Commit Graph
8897 Commits
Author SHA1 Message Date
bors[bot]andDzmitry Malyshau fa5b6bd6d7 [rs] Merge #377
377: Update with minBufferBindingSize r=cwfitzgerald a=kvark

Depends on https://github.com/gfx-rs/wgpu/pull/726
Also reverts #373 : buffer bindings now have to include at least one element of an unsized struct portion, so they can't be zero-sized.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-17 04:19:47 +00:00
Dzmitry Malyshau 24a4faf065 [rs] Update with minBufferBindingSize 2020-06-17 00:13:47 -04:00
bors[bot]andDzmitry Malyshau fc2dd481b2 Merge #726
726: Basic support for minBufferBindingSize r=cwfitzgerald a=kvark

**Connections**
Has basic (partial) implementation of https://github.com/gpuweb/gpuweb/pull/678
wgpu-rs update - https://github.com/gfx-rs/wgpu-rs/pull/377

**Description**
This change allows users to optionally specify the expected minimum binding size for buffers. We are then validating this against both the pipelines and bind groups.
If it's not provided, we'll need to validate at draw time - this PR doesn't do this (focus on API changes first).
It also moves out the `read_spirv`, since wgpu-types wasn't the right home for it ever.

**Testing**
Tested on wgpu-rs examples

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-17 03:25:27 +00:00
Dzmitry Malyshau 8e0ae1cf06 Convert glsl module line endings to Unix 2020-06-16 22:54:26 -04:00
Dzmitry Malyshau d55cd79a21 Minor improvements, expose Handle, move glsl module inside the dir. 2020-06-16 22:54:26 -04:00
bors[bot]andOptimisticPeach 4043a10fbc [rs] Merge #375
375: Add water example.  r=kvark,cwfitzgerald a=OptimisticPeach

Solves #329, water example requested by @kvark.

I tuned it to my personal preference of visuals, however it might be different for you. 

Note: I used https://github.com/ashima/webgl-noise for 3D open simplex noise.

I've commented to explain what most of the things in the rust side of the example mean. However, I'm not 100% sure I did the best job at giving a brief overview, and wouldn't mind someone making sure my terminology/definitions are correct. 

Thanks!
Patrik

Co-authored-by: OptimisticPeach <patrikbuhring@yahoo.com>
2020-06-17 02:02:43 +00:00
bors[bot]andNikita Krupitskas 9e699330a8 Merge #719
719: Implement debug marker support r=kvark a=krupitskas

**Connections**
Closes https://github.com/gfx-rs/wgpu/issues/697

**Description**
Looks like because I've once pushed forward reset branch to my master, previous pull request https://github.com/gfx-rs/wgpu/pull/713
show that there no commits and it automatically was closed :/

**Testing**
Not tested yet


Co-authored-by: Nikita Krupitskas <krupitskas@icloud.com>
2020-06-16 21:34:27 +00:00
Nikita Krupitskas eaf2acb43a Implement debug marker support 2020-06-17 00:05:52 +03:00
Dzmitry Malyshau 89d2afa11b Comparison sampler API (#66) 2020-06-16 14:55:40 -04:00
bors[bot]andRukai 5853fb9557 Merge #727
727: Update gfx-memory to the latest commit r=kvark a=rukai

**Description**
In order to test potential fixes for https://github.com/gfx-rs/wgpu-rs/issues/363 we need wgpu master to be able to easily test changes to gfx-memory.

Assuming we dont have any large regressions I think it could be nice to release wgpu 0.6 before merging this. I dont mind either way though.

**Testing**
I ran some wgpu-rs examples.


Co-authored-by: Rukai <rubickent@gmail.com>
2020-06-16 14:11:28 +00:00
Rukai 71ee7d868a Update gfx-memory to the latest commit 2020-06-16 23:32:35 +10:00
Nikita Krupitskas 6a0a4d6c9d [rs] Add debug markers and debug markers example 2020-06-16 11:13:03 +03:00
OptimisticPeach 2ad472538c [rs] Add water example 2020-06-16 02:33:14 -04:00
Dzmitry Malyshau a27341bd6d Basic support for minBufferBindingSize 2020-06-16 00:34:58 -04:00
bors[bot]andDzmitry Malyshau 6fa9f46fbc [rs] Merge #376
376: Revert "API update for RODS2" r=kvark a=kvark

