Although I don't plan to bring over new assertions wholesale
into the current qualification branch, it's entirely possible
that various minor changes in main will use the new assertions;
having this basic support in the release branch will simplify that.
(This is why I'm adding the includes as a separate pass from
rewriting the individual assertions)
[serialized_for_package] if Package CMO is enabled. The latter kind
allows a function to be serialized even if it contains loadable types,
if Package CMO is enabled. Renamed IsSerialized_t as SerializedKind_t.
The tri-state serialization kind requires validating inlinability
depending on the serialization kinds of callee vs caller; e.g. if the
callee is [serialized_for_package], the caller must be _not_ [serialized].
Renamed `hasValidLinkageForFragileInline` as `canBeInlinedIntoCaller`
that takes in its caller's SerializedKind as an argument. Another argument
`assumeFragileCaller` is also added to ensure that the calle sites of
this function know the caller is serialized unless it's called for SIL
inlining optimization passes.
The [serialized_for_package] attribute is allowed for SIL function, global var,
v-table, and witness-table.
Resolves rdar://128406520
We maintained a counter of the number of pending local archetypes
that had not yet been defined. However, if an instruction that
references a pending local archetype was deleted before the
local archetype was defined, the counter would never decrement.
Before reading the counter value, garbage collect any inserted
placeholders that have no uses. These correspond to pending
local archetypes that are no longer in use and will never be
defined.
LLVM is presumably moving towards `std::string_view` -
`StringRef::startswith` is deprecated on tip. `SmallString::startswith`
was just renamed there (maybe with some small deprecation inbetween, but
if so, we've missed it).
The `SmallString::startswith` references were moved to
`.str().starts_with()`, rather than adding the `starts_with` on
`stable/20230725` as we only had a few of them. Open to switching that
over if anyone feels strongly though.
It's not thread safe and can cause false alarms in case multiple modules exist in different threads. E.g. when building swiftmodules from interfaces.
The leaking check is not important anymore because the builder APIs enforce that instructions are not leaking.
I.e. it's not possible to create an instruction without inserting it into a basic block. Also, it's not possible to remove an instruction from a block without deleting it.
rdar://122169263
Ad-hoc requirements are now obsolete by making `remoteCall`,
`record{Argument, ReturnType}`, `decodeNextArgument` protocols
requirements and injecting witness tables for `SerializationRequirement`
conformances during IRGen.
Decls with a package access level are currently set to public SIL
linkages. This limits the ability to have more fine-grained control
and optimize around resilience and serialization.
This PR introduces a separate SIL linkage and FormalLinkage for
package decls, pipes them down to IRGen, and updates linkage checks
at call sites to include package linkage.
Resolves rdar://121409846
First, "can have an absence of Copyable" is a rather confusing notion,
so the query is flipped to "can be Copyable". Next, it's more robust to
ask if a conformance exists for the TypeDecl to answer that question,
rather than trying to replicate what happens within that conformance
lookup.
Also renames `TypeDecl::isEscapable` to match.
I've renamed the method to `TypeDecl::isNoncopyable`, because the query
doesn't make sense for many other kinds of `ValueDecl`'s beyond the
`TypeDecl`'s. In fact, it looks like no one was relying on that anyway.
Thus, we now have a distinction where in Sema, you ask whether
a `Type` or `TypeDecl` is "Noncopyable". But within SIL, we still
preserve the notion of "move-only" since there is additionally the
move-only type wrapper for types that otherwise support copying.
Deserialization is calling AccessMarkerElimination repeatedly on the
same function.
The bug was introduced here:
commit 872bf40e17
Date: Mon Aug 13 10:24:20 2018
[sil-optimizer] Centralize how we send out serialization notifications.
Where the code that uniques the deserialization callbacks was simply
removed!
As a result, this pass was being invoked a number of times equal to
the number of functions in the module *multiplied* by the number of
functions being deserialized.
