Commit Graph

1326 Commits

Author SHA1 Message Date
John McCall
4e49329393 Make is-reflectable a type descriptor flag instead of a separate field.
Also fix a test for 32-bit compatibility.
2018-02-28 00:38:00 -05:00
John McCall
8d93cee012 Add a superclass reference to class descriptors. 2018-02-28 00:38:00 -05:00
John McCall
fa2518b0c9 Clarify the relationship to the type descriptor fields being filled; NFC. 2018-02-27 16:00:21 -05:00
John McCall
a53f82449c Clarify which type-descriptor-specific flags apply to which types; NFC. 2018-02-27 16:00:21 -05:00
John McCall
dd99536d31 Move the metadata-pattern header into the type context descriptor.
This is yet another waypoint on the path towards the final
generic-metadata design.  The immediate goal is to make the
pattern a private implementation detail and to give the runtime
more visibility into the allocation and caching of generic types.
2018-02-26 12:10:24 -05:00
John McCall
46e5c07818 Remove the now-dead field type vector slot in metadata templates. 2018-02-25 04:04:01 -05:00
Pavel Yaskevich
333f6146fc [IRGen] NFC: Update structure of ClassContextDescriptor representation 2018-02-21 15:26:29 -08:00
Pavel Yaskevich
94017f7ee7 [IRGen] Remove 'FieldNames' field from type context descriptor
All of the information contained by this field (list of property names)
is already encoded as part of the field reflection metadata and
is accessible via `swift_getFieldAt` runtime method.
2018-02-20 18:49:00 -08:00
Pavel Yaskevich
f6be62dfdb [IRGen] Remove lazy field type accessor functions
All of their usages have been replaced with new runtime
`swift_getFieldAt` method.
2018-02-20 18:45:41 -08:00
Doug Gregor
800e3818b7 [Type metadata] @objc protocol conformance requirements don't have arguments.
@objc protocols don't have witness tables. However, both type metadata
(in the nominal type descriptors) and the runtime code to demangle
type names into metadata weren't acknowledging this. Fix type metadata
emission to not count an "extra argument" for @objc protocol
conformance requirements, and teach the runtime code to properly look
for conformances to @objc protocols (through the Objective-C runtime)
and not record witness tables for them.
2018-02-15 14:07:46 -08:00
Joe Groff
953dddd5d3 IRGen/Runtime: Allow mangled type refs to embed "symbolic references" to type context descriptors.
This makes resolving mangled names to nominal types in the same module more efficient, and for eventual secrecy improvements, also allows types in the same module to be referenced from mangled typerefs without encoding any source-level name information about them.
2018-02-10 10:43:47 -08:00
Pavel Yaskevich
478ba2693a [IRGen] Emit type field descriptors for imported structs 2018-02-06 15:50:52 -08:00
Arnold Schwaighofer
d981bb1d96 Mangling: noescape functions will be trivial and no longer compatible with escape function types.
Mangle escapeness as part of the type.

Part of:
SR-5441
rdar://36116691
2018-02-06 08:51:43 -08:00
Mark Lacey
7f805ba2bc Replace classifyAsOptionalType with isOptionalDecl. 2018-02-05 23:59:00 -08:00
Mark Lacey
b4b66bc8e8 Replace getAnyOptionalObjectType with getOptionalObjectType. 2018-02-05 23:59:00 -08:00
Mark Lacey
393eae2fb4 Merge pull request #14364 from rudkx/typealias-for-iuo
Make ImplicitlyUnwrappedOptional<T> an unavailable typealias.
2018-02-02 17:46:49 -08:00
Doug Gregor
b9e91e3934 [ABI] Describe module context and conditional requirements of conformances.
Extend protocol conformance descriptors with two more bits of information:

* For retroactive conformances, add the module in which the conformance
  occurs. This will eventually be used for error reporting/ambiguity
  resolution when retroactive conformances collide.
* For conditional conformances, add the conditional requirements. We need
  these for runtime evaluation of conditional conformances.
2018-02-02 16:41:27 -08:00
Mark Lacey
2008674495 Make ImplicitlyUnwrappedOptional<T> an unavailable typealias.
Also remove the decl from the known decls and remove a
bunch of code referencing that decl as well as a bunch of other
random things including deserialization support.

This includes removing some specialized diagnostics code that
matched the identifier ImplicitlyUnwrappedOptional, and tweaking
diagnostics for various modes and various issues.