This reverts commit edb3610530.
It also update `wgpu` dependency to https://github.com/gfx-rs/wgpu/pull/723

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2020-06-15 22:17:32 +00:00
Dzmitry Malyshau 06f4c122c6 [rs] Revert "API update for RODS2"
This reverts commit edb3610530.
2020-06-15 18:16:14 -04:00
bors[bot]andDzmitry Malyshau d1deae5747 Merge #723
723: Always use DepthStencilReadOnlyOptimal for sampled depth r=kvark a=kvark

**Connections**
This is a follow-up to #712  that reverts the API changes but also fixes our internal logic to work better.

**Description**
It solves the problem of image layouts by always using a RODS layout on depth images, when sampled.

**Testing**
Tested on https://github.com/gfx-rs/wgpu-rs/pull/375


Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2020-06-15 21:45:51 +00:00
Dzmitry Malyshau 58de765961 Always use DepthStencilReadOnlyOptimal for sampled depth 2020-06-15 17:45:19 -04:00
bors[bot]andDzmitry Malyshau 3d9d82dc77 Merge #722
722: Refactor tracking of device last completed submission r=kvark a=kvark

**Connections**
Fixes https://github.com/gfx-rs/wgpu-rs/issues/358

**Description**
We used to track the next submission index in `device.life_guard.submission_index` atomic. This PR changes that to point to the last *done* submission, and also introduces a non-atomic field to keep track of the current/next submission.
This allows us to avoid waiting on the frame semaphore on presentation if the relevant submission is done by then.

**Testing**
Not tested!

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-15 20:58:15 +00:00
bors[bot]andConnor Fitzgerald f340882712 [rs] Merge #369
369: Implement SAMPLED_TEXTURE_ARRAY_NON_UNIFORM_INDEXING r=kvark a=cwfitzgerald

This implements https://github.com/gfx-rs/wgpu/pull/715 in wgpu-rs. I haven't changed the example, as I want to actually think up a better example to use and didn't want that to block this. It will change in the future however.

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2020-06-15 20:44:20 +00:00
Timo de Kort 82573b288b Implement basic SPIR-V back-end (#46) 2020-06-15 16:01:00 -04:00
Connor Fitzgerald 8b60da572f [rs] Implement SAMPLED_TEXTURE_ARRAY_NON_UNIFORM_INDEXING 2020-06-15 15:09:06 -04:00
bors[bot]andRukai 4804fe496b [rs] Merge #373
373: add test case for empty buffer r=kvark a=rukai

Depends on https://github.com/gfx-rs/wgpu/pull/720

Co-authored-by: Rukai <rubickent@gmail.com>
2020-06-15 02:26:27 +00:00
Rukai 4b0534949b [rs] add test case for empty buffer 2020-06-15 12:18:58 +10:00
bors[bot]andConnor Fitzgerald 2dd3439475 Merge #717
717: Implement Capability/Extension Split r=kvark a=cwfitzgerald

## Connections

Follow up to a discussion we had on #wgpu:matrix.org. 

## Description

Splits capabilities and extensions so that extensions are things you ask for and possibly change behavior and capabilities are passively enabled when their extension (if any) are enabled.

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2020-06-15 01:46:14 +00:00
bors[bot]andRukai bb3e179d53 Merge #720
720: Fix validation errors + panics on empty buffers r=kvark a=rukai

**Description**
My previous PR left a vulkan validation error when creating an empty buffer.
This PR fixes that and also fixes a panic preventing the compute-example from running with no elements.

**Testing**
Unit test added in wgpu-rs PR. https://github.com/gfx-rs/wgpu-rs/pull/373
<!--
Non-trivial functional changes would need to be tested through:
  - [wgpu-rs](https://github.com/gfx-rs/wgpu-rs) - test the examples.
  - [wgpu-native](https://github.com/gfx-rs/wgpu-native/) - check the generated C header for sanity.

