Preparation to fix <rdar://problem/18151694> Add Builtin.checkUnique
to avoid lost Array copies.
This adds the following new builtins:
isUnique : <T> (inout T[?]) -> Int1
isUniqueOrPinned : <T> (inout T[?]) -> Int1
These builtins take an inout object reference and return a
boolean. Passing the reference inout forces the optimizer to preserve
a retain distinct from what’s required to maintain lifetime for any of
the reference's source-level copies, because the called function is
allowed to replace the reference, thereby releasing the referent.
Before this change, the API entry points for uniqueness checking
already took an inout reference. However, after full inlining, it was
possible for two source-level variables that reference the same object
to appear to be the same variable from the optimizer's perspective
because an address to the variable was longer taken at the point of
checking uniqueness. Consequently the optimizer could remove
"redundant" copies which were actually needed to implement
copy-on-write semantics. With a builtin, the variable whose reference
is being checked for uniqueness appears mutable at the level of an
individual SIL instruction.
The kind of reference count checking that Builtin.isUnique performs
depends on the argument type:
- Native object types are directly checked by reading the
strong reference count:
(Builtin.NativeObject, known native class reference)
- Objective-C object types require an additional check that the
dynamic object type uses native swift reference counting:
(Builtin.UnknownObject, unknown class reference, class existential)
- Bridged object types allow the dymanic object type check to be
bypassed based on the pointer encoding:
(Builtin.BridgeObject)
Any of the above types may also be wrapped in an optional. If the
static argument type is optional, then a null check is also performed.
Thus, isUnique only returns true for non-null, native swift object
references with a strong reference count of one.
isUniqueOrPinned has the same semantics as isUnique except that it
also returns true if the object is marked pinned regardless of the
reference count. This allows for simultaneous non-structural
modification of multiple subobjects.
In some cases, the standard library can dynamically determine that it
has a native reference even though the static type is a bridge or
unknown object. Unsafe variants of the builtin are available to allow
the additional pointer bit mask and dynamic class lookup to be
bypassed in these cases:
isUnique_native : <T> (inout T[?]) -> Int1
isUniqueOrPinned_native : <T> (inout T[?]) -> Int1
These builtins perform an implicit cast to NativeObject before
checking uniqueness. There’s no way at SIL level to cast the address
of a reference, so we need to encapsulate this operation as part of
the builtin.
Swift SVN r27887
@objc protocols aren't supported with an ObjC runtime, but we still want values of AnyObject type to be word-sized. Handle this by turning the binary "needsWitnessTable" condition into a "dispatch strategy" enum, so we can recognize the condition "has no methods, so neither swift nor objc dispatch" as distinct from either swift or ObjC protocol representations. Assign this dispatch strategy when we lower AnyObject. Should be NFC for the ObjC-enabled build.
(It would also be beneficial for the ObjC-runtime-enabled version of Swift if AnyObject weren't an @objc protocol; that would mean we could give it a canonical protocol descriptor in the standard library, among other things. There are fairly deep assumptions in Sema that AnyObject is @objc, though, and it's not worth disturbing those assumptions right now.)
Reapplying with updates to the runtime unit tests.
Swift SVN r27341
@objc protocols aren't supported with an ObjC runtime, but we still want values of AnyObject type to be word-sized. Handle this by turning the binary "needsWitnessTable" condition into a "dispatch strategy" enum, so we can recognize the condition "has no methods, so neither swift nor objc dispatch" as distinct from either swift or ObjC protocol representations. Assign this dispatch strategy when we lower AnyObject. Should be NFC for the ObjC-enabled build.
(It would also be beneficial for the ObjC-runtime-enabled version of Swift if AnyObject weren't an @objc protocol; that would mean we could give it a canonical protocol descriptor in the standard library, among other things. There are fairly deep assumptions in Sema that AnyObject is @objc, though, and it's not worth disturbing those assumptions right now.)
Swift SVN r27338
Some future-proofing to let us change ErrorType's reference counting in the future, or to use various tagged pointer optimizations in its representation.
Swift SVN r27213
These aren't really orthogonal concerns--you'll never have a @thick @cc(objc_method), or an @objc_block @cc(witness_method)--and we have gross decision trees all over the codebase that try to hopscotch between the subset of combinations that make sense. Stop the madness by eliminating AbstractCC and folding its states into SILFunctionTypeRepresentation. This cleans up a ton of code across the compiler.
