Commit Graph
9013 Commits
Author SHA1 Message Date
bors[bot]andAtul Bhosale f015eb22d8 Merge #149
149: Remove wgpu_buffer_set_sub_data function r=kvark a=Atul9

closes #145 

Co-authored-by: Atul Bhosale <atul1bhosale@gmail.com>
2019-05-03 13:50:39 +00:00
Atul Bhosale 3330509678 Remove wgpu_buffer_set_sub_data function 2019-05-03 07:01:59 +05:30
Atul Bhosale 3585c4481c Merge '.rustfmt.toml' into 'rustfmt.toml' 2019-05-01 23:06:36 +05:30
psincf 10547b937f add serde into scope and fix wgpu-remote 2019-05-01 13:31:56 +02:00
psincf 565e5fdfeb make trait TypedId:From<Id> 2019-05-01 12:15:23 +02:00
psincf f7794fd82b change PhantomData syntax 2019-05-01 12:11:20 +02:00
psincf bbef2b7622 Put new_id! in define_id! 2019-05-01 11:55:28 +02:00
psincf bfc983a8cc Clean code 2019-05-01 11:55:28 +02:00
psincf 44cf64495a Use strong type 2019-05-01 11:55:28 +02:00
bors[bot]andDzmitry Malyshau 1e177d0813 Merge #144
144: Rewrite bindings generation r=grovesNL a=kvark

This PR is shaping wgpu-remote to become distantly usable, introduces an example that tests it on CI.

It also rewrites our binding generation to use `cbindgen.toml` configuration, which goes in line with Gecko use case (even though they are going to have a separate configuration).

The `ClientFactory` is introduced so that we can have multiple clients (say, one page per client) talking to the same server on a dedicated thread within the GPU process.

Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2019-05-01 02:14:57 +00:00
Dzmitry Malyshau 6ac666a5aa Introduce ClientFactory 2019-04-30 15:37:46 -04:00
Dzmitry Malyshau 61fca00f34 Remove wgpu-bindings in favor of configuration scripts 2019-04-30 12:45:05 -04:00
Dzmitry Malyshau 0011d9f4ff Global manager for instance names on the remote client 2019-04-30 09:19:40 -04:00
Dzmitry Malyshau 52ee3e019c Remote example and improved header, tested on CI 2019-04-29 22:41:09 -04:00
bors[bot]andSerhii Plyhun 5ec6f35368 Merge #139
139: Scissors test r=kvark a=snuk182

Closes #138 

Co-authored-by: Serhii Plyhun <snuk188@gmail.com>
2019-04-28 02:28:16 +00:00
bors[bot]andShalokShalom a1908c1873 Merge #142
142: Correct Link r=kvark a=ShalokShalom



Co-authored-by: ShalokShalom <shalokshalom@protonmail.ch>
2019-04-28 02:14:02 +00:00
Serhii Plyhun 092bcc4afb Scissor test 2019-04-28 00:45:37 +02:00
ShalokShalom 3c12edd0b7 Correct Link 2019-04-27 21:05:39 +02:00
bors[bot]andAlexis Sellier 6f810f4f23 Merge #131
131: Support dynamic offsets r=kvark a=cloudhead

Closes #125

Adds support for dynamic offsets when setting a bind group.

I haven't actually tried if this works with my use-case, but it's fairly straight forward. 😅 

Co-authored-by: Alexis Sellier <alexis@monadic.xyz>
2019-04-26 22:34:55 +00:00
Alexis Sellier c468840eee Support dynamic offsets
Closes #125

* Add support for dynamic offsets when setting a bind group.
* Add new binding types: `UniformBufferDynamic` and `StorageBufferDynamic`
2019-04-26 22:20:54 +02:00
bors[bot]andAtul Bhosale ee3025c4d8 Merge #130
130: Format code using 'cargo fmt' r=kvark a=Atul9



Co-authored-by: Atul Bhosale <atul1bhosale@gmail.com>
2019-04-26 16:59:20 +00:00
Atul Bhosale af38d79d79 Format code using 'cargo fmt' 2019-04-26 22:01:47 +05:30
993293f709 Merge #127
127: Track async mapping properly r=grovesNL,swiftcoder a=kvark

Fixes #117, fixes #95, fixes #132 

The change can be logically split into 3 parts:
  1. when `ActiveSubmission` is retired, we now move the mapped buffers into the "ready to map" vector. This was the missing bit that caused mapping to not work previously.
  2. mapping callbacks in Rust wrapper are refactored and they `unmap()` automatically now on users behalf
  3. we wait for idle before destroying the device, which allows us to process all the pending callbacks. This fix gets rid of the dummy submission our compute example used to do.

