When performing multi-threaded IRGen, all of the link libraries used
for autolinking went (only) into the first object file. If the object
files were then placed in a static archive, the autolinking would only
kick in if clients that link the static library reference something in
the first object file. This causes link failures.
Make sure that we put the autolink information into each of the object
files we create when doing multi-threaded IRGen.
Fixes the rest of rdar://162400654.
Upstream LLVM in llvm/llvm-project#139584 changed `DiagnosticOptions`
from being a referenced counted object to just be a reference, not owned
by the `clang::DiagnosticEngine`.
In 0981b71090 (part of #82243), the usages
of the Swift repository were adapted to the new memory model, but it
introduced at least one use-after-free and a potential one around the
usage of Clang in the Clang Importer.
This commit tries to fix the use-after-free in both cases, by returning
a `unique_ptr` to the `clang::DiagnosticOptions`, which makes the
lifetime of the `DiagnosticOptions` match the lifetime of the variable
that uses it (normally a `CompilerInvocation`).
Other cases in 0981b71090 should be safe
because the lifetime of the `DiagnosticOptions` do not seem to propagate
beyond the scope of the functions where they live (but I am not fully
sure about the one in `IDETool/CompilerInvocation.cpp` completely).
This was causing compiler crashes during the test
`Interop/Cxx/stdlib/unsupported-stdlib.swift` which eventually uses
`createClangDriver` and tries to emit a diagnostic, which in some cases
was reading the memory from `DiagnosticOptions` when it was already out
of scope.
This makes sure that whenever we emit `#if defined(__OBJC__)`, the matching `#endif` has a comment `// defined(__OBJC__)`.
This makes both testing and reading the header file easier.
On all platforms except Linux, calling `withLockIfAvailable()`
recursively just returns `nil`. However, our Linux implementation
chooses to abort the process instead. We don't need this
inconsistent/destructive behaviour and can just return `nil` as we do
elsewhere.
Teach SIL type lowering to recursively track custom vs. default deinit status.
Determine whether each type recursively only has default deinitialization. This
includes any recursive deinitializers that may be invoked by releasing a
reference held by this type.
If a type only has default deinitialization, then the deinitializer cannot
have any semantically-visible side effects. It cannot write to any memory
On macOS 26, malloc_zone_statistics() always returns
a malloc_statistics_t with the max_size_in_use field
set to zero. This was an intentional change, so let's
remove calls to this API.
Instead, use the proc_pid_rusage() API and return the
ri_lifetime_max_phys_footprint field of rusage_info_v4.
Finally, remove the llvm::sys::Process::GetMallocUsage()
call on other platforms altogether. It returns the
current value and not the max value, so its misleading.
When replacing expected diagnostic content, we should consider
expected diagnostics targeting the same target, not targeting the line
the expected diagnostic is on.
This adds a lit plugin wrapping the core of update-verify-tests. When
lit is invoked with --update-tests it will call the plugin, which checks
if the failure is due to a -verify mismatch. If it is, it tries to
repair the test. If the source file was originally created by
split-file, the changes are propagated to the parent file.
No tests are added, because I don't want to run nested llvm-lit tests
in Swift's test suite, but the core functionality is tested through
update-verify-tests.py.
`??` operator is overloaded on optionality of its result. When the
first argument matches exactly, the ranking is going to be skewed
towards selecting an overload choice that returns a non-optional type.
This is not always correct i.e. when operator is involved in optional
chaining. To avoid producing an incorrect favoring, let's skip the this
disjunction when constraints associated with result type indicate
that it should be optional.
Simply adding it as a binding won't work because if the second argument
is non-optional the overload that returns `T?` would still have a lower
score.
Resolves: rdar://164201746
Make sure we only ever type-check the `do` body of a `do-catch`
statement when lazily type-checking the bound error type, which we can
do for completion.
rdar://164481242
[cxx-interop] Make ClangDeclExplicitSafety request non-recursive
[cxx-interop] Do not import template type arguments
[cxx-interop] Check template argument safety in ClangDeclExplicitSafety
