I made this change by removing the SILOption and then doing a compile, fix loop. I
purposely did not move around the code to make the refactoring really easy to
see.
Separate out the semantic state for the ‘dynamic’ check (from the
presence of the attribute), and move all of the computation of the
‘dynamic’ bit into the request-evaluator.
In the process, this fixes a bug where implicitly-synthesized initializers
in subclasses of imported classes would not be implicitly made ‘final’.
More groundwork for protocols with superclass constraints.
In several places we need to distinguish between existential
types that have a superclass term (MyClass & Proto) and
existential types containing a protocol with a superclass
constraint.
This is similar to how I can write 'AnyObject & Proto', or
write 'Proto1 & Proto2' where Proto1 has an ': AnyObject'
in its inheritance clause.
Note that some of the usages will be revisited later as
I do more refactoring and testing. This is just a first pass.
The storage kind has been replaced with three separate "impl kinds",
one for each of the basic access kinds (read, write, and read/write).
This makes it far easier to mix-and-match implementations of different
accessors, as well as subtleties like implementing both a setter
and an independent read/write operation.
AccessStrategy has become a bit more explicit about how exactly the
access should be implemented. For example, the accessor-based kinds
now carry the exact accessor intended to be used. Also, I've shifted
responsibilities slightly between AccessStrategy and AccessSemantics
so that AccessSemantics::Ordinary can be used except in the sorts of
semantic-bypasses that accessor synthesis wants. This requires
knowing the correct DC of the access when computing the access strategy;
the upshot is that SILGenFunction now needs a DC.
Accessor synthesis has been reworked so that only the declarations are
built immediately; body synthesis can be safely delayed out of the main
decl-checking path. This caused a large number of ramifications,
especially for lazy properties, and greatly inflated the size of this
patch. That is... really regrettable. The impetus for changing this
was necessity: I needed to rework accessor synthesis to end its reliance
on distinctions like Stored vs. StoredWithTrivialAccessors, and those
fixes were exposing serious re-entrancy problems, and fixing that... well.
Breaking the fixes apart at this point would be a serious endeavor.
It is safe to omit the retain/release dance in the reabstraction thunk
because we know we have an aditional reference outstanding for the
is_escaping verification.
The problem with throwing an objc exception inside a noescape closure is
that we verify the reference count of the closure sentinel. The
reabstraction thunk would increase the reference count call the
implementation function that then throws skipping the decrement.
rdar://40857699
Replace two prominent uses of SubstitutionList, in ConcreteDeclRef and
Witness, with SubstitutionMap. Deal with the myriad places where we
now have substitution maps and need substitution lists (or vice versa)
caused by this change.
Overall, removes ~50 explicit uses of SubstitutionList (of ~400).
closure lifetimes.
SILGen will now unconditionally emit
%cvt = convert_escape_to_noescape [guaranteed] %op
instructions. The mandatory ClosureLifetimeFixup pass ensures that %op's
lifetime spans %cvt's uses.
The code in DefiniteInitialization that handled a subset of cases is
removed.
To mark when a user of it is known to escape the value. This happens
with materializeForSet arguments which are captured and used in the
write-back. This means we need to keep the context alive until after
the write-back.
Follow-up patches to fully replace the PostponedCleanup hack in SILGen
by a mandatory SIL transformation pass to guarantee the proper lifetime
will use this flag to be more conservative when extending the lifetime.
The problem:
%pa = partial_apply %f(%some_context)
%cvt = convert_escape_to_noescape [not_guaranteed] [escaped] %pa
%ptr = %materialize_for_set(..., %cvt)
... write_back
... // <-- %pa needs to be alive until after write_back
We want to treat arguments to ObjC override and protocol conformance thunks like "call results", since they might be called from ObjC code that doesn't fulfill its nullability promises in practice. Fixes SR-7240 | rdar://problem/38675815.
By using this API, we properly forward ownership when we perform this
conversion. Previously if we had a guaranteed bridged error argument, we would
wrap the bridged error argument in an class existential box and then destroy
that box, violating the +0 contract. Instead now we forward the ownership
correctly through the init_existential_ref.
Found while updating SILGen tests for +0 arguments.
rdar://34222540
Previously, we were not verifying these thunks later in the pipeline. This makes
it harder to track down verifier errors in such thunks. Instead, this PR just
moves the thunk verification to right after we generate the thunk body,
i.e. right after the bad code would have been emitted.
