Commit Graph
8323 Commits
Author SHA1 Message Date
teoxoy 66d7387f0d [msl] refactor chain of if blocks to match 2024-05-14 17:57:04 +02:00
Connor Fitzgerald 7078b0a061 Fix Subgroup Ops on VK 1.2 Device (#5624) 2024-05-14 14:21:53 +00:00
Nick Guletskii 65d8c94afd Issue SetDrawColorBuffers before clearing buffers in GLES, use clear_buffer_f32_slice instead of clear (#5666)
* Issue SetDrawColorBuffers commands before issuing ClearColor

This is necessary for glClearBuffer calls to work correctly on some machines (e.g. AMD Renoir graphics running on Linux). Without this, glClearBuffer calls are ignored.

* Use clear_buffer_f32_slice instead of gl.clear to suppress WebGL warnings

This fixes the following WebGL warning: "WebGL warning: drawBuffers: `buffers[i]` must be NONE or COLOR_ATTACHMENTi."

When using native OpenGL, it is acceptable to call glDrawBuffers with an array of buffers where i != COLOR_ATTACHMENTi. In WebGL, this is not allowed.

* Run cargo fmt

* Add changes for PR GH-5666 to the CHANGELOG
2024-05-14 05:57:18 -04:00
Kevin Reid 9b70254437 Reduce string allocations related to labels and logging. (#5690) 2024-05-14 04:39:28 -04:00
Jim Blandy ffd96a0a53 [core] Use a for loop in LifetimeTracker triage code.
A `for` loop is less noisy than a `drain`, which requires:

- a `mut` qualifier for a variable whose modified value we never
  consult

- a method name appearing mid-line instead of a control structure name
  at the front of the line

- a range which is always `..`, establishing no restriction at all

- a closure instead of a block

Structured control flow syntax has a fine pedigree, originating in,
among other places, Dijkstrsa's efforts at designing languages in a
way that made it easier to formally verify programs written in
them (see "A Discipline Of Programming"). There is nothing "more
mathematical" about a method call that takes a closure than a `for`
loop. Since `for_each` is useless unless the closure has side effects,
there's nothing "more functional" about `for_each` here, either.
Obsessive use of `for_each` suggests that the author loves Haskell
without understanding it.
2024-05-14 04:38:43 -04:00
Jim Blandy 8981058ad1 [core] In LifetimeTrackers triage code, use more specific names.
Rename `LifetimeTracker::triage_resources`'s `resources_map` argument
to `suspected_resources`, since this always points to a field of
`LifetimeTracker::suspected_resources`.

In the various `triage_suspected_foo` functions, name the map
`suspected_foos`.
2024-05-14 04:38:43 -04:00
Jim Blandy 8d73e5a9cd [core] Refactor LifetimeTracker::triage_resources.
Check whether the resource is abandoned first, since none of the rest
of the work is necessary otherwise.

Rename `non_referenced_resources` to `last_resources`. This function
copes with various senses in which the resource might be referenced or
not. Instead, `last_resources` is the name of the `ActiveSubmission`
member this may point to, which is more specific.

Move the use of `last_resources` immediately after its production.
2024-05-14 04:38:43 -04:00
Jim Blandy 64777d4fd8 [core] Doc fixes for lifetime management, minor typos.
- Document `LifetimeTracker::triage_resources`.

- Fix various typos and bad grammar.
2024-05-14 04:38:43 -04:00
Marijn Suijten 8879733875 [wgpu-hal] Upgrade to ash 0.38 (#5504) 2024-05-13 11:57:44 -04:00
Erich Gubler ca91744955 chore: apply unused_qualifications lint via cargo +1.78.0 fix && cargo fmt 2024-05-13 10:10:54 -04:00
Erich Gubler f1b06f55ce chore: satisfy clippy::assigning_clones 2024-05-13 10:10:54 -04:00
Erich Gubler bfe96d87b2 chore(spv): remove unused MergeInstruction 2024-05-13 10:10:54 -04:00
Erich Gubler 78a208b005 chore: satisfy clippy::empty_docs 2024-05-13 10:10:54 -04:00
Erich Gubler b4193d50db chore: satisfy clippy::multiple_bound_locations 2024-05-13 10:10:54 -04:00
dependabot[bot] 0f751b147a build(deps): bump the patch-updates group across 1 directory with 4 updates (#5698)
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-05-13 04:43:44 -04:00
Andreas Reich 452cf24fa1 Remove unnecessary Cargo.toml dependencies via cargo machete (#5692) 2024-05-12 19:05:00 -04:00
bjorn3 fa48562229 Avoid introducing spurious features for optional dependencies (#5691)
* Avoid introducing spurious features for optional dependencies

If a feature depends on an optional dependency without using the dep:
prefix, a feature with the same name as the optional dependency is
introduced. This feature almost certainly won't have any effect when
enabled other than increasing compile times and polutes the feature list
shown by cargo add. Consistently use dep: for all optional dependencies
to avoid this problem.

