Commit Graph

1894 Commits

Author SHA1 Message Date
David Ungar
0e109445c1 Format 2019-01-20 23:30:37 -08:00
David Ungar
e7882170a7 Use array for the cache. 2019-01-20 23:30:03 -08:00
David Ungar
fe999d9a47 Format. 2019-01-20 22:47:40 -08:00
David Ungar
4ca97abaff Key cache by SourceFile. 2019-01-20 22:46:47 -08:00
David Ungar
a307dc7524 Initialize the cache. 2019-01-20 17:39:12 -08:00
David Ungar
5827ffae45 Format 2019-01-20 16:34:48 -08:00
David Ungar
51bb354328 Move cache to ASTContext 2019-01-20 16:34:18 -08:00
Argyrios Kyrtzidis
c7ac859310 [Parse] Optimize syntax parsing: Speed-up Lexer::lexTrivia()
Introduce ParsedTrivia which is a more efficient structure to use during lexing than syntax::Trivia.
2019-01-17 12:10:27 -08:00
Parker Schuh
f5859ff46e Rename NameAliasType to TypeAliasType. 2019-01-09 16:47:13 -08:00
Pavel Yaskevich
35202ab5b0 [TypeChecker] Always emit a fallback error if type-check failed without producing one
Sometimes constraint solver fails without producing any diagnostics,
it could happen during different phases e.g. pre-check, constraint
generation, or even while attempting to apply solution. Such behavior
leads to crashes down the line in AST Verifier or SILGen which are
hard to diagnose.

Let's guard against that by tracking if solver produced any diagnostics
upon its failure and if no errors were or are scheduled to be produced,
let's produce a fallback fatal error pointing at affected expression.

Resolves: rdar://problem/38885760
2019-01-07 10:42:00 -08:00
Joe Groff
89979137fc Push ArchetypeType's API down to subclasses.
And clean up code that conditionally works only with certain kinds of archetype along the way.
2018-12-12 19:45:40 -08:00
Joe Groff
f1648a1b3e Split subclasses out of ArchetypeType.
Context archetypes and opened existential archetypes differ in a number of details, and this simplifies the overlapping storage of the kind-specific fields. This should be NFC; for now, this doesn't change the interface of ArchetypeType, but should allow some refinements of how the special handling of certain archetypes are handled.
2018-12-12 08:55:56 -08:00
Slava Pestov
1210bb68a4 SIL: SILFunctionTypes don't allow generic signatures where all parameters are concrete 2018-12-08 22:21:12 -05:00
Slava Pestov
06c2e980cb Merge pull request #21133 from slavapestov/lazy-implicit-inits
Lazy synthesis of implicit constructors in non-primary files
2018-12-07 22:45:40 -05:00
Slava Pestov
6c012b2aec AST: Remove some unnecessary LazyResolver * parameters from ASTContext methods 2018-12-07 20:39:27 -05:00
Slava Pestov
aa747dcd81 Remove property behaviors 2018-12-07 20:38:33 -05:00
swift-ci
39161d5b36 Merge pull request #20600 from adrian-prantl/36032653 2018-12-05 17:01:58 -08:00
Adrian Prantl
d63debeb60 Experimental: Extend ClangImporter to import clang modules from DWARF
When debugging Objective-C or C++ code on Darwin, the debug info
collected by dsymutil in the .dSYM bundle is entirely
self-contained. It is possible to debug a program, set breakpoints and
print variables even without having the complete original source code
or a matching SDK available. With Swift, this is currently not the
case. Even though .dSYM bundles contain the binary .swiftmodule for
all Swift modules, any Clang modules that the Swift modules depend on,
still need to be imported from source to even get basic LLDB
functionality to work. If ClangImporter fails to import a Clang
module, effectively the entire Swift module depending on it gets
poisoned.

This patch is addressing this issue by introducing a ModuleLoader that
can ask queries about Clang Decls to LLDB, since LLDB knows how to
reconstruct Clang decls from DWARF and clang -gmodules producxes full
debug info for Clang modules that is embedded into the .dSYM budle.

