This is needed for Windows which does not support cross-module data
references without indirection. By lazy initializing the data, we can
indirect through the IAT for the data pointer and fill in the parent
pointer.
Field offset vectors are always filled out with either zero or the static layout's offset, depending on the metadata initialization strategy. This change means that the static layout's offset will only be non-zero for properties with a statically-known layout. Existing runtimes doing dynamic class layout assign class properties a zero offset if the field offset vector entry is zero and the property is zero-sized. So this effectively brings the compiler into accord with the runtime (for all newly-compiled Swift code, which will eventually be all Swift code because the current public releases of Swift 5 are not yet considered ABI-stable) and guarantees a zero value for the offset everywhere.
Since the runtime will agree with the compiler about the zero value of the offset, the compiler can continue to emit such offset variables as constant. The exception to this rule is if the class has non-fragile ObjC ancestry, in which case the ObjC runtime (which is not aware of this special rule for empty fields) will attempt to slide it along with everything else.
Fixes rdar://48031465, in which the `FixedClassMetadataBuilder` for a class with a legacy-fixed layout was writing a non-zero offset for an empty field into the field offset vector, causing the runtime to not apply the special case and thus to compute a non-zero offset, which it then attempted to copy into the global field offset variable, which the compiler had emitted as a true-constant zero.
This consolidates the various doesClassMetadataRequire*() checks, making
them more managable.
This also adds a forth state, ClassMetadataStrategy::Update. This will be used
when deploying to the new Objective-C runtime. For now it's not plumbed through.
Progress on <rdar://problem/47649465>.
In our initial approach for resolving metadata dependency cycles with classes, non-transitively complete superclass metadata was fetched by the subclass's metadata completion function and passed to `swift_initClassMetadata`. That could mean generating quite a lot of code in the completion function, and so we fairly recently changed it so that `swift_initClassMetadata` instead fetched the superclass metadata via a demangling. Unfortunately, the metadata demangler only fetches _abstract_ metadata by default, and class metadata cannot be considered even non-transitively complete when its superclass reference not at that stage. If the superclass metadata is being completed on one thread, and a subclass is being completed on another, and the subclass installs the incomplete superclass metadata in its superclass field and attempts to register the subclass with the Objective-C runtime, the runtime may crash reading the incompletely-initialized superclass.
The proper fix is to make `swift_initClassMetadata` fetch non-transitively complete metadata for the superclass, delaying completion if that metadata is unavailable. Unfortunately, that can't actually be implemented on top of `swift_initClassMetadata` because that function has no means of reporting an unsatisfied dependency to its caller, and we can no longer simply change its signature without worrying about a small of internal code that might still be using it. We cannot simply perform a blocking metadata request in `swift_initClassMetadata` because it is deeply problematic to block within a metadata completion function. The solution is therefore to add a `swift_initClassMetadata2` which has the ability to report unsatisfied dependencies. That was done in #22386; this patch builds on that by teaching the compiler to generate code to actually use it. It is therefore not safe to use this patch if you might be running on an OS that only provides the old runtime function, but that should be a temporary Apple-internal problem.
Fixes rdar://47549859.
When -enable-anonymous-context-mangled-names is provided, emit mangled
names as part of the metadata of an anonymous context. This will allow
us to match textual mangled names to the metadata.
This is a backward-compatible ABI extension. Part of rdar://problem/38231646/.
Protocol descriptors for resilient protocols relatively-reference
default witness thunks, so when using -num-threads N with N > 1,
we must ensure the default witness thunk is emitted in the same
LLVM module.
Anonymous context descriptors were being treated as non-generic by
IRGen, which lead to problems for (file)private types within generic
types. Emit generic parameters and requirements for anonymous contexts
as well.
The runtime was mostly prepared for this, and the ABI already
accounted for it, so the runtime change is minor---it only affected
building a demangle tree from metadata.
Fixes rdar://problem/46853806.
We were wastefully emitting an accessor if a field had a type, for
example if my field type was (() -> (X, Array<Y>>) we would force
the emission of a function to construct (() -> (X, Array<Y>)) even
though all we care about is the type metadata for X and Y.
Conversely, we would skip the field type if it contained an
archetype, even if it otherwise contained metadata that we need
to force to emit, for instance something like (T, X) where T is
a generic parameter and X is a nominal type.
A final side effect is we no longer try to emit type metadata for
one-element tuples when emitting enum payload metadata, which is
something I want to assert against.
Start emitting associated conformance requirement descriptors for
inherited protocols, so we have a symbol to reference from resilient
witness tables and mangled names in the future.
On Windows the image format does not support cross-image absolute
data symbol references. One case where we emit these is in class
metadata, because the value witness table always points at the
value witness table for Builtin.NativeObject, defined in the
runtime.
Instead, fill in the value witness table at runtime when doing
singleton metadata initialization.
Another change that will come later is to force use of singleton
metadata initialization on Windows, even if the class is otherwise
completely fixed.
IRGen always just emits a simple implementation that immediately
calls swift_relocateClassMetadata(); so allow the function to be
null in this case to save on code size.
When a (file)private entity occurs inside a generic context, we still need
information about the genericity of the enclosing context to demangle
to metadata. Emit complete context descriptors for parents of anonymous
contexts.
Fixes rdar://problem/46109026.
The current representation of an associated conformance in a witness
tables (e.g., Iterator: IteratorProtocol within a witness table for
Sequence) is a function that the client calls.
Replace this with something more like what we do for associated types:
an associated conformance is either a pointer to the witness table (once
it is known) or a pointer to a mangled name that describes that
conformance. On first access, demangle the mangled name and replace the
entry with the resulting witness table. This will give us a more compact
representation of associated conformances, as well as always caching
them.
