The previous matmul is using 1-dimensional shared array. This PR
adds the 2-dimensional shared array. The perf result shows that:
1. For 1-dimensional shared array, enabe-robustness wil bring almost 2x
regression both for matmulFloat and matmulVec4 on Intel CFL.
2. For 2-dimensional shared array, enabe-robustness will bring almost 2x
regression on matmulFloat. But have little impact on matmulVec4 on Intel
CFL.
Tested on Intel_R_UHD_Graphics_630.
shader enable robustness disable robustness
MatMulFloatOneDimSharedArray 5383 us 3105 us
MatMulFloatTwoDimSharedArray 4788 us 2608 us
MatMulVec4OneDimSharedArray 3070 us 1743 us
MatMulVec4TwoDimSharedArray 1840 us 1802 us
Bug: dawn:594
Change-Id: Ia29a78cf70649ef8d3ba8476db1ad4d6ded80840
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50481
Reviewed-by: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiajia Qin <jiajia.qin@intel.com>
Added in https://dawn-review.googlesource.com/c/tint/+/50742
Bindings may not collide except on the D3D12 backend where we use
the register offset as the binding number. Offsets may alias if
they are for different HLSL register types (buffer, texture, etc.)
Bug: tint:797
Change-Id: I020e3b5cd1d35fdf5678b587225507e0cf24f5b3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50801
Commit-Queue: Austin Eng <enga@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Auto-Submit: Austin Eng <enga@chromium.org>
Reviewed-by: Ben Clayton <bclayton@google.com>
Certain backends performs transformations on the tint::Program in the ShaderModule.
These transformations may raise diagnostics.
Tint diagnostics generated from an underlying tint::Source::File need that File kept alive.
Change-Id: I32deceb4cedb419e914631d342884fc27972b5cb
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50581
Auto-Submit: Ben Clayton <bclayton@google.com>
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Skip these tests on D3D12 because it is not implemented yet.
Bug: dawn:547
Change-Id: I0a009b00bd26bb04794bb627c53f4540462f1ca7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50243
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Yunchao He <yunchao.he@intel.com>
Bug: dawn:783
Change-Id: I3c7f0ffc3f45a0d67b411a39342e89c710604d54
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50244
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Stephen White <senorblanco@chromium.org>
This is more consistent with the other backends that have a
CommandRecordingContext.
Bug: dawn:780
Change-Id: Icb689d00cc4d873a489fc4d4c247feeb15f8797c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50120
Commit-Queue: Stephen White <senorblanco@chromium.org>
Auto-Submit: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Adds UNASSIGNED-CoreValidation-DrawState-InvalidImageLayout (0x4dae5635)
initially, since we can see that it's causing issues with the WebGPU CTS
tests.
Change-Id: I862e3986d59ae590458f9011a16cc824bf752e72
Bug: dawn:785
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50423
Commit-Queue: Brandon Jones <bajones@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
This patch skips CopyTests_T2B/CopyOneRowWithDepth32Float on all
D3D12 backends because it fails on many D3D12 drivers.
See https://crbug.com/dawn/727 for more details.
BUG=dawn:727
TEST=dawn_end2end_tests
Change-Id: I3d2c4cf363772939796a71c7efa165e57ed6ba41
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50280
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
This is now handled inside the SPIR-V sanitizing transform and can be
enabled via a transform config option.
Fixed: tint:753
Change-Id: I897c60dd304dd4643cc905a8ca3001dd963c5fb0
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/50060
Auto-Submit: James Price <jrprice@google.com>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
This CL complete the validation rules for CopyTextureForBrowser by:
- Restrict source texture must have |CopySrc| and |Sampled| usage.
- Restrict destinaton texture must have |CopyDst| and
|RenderAttachment| usage.
- Restrict sample counts of source texture and destination texture
must be 1.
- Restrict source copy origin.z must be 0.
- Restrict CopyTextureForBrowser() can only copy to single slice.
A validation unittest is added to check.