Ideally, a PR needs to link to the draft PRs in these projects with relevant modifications.
See https://github.com/gfx-rs/wgpu/pull/666 for an example.
If you can add a unit/integration test here in `wgpu`, that would be best.
-->


Co-authored-by: Rukai <rubickent@gmail.com>
2020-06-15 01:06:46 +00:00
Rukai b0475520fa Fix validation errors + panics on empty buffers 2020-06-15 11:04:36 +10:00
Dzmitry Malyshau 835d28b37c Refactor tracking of device last completed submission 2020-06-14 20:19:40 -04:00
bors[bot]andDzmitry Malyshau ee7e1c7705 [rs] Merge #370
370: API update for RODS2 r=cwfitzgerald a=kvark

Reflects upstream changes in https://github.com/gfx-rs/wgpu/pull/716
I'm not too happy about the API just yet, has another idea that I'm going to discuss with the group on Monday, hopefully. But we need to get going and not block other wgpu-rs changes, so this should land in some way.

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2020-06-14 13:25:59 +00:00
Dzmitry Malyshau 6cf328f326 [rs] API update for RODS2 2020-06-14 09:25:13 -04:00
Connor Fitzgerald 5a7fb4c1e5 Implement Capability/Extension Split 2020-06-14 02:32:54 -04:00
Timo de Kort f9fa281da5 Update integers to float32 (#64) 2020-06-13 13:05:30 -04:00
bors[bot]andDzmitry Malyshau ad34c37127 Merge #716
716: Add RODS texture binding property r=cwfitzgerald a=kvark

This is a follow-up to #712, which I screwed up right before it got in. Fortunately, we still need the code that went in, we just need a little bit more :)

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-13 02:19:35 +00:00
Dzmitry Malyshau c038c206e7 Add RODS texture binding property 2020-06-12 21:26:19 -04:00
bors[bot]andDzmitry Malyshau 0357dd80af Merge #712
712: RODS part 2 r=cwfitzgerald a=kvark

**Connections**
Unblocks https://github.com/gfx-rs/wgpu-rs/issues/359
Is a follow-up to RODS part 1 - #685

**Description**
There is a few things in here.

Thing 1: Stronger assertions on the load/store ops of the depth & stencil.

### Thing 2: rewritten tracking of render attachments
Background: from the usage tracker point of view, each subresource can be in either "extend" mode, where it accumulates a single combined usage, or in the "replace" mode, where it goes from one usage to another, producing the relevant transitions on the way.

The problem turned out to come from the fact that the render pass attachments were always tracked in "replace" mode. This is needed because their track don't have a single state: render pass itself encodes a transition of attachments. However, it also means that there was no way to "extend" the usage in RODS scenarios...

So I could see two ways to address this:
  - re-achitecture the tracking a bit more in general, representing it as a sequence of merges.
  - introduce the "prepend()" tracking operator that's *only* used for render pass attachments

I opted for the latter as it seems much less intrusive. The render pass attachments accumulate their usage like everything else in the "extend mode". But right before we are inserting the transitions (between the active command buffer and the pass), we turn the tracking of the attachments from "extend" into "replace" mode by installing the "first" usage according to what we expect it to be.

### Thing 3: missing API for RODS bind groups
The original RODS design missed a problem with Vulkan image layouts. When creating a bind group, one has to specify what layout the image will be in. We always used `ShaderReadOnlyOptimal` until now for texture views. However, in RODS scenarios this has to be `DepthStencilReadOnlyOptimal`. Luckily, it's compatible with sampling from the shader, but we still need to know about this when creating the bind group.

**Testing**
Tested on the modified water example provided in https://github.com/gfx-rs/wgpu-rs/issues/359#issuecomment-642167269 

Added a few tests to the buffer implementation of the new `prepend()` operator.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-12 23:58:36 +00:00
Dzmitry Malyshau 8ebfdb0c34 Rewrite the render attachment tracking using the new prepend() operation. 2020-06-12 19:54:50 -04:00
bors[bot]andConnor Fitzgerald 373a1c2bf0 Merge #715
715: Implement SAMPLED_TEXTURE_ARRAY_NON_UNIFORM_INDEXING r=kvark a=cwfitzgerald

## Connections

Closes #483. Extends support for #106.

## Description

This forwards the descriptor indexing features gfx-hal up to wgpu. 

Note this PR also changes the name of the `TEXTURE_BINDING_ARRAY` to `SAMPLED_TEXTURE_BINDING_ARRAY` to be more consistent with extensions.

