glsl location qualifier for opengl ES 3.0 2021-06-02 12:36:04 -04:00
bors[bot]andDzmitry Malyshau 6566a6ce49 [rs] Merge #920
920: Bunnymark example r=grovesNL a=kvark



Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-06-02 16:20:43 +00:00
Sean Moulton 9bdc65e5f7 [rs] Add include_wgsl!() macro (#922)
* Add include_wgsl macro

* Add include_wgsl macro

* Change ShaderFlags to all
2021-06-01 22:31:02 -04:00
João Capucho 3699034ccc [glsl-in] Fix for if being in the wrong body 2021-06-01 21:12:46 -04:00
João Capucho 4cb036b335 [glsl-in] Fix op assignments 2021-06-01 21:12:46 -04:00
João Capucho 344acc3121 [glsl-in] Reenable snapshot tests 2021-06-01 21:11:50 -04:00
João Capucho 679b6a308f [glsl-in] Fix emitting issues 2021-06-01 21:11:50 -04:00
João Capucho 027634451d [glsl-in] Collect entry point arguments usage 2021-06-01 21:11:50 -04:00
Dzmitry Malyshau e98f36245b [rs] Framework and README update 2021-06-01 09:54:44 -04:00
Dzmitry Malyshau f438f0b920 [rs] Bunnymark example 2021-06-01 09:19:19 -04:00
bors[bot]andDzmitry Malyshau 6e680d96a4 [rs] Merge #921
921: Update wgpu to eadaa1b7d8 r=kvark a=kvark

Picks up https://github.com/gfx-rs/wgpu/pull/1419, https://github.com/gfx-rs/wgpu/pull/1417, and https://github.com/gfx-rs/wgpu/pull/1403

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-06-01 13:10:49 +00:00
Dzmitry Malyshau ac6276dfb5 [rs] Update wgpu to eadaa1b7d8 2021-06-01 09:02:19 -04:00
bors[bot]andDzmitry Malyshau eadaa1b7d8 Merge #1419
1419: Skip the cube compatibility on a downlevel flag r=cwfitzgerald a=kvark

**Connections**
Fixes #1418

**Description**
Adds a downlevel flag for cube arrays, checks for it.

**Testing**
Untested, but should work.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-06-01 12:44:53 +00:00
Dzmitry Malyshau beb88b0a88 Skip the cube compatibility on a downlevel flag 2021-06-01 08:29:44 -04:00
bors[bot]andDzmitry Malyshau 9cd01635d7 Merge #1417
1417: Split the tracker into stateful/stateless to reduce the overhead r=cwfitzgerald a=kvark

**Connections**
Implements https://github.com/gfx-rs/wgpu/issues/1413#issuecomment-851604344
Reduces the overhead for resource tracking in the Animometer benchmark by up to 50%.

**Description**
We used to use the full tracker set on the usage scopes associated with compute/render passes. A resource tracker has 2 responsibilities: ensuring the resource is held alive, and validating and recording the state transitions. This PR exploits the fact that the latter responsibility is only applicable for buffers and textures. So doing all the lifetime tracking for a pass is a waste: we can instead just attach the lifetimes to the parent command buffer, straight.

In the Animometer benchmark, there is one large buffer, and thousands of bind groups pointing to different offsets into it. The old code would fill up the pass tracker with those bind groups, and then merge it into the command buffer tracker. The new code would just fill up the command buffer tracker instead. Since there is only one buffer, the pass tracking becomes much lighter.

**Testing**
Untested. It would be nice to have some benchmarks here, possibly after #1397 ?

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2021-06-01 06:45:05 +00:00
bors[bot]andDzmitry Malyshau c332c7fffe Merge #1403
1403: Trace render passes before they are ran r=cwfitzgerald a=kvark

**Connections**
Follow-up to #1245 

**Description**
We want the API traces to reproduce issues. Previously, we moved all the resource creation to happen after the trace is written.
This PR follow-up by also recording the render pass commands even in the case where one of them triggers a validation error.

Another big part of the PR now is handling the command buffers that failed for some reason. The submission is skipping them now instead of erroring the whole thing.

**Testing**
Tested manually on wgpu-rs examples

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2021-06-01 06:31:28 +00:00
Lucien Greathouse ba041e4451 Expand 0.8.0 Changelog (#1420)
* humble beginnings

* Current progress

* More precise NonZeroU32 change

* Fix typo

* Changelog enough to get my game to compile again

* Fix type change line
2021-06-01 02:03:59 -04:00
Dzmitry Malyshau 7b7e72b1c2 Split the tracker into stateful/stateless to reduce the overhead 2021-05-31 14:18:44 -04:00
Dzmitry Malyshau 61bfb29963 Don't consider discard and return inside uniform control flow 2021-05-31 00:49:25 -04:00
Fredrik Norén 58e5b2565d Validate matrix component-wise ops correctly 2021-05-30 11:03:54 -04:00
Wumpf f62d0d5edc Native Extension: Expose fill buffer & clear image (#1335)
* Add command buffer fill buffer method