* Add changelog entry
2024-05-12 08:45:35 +02:00
bjorn3 ebbf901304 Remove unused num-traits dependency of naga (#5689) 2024-05-11 16:26:49 -04:00
Leo Kettmeir d0a5e48aa7 chore: backport deno changes (#5686) 2024-05-11 09:29:09 +00:00
Connor Fitzgerald 2f4522714c Remove #[doc(inline)] to Prevent Doc Spam 2024-05-10 14:18:15 -04:00
dependabot[bot]andConnor Fitzgerald f4b8b15df3 build(deps): bump the patch-updates group across 1 directory with 26 updates (#5673)
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2024-05-09 14:26:38 -04:00
dependabot[bot] 3b6112d45d build(deps): bump crate-ci/typos from 1.20.10 to 1.21.0 (#5672)
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-05-06 15:08:37 -04:00
Xiaopeng LiandJim Blandy d5d683d3c4 Clean up weak references to texture views and bind groups (#5595)
* Clean up weak references to texture views

* add change to CHANGELOG.md

* drop texture view before clean up

* cleanup weak ref to bind groups

* update changelog

* Trim weak backlinks in their holders' triage functions.

---------

Co-authored-by: Jim Blandy <jimb@red-bean.com>
2024-05-06 12:53:03 +02:00
Valaphee The Meerkat 8084eb6c8f Seperate GLES and GL version selection (#5657)
* Seperate GLES and GL version selection

* Add info when gles minor version is specified and not used
2024-05-04 21:42:45 +02:00
Jim Blandy 0d70a7361d [core] Properly recycle Device::temp_suspected.
Change `Device::untrack` to properly reuse the `ResourceMap` allocated
for prior calls. The prior code tries to do this but always leaves
`Device::temp_suspected` set to a new empty `ResourceMap`, leaving the
previous value to be dropped by `ResourceMap::extend`.

Change `ResourceMap::extend` to take `other` by reference, rather than
taking it by value and dropping it.
2024-05-04 00:28:46 -04:00
Jim Blandy 5f464688e6 [core] Don't check for uniquely referenced resources at submission.
Remove unreachable code from `Global::queue_submit` that checks
whether the resources used by the command buffer have a reference
count of one, and adds them to `Device::temp_suspected` if so.

When `queue_submit` is called, all the `Arc`s processed by this code
have a reference count of at least three, even when the user has
dropped the resource:

- `Device::trackers` holds strong references to all the device's
  resources.

- `CommandBufferMutable::trackers` holds strong references to all
  resources used by the command buffer.

- The `used_resources` methods of the various members of
  `CommandBufferMutable::trackers` all return iterators of owned
  `Arc`s.

Fortunately, since the `Global::device_drop_foo` methods all add the
`foo` being dropped to `Device::suspected_resources`, and
`LifetimeTracker::triage_suspected` does an adequate job of accounting
for the uninteresting `Arc`s and leaves the resources there until
they're actually dead, things do get cleaned up without the checks in
`Global::queue_submit`.

This allows `Device::temp_suspected` to be private to
`device::resource`, with a sole remaining use in `Device::untrack`.

Fixes #5647.
2024-05-03 17:00:07 -04:00
Valaphee The Meerkat 565a0310e9 Fix OpenGL non-srgb on Linux (#5642) 2024-05-03 22:29:27 +02:00
Jim Blandy 4a07ab2b57 [core] Ensure all lock guards are properly nested.
The lock analyzers in the `wgpu_core::lock` module can be a bit
simpler if they can assume that locks are acquired and released in a
stack-like order: that a guard is only dropped when it is the most
recently acquired lock guard still held. So:

- Change `Device::maintain` to take a `RwLockReadGuard` for the device's
  hal fence, rather than just a reference to it.

- Adjust the order in which guards are dropped in `Device::maintain`
  and `Queue::submit`.

