to being a semantic assign. The functional difference here is only
for unowned pointers, where now the conversion from strong to unowned
is implicit in the assign.
The logic behind this is that 'assign' is really part of SILGen, that
is only moved later to make it easier to make it be flow sensitive.
Since it is part of SILGen and will be using some of the type lowering
functionality used by SILGen, we should play by its rules.
No functionality change.
Swift SVN r7507
This is particularly convenient for dealing with generated
code (e.g. as produced by the SIL printer), but it's also
quite defensible as a way for users to logically group
related attributes and split unrelated ones out, should
users find themselves using a lot of verbose attributes
(and prior experience with e.g. user-defined attributes
is that they definitely do sometimes turn into walls of
text...).
Swift SVN r7376
description subsumes most of it. It was amusing that I thought (back in the day) as
structs as a special case of unions, but this isn't pertinent any more.
Rename the expressions.rst file. It does have some useful stuff, but it should be merged
into other docs at some point.
Swift SVN r7359
This documents the current state of affairs, where "Int?[]" is allowed but
"Int[]?" is not. I'll update this again if/when we reach a new decision.
Swift SVN r7269
Now that we don't have formatForPrinting anymore, a terser name for
debugFormat makes more sense. Also, the new name isn't freighted with
so much implication about the end-user's activity.
Swift SVN r7256
Jordan pointed out that we already have these multi-parameter
format(...) functions that also return Streamables; it doesn't make
much sense to spell them differently.
Swift SVN r7255
These value witness table entries will be conditionally available for types that support specialized union representation through extra inhabitants and/or spare bits and for union value witnesses:
- storeExtraInhabitant, to store an extra inhabitant representation;
- getExtraInhabitantIndex, to recognize an extra inhabitant representation;
- getUnionTag, to get a union's discriminator; and
- inplaceProjectUnionData, to extract the value in place from a union.
This just sets up the enumerators and related IR types and mangling; nothing emits these witnesses yet.
Swift SVN r7234
In designing the runtime model for generic unions with John, we decided we're going to need very different SIL semantics for destructuring address-only unions, so I don't think trying to stretch switch_union to cover all unions makes any sense anymore.
Swift SVN r7225
We never really discussed this and it doesn't really buy us much. If we
want to have a compact way to import many things, it may not even end
up looking like this.
Swift SVN r7015
The current implementation of dealloc_stack in IR-gen is a
no-op, but that's very much wrong for types with non-trivial
local allocation requirements, e.g. archetypes. So we need
to be able to do non-trivial code here. However, that means
modeling both the buffer pointer and the allocated address
in SIL.
To make this more type-safe, introduce a SIL-specific
'[local_storage] T' type that represents the required
allocation for locally storing a T. alloc_stack now returns
one of those in additon to a *T, and dealloc_stack expects
the former.
IR-gen still implements dealloc_stack as a no-op, but
that's now easy to fix.
Swift SVN r6937
Have project_existential return $*This instead of $Builtin.OpaquePointer, and have protocol_method do the same for methods of opaque protocols. This makes it easier for passes to reason about the semantics of the projected address, since it's restricted by the semantics of SIL addresses.
Swift SVN r6872
We need to handle three cases:
- If a protocol conformance has no associated types, or the associated type witnesses all have statically resolvable metadata, we can expose a *direct* witness table symbol.
- If a protocol conformance has associated types with runtime-instantiated metadata, we need to gate the witness table behind a *lazy* initializer function to fill in the metadata fields.
- If a protocol conformance has associated types where the type or one of its conformances are *dependent* on its parent's generic parameters, we need to instantiate multiple witness tables at runtime.
Swift SVN r6805
Modules can be in either 'Raw' or 'Canonical' form, with different invariants on each. We don't actually distinguish those invariants yet, but this patch adds the field to SILModule and adds a "sil_stage" declaration to SIL printer/parser syntax.
Swift SVN r6793
Also, update LangRef.
Note that an explicit "import module" has been left out for now, since
it's not strictly necessary and "module" isn't a keyword yet.
Swift SVN r6786
We haven't fully updated references to union cases, and enums still are not
their own thing yet, but "oneof" is gone. Long live "union"!
Swift SVN r6783
the debug info, to aid the debugger in figuring out the implementation
language of a given type in multi-language environments.
FYI, where applicable, we also emit the
DW_AT_APPLE_runtime_class(DW_AT_lang_Swift) attribute.
The demangler accepts type names with the "_Tt" prefix without the --type
switch.
Swift SVN r6714
Standardize on the more-common "superclass" and "subclass" terminology
throughout the compiler, rather than the odd mix of base/derived and
super/sub.
Also, have ClassDecl only store the Type of the superclass. Location
information will be part of the inheritance clause for parsed classes.
Swift SVN r6687
This makes long literals like 1_000_000_000 or 0x7FFF_FFFF_FFFF_FFFF much easier to read, and has precedent in Perl, OCaml, Python, Ruby, Rust, ... Fixes <rdar://problem/14247571>.
Swift SVN r6681
This way we don't need to deal with the inaccuracy of decimal float literals. While we're here, modify the in-memory representation of IntegerLiteralInst and FloatLiteralInst to save the word array of the APInt instead of round-tripping through plain text.
Swift SVN r6676