BUG=dawn:465
Change-Id: I5e645a4b69edeaf97ce1231bd7c8036027524ba8
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49306
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Below are the list of all the individual changes, which are a good order
in which to review this CL.
Core changes:
- Change the tracking in the frontend to produce a synchronization
scope per dispatch instead of per compute pass. Some bindgroups might
not be part of any synchronization scopes so we also track all the
referenced resources on the side so they can be checked during
Queue::Submit validation.
- Fix clearing in the GL and Metal backends to use the per-dispatch
synchronization scopes.
- Fix the Vulkan backend to use the per dispatch synchronization scopes
to produce the correct pipeline barriers. This allows the removal of
previous logic that was subtly incorrect for Indirect buffer. This
allows the merging of the Compute and Render DescriptorSetTracker into
a single small helper class.
- D3D12 changes are similar to Vulkan, but the simplification is just a
the suppression of a branch with a lot of code in
BindGroupStateTracker.
Test changes:
- Fixup all the ResourceUsageTracking tests to follow the WebGPU spec
for synchronization scopes (fixing a lot of TODOs).
- Add additional tests checking that Indirect buffers are not allowed
to be used as a writeable storage in the same synchronization scope.
- Add tests for Queue::Submit validation correctly taking into account
resources that are bound but unused in compute passes.
- Add an end2end test for using a buffer as Indirect and Storage at the
same time in a DispatchIndirect, which would previously produce
incorrect barriers in the Vulkan and D3D12 backends.
Other small changes (that I was to lazy to put in a different CL):
- Add the utils::MakePipelineLayout helper function.
- Fix Indirect not being in the list of readonly buffer usages (caught
by a test added in this CL).
Bug: dawn:632
Change-Id: I77263c3535a4ba995faccbf26255da9a2f6ed3b5
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49887
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
This workaround was only necessary in the Metal backend but it was easy
to put it in the frontend so it was put there. However, this simplification
gets in the way of a validation change in the follow-up CL. So we move
the nooping to the Metal backend.
Bug: dawn:632
Change-Id: I0a9957a1bc9bdd7867eb8c60fcab710832c13af2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49886
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This change is a preparation for making the compute pass track the
synchronization scope usages per dispatch instead of for the whole pass.
This CL just separates the Compute and RenderPassResourceUsage types.
This requires making the difference between
SyncScope/ComputePass/RenderPass ResourceUsageTracker instead of having
a single combined tracker.
This change also duplicates SetBindGroup by removing the common handling
in ProgrammablePassEncoder and putting it in ComputePassEncoder and
RenderEncoderBase. This is necessary because the UsageTracker types are
now split, but it will also help have different handling of SetBindGroup
for compute and render in follow-up CLs.
There are no functional changes.
Bug: dawn:632
Change-Id: I482c04483d8b734fb10e44e717071eedcff2f15f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49884
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This makes error messages from command buffers more useful because they
keep the whole stack trace instead of just showing that the error was
created in the CommandBuffer::Finish call.
Bug: dawn:632
Change-Id: I23e66045c3caa1ad086003a04eed78c40aefc562
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49885
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This change is a preparation for making the compute pass track the
synchronization scope usages per dispatch instead of for the whole pass.
- Split the tracking of render and compute passes usages.
- Rename PassTextureUsage to TextureSubresourceUsage since is it not
per-pass
- Add SyncScopeResourceUsage as a subclass of PassResourceUsage to
start modifying some of the code to work with synchronization scopes
(even if syncscope == pass at the moment).
There are no functional changes.
Bug: dawn:632
Change-Id: Ieeb6d70a44dc1c726f26989eebcd87e63e732785
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49883
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This patch implements CallbackTask and CallbackTaskManager to store
the callbacks of Create*PipelineAsync().
In the futureCallbackTaskManager will manage all the callbacks that
should be called in Device.Tick().
BUG=dawn:529
Change-Id: I6ad4352371eb44515bc2d85cdc68220c9b758b8e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49060
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Mip dimension should be greater than or equal to 1, while width >> level
may lead to 0.