Fixes #5610.
2024-05-03 13:06:00 -04:00
Jim Blandy 7c842a3b48 [core] Implement downgrade for the lock module's RwLocks.
Implement `RwLockWriteGuard::downgrade` for `lock::vanilla` and
`lock::ranked`, to downgrade a write lock to a read lock.
2024-05-03 13:06:00 -04:00
Jim Blandy 651214ef74 [core] Refactor lock::ranked to use LockStateGuard.
Rather than implementing `Drop` for all three lock guard types to
restore the lock analysis' per-thread state, let lock guards own
values of a new type, `LockStateGuard`, with the appropriate `Drop`
implementation. This is cleaner and shorter, and helps us implement
`RwLock::downgrade` in a later commit.
2024-05-03 13:06:00 -04:00
Jim Blandy fb3b33d092 [wgpu] Clarify documentation for Queue transfer methods.
Explain more clearly that the `write_buffer`, `write_buffer_with`, and
`write_texture` methods on `wgpu::Queue` do not immediately submit the
transfers for execution. Provide sample code for flushing them.
2024-05-01 17:42:00 -07:00
Jim Blandy e5201a7267 When a #[gpu_test] test fails, print its source location.
Use `std::panic::Location` to record the source location of each
`#[gpu_test]` test, and if it fails, include that in the error output.
This is not essential, but it should make working with failures a bit
more comfortable.
2024-05-01 17:08:22 -07:00
Samson 5fa537bfd9 Cast ptr to Device not Surface (#5640) 2024-04-30 11:39:56 -04:00
Elabajaba bdf30fa45a fix halmark (#5635) 2024-04-29 20:45:37 +00:00
Jim Blandy b765eeb474 [core] Removed outdated safety comments from no-longer-unsafe fns. (#5633) 2024-04-29 13:34:06 -04:00
Andreas Reich 08efa72a83 move shader compilation error query out of 0.20 changelog - it didn't get released with it 2024-04-29 14:29:18 +02:00
Connor Fitzgerald 90e7060d0d Fix Failure Case for MacOS 14.3 2024-04-29 12:01:05 +02:00
stefnotch f874ed061c Add get_compilation_info (#5410)
* Add get_compilation_info API

* Rename glsl ParseError to ParseErrors

* Document ParseError label order

* Update line_position to count UTF-8 bytes
2024-04-29 11:35:36 +02:00
dependabot[bot] 8597367d55 build(deps): bump crate-ci/typos from 1.20.9 to 1.20.10 (#5623)
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-04-29 02:50:12 +00:00
Connor Fitzgerald 4521502da6 Release v0.20.0 (#5619) wgpu-types-v0.20.0 wgpu-v0.20.0 wgpu-core-v0.20.0 d3d12-v0.20.0 naga-v0.20.0 naga-cli-v0.20.0 wgpu-info-v0.20.0 v0.20.0 wgpu-hal-v0.20.0 2024-04-28 18:06:35 -04:00
dependabot[bot]andConnor Fitzgerald 05cfdececc build(deps): bump the patch-updates group with 5 updates (#5605)
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2024-04-28 17:44:12 -04:00
Erich Gubler e8e33ede11 refactor(naga): remove extraneous leading path qualifiers (#5612)
Implemented by `cargo fix -p naga --lib`.
2024-04-28 16:33:13 +02:00
Jim Blandy 1ea96391ea [hal] Document Api::Fence, its users, and its Vulkan impl. (#5618)
Also, rename `wgpu_hal::vulkan::Fence::check_active` argument to
`last_completed`, and explain better its rationale.
2024-04-28 05:46:06 +00:00
Jim Blandy 4af2e7b8fb [hal] Document Queue::submit ordering guarantees a bit. 2024-04-27 22:14:23 -07:00
Jim Blandy a2cd2b92b3 [hal] Fix cargo doc --document-private-items. Check in CI. (#5617) 2024-04-27 20:21:29 +00:00
vero 9eb1b71d33 Add some more changelog entries (#5611)
* Add some more changelog entries

* Address feedback
2024-04-26 15:06:34 +02:00
vero 739905e73c Extract a naga pub create_validator function for use in Bevy (#5606) 2024-04-26 10:19:45 +02:00
Jim Blandy d4f30638b7 [core] Fix caller location tracking in lock::ranked.
Properly track the location of callers of `Mutex::lock`,
`RwLock::read`, and `RwLock::write`, for use in panic messages.
2024-04-25 12:46:05 -07:00
teoxoy 55c9d69ba0 [naga] fix the way we adjust constant initializers when processing overrides
This fixes 2 issues:
- we used to index `adjusted_global_expressions` with the handle index of the constant instead of its initializer
- we used to adjust the initializer multiple times if the arena contained multiple `Expression::Constant`s pointing to the same constant
2024-04-25 11:04:09 -07:00
Samson 5735f85720 CreateBindGroup validation error on device mismatch (#5596)
* Fix cts_runner command invocation in readme

* Remove assertDeviceMatch from deno_webgpu in createBindGroup

This should be done as verification in wgpu-core.

* Add device mismatched check to create_buffer_binding

* Extract common logic to create_sampler_binding

* Move common logic to create_texture_binding and add device mismatch check
2024-04-25 12:17:00 +02:00