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
2019-04-25 13:36:59 +00:00
Dzmitry Malyshau 06d34830d4 Update OSX CI version to 9.4 2019-04-25 08:29:53 -04:00
Dzmitry Malyshau 8578643f38 Rename DestroyedResources to PendingResources 2019-04-25 08:29:40 -04:00
Dzmitry Malyshau 41141b1ae0 Wait for fences instead of device idle 2019-04-24 22:42:49 -04:00
Dzmitry Malyshau aa1f876de3 Assert on a mapping buffer being used 2019-04-23 17:26:31 -04:00
Dzmitry Malyshau 326af8cce6 Do mapping transitions earlier. 2019-04-23 17:22:53 -04:00
Dzmitry Malyshau 7200f727e2 Expose device polling 2019-04-23 16:56:17 -04:00
Dzmitry Malyshau e2dcac91d2 Round up mappable buffer size to non-coherent atom, if needed 2019-04-23 16:27:39 -04:00
bors[bot]andpsincf 0a60c82c43 Merge #128
128: Fix winit event loop in the triangle example r=kvark a=psincf

It's the refactored https://github.com/gfx-rs/wgpu/pull/122 PR

Co-authored-by: psincf <44228825+psincf@users.noreply.github.com>
2019-04-22 21:13:09 +00:00
psincf e70b01ccc0 Fix winit event loop in the triangle example 2019-04-22 22:59:25 +02:00
Dzmitry Malyshau 1f4d67aeb2 Track async mapping properly 2019-04-22 13:57:35 -04:00
bors[bot]andMarkus Ginthör 4431df63d8 Merge #124
124: Configure Swapchain with supported composite_alpha r=kvark a=silphendio

CompositeAlpha::Inherit (default) is not supported on every graphics card.
Relevant : https://github.com/gfx-rs/gfx/issues/2507

Co-authored-by: Markus Ginthör <markus.ginthoer@gmx.at>
2019-04-22 16:35:17 +00:00
Markus Ginthör 89c5564133 Configure Swapchain with supported composite_alpha 2019-04-12 17:44:44 +02:00
bors[bot]andJoshua Groves f5ad76726b Merge #121
121: Implement index formats r=kvark a=grovesNL

I decided to implement index formats to try to catch up on some of the recent changes. The idea is to either source the index format from the last bound pipeline, or switch the index format based on a newly bound pipeline, through the following:

- During `set_index_buffer`, use the index format from the render pass, and cache the bound index buffer ID and offset
- During `set_pipeline`, check if the index format has changed from the previous pipeline, and rebind the index buffer using the new index format if necessary

Some things I wasn't sure about:

- Is the usage of `get_with_extended_usage` correct here?
- Are the caches located in the correct place (i.e. `RenderPassContext`)?

Co-authored-by: Joshua Groves <josh@joshgroves.com>
2019-04-07 11:29:41 +00:00
Joshua Groves 7f7bdd1ba1 Implement index formats 2019-04-07 08:15:05 -02:30
bors[bot]andDzmitry Malyshau 20841859d1 Merge #119
119: Fix waiting on frame semaphores, plus check the epochs r=grovesNL a=kvark

Fixes #88 

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2019-04-04 12:32:19 +00:00
Dzmitry Malyshau 83a83f33e7 Track epochs of the frame semaphores to wait for 2019-04-03 17:48:26 -04:00
bors[bot]andNathan Youngman 3bf6992e75 Merge #118
118: readme: revise supported platforms r=kvark a=nathany

Should this say D3D11?

Co-authored-by: Nathan Youngman <4566+nathany@users.noreply.github.com>
2019-04-02 16:31:36 +00:00
Nathan Youngman acc8443b5e readme: revise supported platforms 2019-04-02 08:35:39 -06:00
bors[bot]andDzmitry Malyshau 81ff6fac05 Merge #116
116: Validation bits from 0.2 r=kvark a=kvark

Sibling of #114 

Co-authored-by: Dzmitry Malyshau <dmalyshau@mozilla.com>
2019-04-01 00:08:02 +00:00
Dzmitry Malyshau c283a8b115 Update changelog and versions 2019-03-31 20:04:39 -04:00
Dzmitry Malyshau a3ffa7072a Fix the triangle examples 2019-03-31 20:03:24 -04:00
Dzmitry Malyshau a8a6b45c9a Rename expectation to LayoutChange 2019-03-31 20:00:52 -04:00
Dzmitry Malyshau 5e041ab239 Validate resource types against the bind group layout 2019-03-31 19:59:34 -04:00
Dzmitry Malyshau b96983a56b Fix submission tracking 2019-03-31 19:59:21 -04:00
Dzmitry Malyshau 006d40a4b5 Track bind group compatibility across the whole pipeline 2019-03-31 16:04:37 -04:00
Dzmitry Malyshau 5923deff6e Validate bind group compatibility 2019-03-31 16:03:08 -04:00
Dzmitry Malyshau 332305012c Blend color and the first bits of validation 2019-03-31 15:52:32 -04:00