Speculatively fixing this to rule out potential miscompiles.
The compiler needs to know if a reference is being materialized out of
thin air. The proper way to do that is with the Unmanaged API.
Under the hood, this forces the reference into an "unowned(unsafe)"
variable which the reference must be reloaded from. That tells the
compiler that it can't optimize some seemingly unrelated object which
the reference may happen to refer to at runtime.
/// Warning: Casting from an integer or a pointer type to a reference type
/// is undefined behavior. It may result in incorrect code in any future
/// compiler release. To convert a bit pattern to a reference type:
/// 1. convert the bit pattern to an UnsafeRawPointer.
/// 2. create an unmanaged reference using Unmanaged.fromOpaque()
/// 3. obtain a managed reference using Unmanaged.takeUnretainedValue()
/// The programmer must ensure that the resulting reference has already been
/// manually retained.
This code is exercised on pretty much every access to ManagedBuffer — let’s keep it as simple as possible.
The overflow case is not expected to ever occur here; this code is used to align offsets within objects, it isn’t expected to get inputs anywhere near Int.max.
Reduces the number of _ContiguousArrayStorage metadata.
In order to support constant time bridging we do need to set the correct
metadata when we bridge to Objective-C. This is so that the type check
succeeds when bridging back from Objective-C to reuse the storage
instance rather than bridging the elements.
To support dynamically setting the `_ContiguousArrayStorage` element
type i needed to add support for optimizing `alloc_ref_dynamic`
throughout the optimizer.
Possible future improvements:
* Use different metadata such that we can disambiguate native Swift
classes during destruction -- allowing native release rather then unknown
release usage.
* Optimize the newly added semantic function
getContiguousArrayStorageType
rdar://86171143
This abbreviation for "if and only if" is confusing to those not coming
from a background in formal mathematics, and is frequently reported as a
type by developers reading the documentation.
This commit also changes doc comments in internal and private members,
which don't become part of the public documentation, because omitting
"iff" everywhere makes it much easier to check for any later changes
that reintroduce it.
Adds two new IRGen-level builtins (one for allocating, the other for deallocating), a stdlib shim function for enhanced stack-promotion heuristics, and the proposed public stdlib functions.
And set this option in various presets for buildbots.
Don't enable the checks by default because when linking against the OS library (which does not support COW checking) it will result in unresolved symbol errors.
So far it was handled by an availability checks against 9999 (which was a hack), but this does not work anymore.
Note, all this is only relevant for assert builds of the stdlib.
rdar://83673798
Commit the platform definition and build script work necessary to
cross-compile for arm64_32.
arm64_32 is a variant of AARCH64 that supports an ILP32 architecture.
This is only relevant for assert builds of the library.
Fixes an undefined symbol error if an executable, compiled with an assert build of the stdlib, is back deployed on a platform which does not support lazy symbol binding.
rdar://problem/71201102
Use the new builtins for COW representation in Array, ContiguousArray and ArraySlice.
The basic idea is to strictly separate code which mutates an array buffer from code which reads from an array.
The concept is explained in more detail in docs/SIL.rst, section "Copy-on-Write Representation".
The main change is to use beginCOWMutation() instead of isUniquelyReferenced() and insert endCOWMutation() at the end of all mutating functions. Also, reading from the array buffer must be done differently, depending on if the buffer is in a mutable or immutable state.
All the required invariants are enforced by runtime checks - but only in an assert-build of the library: a bit in the buffer object side-table indicates if the buffer is mutable or not.
Along with the library changes, also two optimizations needed to be updated: COWArrayOpt and ObjectOutliner.
In an assert built of the library, store an extra boolean flag (isImmutable) in the object side-buffer table.
This flag can be set and get by the Array implementation to sanity check the immutability status of the buffer object.
Returns `true` if `T.Type` is known to refer to a concrete type. The
implementation allows for the optimizer to specialize this at -O and
eliminate conditional code.
Includes `Swift._isConcrete<T>(T.Type) -> Bool` wrapper function.
the builtin.globalStringTablePointer to the new OSLog overlay.
Modify the new OSLog implementation to use this SPI instead of
`withCString` to pass the (compiler-generated) format string to
the C os_log_impl ABI.