Bug: dawn:547
Change-Id: Ib3dfb9fbdbed0e922df6efa366598eff0ca10df2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49506
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Note that a slice somehow means a subresource on D3D12. There
are mip slice, array slice, and plane/aspect slice in D3D12.
We reuse the term "slice" for multiple depth of a 3D texture,
although one single depth slice of multiple depth slices is
not a separate subresource of a 3D texture (all these depth
slices for one mip are a separte subresource in 3D texture).
For the reason above, this change also renames "slice" to
"layer" in some functions if "slice" is a layer in that
function. Because a layer is definitely a subresource but a
slice may not be (like a single depth slice of a 3D texture).
Bug: dawn:547
Change-Id: I88b8120ef7f73bfc261fc225f4242924da221654
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49240
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
This patch fixes the incorrect return status of Create*PipelineAsync
when device is lost by explicitly calling and clearing all the
Create*PipelineAsync callbacks in DeviceBase::HandleError() when the
device is lost.
BUG=dawn:529
TEST=dawn_end2end_tests
Change-Id: I67a8047b2e5a54f6f85c5a4cbcf420b744ac0d5c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49080
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Remove d3d12.li, dxgi.lib and d3d11.lib dependencies when targeting UWP.
Add dxgi.lib only for DXGIGetDebugInterface1 in debug build when targeting UWP.
Use DXGIGetDebugInterface1 only in debug build when targeting UWP.
Bug: dawn:766
Change-Id: I5fa53dbb257acf604836f861f75a122a7d417e7c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49040
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: 陈俊嘉 <cjj19970505@live.cn>
This patch removes an incorrect ASSERT(IsComplete()) in the
destructor of AsyncWaitableEvent because when the destructor of
AsyncWaitableEvent is called, the async task attached to mFuture
may not be completed and the status of mFuture may not be 'ready'.
In fact in C++14 we can always guarantee the attached async task
is completed after the destruction of mFuture.
BUG=dawn:529
Change-Id: I2f28246beb025a0d39dd432a404c0b04aed17249
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48900
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
GBV's ability to patch shaders with validation code
was hanging several end2end tests. Since disabling
GBV shader patching would render partial validation
ineffective, partial validation was re-mapped to have
GBV disabled while full validation enables GBV by default.
End2end tests now default to partial validation so
tests that would otherwise not run with GBV can still
have some coverage.
BUG=dawn:592
Change-Id: Iafb271b5ee8862592152fa45c34f188d64432e60
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46725
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Bryan Bernhart <bryan.bernhart@intel.com>
Auto-Submit: Bryan Bernhart <bryan.bernhart@intel.com>
This avoids breaking dependents when they use a version of clang that's
different than the one being tested on Dawn CQ.
Also re-enable -Wdeprecated-copy(-dtor)
Bug: chromium:1201858
Change-Id: I09763fdb98e7dba53b4d9fa66fa4c606620a0078
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49100
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Nico Weber <thakis@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Remove the following 2 function declarations since their function
bodies are no longer available.
1. ResultOrError<Ref<RenderPipelineBase>> CreateRenderPipeline(
const RenderPipelineDescriptor* descriptor);
in src/dawn_native/Device.h
2. BufferUploader* GetBufferUploader() const;
in src/dawn_native/vulkan/DeviceVk.h
BUG=none
Change-Id: I55a92a4acb3af3ea4807bfd0b58a0246d0b28cc9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/49061
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Add necessary cflags and cflags_cc for winrt compilation.
Add 'dawn_is_winuwp'.
Set 'dawn_enable_vulkan' and 'dawn_supports_glfw_for_windowing' when compiling for UWP.
Link d3d12, d3d11 and dxgi stub libs when compiling for UWP.
Use LoadPackagedLibrary instead of LoadLibraryA in DynamicLib when compiling for UWP.
Swapchain related changes will be in another commit.
Bug: dawn:766
Change-Id: I1210798a21cc175bab77281403d262d4bfb02d99
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48480
Commit-Queue: 陈俊嘉 <cjj19970505@live.cn>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
This CL adds two helpers for more ergonomic processing of
ChainedStructs.