This initial version does not contain any advanced functionality at
all, it merely produces an empty ModuleDecl. Intertestingly, even this
is a considerable improvement over the status quo. LLDB can now print
Swift-only variables in modules with failing Clang depenecies, and
becuase of fallback mechanisms that were implemented earlier, it can
even display the contents of pure Objective-C objects that are
imported into Swift. C structs obviously don't work yet.

rdar://problem/36032653
2018-12-05 13:54:13 -08:00
Adrian Prantl
ff63eaea6f Remove \brief commands from doxygen comments.
We've been running doxygen with the autobrief option for a couple of
years now. This makes the \brief markers into our comments
redundant. Since they are a visual distraction and we don't want to
encourage more \brief markers in new code either, this patch removes
them all.

Patch produced by

      for i in $(git grep -l '\\brief'); do perl -pi -e 's/\\brief //g' $i & done
2018-12-04 15:45:04 -08:00
Azoy
65217a71de Cache bool builtin init
remove newline
2018-12-02 22:38:24 -06:00
Azoy
a2dafc3d26 Remove _getBool 2018-12-01 18:51:25 -06:00
John McCall
5553224fd4 Support the explicit representation of self-conformances.
Big, but actually NFC because we're never actually creating them.
2018-11-15 16:42:03 -05:00
Pavel Yaskevich
53417f4aba [AST] Propagate @autoclosure flag to parameter decl and type flags 2018-11-10 11:59:28 -08:00
Slava Pestov
9c50867370 Move RemoteAST's RemoteASTTypeBuilder to lib/AST/ASTDemangler.cpp 2018-11-02 18:19:40 -04:00
John McCall
cf511445e2 Basic support for Builtin.IntegerLiteral. 2018-10-31 18:42:34 -04:00
John McCall
4ea01224fd [NFC] Improve the interfaces for integer widths and parsing integers 2018-10-31 15:49:40 -04:00
Doug Gregor
599e07e5d9 [Type checker] Keep the type checker alive as long as the ASTContext is.
It is possible for the SIL optimizers, IRGen, etc. to request information
from the AST that only the type checker can provide, but the type checker
is typically torn down after the “type checking” phase. This can lead to
various crashes late in the compilation cycle.

Keep the type checker instance around as long as the ASTContext is alive
or until someone asks for it to be destroyed.

Fixes SR-285 / rdar://problem/23677338.
2018-10-10 16:44:42 -07:00
Joe Groff
5414747873 Merge pull request #19613 from jckarter/enable-consuming
Make `__consuming` meaningful for code generation.
2018-10-01 10:06:29 -07:00
Jordan Rose
36ca10f6f2 Merge pull request #19609 from jrose-apple/your-code-is-programmed-not-to-love-me-but-you-cant-pretend
[Stats] Track NumASTBytesAllocated continuously
2018-09-28 20:03:22 -07:00
Joe Groff
9c5432c0dd Make __consuming meaningful for code generation.
Previously, the `__consuming` decl modifier failed to get propagated to the value ownership of the
method's `self` parameter, causing it to effectively be a no-op. Fix this, and address some of the
downstream issues this exposes:

- `coerceCallArguments` in the type checker failing to handle the single `__owned` parameter case
- Various places in SILGen and optimizer passes that made inappropriate assertions that `self`
  was always passed guaranteed
2018-09-28 14:09:59 -07:00
Jordan Rose
91d1abf8b0 [Stats] Track NumASTBytesAllocated continuously
...so it can be used with flamegraphs.
2018-09-28 13:43:36 -07:00
Slava Pestov
ce770cdf4e AST: Introduce GenericSignature::forEachParam()
This replaces the inefficient pattern:

  for (auto param : sig->getGenericParams()) {
    if (sig->isCanonicalTypeInContext(param)) {
      ...
    } else {
      ...
    }
  }
2018-09-27 21:28:36 -07:00
Slava Pestov
3b60ae153d AST: Rename AnyFunctionType::Param::getType() to getOldType() 2018-09-26 11:05:23 -07:00
Slava Pestov
4975f3a15f AST: Remove FunctionType::getInput() 2018-09-26 11:05:23 -07:00
Slava Pestov
3554da1a18 AST: Redo FunctionType canonicalization and profiling
Use the parameter list instead of the input type.
2018-09-26 11:05:23 -07:00
Arnold Schwaighofer
7ba31e36e6 SIL: Profile the witnesMethodConformance as part of the SILFunctionType
We use to return a SILFunctionType with a wrong witness method protocol
conformance e.g when there were base protocols.