I couldn't quite eliminate AbstractCC's information from AST function types, since SIL type lowering transiently created AnyFunctionTypes with AbstractCCs set, even though these never occur at the source level. To accommodate type lowering, allow AnyFunctionType::ExtInfo to carry a SILFunctionTypeRepresentation, and arrange for the overlapping representations to share raw values.
In order to avoid disturbing test output, AST and SILFunctionTypes are still printed and parsed using the existing @thin/@thick/@objc_block and @cc() attributes, which is kind of gross, but lets me stage in the real source-breaking change separately.
Swift SVN r27095
"Autoclosure" is uninteresting to SIL. "noescape" isn't currently used by SIL and we shouldn't have it until it has a meaningful effect on SIL. "throws" should be adequately represented by a SIL function type having an error result.
Swift SVN r27023
The set of attributes that make sense at the AST level is increasingly divergent from those at the SIL level, so it doesn't really make sense for these to be the same. It'll also help prevent us from accidental unwanted propagation of attributes from the AST to SIL, which has caused bugs in the past. For staging purposes, start off with SILFunctionType's versions exactly the same as the FunctionType versions, which necessitates some ugly glue code but minimizes the potential disruption.
Swift SVN r27022
Fix a layout discrepancy when an associated type had @objc protocol constraints, in which case WitnessTableLayout would reserve a slot for the witness table, but WitnessTableBuilder would never emit it. Also add assertions to WitnessTableBuilder that check that the witness table we're building follows the claimed WitnessTableLayout. Fixes rdar://problem/20418117.
Swift SVN r26953
We both mixed up dest and src, and begin and end. Oops. Noticed by inspection. I also noticed the (wrong) code was duplicated in two places, so factor those together.
Swift SVN r26924
As part of this, re-arrange the argument order so that
generic arguments come before the context, which comes
before the error result. Be more consistent about always
adding a context parameter on thick functions, even
when it's unused. Pull out the witness-method Self
argument so that it appears last after the error
argument.
Swift SVN r26667
Provide a special single-ObjC-refcounted type info for error existentials, and lower the existential box instructions to their corresponding runtime calls.
Swift SVN r26469
Getting the protocols of an arbitrary type doesn't make sense, so start phasing this out by introducing specialized entry points that do make sense:
- get the inherited protocols of a ProtocolDecl
- get the conforming protocols for an associated type or generic
type parameter
- (already present) ask for the protocols to which a nominal type conforms
Swift SVN r26411
The deallocating parameter convention is a new convention put on a
non-trivial parameter if the caller function guarantees to the callee
that the parameter has the deallocating bit set in its object header.
This means that retains and releases do not need to be emitted on these
parameters even though they are non-trivial. This helps to solve a bug
in +0 self and makes it trivial for the optimizer to perform
optimizations based on this property.
It is not emitted yet by SILGen and will only be put on the self
argument of Deallocator functions.
Swift SVN r26179
This reduces the size of protocoll witnesses in the dylib by 12%, resulting in an overall dylib code size reduction of 1.7%.
And there are a few performance improvements for -Onone: CaptureProp: +34%, RangeAssignment: +16%.
Swift SVN r25269
with more explicit/semantic conversions in and out.
Using a PointerUnion with overlapping pointer types
is both error-prone and pretty close to illegible.
Swift SVN r24707
OpaqueStorageTypeInfo uses iNNN types that don't always have the correct alloc size for an expected
size at the LLVM level. This needs to be fixed before my partial apply closure fixes can hold.
Swift SVN r24551
In order to deal with generic indirect value captures, we need to be able to bind their type metadata in the partial apply forwarder and heap object destructor to have access to the value operations for that type. NecessaryBindings gives us a way to do that and clean up our ad-hoc polymorphic argument forwarding we had before. N(intended)FC yet, aside from some harmless reordering of operations, since we also need to implement NonFixedOffsets for heap objects to be fully operational.
Swift SVN r24526
When dealing with multiple levels of generic parameters, the mapping
from potential archetypes down to actual archetypes did not have
access to the archetypes for outer generic parameters. When same-type
requirements equated a type from the inner generic parameter list with
one from the outer generic parameter list, the reference to the outer
generic parameter list's type would remain dependent. For example,
given:
struct S<A: P> {
init<Q: P where Q.T == A>(_ q: Q) {}
}
we would end up with the dependent type for A (τ_0_0) in the same-type
constraint in the initializer requirement.
