- Move RefCounted to common (from dawn_native) so that we can use
it from additional places.
- Use EXPECT_ macros instead of ASSERT_ in RefCounted tests for
improved logging on failures.
- Add a missing test for Ref::Detach.
- Plug memory leak in RaceOnReferenceRelease
Change-Id: Iaa7b11b5a6fa146e3c322143279a21a4ac027547
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19903
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Rafael Cintron <rafael.cintron@microsoft.com>
This reverts commit 6d9e4f8076b645c557453f4b566bf9c38b4a51eb.
Reason for revert: Breaks the roll in Chromium, gpu_sampler.cc must first be
fixed to use the new undefined value, before this can be landed.
Original change's description:
> Add ComparisonSampler binding type and validation tests
>
> Bug: dawn:367
> Change-Id: Iba1d3d03f6247a356b6f3fabfe7a7ba3c0753171
> Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/18423
> Reviewed-by: Austin Eng <enga@chromium.org>
> Commit-Queue: Austin Eng <enga@chromium.org>
TBR=cwallez@chromium.org,kainino@chromium.org,enga@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: dawn:367
Change-Id: Ic071a601df2063bd2da5388b2e75c1a121924a69
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19983
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This CL adds type determinater support for Normalize.
Bug: tint:5
Change-Id: Idbea41942172f8404122b7a756ccf1ec5aad2af8
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19950
Reviewed-by: David Neto <dneto@google.com>
This CL adds type determiner support for the Exp, Exp2, Log, Log2, Sqrt
and InverseSqrt GLSL methods.
Bug: tint:5
Change-Id: I3b9e799a8ebe5e8c96c1daf07131feae40e0c54e
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19949
Reviewed-by: David Neto <dneto@google.com>
This CL adds support for the GLSL trig functions.
* sin
* cos
* tan
* asin
* acos
* atan
* sinh
* cosh
* tanh
* asinh
* acosh
* atanh
Bug: tint:5
Change-Id: I38c8bf45c3aeeef81711de3a3eca6a9339af146c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19948
Reviewed-by: David Neto <dneto@google.com>
This CL adds support to the type determiner for the GLSL Radians and
Degrees methods. We use the general float calls for this because we only
support FP32 and have plans for FP16. Other floating point sizes are not
supported but we have no support for them..
Bug: tint:5
Change-Id: I38f0551ce5f9ed7dd31496f13880697cd1f21ba4
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19947
Reviewed-by: David Neto <dneto@google.com>
This CL adds Fract into the type determiner.
Bug: tint:5
Change-Id: I616d3259ca903b82339f49317f27966265c3f1d5
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19946
Reviewed-by: David Neto <dneto@google.com>
This CL adds Ceil support to the type determiner
Bug: tint:5
Change-Id: Iba785dbea412cc94f9b0faa3e6d38f02453d5215
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19945
Reviewed-by: David Neto <dneto@google.com>
This CL adds Floor into the type determination
Bug: tint:5
Change-Id: I2b9013e073458fbe9f8a9ff657582a715f91c480
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19944
Reviewed-by: David Neto <dneto@google.com>
This CL adds FSign to the type determination.
Bug: tint:5
Change-Id: I8e3a7e94f367c4aaa510e72f774097946382a968
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19943
Reviewed-by: David Neto <dneto@google.com>
This CL adds FAbs to the type determinator.
Bug: tint:5
Change-Id: I9793f5d469eb5a70f2a024c325e098ed9ca8ea7c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19942
Reviewed-by: David Neto <dneto@google.com>
This CL adds type determination for the Trunc method.
Bug: tint:5
Change-Id: I4f23dad7725f1f076fbbc7da8687de42a02cc482
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19941
Reviewed-by: David Neto <dneto@google.com>
This CL adds support for the GLSL RoundEven command in type
determination.
Bug: tint:5
Change-Id: I956ea9532785a42fe51a24b5121ce47cfbc78242
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19940
Reviewed-by: David Neto <dneto@google.com>
This CL adds the type determination and builder code to support
outputting the GLSL Round call.
