We are leaving this as an open part of the design space. In the mean time if
people need a +0 parameter, they can use __shared with sending.
rdar://129116182
TLDR: This makes it so that we always can parse sending/transferring but changes
the semantic language effects to be keyed on RegionBasedIsolation instead.
----
The key thing that makes this all work is that I changed all of the "special"
semantic changes originally triggered on *ArgsAndResults to now be triggered
based on RegionBasedIsolation being enabled. This makes a lot of sense since we
want these semantic changes specifically to be combined with the checkers that
RegionBasedIsolation turns on. As a result, even though this causes these two
features to always be enabled, we just parse it but we do not use it for
anything semantically.
rdar://128961672
We want to ensure that functions/methods themselves do not have sending mangled
into their names, but we do want sending mangled in non-top level positions. For
example: we do not want to mangle sending into a function like the following:
```swift
// We don't want to mangle this.
func test(_ x: sending NonSendableKlass) -> ()
```
But when it comes to actually storing functions into memory, we do want to
distinguish in between function values that use sending vs those that do not
since we do not want to allow for them to alias. Thus we want to mangle sending
into things like the following:
```swift
// We want to distinguish in between Array<(sending T) -> ()> and
// Array((T) -> ()>
let a = Array<(sending T) -> ()>
// We want to distinguish in between a global contianing (sending T) -> () and a
// global containing (T) -> ().
var global: (sending T) -> ()
```
This commit achieves that by making changes to the ASTMangler in getDeclType
which causes getDeclType to set a flag that says that we have not yet recursed
through the system and thus should suppress the printing of sendable. Once we
get further into the system and recurse, that flag is by default set to true, so
we get the old sending parameter without having to update large amounts of code.
rdar://127383107
Teach dependency scanner to report all the module canImport check result
to swift-frontend, so swift-frontend doesn't need to parse swiftmodule
or parse TBD file to determine the versions. This ensures dependency
scanner and swift-frontend will have the same resolution for all
canImport checks.
This also fixes two related issues:
* Previously, in order to get consistant results between scanner and
frontend, scanner will request building the module in canImport check
even it is not imported later. This slightly alters the definition of
the canImport to only succeed when the module can be found AND be
built. This also can affect the auto-link in such cases.
* For caching build, the location of the clang module is abstracted away
so swift-frontend cannot locate the TBD file to resolve
underlyingVersion.
rdar://128067152
This isn't fully implemented yet so it would crash eventually, so instead of
letting the compiler crash put up a proper diagnostic indicating this isn't
yet implemented. rdar://129034189
Updates swift-symbolgraph-extract to parse "-cxx-interoperability-mode"
flags and update the underlying compiler invocation. This fixes a bug
where were are unable to extract the symbol graph from swiftmodules with
transitive cxx modules because we parsed cxx headers as c headers.
rdar://128888548 (Add support for parsing cxx headers)
The reason why we are doing this is that otherwise, we have that the alloc_stack
formed for the result is disconnected and despite the fact that we merge it into
the actor region of the class method, we do not have that the alloc_stack
specifically is marked when we attempt to squelch Please.
This patch fixes that problem by detecting when an alloc_stack is being used as
a temporary for an out parameter and makes the alloc_stack initially isolated as
appropriate. It only does this in the specific cases where we can pattern match
it which in my limited testing has handled everything.
inlining, generic/closure specialization, and devirtualization optimization passes.
SILFunction::canBeInlinedIntoCaller now exlicitly requires a caller's SerializedKind_t arg.
isAnySerialized() is added as a convenience function that checks if [serialized] or [serialized_for_pkg].
Resolves rdar://128704752
When merging SILIsolationInfo for regions, we want to drop
nonisolated(unsafe). This is important since nonisolated(unsafe) should only
apply to the specific "value" that it belongs to, not the entire region.
This creates a problem since in a few places in the code base we initialize a
value (producing a disconnected value) and then initialize it by merging in an
actor isolation. This no longer work since we will then always have
nonisolated(unsafe) stripped, so no values would ever be considered to be
nonisolated(unsafe). After analyzing the use case, I realized that these were
just initialization patterns and in this commit, I added a specific
initialization operation called SILIsolationInfo::initializeTrackableValue and
eliminated those calls to SILIsolationInfo::mergeIsolationRegionInfo.
Since SILIsolationInfo no longer has any merge operation on it, I then
eliminated that code in this commit. This completes the behavior split that I
put into the type system in the last commit. Specifically, I defined a
composition type called SILDynamicMergedIsolationInfo. It represents a
SILIsolationInfo that has been merged... that is why I called it the
DynamicMergedIsolationInfo. It could probably use a better name = (.
This fixes one of the last weird test case that I wrote where we were not letting through valid
nonisolated(unsafe) code.
At the same time, I discovered an additional issue (which can be seen in the
TODOs in this commit), where we are being too conservative around a non-Sendable
class var field. I am going to fix that in the next commit.
rdar://128299305
DISCUSSION: The analysis itself is unable to emit errors. So we achieve the same
functionality by in such cases emitting a partition op that signals to our user
that when they process that partition op they should emit an "unknown pattern"
error at the partition op's instructions.
I have wanted this for a long time, but I never got around to it.
Specifically, I introduced a new composition type called
SILDynamicMergedIsolationInfo that just contains a
SILIsolationInfo. Importantly, whenever one merges a SILIsolationInfo with
another SILIsolationInfo, one gets back a SILDynamicMergedIsolationInfo.
The reason why I am doing this is that we drop nonisolated(unsafe) when merging
so I want to ensure that parts of the code that use merging (where the dropping
occurs) and normal SILIsolationInfo where we do not want to merge is
distinguished.
I made sure we match what we get without region isolation by turning off region
isolation in one of the test runs on the test for this.
There is one problem where for non-final classes with nonisolated(unsafe) var
fields, we currently do not properly squelch since I need to do more
infrastructure work. I am going to do that in the next commit.
rdar://128299305
The design change here is that instead of just initializing the regionInfo with
disconnected, we set it as .none and if we see .none, just return a newly
construct disconnected isolation region info when getIsolationRegionInfo() is
called.
This enables us to provide a setIsolationRegionInfo() helper for
RegionAnalysisValueMap::getTrackableValue that does not perform a merge. This is
important since for nonisolated(unsafe), we want to not have nonisolated(unsafe)
propagate through merging. So if we use merging to initialize the internal
regionInfo state of a SILIsolationInfo, we will never have a SILIsolationInfo
with that bit set since it will be lost in the merge. So we need some sort of
other assignment operator. Noting that we should only compute a value's
SILIsolationInfo once in RegionAnalysisValueMap before we cache it in the map,
it made sense to just represent it as an optional that way we can guarantee that
the regionInfo is only ever set exactly once by that routine.
The computation that determined whether an access to a `let` instance
property within a constructor should be an initialization conflated the
cases of "we don't have a base expression" and "the base expression is
not something that could be `self`", and incorrectly identified rvalue
bases as being "initializable". Make the interface properly separate
out these cases, so we don't turn an lvalue into an rvalue access.
Fixes rdar://128661833.