## Testing

Texture-array example was simply extended to use descriptor indexing. This works on all platforms that the feature is supported. (thanks @stararawn)



Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2020-06-12 23:53:37 +00:00
Connor Fitzgerald bfcc369bc7 Implement SAMPLED_TEXTURE_ARRAY_NON_UNIFORM_INDEXING 2020-06-12 19:52:10 -04:00
bors[bot]andDzmitry Malyshau 91c4849be9 [rs] Merge #357
357: Render bundle support, use in msaa-line example r=cwfitzgerald a=kvark



Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-12 23:07:36 +00:00
Dzmitry Malyshau 5e2d0d7680 [rs] Render bundle support, use in msaa-line example 2020-06-11 22:59:40 -04:00
bors[bot]andDzmitry Malyshau b0b99d89b7 Merge #714
714: Track buffers and textures in the future submissions r=cwfitzgerald a=kvark

**Connections**
Fixes https://github.com/gfx-rs/wgpu-rs/issues/362

**Description**
When we do `write_buffer` or have it mapped on creation, we record the copy command in a "pending command buffer", which is only actually executed on the next submission. We adjust the submission index on the buffer accordingly.
If the buffer is dropped *before* the next submission, and there happens to be a `maintain()` step, there is code that checks if the buffer-associated submission is one of the currently executing ones. If it doesn't fine an active one, it considers the last-used submission to already be done with, which is precisely what is not happening here, since the submission hasn't been scheduled yet.

The solution here is to always defer suspecting buffers and textures for deletion till the next submission. The buffers and textures are the only objects that may have "future" submission index, due to `write_buffer` and `write_texture`.

**Testing**
Confirmed on Linux.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-12 02:15:11 +00:00
Dzmitry Malyshau e843067018 Track buffers and textures in the future submissions 2020-06-11 20:45:53 -04:00
bors[bot]andConnor Fitzgerald 55eec80920 [rs] Merge #360
360: Implement TEXTURE_BINDING_ARRAY extension r=kvark a=cwfitzgerald

This extends https://github.com/gfx-rs/wgpu/pull/711 into wgpu-rs.

Notable changes: 
- Added an example showing off both this extension and the future descriptor indexing extension.
- Changed the framework so there is a static function showing what extensions you need. This is provided by the trait,, which defaults to no extensions, so existing examples and new ones don't have to care about extensions.

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2020-06-11 20:37:46 +00:00
Connor Fitzgerald 200025e3cc [rs] Implement TEXTURE_BINDING_ARRAY extension 2020-06-11 16:36:53 -04:00
bors[bot]andDzmitry Malyshau 475696af59 Merge #286
286: RenderBundle support r=cwfitzgerald a=kvark

Implements the API of https://github.com/gpuweb/gpuweb/pull/301

The general concept here is having re-usable command streams, which seems much desired given that our command buffers are not reusable.

Currently, only "software" render bundles are supported. That means, they are just smaller chunks of render pass commands, not backed by any driver object.

TODO:
- [x] https://github.com/gfx-rs/wgpu-rs/pull/357
- [x] https://github.com/gfx-rs/wgpu-native/pull/37
- [x] figure out the lifetime solution

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2020-06-11 19:46:31 +00:00
Dzmitry Malyshau 7aa5753701 Re-architecture the bundles using normalized command streams.
This is a major change in how the bundles are implemented. Instead of
transparently injecting them into the pass command stream, we are now
treating bundles as first-class API objects and API tracing them
accordingly. The bundle contains a normalized command stream that is
very easy to inject into a native command buffer multiple times.
2020-06-11 15:34:45 -04:00
Dzmitry Malyshau afc4517db1 Render bundle reset states, updated descriptors 2020-06-11 15:28:55 -04:00
Dzmitry Malyshau fbf35e43a6 Render bundles 2020-06-11 15:28:55 -04:00
bors[bot]andLachlan Sneff 9842d2065f [rs] Merge #366
366: Fix buffer mapping examples r=cwfitzgerald a=lachlansneff

Fixes #364 

Co-authored-by: Lachlan Sneff <lachlan.sneff@gmail.com>
2020-06-11 16:09:55 +00:00
Lachlan Sneff 5216fd3285 [rs] Fix buffer mapping examples 2020-06-11 12:07:47 -04:00