Emit the "[loop]" attribute on "for" and "while" so that FXC does not
attempt to unroll them. This is to work around an FXC bug where it fails
to unroll loops with gradient operations.
FXC ostensibly unrolls such loops because gradient operations require
uniform control flow, and loops that have varying iterations may
possibly not be uniform. Tint will eventually validate that control flow
is indeed uniform, so forcing FXC to avoid unrolling in these cases
should be fine.
Bug: tint:1112
Change-Id: I10077f8b62fbbb230a0003f3864c75a8fe0e1d18
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/69880
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
Failures in the following tests are preventing Dawn from rolling into
Chromium due to errors when run against D3D12_NVIDIA_Quadro_P400
ComputeDispatchTests.IndirectBasic
ComputeDispatchTests.IndirectOffset
ComputeDispatchTests.MaxWorkgroups
Bug: dawn:1196
Change-Id: Ie7188384656307ede0f467ead6fba9289d9db655
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/69920
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Brandon Jones <bajones@chromium.org>
Implements MultiplanarExternalTextureTransform to allow transforming a
texture_external binding into two texture_2d<f32> bindings and a uniform
buffer binding. Transforms textureSampleLevel and textureLoad calls with
a texture_external parameter into custom functions that can handle both
single-plane RGBA or bi-planar YUV. Includes tests.
Bug: dawn:1082
Change-Id: Icb6d8b0f3773feca01c833171f07230c3531f3aa
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68620
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
This patch supports [[num_workgroups]] on D3D12 for DispatchIndirect
by appending the values of [[num_workgroups]] at the end of the
scratch buffer for indirect dispatch validation and setting them as
the root constants in the command signature.
With this patch, for every DispatchIndirect call:
- On D3D12:
1. Validation enabled, [[num_workgroups]] is used
The dispatch indirect buffer needs to be validated, duplicated and
be written into a scratch buffer (size: 6 * uint32_t).
2. Validation enabled, [[num_workgroups]] isn't used
The dispatch indirect buffer needs to be validated and be written
into a scratch buffer (size: 3 * uint32_t).
3. Validation disabled, [[num_workgroups]] is used
The dispatch indirect buffer needs to be duplicated and be written
into a scratch buffer (size: 6 * uint32_t).
4. Validation disabled, [[num_workgroups]] isn't used
Neither transformations or scratch buffers are needed for the dispatch
call.
- On the other backends:
1. Validation enabled,
The dispatch indirect buffer needs to be validated and be written
into a scratch buffer (size: 3 * uint32_t).
2. Validation disabled,
Neither transformations or scratch buffers are needed for the dispatch
call.
BUG=dawn:839
TEST=dawn_end2end_tests
Change-Id: I4105f1b2e3c12f6df6e487ed535a627fbb342344
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/68843
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
This patch fixes a bug in the generation of expected data in CopyTests.
In function GetExpectedTextureData() we intend to initialize every byte
of the expected texture data, so we need to calculate the index of the
expected texture data with "bytesPerRow" instead of "texelBlocksPerRow".
BUG=chromium:1161355
TEST=dawn_end2end_tests
Change-Id: I05bcab629fc961c7425c10fcd534c451af199712
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/69600
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Found while investigating Swiftshader support. Some Vulkan loaders
fail to create an instance if VK_ICD_FILENAMES is empty string
rather than entirely absent. It was set to empty string because
Dawn did not distinguish between nonexistent environment variables
and the empty string. This CL adds distinguishing between the two,
including tests for the behavior.
Bug: chromium:1266550
Change-Id: I1680a281f62e6b340009e01da65db9d485e2975e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/69520
Auto-Submit: Austin Eng <enga@chromium.org>
Reviewed-by: Loko Kung <lokokung@google.com>
Commit-Queue: Austin Eng <enga@chromium.org>
- Renames ProgrammablePassEncoder to just ProgrammableEncoder since it is also used in RenderBundleEncoder which is not a "pass"
- Adds testing infrastructure to further test device errors
- Ensures AttachmentStates are de-reffed when encoder objects are destroyed for proper cleanup
- Makes sure that both encoded and partial encoded commands are freed at destruction
Bug: dawn:628
Change-Id: Id62ab02d54461c4da266963035e8666799f61e9a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/68461
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Loko Kung <lokokung@google.com>
This adds overloads for WireCmd [De]serialization that don't take the
object id provider/resolvers and produce a fatal error as soon as an
object is encountered.
Bug: dawn:1186
Change-Id: I13e796a5d8f59c26279b9079d4496390506c739a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/68941
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Loko Kung <lokokung@google.com>
The new test can pass without further implementation changes on
D3D12 because:
- stencil aspect shares the same resource state with depth:
D3D12_RESOURCE_STATE_DEPTH_READ.
- stencil flag (D3D12_DSV_FLAG_READ_ONLY_STENCIL) for readonly
DepthStencilView descriptor has already been added into Dawn
in previous patch.
Bug: dawn:485
Change-Id: I87aaaac0f01744d3533f2d97f987ee23ad0a1f53
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/69300
Commit-Queue: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Add a new 'Target' to the ast::CallExpression, which can be either an
Identifier or Type. The Identifier may resolve to a Type, if the Type is
a structure or alias.