rdar://40149340
2018-09-21 13:23:27 -07:00
Doug Gregor
32fd274f5e Merge pull request #19391 from DougGregor/assoc-conformance-default-witnesses
[ABI] Associated conformance defaults
2018-09-19 13:15:46 -07:00
Doug Gregor
fde7eb5f3d [AST] Record default associated conformance witnesses directly.
Referencing associated conformance witnesses via the index of the
requirement in the requirement signature is inconvenient. Reference
them by CanType/ProtocolDecl* instead.
2018-09-19 10:53:46 -07:00
Jordan Rose
e1901fdad1 Replace explicit 'delete' calls with std::unique_ptr in AST (#19366) 2018-09-18 19:41:11 -07:00
Doug Gregor
89f3da6d28 [Sema] Start recording default associated conformances.
When forming the default witness table for a resilient protocol, look
for default associated conformances as well. These are identified by
associated conformance requirements whose root associated type has a
default type. For such requirements, we look for a conformance and
record it as a default associated conformance.

Emit a warning in cases where we don't find such a conformance,
because the associated type and conformance *may* have been added
with the intent of being resilient, and we can't know. This warning
might be a terrible idea, but it is only enabled under
-enable-resilience (which itself is hidden) and fires in one only
place in the standard library (which seems legitimate), so we'll try
it for now.
2018-09-18 15:49:43 -07:00
Doug Gregor
2ef9363bd1 [ABI] Add default associated type witnesses to resilient protocols.
When an associated type witness has a default, record that as part of
the protocol and emit a default associated type metadata accessor into the
default witness table. This allows a defaulted associated type to be
added to a protocol resiliently.

This is another part of rdar://problem/44167982, but it’s still very
limiting because the new associated type cannot have any conformances.
2018-09-15 22:04:46 -07:00
Doug Gregor
4f544f965a [AST] Kill rarely-used DenseMap from ProtocolDecl.
Put a DenseMap onto the ASTContext’s implementation instead.
2018-09-15 22:04:46 -07:00
Jordan Rose
ac4a92a968 Merge pull request #19292 from jrose-apple/get-out-the-map
Minor improvements to the use of StringMap/StringSet
2018-09-14 10:52:59 -07:00
Jordan Rose
245b06b832 Use llvm::StringSet instead of StringMap where appropriate
That is, where we aren't using the value for anything.

No functionality change.
2018-09-13 15:15:24 -07:00
Xi Ge
033732b1e5 ASTVerifier: avoid verifying IterableDeclContext if it has unparsed members. 2018-09-12 14:11:41 -07:00
Xi Ge
e0c1fea21b Fix an unused variable warning in no-assertion build. 2018-09-11 16:30:57 -07:00
Doug Gregor
ac37747c78 [Type checker] Break isObjC/getObjCSelector cycle.
When we're marking a declaration as @objc and recording it in the
class and source-file lookup tables (for @objc collision detection),
don't cause a cycle by querying `getObjCSelector()`. This is somewhat
of a hack: a better long-term approach would be to move the recording
much later, and request'ify the name computation. That'll be follow-up
work.
2018-09-10 17:30:17 -07:00
Xi Ge
24b0eac9a4 Parser: parse members in extension decls incrementally. 2018-09-05 17:00:39 -07:00
Slava Pestov
542ed54c79 Merge pull request #19042 from slavapestov/match-call-arguments
Refactor matchCallArguments() to take arrays of AnyFunctionType::Param instead of argument tuples
2018-08-29 17:16:05 -07:00
Xi Ge
3510f082e3 Merge pull request #19035 from nkcsgexi/refactor-graph-viz
[Evaluator] Refactor evaluator's GraphViz output to the destructor of the ASTContext. NFC
2018-08-29 10:49:54 -07:00