Now, notify the ArchetypeBuilder of outer generic signatures (and,
therefore, outer generic parameters), so that it has knowledge of the
mapping from those generic parameters to the corresponding
archetypes. Use that mapping when translating potential archetypes to
real archetypes. Additionally, when a potential archetype is mapped to
a concrete type (via a same-type constraint to a concrete type),
substitute archetypes for any dependent types within the concrete
type.
Remove a bunch of hacks in the compiler that identified dependent
types in "strange" places and tried to map them back to
archetypes. Those hacks handled some narrow cases we saw in the
standard library and some external code, but papered over the
underlying issue and left major gaps.
Sadly, introduce one hack into the type checker to help with the
matching of generic witnesses to generic requirements that follow the
pattern described above. See ConstraintSystem::SelfTypeVar; the proper
implementation for this matching involves substituting the adoptee
type in for Self within the requirement, and synthesizing new
archetypes from the result.
Fixes rdar://18435371, rdar://18803556, rdar://19082500,
rdar://19245317, rdar://19371678 and a half dozen compiler crashers
from the crash suite. There are a few other radars that I suspect this
fixes, but which require more steps to reproduce.
Swift SVN r24460
IRGen uses a typedef, SpareBitVector, for its principal
purpose of tracking spare bits. Other uses should not
use this typedef, and I've tried to follow that, but I
did this rewrite mostly with sed and may have missed
some fixups.
This should be almost completely NFC. There may be
some subtle changes in spare bits for witness tables
and other off-beat pointer types. I also fixed a bug
where IRGen thought that thin functions were two
pointers wide, but this wouldn't have affected anything
because we never store thin functions anyway, since
they're not a valid AST type.
This commit repplies r24305 with two fixes:
- It fixes the computation of spare bits for unusual
integer types to use the already-agreed-upon type
size instead of recomputing it. This fixes the
i386 stdlib build. Joe and I agreed that we should
also change the size to use the LLVM alloc size
instead of the next power of 2, but this patch
does not do that yet.
- It changes the spare bits in function types back
to the empty set. I'll be changing this in a
follow-up, but it needs to be tied to runtime
changes. This fixes the regression test failures.
Swift SVN r24324
IRGen uses a typedef, SpareBitVector, for its principal
purpose of tracking spare bits. Other uses should not
use this typedef, and I've tried to follow that, but I
did this rewrite mostly with sed and may have missed
some fixups.
This should be almost completely NFC. There may be
some subtle changes in spare bits for witness tables
and other off-beat pointer types. I also fixed a bug
where IRGen thought that thin functions were two
pointers wide, but this wouldn't have affected anything
because we never store thin functions anyway, since
they're not a valid AST type.
Swift SVN r24305
storage for arbitrary values.
A buffer doesn't provide any way to identify the type of
value it stores, and so it cannot be copied, moved, or
destroyed independently; thus it's not available as a
first-class type in Swift, which is why I've labelled
it Unsafe. But it does allow an efficient means of
opaquely preserving information between two cooperating
functions. This will be useful for the adjustments I
need to make to materializeForSet to support safe
addressors.
I considered making this a SIL type category instead,
like $@value_buffer T. This is an attractive idea because
it's generally better-typed. The disadvantages are that:
- it would need its own address_to_pointer equivalents and
- alloc_stack doesn't know what type will be stored in
any particular buffer, so there still needs to be
something opaque.
This representation is a bit gross, but it'll do.
Swift SVN r23903
isn't used yet, but will be for modeling the self argument passed to an
address-only witness implementation. NFC since all this code is dead :-)
Swift SVN r23857
Previously, we were not respecting the representation of the existential
metatype and were treating all existential metatypes as if the metatype
was a thick metatype. Instead now we properly grab the instance of the
class from the existential and then query the runtime for the
objc_class. This is done via the new entrypoint
emitHeapMetadataRefForUnknownHeapObject.
I also modified emitHeapMetadataRefForHeapObject to use
emitHeapMetadataRefForUnknownHeapObject instead of
emitLoadOfObjCHeapMetadataRef since the latter does not properly handle
tagged pointers. This bug was found on inspection when Joe and I were
talking about this change.
rdar://18841292
Swift SVN r23308
Move the uniquing information for ForeignTypeMetadata behind the address point so we can share the layout between foreign classes and the existing layout for struct and enum metadata. Emit metadata records for imported structs and enums as foreign metadata candidates, and dynamically unique references to the metadata by calling swift_getForeignTypeMetadata.
Swift SVN r23081
As a stopgap till IRGen properly emits indirected class references, support direct class reference protocol conformance records. This should get us to the point we can replace the dlsym hack with the new implementation.
Swift SVN r23073
