Commit Graph

5116 Commits

Author SHA1 Message Date
Alexis Laferrière
552d3a4984 [Sema] Restrict reexported SPIs to modules with an export_as relationship
@_exported exports SPIs only when the exported module defines export_as
pointing to the exporter module. Other reexports do not reexport SPIs.
This is to prevent SPI reexporting to get out of hands with the wide
reexports of the Objective-C world.

rdar://102335473
2022-11-14 13:17:55 -08:00
swift-ci
c6a647ae06 Merge remote-tracking branch 'origin/main' into rebranch 2022-11-13 20:33:41 -08:00
Doug Gregor
51fcde7b76 Merge pull request #62086 from DougGregor/macro-decl 2022-11-13 20:25:48 -08:00
Doug Gregor
5ab6b72604 [Macros] Turn Macro into a declaration node.
Although the declaration of macros doesn't appear in Swift source code
that uses macros, they still operate as declarations within the
language. Rework `Macro` as `MacroDecl`, a generic value declaration,
which appropriate models its place in the language.

The vast majority of this change is in extending all of the various
switches on declaration kinds to account for macros.
2022-11-13 12:21:29 -08:00
swift-ci
9321d03c14 Merge remote-tracking branch 'origin/main' into rebranch 2022-11-12 10:34:14 -08:00
Slava Pestov
1e00c3107a Serialization: Support PackType and PackExpansionType 2022-11-12 02:13:54 -05:00
swift-ci
ca8635e2f6 Merge remote-tracking branch 'origin/main' into rebranch 2022-11-11 12:03:33 -08:00
Alexis Laferrière
b526205f42 Merge pull request #62044 from xymus/module-alias-disambiguate-fix
[ModuleInterface] Support loading an aliased module with an underlying module
2022-11-11 11:55:44 -08:00
swift-ci
059746396a Merge remote-tracking branch 'origin/main' into rebranch 2022-11-11 11:14:29 -08:00
Alexis Laferrière
5fa31c68fe Merge pull request #62029 from xymus/reexport-spi
[Sema] @_exported imports reexport @_spi decls as they reexport APIs
2022-11-11 11:06:22 -08:00
Alexis Laferrière
c0abde01a2 [Sema] @_exported imports export @_spi decls too
Enable transitive imports of all SPI groups through @_exported imports.
This brings to SPI the same behavior that we have for API.

```
// Module A
@_spi(S) public func foo() {}

// Module B
@_exported import A

// Module C
@_spi(S) import B

foo() // SPI imported through the reexport of A from B
```

rdar://101566534
2022-11-10 16:46:14 -08:00
Alexis Laferrière
a1270f66cc [Serialization] Support loading an aliased module with an underlying module
Importing a module using an alias (set with -module-alias) fails if the
imported module has an underlying clang module.

rdar://102217790
2022-11-10 16:30:35 -08:00
swift-ci
e0e346ae57 Merge remote-tracking branch 'origin/main' into rebranch 2022-11-10 07:15:27 -08:00
Michael Gottesman
78ffc87c70 [move-only] Make sure we can properly deserialize deinits from structs/enums.
We previously had a cast<ClassDecl> here. I changed it to NominalTypeDecl and
added tests for this.

rdar://102173184
2022-11-09 19:40:04 -08:00
Alexis Laferrière
3ca1de0a0f [Sema] Clean up lookupImportedSPIGroups 2022-11-08 16:35:43 -08:00
swift-ci
5a30218b8e Merge remote-tracking branch 'origin/main' into rebranch 2022-11-07 22:54:10 -08:00
Michael Gottesman
5406131e8c [move-only] Emit a nice error if a move only type is imported into a module that does not have experimental-move-only enabled.
This just prevents user confusion when they forget to do this since weird
behavior /could/ result.