Fixes most of rdar://problem/37121121, among other things.
2018-02-02 08:35:53 -08:00
Joe Groff
a7a3b17597 Replace nominal type descriptors with a hierarchy of context descriptors.
This new format more efficiently represents existing information, while
more accurately encoding important information about nested generic
contexts with same-type and layout constraints that need to be evaluated
at runtime. It's also designed with an eye to forward- and
backward-compatible expansion for ABI stability with future Swift
versions.
2018-01-29 16:19:25 -08:00
Arnold Schwaighofer
78b3747c27 IRGen: Don't mark metadata accessors that pass arguments indirectly through temp memory with readnone
This should fix the -Osize bots.

rdar://36683060
2018-01-23 13:57:44 -08:00
Doug Gregor
902ea27a67 [IRGen] Consistently record runtime metadata for foreign types.
All Swift-defined nominal types have their runtime metadata recorded
in a special section, so it can be found later. This recording is
suppressed when that type is stated to conform to a protocol, because
the runtime can find nominal types in either place.

Imported types would get their conformances recorded, but would not
get recorded in the runtime metadata record otherwose. Therefore, the
runtime would not be able to find such types by name.

For any foreign type whose metadata we emit, make sure that metadata
can be found by a runtime lookup.
2018-01-20 22:06:33 -08:00
Doug Gregor
30ecd8ed7f [IRGen] Always emit access functions into nominal type descriptors. 2018-01-20 15:33:26 -08:00
Doug Gregor
ca76c76272 [IRGen] Maintain the prior use of isForeign() for identifying foreign classes. 2018-01-19 23:31:56 -08:00
Doug Gregor
817c0b368e [WIP] Emit nominal type access functions for imported types.
Emit nominal type access functions for imported types. These access
functions work with non-unique metadata references, so they perform
uniquing through the runtime on first access.

Fixes rdar://problem/36430234.
2018-01-19 23:11:50 -08:00
Doug Gregor
eb4f9a3d4c [ABI] Reserve space in the protocol descriptor for the superclass.
Extend protocol descriptors with a field for the superclass bound of the
protocol itself. This carves out space in the ABI for

    class C { }
    protocol P : C { ... }

although the feature is not yet implemented.
2018-01-19 14:03:24 -08:00
Greg Parker
91bcf1ec87 [IRGen] Fix some is-stable-ABI marker bits. (#14004)
* Don't emit "stable ABI masquerading as legacy ABI" bit state yet.
* Don't unconditionally emit bit value 1 in generic class templates.

rdar://36612173
2018-01-18 23:24:00 -08:00
Slava Pestov
d4285266ee Merge pull request #13845 from slavapestov/class-dispatch-thunks-irgen
Generate class dispatch thunks in IRGen
2018-01-16 11:48:07 -08:00
Doug Gregor
cb5780f8b5 [IRGen] Don't use "readnone" for calls to metadata accessors with buffers.
When we allocate and pass a buffer to the nominal type access function,
mark the call as "inaccessiblemem_or_argmemonly" rather than "readnone".
2018-01-15 22:08:37 -08:00
Slava Pestov
c13ccde39f IRGen: Refactor emitVirtualMethodValue() 2018-01-15 21:10:05 -08:00
Doug Gregor
644226838b [ABI] Pass only three direct args to nominal type access functions.
The nominal type access functions took all of the generic arguments
directly, which is hard to call from the runtime. Instead, pass up to
three generic arguments directly (because it’s good for code size), and put the rest into an array.
2018-01-13 22:33:46 -08:00
John McCall
52bb547a7e Merge pull request #13866 from rjmccall/accessor-decl
Split AccessorDecl out from FuncDecl.  NFC.
2018-01-12 17:02:35 -05:00
swift-ci
9b58ed0ad7 Merge pull request #13884 from DougGregor/runtime-substitution-mangled-to-metadata 2018-01-12 13:43:10 -08:00
John McCall
7f0f8830cd Split AccessorDecl out from FuncDecl. NFC.
This has three principal advantages:

- It gives some additional type-safety when working
  with known accessors.

- It makes it significantly easier to test whether a declaration
  is an accessor and encourages the use of a common idiom.