1. FindInChain(): Iterates through the chain and automatically
casts the ChainedStruct into the appropriate child type before
returning.
2. ValidateSTypes(): Verifies that the chain only contains structs
with sTypes from a pre-defined set. This also allows the caller
to specify one-of constraints.
3. ValidateSingleSType(): Verifies that the chain contains a
single struct with a specific sType or is an empty chain. This
is a common case of |ValidateSTypes()| and is separated out as
a fast-path.
Change-Id: I938df0bf2a9b1800b1105fb7f80fbde20bef8ec8
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47680
Commit-Queue: Brian Ho <hob@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Previously this was being indirectly supplied by <memory>, which has
been removed. Clang happily still finds std::forward, but GCC does
not.
BUG=dawn:768
Change-Id: I87e6bb6aa83bc7f3253c58e32fa0ba977752858c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48820
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
- Add vkCmdFillBuffer in ResolveQuerySet to clear the buffer to 0s for
these unavailable queries if the buffer has been initialized or fully
used which won't been initialized with 0s again.
- Because vkCmdFillBuffer has driver issue on Intel Windows, Skip some
affected cases.
- Remove unsafe api checking from Occlusion Query.
Bug: dawn:434
Change-Id: Ib34f81d93b0de8f08f0eeebf3c8a967eeb5ecefb
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48320
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This will help detect cases where the mapped data is used after it is
freed, in particular in WebGPU tests around the interaction of mapping
and GC.
Bug: chromium:971949
Change-Id: I820d9885d39379fbc95c6504b9a4151053768d93
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48382
Reviewed-by: Brandon Jones <bajones@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This patch moves the initialization of DXC libraries from
BackendD3D12::GetOrCreateXXX() to Device::Initialize() so
that we don't need to deal with those lazy initializations
in the asynchronous path of Create*PipelineAsync().
BUG=dawn:529
Change-Id: I792847c138738ae8c300d7a1bf3d23fc8fecd746
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48580
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
This test was passing on SPIRV-Cross due to SPRIV-Cross's
overreporting of binding information per entry point. On Tint it was
failing, because the resources in the vertex shader were not being
referenced, so not being reported. I have corrected the values passed
to the utility code that generates the shaders, so that it generates
the needed references.
I have also slightly tweaked the test, so it will distinguish between
failure states better.
BUG=tint:716
Change-Id: If88f600a36cba8b580d888e902b8d8102e05bd10
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48520
Auto-Submit: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
unique_ptr's destructor sets itself to null and frees its owned
memory. This is a problem because for the slab allocator, the
member variable holding the unique_ptr is inside the freed memory.
Bug: skia:10501
Change-Id: I41179261041fe415bb2af3667114b079f61b3c7b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48100
Auto-Submit: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Use tint::transform::DataMap for inputs as well as outputs.
This allows tint to nest transforms inside each other (e.g. embedding
transforms inside sanitizers), and still having a consistent way to pass
data in and out of these transforms, regardless of nesting depth.
Transforms can also now be fully pre-built and used multiple times as
there is no state held by the transform itself.
Bug: tint:389
Change-Id: If1616c77f2776be449021a32f4a6b0b89159aa2a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48060
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Auto-Submit: Ben Clayton <bclayton@google.com>
Previously all uses of reentrant object creation in Dawn native
needed to manually AcquireRef. Change them to use CreateFooInternal that
returns a ResultOrError<Ref<>> and are renamed to CreateFoo.
Bug: dawn:723
Change-Id: Ifcda3659d02cc5a4c63c248dc53af7fee7c4a61d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46626
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
As of https://github.com/gpuweb/gpuweb/pull/1352 the spec indicates that
GPURenderPassColorAttachmentDescriptor and
GPURenderPassDepthStencilAttachmentDescriptor should use .view rather
than .attachment to indicate the TextureView associated with the
render pass attachment.