rdar://102062737
2022-11-07 16:11:46 -08:00
Erik Eckstein
e292d17672 llvm::array_lengthof -> std::size 2022-11-04 20:44:19 +01:00
Erik Eckstein
54fe1304a1 replace LLVM_NODISCARD -> [[nodiscard]]
This is possible because we are now compiling with the C++17 standard.
2022-11-04 20:44:18 +01:00
Doug Gregor
0cb2746c49 Keep track of source files created for macro expansions and such.
Introduce a new source file kind to describe source files for macro
expansions, and include the macro expression that they expand. This
establishes a "parent" relationship

Also track every kind of auxiliary source file---whether for macro
expansions or other reasons---that is introduced into a module, adding
an operation that allows us to find the source file that contains a
given source location.
2022-11-01 08:03:26 -07:00
Alexis Laferrière
47b29b68db Merge pull request #61649 from xymus/index-swiftinterfaces
[Index] Force indexing of system modules to read only from swiftinterfaces
2022-10-31 14:18:45 -07:00
Richard Wei
4ce1ebb120 [Macros] Support user-defined macros as compiler plugins (#61734)
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
2022-10-31 14:03:25 -07:00
Alexis Laferrière
730497e9a3 [Serialization] Add control over adding a loaded module to the in-memory cache 2022-10-31 10:58:57 -07:00
Alexis Laferrière
83cd432785 [Frontend] Intro global control to force loading from swiftinterface
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.
2022-10-31 10:58:54 -07:00
Allan Shortlidge
af032afb9d Serialization: Break the cycle between Serialization and SymbolGraphGen.
Push the top level logic for writing out swiftmodules and associated files into the frontend library which has access to all the necessary dependencies.
2022-10-29 21:37:09 -07:00
Alexis Laferrière
41ec454129 [Serialization] Update error message on deserialization failure
This adds information about whether the module was built from source or
from a swiftinterface, whether it's resilient or not, and tweaks the
format.
2022-10-27 18:51:28 -07:00
Alexis Laferrière
2854c1b3cb [Serialization] Write in the swiftmodule if it's built from a swiftinterface
This information will allow us to distinguish swiftmodule built from
source vs swiftinterface.
2022-10-27 18:51:28 -07:00
Slava Pestov
9dfb6ea067 Merge pull request #61675 from slavapestov/element-generic-environment
AST: Opened element generic environments
2022-10-22 00:58:26 -04:00
Slava Pestov
68514b10b5 AST: Introduce ElementArchetypeType 2022-10-21 21:55:35 -04:00
Richard Wei
56e7cce809 [Macros] Parse MacroExpansionExpr and MacroExpansionDecl
Introduce `MacroExpansionExpr` and `MacroExpansionDecl` and plumb it through. Parse them in roughly the same way we parse `ObjectLiteralExpr`.

