Reset the visited state of nodes before traversing to determine a
parameters requirements with respect to the function return value,
otherwise we may not capture a parameter->retval relationship
correctly.
Fixed: tint:1822
Change-Id: I9802a89fe1c8331b2f9dae594ccb045f339396fd
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/118080
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Use expect_compound_statement() in all the places that use
compound_statement in the WGSL grammar.
Handle attributes on statements inside Resolver::StatementScope, so
that the logic can be reused for the various places where block
statements are used. This will also make it easier to reuse this logic
when we allow these attributes on other types of statement in the
future.
Add an `EmitBlockHeader()` helper to the WGSL writer to reuse the
logic for emitting attributes on block statements for all the places
that use them.
Bug: tint:1809
Change-Id: Iac3bb01f5031e6134c1798ddafdad080412c8bef
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/118000
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: James Price <jrprice@google.com>
- It doens't know which operator == to use when there is using
EnumClassBitmask that looks like (a & b) == 0. Instead use just
the form (a & b).
- It doesn't do automatic capture of constexpr variables in lambdas so
turn a couple constexpr into regular const.
- It (correctly) warns that if constexpr (std::is_constant_evaluated())
is always true, so remove the constexpr keyword.
Bug: None
Change-Id: If7857abd1c30acb0736557844ff13f32a19d54cf
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/117888
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Corentin Wallez <cwallez@chromium.org>
Switch the default severity for all uniformity violations to errors.
We now have an opt-out mechanism for deriviative operations (via
diagnostic filters), and a `workgroupUniformLoad()` builtin for
compute shaders.
Bug: tint:1809
Change-Id: I666c706d6195ca0d24ead14c4709e7f646bfcc64
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/117741
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Add additional `RequiredToBeUniform` nodes for each severity
level. When processing a call to a derivative builtin, look up the
severity from the semantic info for that AST node, and add an edge to
the corresponding `RequiredToBeUniform` node.
Propagate the severities to the callsite and parameter tags for a
function that contains a builtin.
Traverse that graph from each `RequiredToBeUniform` node to look for
violations at each severity level, starting with the most severe. Only
stop the analysis if an error is found, otherwise report the violation
and keep going.
Bug: tint:1809
Change-Id: I4ac838e85da3f4fb3d63f4892dce7f12b096f74b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/117602
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Each sem::Node has a map which stores the diagnostic modifications
applied to that node. The sem::Info class provides a query to get the
diagnostic severity for a given AST node, by walking up the semantic
tree to find the tightest diagnostic severity modification. The
default severity is used if it was not overridden.
This allows components outside of the Resolver/Validator to determine
the diagnostic severity while walking the AST, which is required for
the uniformity analysis.
Bug: tint:1809
Change-Id: I4caf99d7412fb22fb1183b2c8cfde349da2fefd3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/117601
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
The Resolver parses the diagnostic rule and sets the updated severity
in a ScopeStack, which is stored in the Validator.
Automatically generate the diagnostic rule enum and its parsing logic
using intrinsics.def.
Add a "chromium_unreachable_code" diagnostic rule to test this.
Bug: tint:1809
Change-Id: Ia94db4321b8019f01d31a84da0fda25dfdf72f5c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/117566
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Flip the diagnostics so that the trigger location is on the builtin
that requires uniformity.
We also now show the place at which control flow diverges regardless
of where it is in the function call stack.
Change-Id: Id739a137b9011c900649b74165a6600a95d87ca4
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/116691
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: James Price <jrprice@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Accept any type in the intrinsics definition, and then manually
validate that there are no atomics in the type. Add manual E2E tests
for composite types.
Use the BuiltinPolyfill transform to implement it for all backends.
Update the uniformity analysis with special-case tags for the builtin.
Fixed: tint:1780
Change-Id: I95786dff4df70a0b16ed1c53b853b5d0ec6bc501
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/114862
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: James Price <jrprice@google.com>
Kokoro: James Price <jrprice@google.com>
For logical binary expressions that can be short-circuited, if the rhs
tree contained a mix of constant and runtime expressions, we would
erroneously mark the node as runtime, although some of its children were
resolved as kNotEvaluated. This would then fail during backend
generation.
This is a fork of 115820, addressing review comments, as amaiorano is OOO this week.
Bug: chromium:1403752
Bug: tint:1581
Change-Id: I18682c7fe1db092d280390881ff86b3c0db23e9b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/116020
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Auto-Submit: Ben Clayton <bclayton@google.com>
This change is a necessary to support workgroupUniformLoad in a
following patch. Otherwise, there is no change to the set of shaders
that are accepted or rejected by the analysis.
We now distinguish between uniformity requirements on the contents of
a pointer parameter versus the pointer value itself when generating
tags for function parameters.
Whilst processing an expression, if we see a sem::Load node we pass a
flag down through child expressions to indicate that we will be
loading from the result. When processing an identifier expression, we
can then select between adding an edge to the source of the
pointer/reference versus the contents of the root identifier that it
corresponds to.
Since the pointers passed to atomic builtins can be uniform, we
special-case them to capture the fact that their return value is
always considered non-uniform.
The arrayLength builtin no longer needs special-casing.
Added many tests to cover various cases that are now captured
differently in the graph. There are two cases that are disabled as
they require variable pointers to trigger the uniformity violation.
Bug: tint:1780
Change-Id: I03edb65f22a6ffb0e7daf8b2f590f5de898e6262
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/114861
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: James Price <jrprice@google.com>
Reviewed-by: David Neto <dneto@google.com>
Variables declared in for-loop initializers were not being tracked
properly across iterations as a check was wrongly determining them to
be declared inside the loop body.
Also fixes an issue where variables declared in for-loop initializers
were still considered to be in scope after loop exit.
Change-Id: I2ce3a519be45c8daba31bf00e8b2614f0bd6a2de
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/114364
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Auto-Submit: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: James Price <jrprice@google.com>