* Added test for buffer fill

* Added image clear + test

* image/buffer clear methods are now pure clear to zero

* fix cargo clippy & int type issues

* clear buffer alignment is now given by COPY_BUFFER_ALIGNMENT

* joined buffer/image clear player tests

* ImageSubresourceRange is now used in TextureViewDescriptor
2021-05-30 11:03:33 -04:00
Aron Granberg 37724102a3 Improve wgsl lexer error messages.
Better errors for Unexpected, BadInteger, BadFloat, BadTexture, BadTypeCast, UnknownScalarType, UnknownStorageClass, UnknownAttribute, UnknownBuiltin, UnknownShaderStage, UnknownStorageFormat and UnknownConservativeDepth, ZeroStride, ZeroSizeOrAlign and UnknownType.

Also adds lexer::capture_span.

Also fixes some validation for texture sample types and and issue that cauld cause e.g. the type `f33` to be parsed as `f32`.
2021-05-30 11:01:00 -04:00
João Capucho 589f02fad4 [glsl-in] Add implicit conversions tests 2021-05-29 19:38:56 -04:00
João Capucho b9e641c310 [glsl-in] Expression implicit conversions 2021-05-29 19:38:56 -04:00
bors[bot]andJoshua Groves cb18541221 [rs] Merge #918
918: Specify latest gfx release branch in README r=kvark a=grovesNL

Specify latest gfx release branch in README, to hopefully avoid people checking out `hal-0.2` then wondering why the patch doesn't seem to do anything.

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2021-05-29 23:35:06 +00:00
Joshua Groves 2a42e73eef [rs] Specify latest gfx release branch in README 2021-05-29 14:56:56 -02:30
Dzmitry Malyshau 4b5ab528ac Comparison sampling tests 2021-05-28 22:43:39 -04:00
Dzmitry Malyshau d3fbb30d46 [wgsl-out] Improve texture sampling call 2021-05-28 22:43:39 -04:00
Dzmitry Malyshau 8a1d883ba5 [wgsl-in] support textureSampleCompareLevel 2021-05-28 22:43:39 -04:00
Dzmitry Malyshau 1366546645 [glsl-out] improve array types 2021-05-28 22:18:51 -04:00
Dzmitry Malyshau d1e2ac57fe New snapshot test for global variables 2021-05-28 22:18:51 -04:00
Dzmitry Malyshau 0244853d4c [wgsl-in] Support workgroup storage class 2021-05-28 22:18:51 -04:00
Jim Blandy 14908e5126 [spv-out]: Don't pass in the type arena where it's not needed. 2021-05-28 00:47:25 -04:00
Jim Blandy 48363fcef1 Don't accept unsized types as function arguments. 2021-05-26 12:27:29 -04:00
Jim Blandy 398d3545b3 Include the type arena in ResolveContext, since it's always needed. 2021-05-26 10:35:59 -04:00
Dzmitry Malyshau badf99653a [spv-in] refactor the matrix stride check 2021-05-26 00:15:09 -04:00
João Capucho f57739d30b [glsl-in] Provide default interpolation 2021-05-25 13:52:37 -04:00
João Capucho 9464a81ae2 [glsl-in] Fix parameter qualifiers
Previously the types where changed in a later phase now they are changed
right when the argument is added, this makes the implementation slightly
less spec compliant because the following code will compile whilst it shouldn't

```glsl
void test(float a) {}
void test(out float b) {}
```
2021-05-25 13:52:37 -04:00
João Capucho 2b263a1851 [glsl-in] Cast lod value in textureLod 2021-05-25 13:52:37 -04:00
João Capucho 01eb19e29f [glsl-in] Fix vector and matrix constructor
Matrix and vector constructors expected a vector (in the case of the
matrix) or a scalar (in both cases). Now they handle resizing such that
the following code now works

```glsl
mat3 a;
mat4(a); // This would return a validation error because it emitted a cast
```
2021-05-25 13:52:37 -04:00
João Capucho 9868441ebc [glsl-in] Fix storage access 2021-05-25 13:52:37 -04:00
João Capucho 28a1ee4354 [glsl-in] Don't read builtin if it isn't available 2021-05-25 13:52:37 -04:00
João Capucho e7308e1903 [glsl-in] fix validation error of entry result 2021-05-25 13:52:37 -04:00
Jim Blandy 99fbb34bd6 Require dynamically sized arrays to appear in 'storage' memory.
WGSL requires that runtime-sized arrays appear only as the last member of a
structure in in the `storage` storage class. It seems to me that Naga should
enforce this restriction on its own IR as well.
2021-05-25 00:02:08 -04:00