For now, the mangled name is a sham: it’s a mangled relative reference to
the existing witness table accessors, not a true mangled name. In time,
we’ll extend the support here to handle proper mangled names.
Part of rdar://problem/38038799.
TargetGenericParamRef is a specialized structure used to describe the
subject of a generic requirement, e.g., the “T.Assoc” in “T.Assoc: P”.
Replace it with a mangled name, for several reasons:
1) Mangled type names are also fairly concise, can often be shared, and
are a well-tested path
2) Mangled type names can express any type, which might be useful in the
future
3) This structure doesn’t accommodate specifically stating where the
conformances come from (to extract associated type witnesses). Neither
can mangled names, but we’d like to do that work in only one place.
This change exposed an existing bug where we improperly calculated the
generic parameter counts for extensions of nested generic types. Fix that
bug here (which broke an execution test).
Even if we have a constant value, we might be emitting a legacy layout
that can be updated in place by newer runtimes. In this case, clients
cannot assume the field offsets are constant, and the globals cannot
be constant either.
Part of <rdar://problem/17528739>.
Simplify calls to getAddrOfLLVMVariableOrGOTEquivalent() and
getAddrOfLLVMVariable() by moving the computation of the alignment and
default type into LinkEntity.
Co-authored-by: Joe Groff <jgroff@apple.com>
A few utility methods would bypass computing the TypeInfo for a tuple,
but they were only used in assertions or used in places where I can't
imagine tuples coming up often enough for it to matter.
Rather than rely on the metadata initialization function to compute and
fill in the superclass, use the mangled superclass name to construct the
superclass metadata.
The superclass descriptor reference in class context descriptors is only used
for metadata bound computations when the superclass is resilient. Only
include the superclass descriptor reference when the class has a resilient
superclass, using a trailing record. It’s a tiny space savings for
classes that don’t have resilient superclasses.
Encode default associated type witnesses using a sentinel prefix byte
(0xFF) in the mangled name rather than as a second low bit on the
reference. Align all of the mangled names used for type references to
2 bytes (so we get that low bit regardless) and separate the symbol
names for default associated type witnesses vs. other kinds of
metadata or reflection metadata.
Indicate whether a particular associated type witness is a default (whose
mangled name is relative to the protocol) vs. being supplied as part of the
conformance (whose mangled name is relative to the conforming type). The
use of pointer identity to distinguish these cases can fail due to the
coalescing of these linker symbols.
Rather than storing associated type metadata access functions in
witness tables, initially store a pointer to a mangled type name.
On first access, demangle that type name and replace the witness
table entry with the resulting type metadata.
This reduces the code size of protocol conformances, because we no
longer need to create associated type metadata access functions for
every associated type, and the mangled names are much smaller (and
sharable). The same code size improvements apply to defaulted
associated types for resilient protocols, although those are more
rare. Witness tables themselves are slightly smaller, because we
don’t need separate private entries in them to act as caches.
On the caller side, associated type metadata is always produced via
a call to swift_getAssociatedTypeWitness(), which handles the demangling
and caching behavior.
In all, this reduces the size of the standard library by ~70k. There
are additional code-size wins that are possible with follow-on work:
* We can stop emitting type metadata access functions for non-resilient
types that have constant metadata (like `Int`), because they’re only
currently used as associated type metadata access functions.
* We can stop emitting separate associated type reflection metadata,
because the reflection infrastructure can use these mangled names
directly.
If a class has a backward deployment layout:
- We still want to emit it using the FixedClassMetadataBuilder.
- We still want it to appear in the objc_classes section, and get an
OBJC_CLASS_$_ symbol if its @objc.
- However, we want to use the singleton metadata initialization pattern
in the metadata accessor.
- We want to emit metadata for all field types, and call the
swift_updateClassMetadata() function to initialize the class
metadata.
For now, this function just performs the idempotent initialization of
invoking a static method on the class, causing it to be realized with
the Objective-C runtime.
For a resilient protocol that has defaulted associated types, emit
default associated conformance witnesses that compute associated
conformances based on that default witness.
This completes the implementation of resilience protocols that
add new, defaulted associated types, rdar://problem/44167982.
For a resilient conformance, emit the associated conformance accessor
functions into the resilient witness table (keyed on the associated
conformance descriptor) rather than in the fixed part of the witness
table. This is another part of resilience for associated conformances,
and a step toward defaults for associated conformances.
Associated conformance descriptors are aliases that refer to associated
conformance requirements within a protocol descriptor’s list of
requirements. They will be used to provide protocol resilience against
the addition of new associated conformance requirements (which only makes
sense for newly-introduced, defaulted associated types).
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.
SILWitnessTable::Entry already contains a superset of what was supported
by SILDefaultWitnessTable::Entry, the latter of which only had “no entry”
and “method” states. Make SILDefaultWitnessTable::Entry an alias for
SILWitnessTable::Entry, and unify all of the parsing/printing/
(de)serialization logic.
Generic parameter references, which occur in generic requirement
metadata, were hardcoding associated type indices. Instead, use
relative references to associated type descriptors and perform the
index calculation at runtime.
Associated types can now be reordered resiliently (without relying on
sorting), which is the first main step toward rdar://problem/44167982.
Introduce an alias that refers one element prior to the start of a
protocol descriptor’s protocol requirements. This can be subtracted from
an associated type descriptor address to determine the offset of the
associated type accessor within a corresponding witness table. The code
generation for the latter is not yet implemented.
