Serialize derivative function configurations per module.
`@differentiable` and `@derivative` attributes register derivatives for
`AbstractFunctionDecl`s for a particular "derivative function configuration":
parameter indices and dervative generic signature.
To find `@derivative` functions registered in other Swift modules, derivative
function configurations must be serialized per module. When configurations for
a `AbstractFunctionDecl` are requested, all configurations from imported
modules are deserialized. This module serialization technique has precedent: it
is used for protocol conformances (e.g. extension declarations for a nominal
type) and Obj-C members for a class type.
Add `AbstractFunctionDecl::getDerivativeFunctionConfigurations` entry point
for accessing derivative function configurations.
In the differentiation transform: use
`AbstractFunctionDecl::getDerivativeFunctionConfigurations` to implement
`findMinimalDerivativeConfiguration` for canonical derivative function
configuration lookup, replacing `getMinimalASTDifferentiableAttr`.
Resolves TF-1100.
`@differentiable` attribute on protocol requirements and non-final class members
will produce derivative function entries in witness tables and vtables.
This patch adds an optional derivative function configuration
(`AutoDiffDerivativeFunctionIdentifier`) to `SILDeclRef` to represent these
derivative function entries.
Derivative function configurations consist of:
- A derivative function kind (JVP or VJP).
- Differentiability parameter indices.
Resolves TF-1209.
Enables TF-1212: upstream derivative function entries in witness tables/vtables.
In order to allow this, I've had to rework the syntax of substituted function types; what was previously spelled `<T> in () -> T for <X>` is now spelled `@substituted <T> () -> T for <X>`. I think this is a nice improvement for readability, but it did require me to churn a lot of test cases.
Distinguishing the substitutions has two chief advantages over the existing representation. First, the semantics seem quite a bit clearer at use points; the `implicit` bit was very subtle and not always obvious how to use. More importantly, it allows the expression of generic function types that must satisfy a particular generic abstraction pattern, which was otherwise impossible to express.
As an example of the latter, consider the following protocol conformance:
```
protocol P { func foo() }
struct A<T> : P { func foo() {} }
```
The lowered signature of `P.foo` is `<Self: P> (@in_guaranteed Self) -> ()`. Without this change, the lowered signature of `A.foo`'s witness would be `<T> (@in_guaranteed A<T>) -> ()`, which does not preserve information about the conformance substitution in any useful way. With this change, the lowered signature of this witness could be `<T> @substituted <Self: P> (@in_guaranteed Self) -> () for <A<T>>`, which nicely preserves the exact substitutions which relate the witness to the requirement.
When we adopt this, it will both obviate the need for the special witness-table conformance field in SILFunctionType and make it far simpler for the SILOptimizer to devirtualize witness methods. This patch does not actually take that step, however; it merely makes it possible to do so.
As another piece of unfinished business, while `SILFunctionType::substGenericArgs()` conceptually ought to simply set the given substitutions as the invocation substitutions, that would disturb a number of places that expect that method to produce an unsubstituted type. This patch only set invocation arguments when the generic type is a substituted type, which we currently never produce in type-lowering.
My plan is to start by producing substituted function types for accessors. Accessors are an important case because the coroutine continuation function is essentially an implicit component of the function type which the current substitution rules simply erase the intended abstraction of. They're also used in narrower ways that should exercise less of the optimizer.
The only non-trivial bit is making the DiagnosticEngine parameter to swift::performLLVM required. No callers were taking advantage of this parameter allowing NULL.
Rather than registering individual IRGen passes
when we want to execute them, store function
pointers to all the pass constructors on the
ASTContext. This will make it easier to requestify
the execution of pass pipelines.
As part of this, we have to change the type export rules to
prevent `@convention(c)` function types from being used in
exported interfaces if they aren't serializable. This is a
more conservative version of the original rule I had, which
was to import such function-pointer types as opaque pointers.
That rule would've completely prevented importing function-pointer
types defined in bridging headers and so simply doesn't work,
so we're left trying to catch the unsupportable cases
retroactively. This has the unfortunate consequence that we
can't necessarily serialize the internal state of the compiler,
but that was already true due to normal type uses of aggregate
types from bridging headers; if we can teach the compiler to
reliably serialize such types, we should be able to use the
same mechanisms for function types.
This PR doesn't flip the switch to use Clang function types
by default, so many of the clang-function-type-serialization
FIXMEs are still in place.
The `@noDerivative` attribute marks the non-differentiability parameters of a
`@differentiable` function type. All parameters except those marked with
`@noDerivative` are differentiability parameters.
For example, `@differentiable (Float, @noDerivative Float) -> Float` is only
differentiable with respect to its first parameter.
The `@noDerivative` attribute is represented as a
`SILParameterDifferentiability` bit on `SILParameterInfo`.
Add round-trip serialization tests.
Resolves TF-872.
Motivation: `GenericSignatureImpl::getCanonicalSignature` crashes for
`GenericSignature` with underlying `nullptr`. This led to verbose workarounds
when computing `CanGenericSignature` from `GenericSignature`.
Solution: `GenericSignature::getCanonicalSignature` is a wrapper around
`GenericSignatureImpl::getCanonicalSignature` that returns the canonical
signature, or `nullptr` if the underlying pointer is `nullptr`.
Rewrite all verbose workarounds using `GenericSignature::getCanonicalSignature`.
We recently added some checking to ensure that a method override's
generic signature does not have any generic requirements not
satisfied by the base method.
Loosening requirements in the other direction was allowed, because
it means the derived method can be called on potentially more types
than the base method.
However, if the generic signatures don't match, a thunk must be
emitted. While we correctly determined whether a thunk should be
emitted, the thunk had the wrong generic signature, and therefore
the wrong calling convention, which would cause crashes at runtime.
Fixes <rdar://problem/57429775>.
This reverts commit e805fe486e, which reverted
the change earlier. The problem was caused due to a simultaneous change to some
code by the PR with parsing and printing for Clang function types (#28737)
and the PR which introduced Located<T> (#28643).
This commit also includes a small change to make sure the intersecting region
is fixed: the change is limited to using the fields of Located<T> in the
`tryParseClangType` lambda.