Bug: dawn:762
Change-Id: I70d615e19d8e7aae5b26aa5965c7109289ab868b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47902
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Brandon Jones <bajones@chromium.org>
Tint now sorts struct members with entry point IO attributes.
Bug: tint:710
Change-Id: I7bb37cab2f101fbcf55dd8b42ee3b3af1250bf1a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47840
Commit-Queue: Brandon Jones <bajones@chromium.org>
Auto-Submit: James Price <jrprice@google.com>
Reviewed-by: Brandon Jones <bajones@chromium.org>
This reverts commit 5c9b6a8f81.
The issue has now been fixed in Tint and rolled into Dawn.
Fixed: dawn:758
Change-Id: Id7deb6cea9f22ee8c5f345a28d2d37499b47d692
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/48080
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: James Price <jrprice@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Matches most recent spec changes. setBlendColor has been marked as
deprecated.
Bug: chromium:1199057
Change-Id: I4584ce789bd7d14401244509d5ada62a46236a5d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47901
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Brandon Jones <bajones@chromium.org>
Several of the enum names have changed recently. Update them to match
the spec and mark the older ones as deprecated.
BUG: chromium:1199057
Change-Id: I7a29588dd18b8fb738773c2478b173093f2aa834
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47860
Commit-Queue: Brandon Jones <bajones@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
A Dawn CL landed at the same time as a Tint roll in Dawn, causing
failures in a large number of D3D12 use_tint_generator. Disable testing
of that configuration entirely while we investigate the issue and fix
it.
Bug: dawn:758
Change-Id: I6c897efc637821a20281c91668889ae95c1c2b1a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47774
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Changes I/O to use function parameters and return values, removes
unnecessary "-> void" return types, and changes "const" to "let".
BUG: dawn:755
Change-Id: Iabbfcc280fae37d73cba6a2f7e2215ed579a04e0
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47700
Reviewed-by: Brandon Jones <bajones@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Brandon Jones <bajones@chromium.org>
Fixes issues with OpenGL when using the newer style of WGSL I/O that
relies on function parameters and return values. Also added several
end2end tests to cover different variants of that style of I/O.
BUG: dawn:755
Change-Id: I73b4b07ce924ce22e917bf399be705dc67519037
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47701
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Refactors support for toggle control flags, --enable-toggles= &
--disable-toggles, from the end2end tests into a utility class to make
them available elsewhere. The unittests now uses this utility class to
parse toggle control flags.
For some toggles, like 'use_tint_generator', the unittests are known to
be broken with them turned on. They will be fixed in subsequent CLs.
BUG=dawn:756
Change-Id: Ic283e8a82bedcbf255258cca3e62f58c79d3857b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47740
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Ryan Harrison <rharrison@chromium.org>
Use the same pattern for all conditionals: single-call, then row-by-row,
2D case, then 3D.
Bug: dawn:684
Change-Id: I410183815299e9ec2d90790809a056dd578771e3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47320
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
We augment the DoTexSubImage() call for the desktop GL fast path
for compressed textures (upload in a single call where possible).
Bug: dawn:684
Change-Id: Id67c39b1efbc8b435b58064cad66a55c153ce675
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47240
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Also drive-by fixes some other deprecated constructs (const -> let, and
a disabled test having ancient WGSL).
Bug: dawn:755
Change-Id: I924dfbcbd0a7d0478f3e9b3766585751a0392499
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47620
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Ben Clayton <bclayton@google.com>
Depth/stencil formats are invalid for 1D and 3D texture.
This change adds this validation rule, and adds a validation
test for it.
Bug: dawn:730
Change-Id: Idac6d1bf7b8c7261eb7b4b59504de10e13d049cc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47200
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
This will help experiment using dawn_wire for remoting WebGPU to render
on the screen.
Bug: None
Change-Id: I9a60ff8c3889ec917f6fd56e4cbb1ffef639748d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47621
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Move texture uploads into CommandBufferGL::DoTexSubImage() and use it
for both CommandBuffer CopyBufferToTexture and
QueueGL::WriteTextureImpl(). On the CB side, For now this is only used
for compressed ES textures. Desktop GL has a fast-path for compressed
textures that isn't currently implemented.
