The `differentiability_witness_function` instruction looks up a
differentiability witness function (JVP, VJP, or transpose) for a referenced
function via SIL differentiability witnesses.
Add round-trip parsing/serialization and IRGen tests.
Notes:
- Differentiability witnesses for linear functions require more support.
`differentiability_witness_function [transpose]` instructions do not yet
have IRGen.
- Nothing currently generates `differentiability_witness_function` instructions.
The differentiation transform does, but it hasn't been upstreamed yet.
Resolves TF-1141.
It's done by first retrieving all generic parameters from each solution,
filtering boundings into distrinct set and diagnosing any differences.
For example:
```swift
func foo<T>(_: T, _: T) {}
func bar(x: Int, y: Float) {
foo(x, y)
}
```
Static-linked libraries could add symbols to the final tbd file. We need
this flag to specify additional module names to collect symbols from.
rdar://59399684
In fast-completion, a function body can be replaced with another function
body parsed from a new buffer. In such cases, during typechecking the
expressions in the *new* function body, a source location range check in
UnqualifiedLookup didn't work well because they are not from the same
buffer.
This patch workaround it by skipping the source range checks and returns
'success' in such cases.
rdar://problem/58881999
This will be used for compiler-driven type erasure for dynamic
replacement of functions with an opaque return type. For now, just
parse the attribute and ignore it.
The logic here used to consist of a couple of ad-hoc checks,
followed by a general assumption that if something had already
been emitted, it could be referenced directly, whereas everything
else had to go through a GOT entry.
This is way too conservative. Instead, let's try to correctly
calculate what translation unit an entity is going to end up in.
* Update DAK_Available name to match @available keyword
Update expected errors for @available attribute tests
* Update available attribute keyword name in DeclAttrKeyword documentation comment
When a module extends a type from another module, serialize those symbols into
separated files dedicated to those extended modules. This makes it easier to
ingest and categorize those symbols under the extended module if desired.
rdar://58941718
Extracted implementation of SpecializedGenericStructMetadataBuilder into
SpecializedGenericNominalMetadataBuilderBase, a CRTP with a template
template argument for the CRTP superclass and a template argument for
the implementation. That new type is now subclassed by
SpecializedGenericStructMetadataBuilder. Additionally, this new type is
also subclassed by the newly added SpecializedGenericEnumMetadataBuilder
which is responsible for build the prespecialization of generic enum
metadata.
rdar://problem/56960887
Add `SILFunctionType::getAutoDiffTransposeFunctionType`.
It computes the transpose `SILFucntionType` for an original `SILFunctionType`,
given:
- Linearity parameter indices
- Transpose function generic signature (optional)
- Other auxiliary parameters
Add doc comments explaining typing rules, preconditions, and other details.
Add `isTranspose` flag to `autodiff::getConstrainedDerivativeGenericSignature`.
Partially resolves TF-1125.
Unblocks TF-1141: upstream `differentiability_witness_function` instruction.
Changes:
* Allow optimizing partial_apply capturing opened existential: we didn't do this originally because it was complicated to insert the required alloc/dealloc_stack instructions at the right places. Now we have the StackNesting utility, which makes this easier.
* Support indirect-in parameters. Not super important, but why not? It's also easy to do with the StackNesting utility.
* Share code between dead closure elimination and the apply(partial_apply) optimization. It's a bit of refactoring and allowed to eliminate some code which is not used anymore.
* Fix an ownership problem: We inserted copies of partial_apply arguments _after_ the partial_apply (which consumes the arguments).
* When replacing an apply(partial_apply) -> apply and the partial_apply becomes dead, avoid inserting copies of the arguments twice.
These changes don't have any immediate effect on our current benchmarks, but will allow eliminating curry thunks for existentials.
* [CSDiagnostics] Offer a fix-it to insert a return type when returning from a void function
* [CSDiagnostics] Make sure the function name is not empty
The function name will be empty in some cases, for example for property setters. In cases where the function name is empty, skip the note and fix-it.
* [Test] Update existing diagnostics
Remove the option to switch off nested types tables. In a world where
re-entrant direct lookup will cause deserialization to fail (or worse),
disabling these tables will only lead to further instability in the
compiler.
calls over arrays created from array literals. This enables optimizing
further the output of the OSLogOptimization pass, and results in
highly-compact and optimized IR for calls to the new os log API.
<rdar://58928427>