- It saves a small amount of memory in both FuncDecl and its
  serialized form.
2018-01-12 14:20:27 -05:00
Slava Pestov
9ff97367df AST: Rename hasFixedLayout() to isResilient() and flip polarity
In IRGen we call this isResilient() already in IRGen, and it's
more consistent to call it the same thing everywhere.
2018-01-11 21:57:42 -08:00
Doug Gregor
3405852ef0 [Runtime] Find metadata for substitutions into demangled dependent members.
Extend the protocol descriptor with a (space-separated) list of associated
type names, in the order of their requirements. Use this information in
the runtime to support lookup of associated type witnesses by name when
mapping a mangled name to a type and substituting generic parameters.
2018-01-11 16:04:53 -08:00
swift-ci
0eb3ff53a8 Merge pull request #13874 from DougGregor/runtime-tuple-metadata-opt 2018-01-11 12:22:50 -08:00
Doug Gregor
93cd5e8270 [ABI] Make TargetTupleTypeFlags size_t sized and move “labels” bit.
Addresses feedback from @rjmccall.
2018-01-11 11:08:12 -08:00
Doug Gregor
adf38f5486 [Runtime] Pack element count into the flags of swift_getTupleTypeMetadata.
Reduces the # of parameters we need to pass to this runtime API.
2018-01-11 10:06:57 -08:00
Slava Pestov
3dae007465 IRGen: Remove FieldAccess::NonConstantIndirect 2018-01-10 21:32:54 -08:00
Slava Pestov
698881d5c6 IRGen: Protocol dispatch thunks
We want to be able to re-order existing protocol requirements
and add new protocol requirements with default implementations.

Enable this by wrapping the witness table lookup inside a
thunk and calling the thunk, instead of open-coding the
witness table lookup directly in client code.
2018-01-10 13:54:21 -08:00
Slava Pestov
2bd4f9bf66 IRGen: Fix formatting 2018-01-10 13:48:30 -08:00
Doug Gregor
0f4963dba8 [IRGen] Put references to all emitted protocols into a special section.
Introduce a new section that contains (relative) references to all of the
Swift protocol descriptors emitted into this module. We'll use this to
find protocol descriptors by name.
2018-01-09 16:10:18 -08:00
Doug Gregor
5853cf3c87 [Runtime] Teach swift_getTupleTypeMetadata() to copy "labels" string.
Introduce a flags parameter to swift_getTupleTypeMetadata(). Add a flag
stating when the "labels" parameter points into nonconstant memory, in
which case we need to make a copy of the string before adding an entry
into the concurrent map.
2018-01-05 23:11:21 -08:00
Pavel Yaskevich
fc64e34c8e Merge pull request #13701 from xedin/rdar-36278686
[Runtime/ABI] Optimize number of function metadata ABI endpoints
2018-01-05 21:47:13 -08:00
Pavel Yaskevich
0f071848f3 [Runtime/ABI] NFC: Refactor function type metadata generation 2018-01-05 13:38:20 -08:00
Pavel Yaskevich
86de71ee6d [Runtime/ABI] Remove swift_getFunctionTypeMetadata{1-3}WithFlags
Since it's not very common to use such ABI endpoints, let's remove
them and use the most general one `swift_getFunctionTypeMetadata`
instead when function parameters have flags attached to them.

Resolves: rdar://problem/36278686
2018-01-04 16:32:51 -08:00
Pavel Yaskevich
8409cdb8a2 [Runtime/ABI] Add swift_getFunctionTypeMetadata0 function
This ABI endpoint is used to retrieve metadata about functions
without parameters. Which is very common use-case and it
makes sense to save some code size for that.
2018-01-04 16:32:51 -08:00
Doug Gregor
afc5cf5bfb Merge pull request #13734 from DougGregor/nominal-type-refs
[Runtime] Never use type metadata references in conformance records.
2018-01-04 16:02:12 -08:00
Mark Lacey
55eb741fde Fix build failure due to race between two PRs.
The first of the following commits did some renames that result in the
second to fail, but they were both in CI being tested simultaneously,
passed, and were merged.

12a774abec
93af58dead
2018-01-04 12:42:27 -08:00
Doug Gregor
6118d86032 [Runtime] Introduce equality operation for nominal type descriptors.
Nominal type descriptors are not always unique, so testing them via pointer
equality is not correct. Introduce an "isEqual()" operation for
nominal type descriptors that performs the appropriate equality check,
using pointer equality when possible, and falling back to string
comparisons of the mangled type name when it is not possible.

Introduce a "nonunique" flag into nominal type descriptors to describe
when they are, in fact, not unique. The only nonunique nominal type
descriptors currently come from Clang-imported types; all
Swift-defined types have unique nominal type descriptors. Use this
flag to make the aforementioned operation efficient in the "unique"
case.

Use the new isEqual() operation for protocol conformance lookup, and
make sure we're caching results based on the known-canonical nominal
type descriptor.
2018-01-04 12:36:34 -08:00