Bug: dawn:684
Change-Id: I4da02e9c96c13fd71d133778168a5597efa7b59a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47123
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Stephen White <senorblanco@chromium.org>
The HLSL writer issue is now fixed.
Fixed: tint:686
Change-Id: I6d4709652d1ca46971a4ad2d150b21ad954ec055
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46450
Auto-Submit: Ben Clayton <bclayton@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Replace TODOs for those tests that now fail with another issue.
Fixed: tint:682
Change-Id: I2665f0d4f495e4efc5adef4cf46a965df7c8936c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46451
Auto-Submit: Ben Clayton <bclayton@google.com>
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Now that we have a real extension type (PrimitiveDepthClampingState),
we can replace the dummy extensions used in the WireExtensionTests
with the real deal.
Change-Id: I43dfa66f2cc909b5fc95b55382d3cb53ce022a0f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47300
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Brian Ho <hob@chromium.org>
Tint diagnostics hold a raw pointer to the tint::Source::File, so that source snippets can be printed.
The tint::Source::File pointer was taken from a stack allocation, and the diagnostics were left with a dead pointer.
Have the ShaderModuleParseResult keep the tint::Source::File alive with a PIMPL unique_ptr.
Change-Id: I80d77516afd4012e15992821e060ef3038eba1c1
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47422
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Constructing a float2x2 from a float4 in a ConstantBuffer by extracting individual scalars with the array-index operator appears to produce invalid DXBC.
Tweak the test so that we use .xy, .zw swizzles instead.
Bug: tint:681
Change-Id: I027f8df96f2a5fda355668b35773d79f4fcdc08b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47421
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Tint only reflects the bindings that are statically used so before this
fix it would produce an empty pipeline layout. Fix this by statically
referencing the sampler binding in the shader.
Fixed: dawn:754
Change-Id: I381acd9adb7d1e80ccf0642b5189750a5a43de25
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47420
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Validation is now automatically performed, and the Validator class is
now redundant and will soon be removed from Tint.
Bug: tint:642
Change-Id: I9ddbe523d85e2a4dfeff6d08324a6ca368d6cc74
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47124
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
The various backends hit UNREACHABLE() during pipeline creation if
depthStencil.depthCompare (or likewise for stencil) are set to
undefined.
Bug: chromium:1195694
Change-Id: Ibf4d8d47b4c98343dce3caccdf79ee90c0de899f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46863
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
This change also marks 3D view creation as Unsafe API.
BUG: dawn:547
Change-Id: Icdb7b48f19054d70258363f6d58ded957be72b70
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46723
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Test is flaky on the NV bot so it's being disabled
as I investigate.
BUG=dawn:733
Change-Id: I9b721f05665cefe8a6eee53fe691f83071484965
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47260
Reviewed-by: Corentin Wallez <cwallez@google.com>
Commit-Queue: Bryan Bernhart <bryan.bernhart@intel.com>
After we applied __VA_ARGS__ in macros and use {x, y} and {w, h}
to replace (x, y) and (w, h) in EXPECT_TEXTURE_RGBA8_EQ and
EXPECT_TEXTURE_FLOAT_EQ, we can use the more general macro
EXPECT_TEXTURE_EQ. Then these two macros can be removed. Austin
has already put a TODO for this change but didn't do that.
utils::MakeOrigin and utils::MakeExtent are removed because they
are not needed.
In addition, this change removes 0 in callers for parameter level,
because level's default values is 0 in helpers implementation.
BUG: dawn:748
Change-Id: Iece4db7a8ed1d47b57988412f1c897205e7403d4
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47100
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
The previous CL had no way of communicating disagnostic messages from
Tint to the ShaderModule in the event that it failed to validate. This
change ensures that messages generated during validation aren't dropped
on the floor and can be queried from the failed modules CompilationInfo.