The Resolver now resolves the CallExpression target to one of the
following sem::CallTargets:
* sem::Function
* sem::Intrinsic
* sem::TypeConstructor
* sem::TypeCast
This change will allow us to remove the type tracking logic from the WGSL
parser, which is required for out-of-order module scope declarations.
Bug: tint:888
Bug: tint:1266
Change-Id: I696f117115a50981fd5c102a0d7764641bb755dd
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68525
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Change all the texture types to the GLSL equivalents. Note that some
types don't actually exist in GLSL ES, e.g., 1D textures, but this
will be handled later.
Change all the texture functions from HLSL-style to GLSL (e.g.,
texture.Sample(...) -> texture(texture, ...). Note that depth
comparison functions are probably wrong.
Implement writeonly storage texture type and functions.
Samplers are skipped entirely in the GLSL backend, with the assumption
that they will already have been combined into GLSL-style combined
samplers and textures by the client code.
Move the SingleEntryPoint transform above the RemovePhonies pass. This
ensures that texture variables are not optimized out. (Otherwise some
tests produce valid but not very useful results.)
Add the builtin-functions to the GLSL keywords list for renaming.
They're not keywords, but they can't be identifiers either.
Bug: tint:1298, tint:1299
Change-Id: I86c4547fcdd1eba80be98f6c05b939f345fd4c3e
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/69200
Commit-Queue: Stephen White <senorblanco@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Replace all calls to create<ast::TypeConstructorExpression>() with
Construct(). TypeConstructorExpression will be folded into
ast::CallExpression, but the Construct() call signature will remain
identical.
Bug: tint:888
Change-Id: Ifb28a90ccf5184c8090c2e32fa8c82f3996dfa33
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/69108
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
ast::CallExpression will soon encompase function & intrinsic calls,
along with type-constructors and type-casts. The latter two cannot be
wrapped with a CallStatement, so change
ProgramBuilder::WrapInStatement() to always assign the expression to a
temporary.
Fix the few places that actually relied on this behavior to use
CallStmt() explicitly.
Bug: tint:888
Change-Id: I48b8a2be73128df9cd2b4bdcc00ae81c4a872359
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/69104
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: James Price <jrprice@google.com>
Pipeline overidable constants are not compile-time constant.
If a module-scope const has an [[override]] decoration, do not assign
the constant value to it, as this will propagate, and the constant value
may become inlined in places that should be overridable.
Also: Rename sem::GlobalVariable::IsPipelineConstant() to
IsOverridable() to make it clearer that this is not a compile-time known
value. Add SetIsOverridable() so we can correctly set the
IsOverridable() flag even when there isn't an ID.
Change-Id: I5ede9dd180d5ff1696b3868ea4313fc28f93af4b
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/69140
Auto-Submit: Ben Clayton <bclayton@google.com>
Reviewed-by: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
- Ensures that the D3D12 buffer destruction implementation no longer handles unmapping as that is now handled in the frontend.
- Moves around when the frontend buffer state is set to MappedAtCreation so that it only happens when the staging buffer (when necessary) is created successfully. Note that if the staging buffer was not created successfully, we will never return that buffer and would instead return an error buffer anyways. We need the state to be Unmapped though since it needs to be properly destroyed.
Bug: chromium:1265923, dawn:628
Change-Id: I62f98f0f1379a9cf0af565adfb8256ffe592b1ad
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/68880
Commit-Queue: Loko Kung <lokokung@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Literals are now expressions, so in keeping with all the other
expression types, suffix the class name with Expression.
I'm not overly keen on requiring everything to have an Expression
suffix, but consistency is better than personal preference.
Note: this should have been part of 30848b6, but I managed to drop
this change instead of squashing it. Opps.
Bug: tint:888
Change-Id: Idfaf96abe165a6bf5028e60a160e7408aa2bf9db
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68943
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
Literals are now expressions, so in keeping with all the other
expression types, suffix the class name with Expression.
I'm not overly keen on requiring everything to have an Expression
suffix, but consistency is better than personal preference.
Bug: tint:888
Change-Id: Ida1f1b98c71d5d0fe6399bca938010cd5d3a00c3
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68840
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
Just make Literal an expression. The ScalarConstructorExpression
provides no real value, aside from having a ConstructorExpression base
class that's common between ScalarConstructorExpression and
TypeConstructorExpression. TypeConstructorExpression will be folded into
CallExpression, so this hierarchy will serve no purpose.
First step in resolving the parser ambiguity of type-constructors vs
type-casts vs function calls.
Bug: tint:888
Change-Id: I2585d5ddbf6c0619a8f24c503e61ebf27c182ebe
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68524
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
In HLSL code, if a switch statement has only a default case, FXC will
effectively ignore the code in that case. In this change, we detect this
and work around it by emitting the code in the default block without the
switch.
Bug: tint:1188
Change-Id: I69b405cdb4c669fb093eb49aa138923419dcf8f8
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/68440
Kokoro: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Ben Clayton <bclayton@chromium.org>
