The resolver has sensible places to dispatch to ArrayConstructor() and
StructureConstructor(), so don't do additional, unnecessary type
dispatches.
Change-Id: Id4edb1fc0dedf431d94507b1c3378bca164746a4
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97588
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
In preparation for array constructors that can infer type and count
based on constructor arguments.
Bug: tint:1628
Change-Id: I9f12d7a30de232cf0d34ed7e1a356dd5b92d26d7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97587
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
This should be immutable once the sem tree is built.
Add a hacky const_cast in Resolver for now, as this is preferable to
having some things non-const in sem.
Bug: tint:745
Change-Id: I67ebce76730347c9543875ab8e1c21a47d71fd56
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97584
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Instead of a C-style array.
Change-Id: Ieb3e9811cddf505bfd3b484b7f265459f7b0e66b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97583
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Add Resolver::ConcreteType() to determine the concrete type for an
abstract-numeric (or composite of abstract).
Will be used to recursively infer the concrete types of abstract arrays.
Bug: tint:1628
Change-Id: Ia26b778abc827b531848b346f3e36938ad1a0470
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97582
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
This is a public API definition of a program-unique override identifier.
Bug: tint:1155
Change-Id: I6e55d43208e72a7a316557a89e2169d1b952f9bf
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/97006
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Use the new vector type in some of the hot code paths of the resolver.
Bug: tint:1613
Change-Id: Ie56d8c96f73c9112f37934ad67e588513aafb982
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/96282
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Use the new utils::Bitset to replace the much slower unordered_map.
Bug: tint:1613
Change-Id: I80503d2b897f754408c4ce14a221210d6abc18f7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/96403
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: dan sinclair <dsinclair@google.com>
If an abstract-vector or abstract-matrix is indexed with a non-constant index expression, then the resulting value is non-constant, and so cannot be abstract.
In this situation the materialization cannot be done post-index, so materialization must happen on the object before indexing.
Bug: chromium:1345468
Change-Id: I9f29dc40301779a7ff8f173724374bd845a3a5b9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/96684
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Add a 'NodeID' to each ast::Node which is the sequentially allocated
index of the node. Use this in sem::Info to map the AST node to the
semantic node, instead of using a std::unordered_map.
Optimised very hot code by entirely eliminating map lookups, and
dramatically reducing cache misses (lookups are usually sequentually
ordered).
Timings running
'webgpu:shader,execution,expression,call,builtin,atan2:f32:inputSource="const";vectorize="_undef_"'
with dawn/node, using SwiftShader:
Without change: 3.22647107s
With change: 3.10578879s
Bug: tint:1613
Change-Id: I22ec48d933b2e5f9da04494bff4e979e6f7b1982
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/96140
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Keep track of the earliest evaluation point for an expression.
Required to properly track what can be assigned to a `const`, `override`, `let`, `var`.
Bug: tint:1601
Bug: chromium:1343242
Change-Id: I301eec21b71e9036dc1bf6c9af8079317d724762
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95949
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Use the @const annotations to more efficently dispatch to the right
method of the ConstEval class.
Reduces a whole lot of dynamic casting logic.
Change-Id: I6791aac51b935b46d63af29abd1e577b3306a0ff
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95950
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Extract out the methods of Resolver::EvaluateXXXValue() to a new
tint::resolver::ConstEval class.
Removes more bloat from Resolver, and creates a centralized class for
constant evaluation, which can be referred to by the IntrinsicTable.
Change-Id: I3b58882ef293fe07f019ad2138a7e9dbbac8de53
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95951
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Restructure the logic so there's less, pointless dynamic casting, and the complexity is reduced.
This alters the order in which variables are validated, hence the change of test.
Change-Id: I9a3120c0278faa5ac9f1db65eeb71a8e4a705596
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95948
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Materialize() was re-evaluating the constant values for the incoming
semantic expression, despite this already being evaluated. Just use the
sem::Expression::ConstantValue().
resolver.cc already has all the semantic pointers, so pass them in
instead of pointlessly hitting the ast -> sem map.
Change-Id: If2bc7cd10f79079fb811e9d83c5150dd3c0c244c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95764
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
No builtins are implemented as `@const` yet, but validation handles this already.
Bug: chromium:1341472
Change-Id: Id85893345299ba3414e2d15b85dd071c326f481d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95762
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Materialize() can return nullptr, if there's an error.
Check the returned pointer for nullptr, before continuing on to using the pointer.
Bug: chromium:1341313
Change-Id: Ib7c9e593fbf2bb4374305c341c2b04e34e7487e9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/95761
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Change sem::Constant to be an interface to the constant data. Implement
this so that zero-initialized data doesn't need to allocate the full
size of the type.
This also makes usage a lot cleaner (no more flattened-list of
elements!), and gives us a clear path for supporting constant
structures if/when we want to support them.
Bug: chromium:1339558
Bug: chromium:1339561
Bug: chromium:1339580
Bug: chromium:1339597
Change-Id: Ifcd456f69aee18d5b84befa896d7b0189d68c2dd
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94942
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@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>
Enable the parsing of 'const'.
Warn on use of module-scope 'let', and automatically replace with 'const'.
