Allow user-defined macros to be loaded from dynamic libraries and evaluated.
- Introduce a _CompilerPluginSupport module installed into the toolchain. Its `_CompilerPlugin` protocol acts as a stable interface between the compiler and user-defined macros.
- Introduce a `-load-plugin-library <path>` attribute which allows users to specify dynamic libraries to be loaded into the compiler.
A macro library must declare a public top-level computed property `public var allMacros: [Any.Type]` and be compiled to a dynamic library. The compiler will call the getter of this property to obtain and register all macros.
Known issues:
- We current do not have a way to strip out unnecessary symbols from the plugin dylib, i.e. produce a plugin library that does not contain SwiftSyntax symbols that will collide with the compiler itself.
- `MacroExpansionExpr`'s type is hard-coded as `(Int, String)`. It should instead be specified by the macro via protocol requirements such as `signature` and `genericSignature`. We need more protocol requirements in `_CompilerPlugin` to handle this.
- `dlopen` is not secure and is only for prototyping use here.
Friend PR: apple/swift-syntax#1022
Intro ASTContext::setIgnoreAdjacentModules to change module loading to
accept load only resilient modules from their swiftinterfaces, ignoring
the adjacent module and any silencing swiftinterfaces errors.
For release-management purposes during development, LLDB's embedded Swift
compiler's version number can sometimes be off-by-one in the last digit
compared to the Swift compiler.
This patch restores the old behavior from before 17183629e4.
rdar://101299168
Previously, when evaluating a `#if canImport(Module, _version: 42)` directive the compiler could diagnose and ignore the directive under the following conditions:
- The associated binary module is corrupt/bogus.
- The .tbd for an underlying Clang module is missing a current-version field.
This behavior is surprising when there is a valid `.swiftinterface` available and it only becomes apparent when building against an SDK with an old enough version of the module that the version in the `.swiftinterface` is too low, making this failure easy to miss. Some modules have different versioning systems for their Swift and Clang modules and it can also be intentional for a distributed binary `.swiftmodule` to contain bogus data (to force the compiler to recompile the `.swiftinterface`) so we need to handle both of these cases gracefully and predictably.
Now the compiler will enumerate all module loaders, ask each of them to attempt to parse the module version and then consistently use the parsed version from a single source. The `.swiftinterface` is preferred if present, then the binary module if present, and then finally the `.tbd`. The `.tbd` is still always used exclusively for the `_underlyingVersion` variant of `canImport()`.
Resolves rdar://88723492
The ObjCMethodLookupTable for protocols was not being serialized and rebuilt on load, so NominalTypeDecl::lookupDirect() on selectors was not working correctly for deserialized types. Correct this oversight.
Change the way swiftmodules built against a different SDK than their
clients are rejected. This makes them silently ignored when the module
can be rebuilt from their swiftinterface, instead of reporting a hard
error.
rdar://93257769
The `VersionTuple` API was changed llvm/llvm-project
219672b8dd06c4765185fa3161c98437d49b4a1b to return `VersionTuple`
from `get*Version` rather than pass in major, minor, and subminor output
parameters. Update uses to the new API.
Note that `getMacOSXVersion` is slightly different in that it returns a
boolean while taking a `VersionTuple` output parameter to match its
previous behaviour. There doesn't seem to be any use that actually
checks this value though, so we should either update the API to return
an `Optional` and actually check it *or* remove the "failure" case and
return a `VersionTuple` like all the others.
Previously, we have an assumption that all equal-joined arguments are alias to the
separate forms, e.g. -tbd-install_name=Foo is an alias to -tbd-install_name Foo. However,
it seems having only the equal-joined version of an argument is possible. This PR adds
support to that scenario.
This normalizes the path so that we always have the mapping in normal
form. This fixes a bug in the cross-module import tracing, allowing us
to finally enable the test on Windows.
When looking for a Swift module on disk, we were scanning all module search paths if they contain the module we are searching for. In a setup where each module is contained in its own framework search path, this scaled quadratically with the number of modules being imported. E.g. a setup with 100 modules being imported form 100 module search paths could cause on the order of 10,000 checks of `FileSystem::exists`. While these checks are fairly fast (~10µs), they add up to ~100ms.
To improve this, perform a first scan of all module search paths and list the files they contain. From this, create a lookup map that maps filenames to the search paths they can be found in. E.g. for
```
searchPath1/
Module1.framework
searchPath2/
Module1.framework
Module2.swiftmodule
```
we create the following lookup table
```
Module1.framework -> [searchPath1, searchPath2]
Module2.swiftmodule -> [searchPath2]
```
We noticed some Swift clients rely on the serialized search paths in the module to
find dependencies and droping these paths altogether can lead to build failures like
rdar://85840921.
