This uses template and macro magic to reflect the fields of a class.
Dawn:
* Reflect the fields of the types that are used by Dawn's stream::Stream<T> specializations, and use tint::ForeachField() to call StreamIn().
Fuzzers:
* Replace tint::fuzzers::DataBuilder::BuildImpl<T> specializations with the new reflection system.
* static_assert that the type is either POD or reflected. Add a specialization for std::optional which was missing.
Move tint::transform::BindingPoints into MultiplanarExternalTexture, as this is only used by MultiplanarExternalTexture.
All this reduces fragility of the struct declarations slipping out of sync with the uses.
Bug: tint:1640
Change-Id: I08729c1c356f1b427e85983efe3c2678fc2ce717
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/101001
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
This transform breaks up matNx2<f32> matrices used in uniform buffers
into column vectors, which fixes std140 layout rules.
Used by the SPIR-V and GLSL backends.
Re-enable tests that were disabled for these cases.
Bug: tint:1632
Change-Id: I596d016582b4189a0b413d762b3e7eabd3504b22
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/100907
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Add a hash implementation to tint::Number.
Add a utils::Hasher specialization for std::variant.
Add an operator!= for Vector. Needed for std::variants.
Drop the need for explicit on Vector constructors from refs.
These all help general usage with STL and util containers.
Change-Id: I1e594edf532e78f531062c534dacaee7616cded5
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/100905
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
STL std::hash<T*> implementations have a tendency of just reinterpreting the
pointer as a size_t. This is fast, but produces a bad hash, as most pointers
tend to be 4 or 16 byte aligned. The lack of entropy in the LSBs causes
clustering of hashmap slots. Use tint::utils::Hasher to get better hashes.
Change-Id: Ife768d573cd1875e746ca9d77a4ac19e43b06aca
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/99281
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Attempting to make a Vector of an incomplete pointer type would trigger an error as the `CanReinterpretSlice` trail magic is wanting to know the base types, which isn't known yet.
Split `CanReinterpretSlice` into class specializations, where the common case of no-cast doesn't look at the base types.
Change-Id: Id016b027b131f7988ccf3cae93622dacb7802a1d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/98140
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Fixes "dxc failed : unable to parse shader model.".
On windows the called program is responsible for splitting arguments
from one joined string. Command line arguments on 'nix systems need to
be passed as separate strings. This CL makes it so that we pass in the
arg separately, and support ignoring empty string args to make writing
this code easier.
Change-Id: Ia9618c2a743f8fdb49913572e2bbfc4bd1519d3a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/98110
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
The elements of the VectorRef is now immutable, but can be
moved, if the caller relinquishes ownership by explicitly using
std::move() at the callsite.
Also add utils::Empty as a way of signalling that a vector should be
constructed with no elements. This is helpful in templated code where
{} cannot be used due to overload ambiguity.
Change-Id: I24a50a13956b0692771a8bc9046336ad46261562
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97842
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Auto-Submit: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Unlike on Windows, where command line argument splitting is done by the
callee, on 'nix systems, the arguments need to be split by the caller.
Fixes issues where validation was different on windows to other
operating systems.
Change-Id: If55738c431586f706e3edf0ac683661f34b53391
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97144
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
* Added a class template argument deduction guide (CTAD) to infer the
`T` and `N` template arguments. This lets you write `Vector{1,2,3}`
instead of `Vector<int, 3>{1,2,3}`. This is important as a mismatch
between the number of constructor arguments and the `N` template
argument can cause silent heap allocations, which we're trying to
avoid. The `T` deduction uses the same smarts as the return-type
deduction of `Switch()`, so:
* `Vector{1, 2.0}` would construct a `Vector<double, 2>`
* `Vector{i32, u32}` would construct a `Vector<const sem::Type*, 2>`
* Removed the Vector(size_t) and Vector(size_t, const T&) constructors.
This is a move away from the std::vector style API, but these are
rarely more efficient than calling Reserve() and Push(), as you remove
the redundant initialization. The main reason for doing this is to
remove ambiguity between `Vector{1}` and `Vector(1)`.
* Added support for covariance conversion
(`Vector<Derived*, N>` -> `Vector<Base*, N>`).
Only supports pointers to `Castable`, as this can only safely work
with single-inheritance.
* Added support for conversion of `Vector<T*, N>` -> `Vector<const T*, N>`.
This will remove pointless vector copies from the sem package.
Bug: tint:1613
Change-Id: I79b9f82d623f90afa14f8ba1613ee49cccceafeb
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97020
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Change tint's `--fxc` flag to take the path of the FXC compiler DLL.
Have tint attempt to validate with both FXC and DXC if `--validate` is
passed.
Fix the 'dirsWithNoPassExpectations' logic which looks like it got
broken with the tint -> dawn merge. It also incorrectly applied
filepath.FromSlash() on windows.
Change-Id: I0f46aa5c21bc48a2abc48402c41f846aff4a8633
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/96800
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
When enabling the SPIR-V reader or SPIR-V writer we were suppressing
-Wnewline-eof, -Wold-style-cast, -Wsign-conversion, and -Wweak-vtables
for the `libtint` cmake target and anything that depended on that
target. Because we'd build all readers/ writers by default this caused
us to never hit those warnings.
A recent change to the cmake build sets the Tint backend based on
the Dawn backend. So, there is a much higher chance of not building
the SPIR-V support if you're on a Mac or Windows machine.
This CL removes the suppression of the warnings, adds specific pragmas
into the SPIR-V reader code which imports the SPIRV-Tools headers
and fixes up the warnings which were then firing due to checking
for the new warnings.
Change-Id: I0d0be6aa3d0b692e939ce8ff924dfb82c82792fc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94901
Reviewed-by: Ben Clayton <bclayton@google.com>
Auto-Submit: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Instead of vector-of-variant.
This:
• Makes it impossible to produce a mix of scalar variant types, which
would make no sense.
• Reduces the size of a Constant, by removing the union-tag from each
element.
Also clean up terminology. Rename 'Constant::Scalar' to
'Constant::Element'. Scalars are well-defined in WGSL, and with the
introduction of abstract-numerics, this no longer makes sense.
Bug: tint:1504
Change-Id: I599aa97ad1ea798b7db8e512a5990ba75827faad
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/91304
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Concatenating strings to use for cache keys is horribly inefficent and very error prone.
Add a UnorderedKeyWrapper helper to allow types to be used as a unordered_map and unordered_set key. Use this for the type_constructor_to_id_ map.
Produces SPIR-V with some duplicate SPIR-V instructions for constructors removed.
Change-Id: Ib072d485ca28bb07f03e979c133cdce1f69ee482
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/88300
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Add a Hash() and Equals() methods to sem::Type.
Have sem::Manager (which should be called sem::TypeManager) derive from
utils::UniqueAllocator. This now uses the Hash() and Equals() for uniquely
constructing semantic types instead of building strings and comparing those.
Bug: tint:1383
Change-Id: I5e3229bd087391ac594d333a0ab4232cfcddf54d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/82743
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: David Neto <dneto@google.com>