Bug: tint:5
Change-Id: I84dadebaf19aee3361fb13b5f32ce1a9f1b0c421
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19923
Reviewed-by: David Neto <dneto@google.com>
This reverts commit f93791ab62965964f413744bb73884dd6b9abc42.
Reason for revert: breaks gl_tests on roll.
Original change's description:
> Special-case GetDefaultQueue in the wire
>
> This makes it so calling GetDefaultQueue always returns the same
> object. It required updating various WireTests to account for the
> additional wire calls.
>
> Bug: dawn:22
>
> Change-Id: I8c74374b7c732b8bb7d0490bbc740dee0d2dface
> Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19726
> Commit-Queue: Corentin Wallez <cwallez@chromium.org>
> Reviewed-by: Austin Eng <enga@chromium.org>
> Reviewed-by: Kai Ninomiya <kainino@chromium.org>
TBR=cwallez@chromium.org,kainino@chromium.org,enga@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: dawn:22
Change-Id: Id2f051b1d4be64a6e16ee8bbe998d72028660334
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19980
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This CL renames the parts of the identifier to segments and then adds a
path() and name() methods to get the respective parts of the identifier.
Change-Id: I1a365df7cb83ac0dd3d6be9ec6ccb83bfc73dd91
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19762
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL moves the ast::Module to be provided to the type determiner
constructor so we can access it for things like the imports.
Change-Id: I110fffe669a8a007461bf84d30d85d70405e37be
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19761
Reviewed-by: David Neto <dneto@google.com>
This CL updates the type determiner test helper to have a ast::Module
and updates all tests to use the module. The tests have all be updated
to consistently use the type determiner provided by the helper as well.
Change-Id: If47a873c439e5a5019e21679cba957d4c762f4e6
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19780
Reviewed-by: David Neto <dneto@google.com>
This CL adds type determination for the call parameters.
Bug: tint:5
Change-Id: I488718bd7a4c1f2304a1c17554b8354d184dc159
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19760
Reviewed-by: David Neto <dneto@google.com>
This CL moves the ast::Module to the constructor of the SPIR-V builder
class. This allows access to things like the ast::Import from the
builder class.
Bug: tint:5
Change-Id: I0d61b164c1274d006e124dde2cc11f41a2385fc5
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19922
Reviewed-by: David Neto <dneto@google.com>
This CL moves the checks for different types into the type class so it
can be used in both the type determinater and the SPIR-V builder.
Change-Id: I9142adaf5fc1d6048792645d7892f8d8900fcf59
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19921
Reviewed-by: David Neto <dneto@google.com>
This CL adds the initial generation of a loop construct into the
spirv-writer. This does not support break or continue yet.
Bug: tint:5
Change-Id: I41f5d9b634a1a4120f880c4143feacc7e58e2147
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19620
Reviewed-by: David Neto <dneto@google.com>
This patch adds validation tests for invisible and unused bindings.
The tests create bindings for compute, fragment and none stage. So,
only some bindings are visible for a particular render/compute pass.
In addition, some visible bindings in bind groups are not used by
pipeline. But resources of all bindings in bind groups (including
invisible/unused bindings) should be tracked, in order to identify
resource usage conflicts.
Bug: dawn:365
Change-Id: Ibb8d45a1a7b69891a17b4949147a16ada4ed6bd8
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/18120
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
This makes it so calling GetDefaultQueue always returns the same
object. It required updating various WireTests to account for the
additional wire calls.
Bug: dawn:22
Change-Id: I8c74374b7c732b8bb7d0490bbc740dee0d2dface
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19726
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
This makes all backends register the default queue at device
initialization time, so that the same queue is returned by
each call to GetDefaultQueue.
All usages of CreateQueue are replaced by GetDefaultQueue
except a couple ones that could use the queue initialized by
DawnTest::SetUp.
A deprecation warning mechanism is added so that users of Dawn
can now that they should upgrade their usage of the API. It also
comes with a backdoor so we can test that they are emitted.