The syntax is gated under `-enable-experimental-feature Macros`.
2022-10-21 01:50:35 -07:00
Becca Royal-Gordon
82d78a384f Merge pull request #60630 from beccadax/at-implementation
Add @_objcImplementation
2022-10-20 17:14:21 -07:00
Erik Eckstein
c078b7677c Don't serialize derived function effects for function with @_semantics("optimize.no.crossmodule") 2022-10-20 09:20:28 +02:00
Adrian Prantl
3cc2831608 Don't require a strict revision match in LLDB.
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
2022-10-19 09:03:13 -07:00
Becca Royal-Gordon
f2a0ab79c7 Add basic Sema support for @_objcImplementation
Does not validate members yet; nor does it emit different metadata.
2022-10-18 17:21:56 -07:00
Hamish Knight
6aa44a1754 [AST] Remove @_typeSequence attribute
This is no longer needed now that we have the
ellipsis spelling.
2022-10-14 15:40:13 +01:00
Hamish Knight
b645e63ce5 [AST] NFC: Refactor GenericTypeParamDecl construction
Add distinct overloads for the parser,
deserialization and code synthesis.
2022-10-14 15:40:12 +01:00
Slava Pestov
36c01e80cc Merge pull request #61537 from hborla/variadic-generics-terminology
[AST] Use consistent variadic generics terminology.
2022-10-13 16:20:43 -04:00
Nate Chandler
ab35362056 [SIL] Added new test_specification instruction.
The new instruction exists only to be used in tests.  The idea is to
specify tests that ought to be run "in the context" of the containing
function.
2022-10-11 17:15:13 -07:00
Holly Borla
c4b946195e [AST] Replace the "type sequence" terminology with "parameter pack". 2022-10-10 16:28:13 -07:00
Holly Borla
9bb837a241 [AST] Rename SequenceArchetype to PackArchetype. 2022-10-10 16:25:26 -07:00
Holly Borla
54f42b5833 Merge pull request #61445 from hborla/pack-shape-requirements
[RequirementMachine] Implement inference and minimization of same-shape requirements.
2022-10-07 14:49:14 -07:00
Slava Pestov
92f1361b82 Merge pull request #61429 from LucianoPAlmeida/opaque-parameters-crash
[AST] Move generic parameter logic for accessor decl from parser to GenericParamListRequest
2022-10-07 16:52:24 -04:00
Luciano Almeida
a95f60e387 [AST] Refactor generic parameter list parse logic for AccessorDecl 2022-10-07 09:43:08 -03:00
Holly Borla
38a2c8218b [Requirement] Rename RequirementKind::SameCount to SameShape. 2022-10-06 20:48:40 -07:00
Erik Eckstein
66f07fe3da Swift SIL: rework function effects and add function side-effects
So far, function effects only included escape effects.
This change adds side-effects (but they are not computed, yet).
It also involves refactoring of the existing escape effects.
Also the SIL effect syntax changed a bit. Details are in docs/SIL.rst
2022-10-05 07:38:11 +02:00
Dario Rexin
210c68d8aa [SILOptimizer] Add prespecialization for arbitray reference types (#58846)
* [SILOptimizer] Add prespecialization for arbitray reference types

* Fix benchmark Package.swift

* Move SimpleArray to utils

* Fix multiple indirect result case

* Remove leftover code from previous attempt

* Fix test after rebase

* Move code to compute type replacements to SpecializedFunction

* Fix ownership when OSSA is enabled

* Fixes after rebase

* Changes after rebasing

* Add feature flag for layout pre-specialization

* Fix pre_specialize-macos.swift

* Add compiler flag to benchmark build

* Fix benchmark SwiftPM flags
2022-09-22 16:29:01 -07:00
Michael Gottesman
c54acc83e2 [move-only] Add the ability to specify a deinit at the SIL level for a move only type.
Specifically, we get an additional table like thing called sil_moveonlydeinit. It looks as follows:

sil_moveonlydeinit TYPE {
  @FUNC_NAME
}

It always has a single entry.
2022-09-20 15:19:31 -07:00
Pavel Yaskevich
bb4b29568c [Serialization] Bump module format due to new @typeWrapperIgnored attr 2022-09-14 17:14:24 -07:00
Hamish Knight
4716f61fba [AST] Introduce explicit actions for ASTWalker
Replace the use of bool and pointer returns for
`walkToXXXPre`/`walkToXXXPost`, and instead use
explicit actions such as `Action::Continue(E)`,
`Action::SkipChildren(E)`, and `Action::Stop()`.
There are also conditional variants, e.g
`Action::SkipChildrenIf`, `Action::VisitChildrenIf`,
and `Action::StopIf`.

There is still more work that can be done here, in
particular:

- SourceEntityWalker still needs to be migrated.
- Some uses of `return false` in pre-visitation
methods can likely now be replaced by
`Action::Stop`.
- We still use bool and pointer returns internally
within the ASTWalker traversal, which could likely
be improved.

But I'm leaving those as future work for now as
this patch is already large enough.
2022-09-13 10:35:29 +01:00
Becca Royal-Gordon
d20e00e531 [NFC] Improve deserialization errors for wrong record kind
There are a bunch of places where we expect to find a record with a particular record ID/kind in the decls_block and have to bail out if we see something we don’t expect. Add an `InvalidRecordKindError` we can use in this situation to produce a useful error message and adopt it in various places.

This change also makes deserialization errors print the path to the invalid file.
2022-09-12 13:35:59 -07:00