Fixed: tint:1580
Change-Id: I214aabca80686dc6b60ae21a7a57fbfb4898ea83
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93786
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
This patch add f16 types and their constructors and conversions in
resolver and intrinsic table. Also implement relating unit tests.
Bug: tint:1473, tint:1502
Change-Id: Ida1336193a72a73959e50e6a3eb12be44c0396b5
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94642
Commit-Queue: Zhaoming Jiang <zhaoming.jiang@intel.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
The WGSL specification changed so that the enforcing of
initializers being only const-expr is no longer done by the
grammar. This is change is required to allow for 'const' expressions to
reference each other.
Bug: tint:1580
Change-Id: Ia95b6a0bc86ce391a38f4248f20898dd9a6cf27f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94683
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: David Neto <dneto@google.com>
The frontends do not currently emit these, nor do the backends currently
handle them.
Bug: tint:1580
Change-Id: I469a5379663c2802145b28a94f5c1e348cc14ff3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94605
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: David Neto <dneto@google.com>
Use the constant evaluated value instead of manually traversing
variables to find the literal value. This is a small step towards
supporting 'const' values for array sizes.
Also make our OOB-byte related error diagnostics consistent.
Bug: tint:1580
Change-Id: Idf9eb22cdbf69d750218c554e9f826c30458c6b9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94600
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Also store expression object in MemberAccessorExpression, as well as the
struct on sem::StructMember.
These are used to implement spir-v reader atomics in a follow-up CL.
Bug: tint:1441
Change-Id: Iea49cfb7f9d2e7898d89d2dac6a16a14022c546f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94523
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
These functions do not return an error - however they might return an
invalid Constant value (if a constant cannot be evaluated), which is
correctly handled by the resolver. So, drop the Result, just emit a
sem::Constant.
Change-Id: I0dd37125c0dba4ba6dcace5ee5f8b1c53cb34d33
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94326
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
If the abstract numeric cannot materialize to the target type, we could
ICE before the validation handles this bad case. To fix, simply hoist
the validation earlier.
Bug: chromium:1337524
Change-Id: Icc603b056900131cfdb029b517f1c3d030b2ecb4
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/94322
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Split the logic out of Resolver::Variable().
Primes the resolver for the introduction of 'const'
Bug: tint:1580
Change-Id: Id67280ed5c8c73a69c62728fb5a81a08f13628a8
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93785
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Split the logic out of Resolver::Variable().
Also correctly allows type inference of module-scoped 'var's.
Fixed: tint:1584
Change-Id: I32eb11f0a847775137fef937da6f4032a3b3c2b9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93784
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Validating sampled and multisampled texture types does not belong in Validator::Variable().
Change-Id: Ie0f2502508c28af6fb6d3f4d7803171d946c511b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93783
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Add the new classes:
* `ast::Let`
* `ast::Override`
* `ast::Parameter`
* `ast::Var`
Limit the fields to those that are only applicable for their type.
Note: The resolver and validator is a tangled mess for each of the
variable types. This CL tries to keep the functionality exactly the
same. I'll clean this up in another change.
Bug: tint:1582
Change-Id: Iee83324167ffd4d92ae3032b2134677629c90079
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93780
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Standarize the messages, using 'let', 'override' or 'var'.
Module-scope 'let' needs to be replaced with 'const', but baby steps.
Bug: tint:1582
Change-Id: I290aede118a30ab0f4294c89ec43005371c87b45
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93446
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
This patch
1. Add F16 literal support in WGSL lexer and parser for both decimal and
hex form. Also fix the f16::Quantize method to deal with subnormal cases
correctly.
2. Fix exactly-representable check for hex f32 literal to deal with
subnormal cases.
3. Implement and fix related unitests for f16 and f32.
Bug: tint:1473, tint:1502
Change-Id: Ia4a7c9144ef9323fb23b2200a64e1ca8afb6c334
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/93100
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: Zhaoming Jiang <zhaoming.jiang@intel.com>
Commit-Queue: David Neto <dneto@google.com>
Attempting to return an abstract-numeric when the function had no return type would trigger an ICE, as an abstract numeric cannot materialize to a void type.
Bug: chromium:1332613
Change-Id: I635ebb8dddb2e7628939607a4f964be62b616745
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/92720
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Add support for @const to builtins in intrinsics.def.
Propagate this flag through to the intrinsic table.
Handle builtins that are @const annotated in the resolver.
Currently no intrinsics are decorated with @const, so there's nothing to
test (yet).
Bug: tint:1504
Change-Id: I172483688617782bd7c58b70e3f38d0222a5d1af
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/92323
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Turn on resolving of abstract-integer and abstract-float types, as well
as materialization to their concrete types.
Bug: tint:1504
Bug: chromium:1330805
Fixed: tint:1572
Change-Id: I09c95406e11b64bb0267fe7b1ed08af986dbd553
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/91840
Commit-Queue: Ben Clayton <bclayton@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
These are tests for when there's no explicit target type for a
materialization. In this case we expect an abstract-int to
materialize to an i32 and an abstract-float to materialize to a
f32.
Fix a bug uncovered where we were creating a transposed matrix.
Bug: tint:1504
Change-Id: Ie69dd7ec47174d3d7bef20315fd3780dade3a325
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/91845
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Ben Clayton
dan sinclair
Zhaoming Jiang
Antonio Maiorano