New DeprecatedAPITests are added that will contain tests for
deprecated APIs, and will also check that deprecation warnings
are produced.
The special casing of GetDefaultQueue in the wire will be done
in a follow-up CL to ease the review. It happens to work through
the regular wire mechanisms at the moment but returns a different
object on each GetDefaultQueue call.
Bug: dawn:22
Change-Id: I78dc1fa474769674278d30040e8d05c658b88360
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19724
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This target was forgotten in the previous cleanup.
Bug: chromium:1064305
Change-Id: Ibf35df4393927ac8cc08d4965b8b8098c36e4bd3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19860
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
https://swiftshader.googlesource.com/SwiftShader/+log/63ed0e445fa5..d25ce8725224
$ git log 63ed0e445..d25ce8725 --date=short --no-merges --format='%ad %ae %s'
2020-04-15 digit Fix LLVM-based Arm64 build.
2020-04-16 bclayton CMakeLists: Enabled more clang warnings.
2020-04-16 bclayton Device/Blitter: Initialize filter3D fields
2020-04-15 digit Fuchsia: Add LLVM 10.0 configuration headers.
2020-04-16 amaiorano CMake: fix PowerVR_Examples build
2020-04-16 amaiorano Don't fail build if Vulkan headers are different
2020-04-15 amaiorano CMake: fix add_subdirectory(SwiftShader) when CMAKE_BUILD_TYPE is set in parent project
2020-04-15 amaiorano CMake: make submodule initialization work for add_subdirectory(SwiftShader)
2020-04-15 amaiorano CMake: fix add_subdirectory(SwiftShader)
2020-04-15 amaiorano CMake: use directory variables consistently
2020-04-15 amaiorano CMake: make root compile options and link libraries private
2020-04-15 bclayton VkSemaphore: Unlock/Lock the mutex outside the blocking_call.
2020-04-16 bclayton Regres: Include a better error message when git add fails.
2020-04-16 bclayton Regres: Avoid dereferencing a nil on process timeout.
2020-04-15 bclayton Make memoryPageSize() functions thread-safe.
2020-04-14 bclayton Vulkan: Move SpecializationInfo to own file.
2020-04-16 bclayton Fix chrome autorollers
2020-04-04 bclayton Reimplement LRUCache, fold away LRUSnapshotCache, add tests.
2020-04-14 amaiorano CMake: split out turbo-cov target
2020-04-14 amaiorano CMake: split out test and benchmark targets
2020-04-08 bclayton VkPipelineCache: Do not publically expose internal mutexes
2020-04-14 bclayton CMake: Fix build when building with `SWIFTSHADER_ENABLE_VULKAN_DEBUGGER`
2020-04-14 sugoi Fix fragments depth values not being clamped
2020-04-08 amaiorano CMake: split out Vulkan sources into separate CMakeLists
2020-04-13 bclayton Device: Rename RoutineCacheT to RoutineCache
2020-04-14 bclayton Device: Don't inherit [Vertex,Setup,Pixel]Processor.
2019-09-09 digit [vulkan] Implement VK_FUCHSIA_external_memory extension.
2020-04-09 capn Reuse Store logic for Modf and Frexp output parameters
2020-04-09 capn Separate Operand store logic from EmitStore()
2020-04-09 capn Only store component count in Operand
2020-04-14 capn Rename size/sizeInComponents to componentCount
2020-04-09 capn Use the type and result ID helpers
2020-04-08 capn Obtain type ID from instruction
2020-04-08 capn Add helper methods for obtaining type and result ID
2020-04-11 bclayton LLVMReactor: Replace deprecated CreateCall overload
2019-11-19 digit [cmake] Add support for CMAKE_SYSTEM_NAME == "Fuchsia"
2020-04-08 capn Rename GenericValue to Operand
2020-04-09 bclayton CMake: Fix linux build with SWIFTSHADER_BUILD_PVR
2020-04-09 bclayton Vulkan/Debug: Fix clang-10 warning
2020-04-09 bclayton third_party/cppdap: Roll forward to 4dcca577
2020-04-08 cwallez Clamp _SNORM formats to -1.0
2020-04-08 sugoi Remove debug only code limiting the number of primitives
2020-04-08 bclayton Regres: Post coverage results even if the test lists have not changed.