This change teaches the serialization to obfuscate the search paths and the deserialization
to recover them. This allows clients to keep accessing these paths without exposing
them when shipping the module to other users.
We've recently added the -experimental-hermetic-seal-at-link compiler flag,
which turns on aggressive dead-stripping optimizations and assumes that library
code can be optimized against client code because all users of the library
code/types are present at link/LTO time. This means that any module that's
built with -experimental-hermetic-seal-at-link requires all clients of this
module to also use -experimental-hermetic-seal-at-link. This PR enforces that
by storing a bit in the serialized module, and checking the bit when importing
modules.
This cleans up 90 instances of this warning and reduces the build spew
when building on Linux. This helps identify actual issues when
building which can get lost in the stream of warning messages. It also
helps restore the ability to build the compiler with gcc.
* Fix unnecessary one-time recompile of stdlib with -enable-ossa-flag
This includes a bit in the module format to represent if the module was
compiled with -enable-ossa-modules flag. When compiling a client module
with -enable-ossa-modules flag, all dependent modules are checked for this bit,
if not on, recompilation is triggered with -enable-ossa-modules.
* Updated tests
Some Swift module versions are beyond the scope of llvm::VersionTuple(), thus we need to sanitize the most
trivial bits so that we can handle most cases in versioned canImport query.
Introduce a new loading restriction that is more strict than the serialization
version check on swiftmodules. Tagged compilers will only load
library-evolution enabled swiftmodules that are produced by a compiler with the
exact same revision id. This will be more reliable in production
environments than using the serialization version which we forgot to
update from time to time. This shouldn't affect development compilers that
will still load any module with a compatible serialization version.
rdar://83105234
Serialize the canonical name of the SDK used when building a swiftmodule
file and use it to ensure that the swiftmodule file is loaded only with
the same SDK. The SDK name must be passed down from the frontend.
This will report unsupported configurations like:
- Installing roots between incompatible SDKs without deleting the
swiftmodule files.
- Having multiple targets in the same project using different SDKs.
- Loading a swiftmodule created with a newer SDK (and stdlib) with an
older SDK.
All of these lead to hard to investigate deserialization failures and
this change should detect them early, before reaching a deserialization
failure.
rdar://78048939
If we are building for ARM64 but we try to import a module with only an ARM64e interface, fall back to importing said interface.
This is the reverse of a similar fallback briefly introduced last year, but removed in #31196.
We should hold off actually building the binary module file until it is imported.
`canImport` queries can happen, for example, during dependency scanning, when we do not wish to have the scanner tool execute any module builds.
Resolves rdar://82603098
The MemoryBuffer loader is used by LLDB during debugging to import binary Swift
modules from .swift_ast sections. Modules imported from .swift_ast sections are
never produced from textual interfaces. By disabling resilience the expression
evaluator in the debugger can directly access private members.
rdar://79462915
Rework Sendable checking to be completely based on "missing"
conformances, so that we can individually diagnose missing Sendable
conformances based on both the module in which the conformance check
happened as well as where the type was declared. The basic rules here
are to only diagnose if either the module where the non-Sendable type
was declared or the module where it was checked was compiled with a
mode that consistently diagnoses `Sendable`, either by virtue of
being Swift 6 or because `-warn-concurrency` was provided on the
command line. And have that diagnostic be an error in Swift 6 or
warning in Swift 5.x.
There is much tuning to be done here.
The dependency scanner's cache persists across different queries and answering a subsequent query's module lookup with a module not in the query's search path is not correct.
For example, suppose we are looking for a Swift module `Foo` with a set of search paths `SP`.
And dependency scanner cache already contains a module `Foo`, for which we found an interface file at location `L`. If `L`∉`SP`, then we cannot re-use the cached entry because we’d be resolving the scanning query to a filesystem location that the current scanning context is not aware of.
Resolves rdar://81175942
It's a known issue that we are using arm64e interfaces contents for the arm64 target,
meaning the encoded module flags are specifying -target arm64e-x-x instead of
-target arm64-x-x. Fortunately, we can tell the target arch from the interface file
name, so we could sanitize the target to use by inferring arch from the file name.
Titled as "// swift-module-flags-ignorable:", this new field contains new
frontend arguments that can be safely ignored by the older version of the compiler.
For compilers that don't know the field at all, all arguments in it are ignored.
rdar://78233352
This mechanism allows the compiler to use a backup interface file to build into a binary module when
a corresponding interface file from the SDK is failing for whatever reasons. This mechansim should be entirely opaque
to end users except several diagnostic messages communicating backup interfaces are used.
Part of rdar://77676064
