In some circumstances, a Swift declaration in module A will depend on
another declaration (usually from Objective-C) that can't be loaded,
for whatever reason. If the Swift declaration is *overriding* the
missing declaration, this can present a problem, because the way
methods are dispatched in Swift can depend on knowing the original
class that introduced the method. However, if the compiler can prove
that the override can still be safely invoked/used in all cases, it
doesn't need to worry about the overridden declaration being missing.
This is especially relevant for property accessors, because there's
currently no logic to recover from a property being successfully
deserialized and then finding out that an accessor couldn't be.
The decision of whether or not an override can be safely invoked
without knowledge of the base method is something to be cautious
about---a mistaken analysis would effectively be a miscompile. So up
until now, this was limited to one case: when a method is known to be
`@objc dynamic`, i.e. always dispatched through objc_msgSend. (Even
this may become questionable if we have first-class method references,
like we do for key paths.) This worked particularly well because the
compiler infers 'dynamic' for any overload of an imported Objective-C
method or accessor, in case it imports differently in a different
-swift-version and a client ends up subclassing it.
However...that inference does not apply if the class is final, because
then there are no subclasses to worry about.
This commit changes the test to be more careful: if the /missing/
declaration was `@objc dynamic`, we know that it can't affect ABI,
because either the override is properly `@objc dynamic` as well, or
the override has introduced its own calling ABI (in practice, a direct
call for final methods) that doesn't depend on the superclass. Again,
this isn't 100% correct in the presence of first-class methods, but it
does fix the issue in practice where a property accessor in a parent
class goes missing. And since Objective-C allows adding property
setters separately from the original property declaration, that's
something that can happen even under normal circumstances. Sadly.
This approach could probably be extended to constructors as well. I'm
a little more cautious about throwing vars and subscripts into the mix
because of the presence of key paths, which do allow identity-based
comparison of overrides and bases.
rdar://problem/56388950
After setting up the .swiftsourceinfo file, this patch starts to actually serialize
and de-serialize source locations for declaration. The binary format of .swiftsourceinfo
currently contains these three records:
BasicDeclLocs: a hash table mapping from a USR ID to a list of basic source locations. The USR id
could be retrieved from the following DeclUSRs record using an actual decl USR. The basic source locations
include a file ID and the results from Decl::getLoc(), ValueDecl::getNameLoc(), Decl::getStartLoc() and Decl::getEndLoc().
The file ID could be used to retrieve the actual file name from the following SourceFilePaths record.
Each location is encoded as a line:column pair.
DeclUSRS: a hash table mapping from USR to a USR ID used by location records.
SourceFilePaths: a hash table mapping from a file ID to actual file name.
BasicDeclLocs should be sufficient for most diagnostic cases. If additional source locations
are needed, we could always add new source location records without breaking the backward compatibility.
When de-serializing the source location from a module-imported decl, we calculate its USR, retrieve the USR ID
from the DeclUSRS record, and use the USR ID to look up the basic location list in the BasicDeclLocs record.
For more details about .swiftsourceinfo file: https://forums.swift.org/t/proposal-emitting-source-information-file-during-compilation
The weak imported flag is now only set if the attribute is unconditionally
weak linked, which is the case when it or one of its parent contexts has a
@_weakLinked attribute.
To correctly handle weak linking based availability with serialized SIL
functions, we need to serialize the actual version tuple when the SIL function
was introduced. This is because the deployment target of the client app can
be older than the deployment target that the original module was built with.
Fixes <rdar://problem/52783668>.
Instead of computing it from the extended type after deserialization --
which is tricky to do, due to potential presence of protocol
compositions -- we obtain the extended nominal directly.
Fixes SR-11227 and linked rdar://problem/53712389.
This reverts commit dc98e17d84b991b6be8b8feb5e0d05aad24f52a4.
I believe this commit was causing test failures on:
IRGen/c_layout.sil
IRGen/existentials.sil
It also recreates the file lib/Serialization/ModuleFormat.h,
which really can't have been intended.
Swift SVN r23732
Adding explicit constructors to Clang-imported structs in the previous commits exposes a latent phase ordering issue between the Clang importer and SIL deserialization. Deserializing the standard library SIL ends up pulling in additional Clang decls which never get type-checked before we attempt to emit their code. Work around this by bringing back the "EagerDeserializedDecls" block in the serialization format, and adding any cross-referenced decls that get referenced in SILSerializeAll mode to it, so that we ensure they're available before SILGen. We also have to type-check external decls after doing so, since when only importing a module, we wouldn't do any type-checking at all otherwise.
Swift SVN r23728
The name Stream didn't seem to be working out as intended; we kept
gravitating back to calling it Generator, which is precedented in other
languages. Also, Stream seems to beg for qualification as Input or
Output. I think we'd like to reserve Stream for things that are more
bulk-character-API-ish.
Swift SVN r13893
Previously, we would just parse vars and subscripts with no definitions,
then let getters and setters be referenced arbitrarily later. This was
problematic for a number of reasons, not least of which, the .sil file
might be invalid.
Instead, change sil to require that a protocol style definition indicate
whether a vardecl/subscript is computed or not, and whether it is both
get-able and set-able, e.g. like "var x : Int { get }". Change the
sil printer to print decls in this form, and change the SILParser to
make SILDeclRef::Func values instead of ::Getter/Setter values.
One thing that this exposed is that we weren't correctly serializing the
accessor state in modules, so accessors would get detatched from their
AbstractStorageDecls when deserialized (and in fact, their ASD never got
deserialized at all in some cases). Fix this in the serialization of
the accessors.
NFC, other than the SIL printer and parser.