2020-04-07 amaiorano CMake: split out all src/OpenGL targets into their own CMakeLists
2020-04-06 amaiorano CMake: split out SwiftShader GL deps into separate files
2020-04-06 amaiorano CMake: clean up dependencies
2020-04-06 bclayton Add SystemBenchmarks.
2020-04-07 bclayton VkCommandBuffer: Remove static Cast() method
Created with:
roll-dep third_party/swiftshader
TBR=enga@chromium.org
Bug: dawn:283
Change-Id: I5326bced6a4863ae780cafdbaf06060ef55aaf27
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19841
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This layer replaces VK_LAYER_LUNARG_standard_validation that is now
deprecated.
Bug: dawn:388
Change-Id: I488c175b997434a8e22deecfba3fc89495969cbc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19722
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
On macOS we can't rely on having the Vulkan loader installed in the
system. So we add the Vulkan loader as an optional dependency of Dawn
and use it on macOS when building Dawn in standalone with Vulkan
support.
Usage of building our own loader might broaden if the loader gains
features that are useful on other OSes. For example the ability to pass
in the "root ICD" entrypoint to the loader so we can have both
Swiftshader and the system driver at the same time.
Bug: dawn:388
Change-Id: I7ade4961cce0463c66846ad17aebf95224f1afcc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19723
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
It's possible for a heap in the residency LRU to outlive the
ResidencyManager. When this happens, some heap in the LRU will be
referencing the LRU head node. On destruction, the outstanding heap
will attempt to access the LRU head node after the memory has been
freed. This commit removes the LinkedList head node from the list
within the LinkedList destructor to fix the bug.
Bug: dawn:387
Change-Id: I13617d1b4e464e1541f989f31caecd4305037019
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19581
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Brandon Jones <brandon1.jones@intel.com>
In the case that a previous submit to the GPU required Dawn to exceed
the residency budget, Dawn must attempt to get back under budget the
next time EnsureCanMakeResident is called. This CL fixes a bug where we
only evicted the current size needed to be resident, which would
incorrectly raise the budget when a submit to the GPU required us to
exceed the budget.
Bug: dawn:193
Change-Id: I04649056e9eecce1fc6b5759c889b52a4f1b4594
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19440
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Brandon Jones <brandon1.jones@intel.com>
Makes it easier to track where and why a test failed.
Bug: none
Change-Id: Icdccde97db68c32c35af6c044a3eacd4d4978b61
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19701
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Accessing the d3d device using COM needlessly
refcounts. This is a particular issue in areas
that frequently access the d3d device like
Populate().
BUG=dawn:155
Change-Id: I24e83093623afd02fa592d8ec0c404b4571c374b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19703
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Bryan Bernhart <bryan.bernhart@intel.com>
GMock generates a lot of code in the destuctor of mock classes, so
factor that in mock_webgpu.cpp instead of compiling the code once for
each file including mock_webgpu.h.
Bug: None
Change-Id: I2b5984e911020eb33457bde9c6a45beaa93b3071
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19725
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
This CL adds binary multiplication generation to the SPIR-V writer.
Bug: tint:5
Change-Id: I668d24035e947c51a9737549fd0841a4e8af1331
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/19700
Reviewed-by: David Neto <dneto@google.com>
DynamicUploader/RingBuffer were incorrectly assuming that they could not
get empty allocation requests. Fix this and add a test.
The test also surfaced a bug in the Metal backend where the command
recording context could be left with a blit encoder open that was not
properly handled on device shutdown.
Bug: chromium:1069076
Change-Id: I9793b37142bd509254ce2894fa9f6208e9a68048
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/19291
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