Fixes rdar://117141871 (Building spends most of its time in
AccessMarkerElimination)
- VTableSpecializer, a new pass that synthesizes a new vtable per each observed concrete type used
- Don't use full type metadata refs in embedded Swift
- Lazily emit specialized class metadata (LazySpecializedClassMetadata) in IRGen
- Don't emit regular class metadata for a class decl if it's generic (only emit the specialized metadata)
- Add a flag to the serialized module (IsEmbeddedSwiftModule)
- Check on import that the mode matches (don't allow importing non-embedded module in embedded mode and vice versa)
- Drop TBD support, it's not expected to work in embedded Swift for now
- Drop auto-linking backdeploy libraries, it's not expected to backdeploy embedded Swift for now
- Drop prespecializations, not expected to work in embedded Swift for now
- Use CMO to serialize everything when emitting an embedded Swift module
- Change SILLinker to deserialize/import everything when importing an embedded Swift module
- Add an IR test for importing modules
- Add a deserialization validation test
Moving the query implementation up to the AST library from SIL will allow
conveniences to be written on specific AST element classes. For instance, this
will allow `EnumDecl` to expose a convenience that enumerates element decls
that are available during lowering.
Also, improve naming and documentation for these queries.
Reformatting everything now that we have `llvm` namespaces. I've
separated this from the main commit to help manage merge-conflicts and
for making it a bit easier to read the mega-patch.
This is phase-1 of switching from llvm::Optional to std::optional in the
next rebranch. llvm::Optional was removed from upstream LLVM, so we need
to migrate off rather soon. On Darwin, std::optional, and llvm::Optional
have the same layout, so we don't need to be as concerned about ABI
beyond the name mangling. `llvm::Optional` is only returned from one
function in
```
getStandardTypeSubst(StringRef TypeName,
bool allowConcurrencyManglings);
```
It's the return value, so it should not impact the mangling of the
function, and the layout is the same as `std::optional`, so it should be
mostly okay. This function doesn't appear to have users, and the ABI was
already broken 2 years ago for concurrency and no one seemed to notice
so this should be "okay".
I'm doing the migration incrementally so that folks working on main can
cherry-pick back to the release/5.9 branch. Once 5.9 is done and locked
away, then we can go through and finish the replacement. Since `None`
and `Optional` show up in contexts where they are not `llvm::None` and
`llvm::Optional`, I'm preparing the work now by going through and
removing the namespace unwrapping and making the `llvm` namespace
explicit. This should make it fairly mechanical to go through and
replace llvm::Optional with std::optional, and llvm::None with
std::nullopt. It's also a change that can be brought onto the
release/5.9 with minimal impact. This should be an NFC change.
When `-unavailable-decl-optimization=complete` is specified obsolete decls
should be preserved because their symbols are still ABI since they are
available to use when targeting deployment targets earlier than the obsoletion
version.
Resolves rdar://110268649
When `-unavailable-decl-optimization=complete` is specified, exclude
unavailable enum cases from the runtime layout of enums with payloads. Without
this, the type metadata for unavailable types may be referenced by enum cases
with unavailable payloads and cause linker failures.
Resolves rdar://107483852
When `-unavailable-decl-optimization=stub` is specified, insert a call to
`_diagnoseUnavailableCodeReached()` at the beginning of the function to cause
it to trap if executed at run time.
Part of rdar://107388493
I've also fixed this so that it should work on instructions that
define multiple values. Someday we'll change all the open_existential
instructions to produce different values for the type dependency and
the value result; today is not that day, though.
Instead of setting the parent pointer to null, set the `lastInitializedBitfieldID` to -1.
This allows to keep the parent block information, even when an instruction is removed from it's list.
The definition of `Lowering::usesObjCAllocator()` was previously implemented in SILGen, which does not match the library membership of the declaration. The implementation of `SILSymbolVisitor` in the SIL library uses this and since SIL is a dependency of SILGen instead of vice-versa this resulted in a linker error.
Specifically, we get an additional table like thing called sil_moveonlydeinit. It looks as follows:
sil_moveonlydeinit TYPE {
@FUNC_NAME
}
It always has a single entry.
Snapshots are copies of a function at a given point in time.
Currently it's only used for running passes repeatedly for performance profiling.
In future it can be used for caching when doing lazy evaluation in the pipeline.
As we do with field indices for struct instructions.
This avoids quadratic behavior in case of enums with lots of cases.
Also: cache field and enum case indices in the SILModule.
The main point of this change is to make sure that a shared function always has a body: both, in the optimizer pipeline and in the swiftmodule file.
This is important because the compiler always needs to emit code for a shared function. Shared functions cannot be referenced from outside the module.
In several corner cases we missed to maintain this invariant which resulted in unresolved-symbol linker errors.
As side-effect of this change we can drop the shared_external SIL linkage and the IsSerializable flag, which simplifies the serialization and linkage concept.
