================
@@ -52,29 +52,36 @@ struct TypeDependenceScope {
enum TypeDependence : uint8_t {
/// Whether this type contains an unexpanded parameter pack
/// (for C++11 variadic templates)
- UnexpandedPack = 1,
+ UnexpandedPack = 1 << 0,
/// Whether this type somehow involves
/// - a template parameter, even if the resolution of the type does not
/// depend on a template parameter.
/// - or an error.
- Instantiation = 2,
+ Instantiation = 1 << 1,
/// Whether this type
/// - is a dependent type (C++ [temp.dep.type])
/// - or it somehow involves an error, e.g. decltype(recovery-expr)
- Dependent = 4,
+ Dependent = 1 << 2,
/// Whether this type is a variably-modified type (C99 6.7.5).
- VariablyModified = 8,
+ VariablyModified = 1 << 3,
/// Whether this type references an error, e.g. decltype(err-expression)
/// yields an error type.
- Error = 16,
+ Error = 1 << 4,
+
+ /// Whether this type contains a placeholder for a type attribute
+ /// awaiting late parsing, e.g. `int *__counted_by(count)` where
+ /// `count` is a later field of the same struct.
+ LateParsedAttr = 1 << 5,
----------------
zmodem wrote:
I obviously don't have as much context as others here, but this does seem
pretty similar to the VariablyModified case.
> I would expect that we have plenty of "if-dependent, do instantation" type
> things all over the place
Wouldn't such checks be done via `Type::isDependentType()` or
`Type::isInstantiationDependentType()` which checks for specific flags?
> and VariablyModified because it is MOSTLY just C, who don't instantiate anyway
fwiw, -fbounds-safety is also mostly C :)
https://github.com/llvm/llvm-project/pull/212906
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