BUG: dawn:746
Change-Id: Ic2551654ca30baab0fb5124a148eb4fcdf4b0f22
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46960
Commit-Queue: Brandon Jones <bajones@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
This patch refactors the implementation of APICreateComputePipelineAsync
as a preparation of the async path of the creation of compute pipeline.
Now the code path of APICreateComputePipelineAsync() includes the following
3 parts:
- When an error occurs in the front-end validations, the callback will be
called at once in the main thread.
- When we can find a proper compute pipeline object in the cache, the
callback will be called at once in the main thread.
- When we cannot find the proper comptue pipeline object in the cache, the
newly-created pipeline object, the callback and userdata will be saved
into the CreatePipelineAsyncTracker, and the callback will be called in
device.Tick(). All the logic mentioned in this section has been put into
one function CreateComputePipelineAsyncImpl(), which will be overrided
by its asynchronous version on all the backends that support creating
pipeline objects asynchronously.
Note that APICreateRenderPipelineAsync is not changed in this patch because
it is now under refactoring to match the current updates in WebGPU SPEC.
BUG=dawn:529
TEST=dawn_end2end_tests
Change-Id: Ie1cf2f9fc8e18c3e6ad723c6a0cefce29a0eb69c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/45842
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Most tests pass, those that don't are skipped.
Change-Id: Ia08373551a48382aceffbbc0e23bd15828ebe219
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46452
Commit-Queue: Austin Eng <enga@chromium.org>
Auto-Submit: Ben Clayton <bclayton@google.com>
Reviewed-by: Austin Eng <enga@chromium.org>
If a texture format supports multisample, it should be renderable.
This change adds this validation rule to fix a bug. It also adds a
validation test in dawn_unittests.
BUG: dawn:731
Change-Id: I33a06cb16367e4e379b29b223ef6b69128baf30f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46840
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
This change removes layer and use origin.z in MACROs
EXPECT_TEXTURE_FLOAT_EQ and EXPECT_TEXTURE_EQ.
It also removes parameter layer in MACROs' implementation
functions AddTextureExpectation and AddTextureExpectationImpl
and use origin.z instead in these functions.
BUG: dawn:747
Change-Id: I0103524074c6c250fa2f85bf9c3abda8c8cd2197
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/47000
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
We do query resetting for each query commands outside render pass,
whether they're rewritten or not, so no longer need to track their
availability on command encoder. The availability on query set is enough
for resolving sparse queries.
But we still need to track query availability on render pass for query
rewrite checking and query resetting per render pass. Because reset
command must be called outside render pass, we need to reset them
together before the beginning render pass based that. Add availability
tracking on pass resource usage tracker (we only need it on render pass
) to facilitate use it in Vulkan backend.
Bug: dawn:434
Change-Id: Ie1b413ff54f62f3b84fe612e4abe45872c387e81
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/45440
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Hao Li <hao.x.li@intel.com>
This change implements 3D texture copy on D3D12 for texture to
buffer and texture to texture copy for full copy upon the entire
3D texture.
It also uses a function named CopyBufferToTexture to wrap
CopyBufferTo3DTexture and CopyBufferTo2DTextureWithCopySplits.
Likewise, it uses another function named CopyTextureToBuffer to
wrap Copy3DTextureToBuffer and Copy2DTextureToBufferWithCopySplits.
BUG: dawn:547
Change-Id: I4293f6ca4d37f604e6f1c10827686d17469a07ca
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46820
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Dawn uses objects with move semantics like Ref<> that feel like they
need to be std::moved in a bunch of places. This warning triggers when
std::move() is used when it has no effect, keep the code less verbose.
Bug: None
Change-Id: I2b11f709e943e7b16024bfde07910577b983aa34
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/46441
Commit-Queue: Stephen White <senorblanco@chromium.org>
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Qin Jiajia
Austin Eng
Ben Clayton
Brandon Jones
Jiawei Shao
James Price
Yan
Corentin Wallez
陈俊嘉
Yunchao He
Bryan Bernhart
Jiang
Hans Wennborg
Ryan Harrison
Brian Ho
Hao Li
shrekshao
Stephen White
Antonio Maiorano