Found while debugging test/SILGen/objc_bridging_any.swift for +0-all-args.
rdar://34222540
- Emit a withoutActuallyEscapingClosure partial apply
This is to convert an @noescape closure to an escaping closure.
This needs to be done in preparation of @noescape closure contexts
becoming trivial.
- Insert escaping to noescape conversions
- Fix SILGen for @noescape
- Postpone closure cleanups to outside the argument scope
- Apply postponement recursively for closures passed to subscripts
- Only skip applying escapeness conversions for Swift thick functions
- Fix parameter convention for noescape closures in thunks
Part of:
SR-5441
rdar://36116691
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.
I don't think there's a way to test this right now because we don't
actually support SwiftValue and so on except under ObjCInterop, but
it's certainly more future-proof.
at AnyObject?.
Also, ensure that @convention(block) functions can be erased to
AnyObject, and teach SILGen how to do this with unchecked_ref_cast.
Fixes a source compatibility suite regression.
Also, fix some logic in the peephole that was incorrectly recognizing
T? -> U? -> AnyObject as a force instead of a request to create NSNull.
In both cases, we ought to still be able to do much better than the
unpeepholed code by just applying the peephole and then merging in NSNull
in the nil case, but in the short term we need to fix this bug.
Fixes rdar://35402853
This takes a series of ManagedValues and combines them together into 1 tuple. It
assumes that all non-trivial ManagedValues are all at +1 or all at +0. It leaves
the verification to the ownership verifier since this would immediately trigger
the ownership verifier when the instruction is created by the SILBuilder.
rdar://34222540
The bridging function that we are calling here takes in the value to be bridged
at +0. We were storing that value via a normal store. This looks like a double
consume since we were destroying the value later. Now we use a store_borrow.
rdar://29791263
Support for @noescape SILFunctionTypes.
These are the underlying SIL changes necessary to implement the new
closure capture ABI.
Note: This includes a change to function name mangling that
primarily affects reabstraction thunks.
The new ABI will allow stack allocation of non-escaping closures as a
simple optimization.
The new ABI, and the stack allocation optimization, also require
closure context to be @guaranteed. That will be implemented as the
next step.
Many SIL passes pattern match partial_apply sequences. These all
needed to be fixed to handle the convert_function that SILGen now
emits. The conversion is now needed whenever a function declaration,
which has an escaping type, is passed into a @NoEscape argument.
In addition to supporting new SIL patterns, some optimizations like
inlining and SIL combine are now stronger which could perturb some
benchmark results.
These underlying SIL changes should be merged now to avoid conflicting
with other work. Minor benchmark discrepancies can be investigated as part of
the stack-allocation work.
* Add a noescape attribute to SILFunctionType.
And set this attribute correctly when lowering formal function types to SILFunctionTypes based on @escaping.
This will allow stack allocation of closures, and unblock a related ABI change.
* Flip the polarity on @noescape on SILFunctionType and clarify that
we don't default it.
* Emit withoutActuallyEscaping using a convert_function instruction.
It might be better to use a specialized instruction here, but I'll leave that up to Andy.
Andy: And I'll leave that to Arnold who is implementing SIL support for guaranteed ownership of thick function types.
* Fix SILGen and SIL Parsing.
* Fix the LoadableByAddress pass.
* Fix ClosureSpecializer.
* Fix performance inliner constant propagation.
* Fix the PartialApplyCombiner.
* Adjust SILFunctionType for thunks.
* Add mangling for @noescape/@escaping.
* Fix test cases for @noescape attribute, mangling, convert_function, etc.
* Fix exclusivity test cases.
* Fix AccessEnforcement.
* Fix SILCombine of convert_function -> apply.
* Fix ObjC bridging thunks.
* Various MandatoryInlining fixes.
* Fix SILCombine optimizeApplyOfConvertFunction.
* Fix more test cases after merging (again).
* Fix ClosureSpecializer. Hande convert_function cloning.
Be conservative when combining convert_function. Most of our code doesn't know
how to deal with function type mismatches yet.
* Fix MandatoryInlining.
Be conservative with function conversion. The inliner does not yet know how to
cast arguments or convert between throwing forms.
* Fix PartialApplyCombiner.
This replaces the '[volatile]' flag. Now, class_method and
super_method are only used for vtable dispatch.
The witness_method instruction is still overloaded for use
with both ObjC protocol requirements and Swift protocol
requirements; the next step is to make it only mean the
latter, also using objc_method for ObjC protocol calls.