Swift SVN r13884
From now on, /any/ changes to SIL or AST serialization must increment
VERSION_MINOR in ModuleFormat.h.
The original intent of VERSION_MAJOR/VERSION_MINOR was that VERSION_MAJOR
would only increment when backwards-incompatible changes are introduced,
and VERSION_MINOR merely indicates whether to expect additional information.
However, the module infrastructure currently isn't forgiving enough to accept
even backwards-compatible changes to the record schemas, and the SIL
serialization design might not be compatible with that at all.
So for now, treat any version number 0.x as incompatible with any other 0.y.
We can bump to 1 when we hit stability.
<rdar://problem/15494343>
Swift SVN r13841
There are some straggling references to the context generic param list, but nothing uses the non-interface param or result types anymore!
Swift SVN r13725
with FuncDecls. This allows us to eliminate special case code for handling
self in various parts of the compiler.
This also improves loc info (debug info and AST info) because 'self' now
has a location instead of being invalid.
I also took the opportunity to factor a bunch of places creating self decls
to use similar patterns and less copy and paste code.
Swift SVN r13196
Making DynamicSelf its own special type node makes it easier to opt-in
to the behavior we want rather than opting out of the behavior we
don't want. Some things already work better with this representation,
such as mangling and overriding; others are more broken, such as the
handling of DynamicSelf within generic classes and the lookup of the
DynamicSelf type.
Swift SVN r13141
We don't currently use this for anything, but if we have the module name
available and easy to access in the bitstream, we can drop the wrapper
around the serialized AST that's put into the binary itself for use by LLDB.
Swift SVN r12921
We don't currently use this for anything, but if we have the module name
available and easy to access in the bitstream, we can drop the wrapper
around the serialized AST that's put into the binary itself for use by LLDB.
Swift SVN r12919
Lower types for SILDeclRefs from the interface types of their referents, dragging the old type along for the ride so we can still offer the context to clients that haven't been weaned off of it. Make SILFunctionType's interface types and generic signature independent arguments of its Derive the context types of SILFunctionType from the interface types, instead of the other way around. Do a bunch of annoying inseparable work in the AST and IRGen to accommodate the switchover.
Swift SVN r12536
This attribute states that all stored properties within the class must
provide initial values. This will allow us to move stored property
initialization into Objective-C's post-allocation initialization hook,
.cxx_construct.
Swift SVN r12228
Make the compiler fully support both UTF-8 and UTF-16 string
literals. A (standard-library-defined) string type (such as String)
that wants UTF-16 string literals should conform to the
BuiltinUTF16StringLiteralConvertible protocol; one that wants UTF-8
string literals should conform to the BuiltinStringLiteralConvertible
protocol.
Note that BuiltinUTF16StringLiteralConvertible inherits from
BuiltinStringLiteralConvertible, so a string type that wants UTF-16
string literals also has to implement support for UTF-8. The UTF-16
entry point is preferred when the compiler knows that UTF-16 is
supported. This only tends to happen when we have a generic parameter
that is required to conform to StringLiteralConvertible, e.g.,
func f<T: StringLiteralConvertible>() {
var t: T = "Hello, World!" // uses UTF-8 entry point
}
because the UTF-8 entry point is the only one guaranteed to be available.
Swift SVN r12014
This is step 1 of implementing the new Container/Sequence/Stream
protocols. See test/Prototypes/Container.swift for the complete picture
of where this is headed.
Swift SVN r11924
with qualifiers on it, we have two distinct types:
- LValueType(T) aka @lvalue T, which is used for mutable values on the LHS of an
assignment in the typechecker.
- InOutType(T) aka @inout T, which is used for @inout arguments, and the implicit
@inout self argument of mutable methods on value types. This type is also used
at the SIL level for address types.
While I detangled a number of cases that were checking for LValueType (without checking
qualifiers) and only meant @inout or @lvalue, there is more to be done here. Notably,
getRValueType() still strips @inout, which is totally and unbearably wrong.
Swift SVN r11727
Lower metatype types as @thin or @thick based on whether the type is static and whether the abstraction pattern allows for a thin metatype. Add a '@thick' attribute and require SIL metatypes to always be annotated with either '@thin' or '@thick' to distinguish them from unlowered metatypes.
Swift SVN r11525
typealias MyInt: ForwardIndex = Int
There is no real reason to allow this; it's just a static_assert that Int
conforms to ForwardIndex, which would be better spelled some other way.
This only applies to concrete typealiases, i.e. those that simply alias an
underlying type. Associated types can still have both inheritance clauses
and a (default) underlying type.
Swift SVN r11481
Previously, cross-references just carried a chain of identifiers and a
top-level module, plus a type to validate against, a generic parameter index,
or an operator fixity. However, referencing "the first generic parameter
of the prefix function ++ that takes a ForwardIndex" requires /all three/
of these filters to unambiguously select the right declaration.
Now, cross-references consist of a chain of trailing records, one for each
link in the path. There are (currently) five kinds of links:
Type: a declaration that cannot have overloads
Value: a declaration that can have overloads (filtered by type)
Extension: filter to decls within extensions on another module
Operator:
- as the first path piece, an operator declaration
- as a later path piece, a fixity filter for operator functions
Generic Param: an indexed generic parameter of the previous result
This should allow us to uniquely cross-reference any Swift declaration we
need to.
Swift SVN r11399
A SpecializedProtocolConformance intentionally contains all of the
information we need to synthesize the type witnesses from the
underlying (generic) conformance. Do so lazily rather than eagerly,
because we won't always need all of them.
As a nice side effect, we no longer need to serialize the witnesses of
these specialized protocol conformances, so we can save some space in
the Swift module file.
Swift SVN r11303