Move the OSLogOptimization pass before constant propagation in
the pass pipeline so that the SPI and the builtin it uses can be
folded to a string_literal instruction.
Update OSLogTests to work with the changes in the implementation.
LLVM r355981 changed various intrinsic functions, including expect,
to require immediate arguments. Swift's _branchHint function has an
expected value that is passed in as an argument, so that it cannot
use LLVM's expect intrinsic. The good news is that _branchHint is only
ever used with immediate arguments, so we can just move the intrinsic
into _fastPath and _slowPath and use those instead of _branchHint.
As was noted in the documentation, the _fastPath and _slowPath names are
confusing but we have passed the point where we can simply rename them.
We could add new names but would still need to keep the old ones around
for binary compatibility, and it is not clear that it is worth the
trouble. I have removed that note from the documentation.
Compilation of code sample snippets was broken due to using the same
identifier as the `type` function itself to store its result resulting
in:
```swift
func printGenericInfo<T>(_ value: T) {
let type = type(of: value)
print("'\(value)' of type '\(type)'")
}
// error: repl.swift:2:16: error: variable used within its own initial
// value
// let type = type(of: value)
// ^
```
Result:
- Snippets are more copy&paste friendly.
- Resolves https://bugs.swift.org/browse/SR-9915
This fixes the Windows platform, where the aligned allocation path is
not malloc-compatible. It won't have any observable difference on
Darwin or Linux, aside from manually allocated memory on Linux now
being consistently 16-byte aligned (heap objects will still be 8-byte
aligned on Linux).
It is unfortunate that we can't guarantee Swift-allocated memory via
Unsafe*Pointer is malloc compatible on Windows. It would have been
nice for that to be a cross platform guarantee since it's normal to
allocate in C and deallocate in Swift or vice-versa. Now we have to
tell developers to always use _aligned_malloc/_aligned_free when
transitioning between Swift/C if they expect their code to work on
Windows.
Even though this fix isn't required today on Darwin/Linux, it makes
good sense to guarantee that the allocation/deallocation paths are
consistent.
This is done by specifying a constant that stdlib can use to round up
alignment, _swift_MinAllocationAlignment. The runtime asserts that
this constant is greater than MALLOC_ALIGN_MASK for all platforms.
This way, manually allocated buffers will always use the aligned
allocation path. If users specify an alignment less than m
round up so users don't need
to pass the same alignment to deallocate the buffer). This constant
does not need to be ABI.
Alternatives are:
1. Require users of Unsafe*Pointer to specify the same alignment
during deallocation. This is obviously madness.
2. Introduce new runtime entry points:
swift_alignedAlloc/swift_alignedDealloc, introduce corresponding
new builtins, and have Unsafe*Pointer always call those. This would
make the runtime API a little more obvious but would introduce
complexity in other areas of the compiler and it doesn't have any
other significant benefit. Less than 16-byte alignment of manually
allocated buffers on Linux is a non-goal.
* [stdlib] Minor documentation revisions
* [docs] Convert 'nonoptional' to 'non-optional'
We're switching to 'non-optional' across the board, as the unhyphenated
form is too easy to read as 'no-noptional'.
Previously we had a single mask for all x86-64 targets which included both the top and bottom bits. This accommodated simulators, which use the top bit, while macOS uses the bottom bit, but reserved one bit more than necessary on each. This change breaks out x86-64 simulators from non-simulators and reserves only the one bit used on each.
rdar://problem/34805348 rdar://problem/29765919
The functions in LibcShims are used externally, some directly and some through @inlineable functions. These are changed to SWIFT_RUNTIME_STDLIB_SPI to better match their actual usage. Their names are also changed to add "_swift" to the front to match our naming conventions.
Three functions from SwiftObject.mm are changed to SPI and get a _swift prefix.
A few other support functions are also changed to SPI. They already had a prefix and look like they were meant to be SPI anyway. It was just hard to notice any mixup when they were #defined to the same thing.
rdar://problem/35863717
* Removing FIXME from methods also marked always/never
* Unavailable/deprecated things don't need inlining
* Trivial implementations
* Enum namespaces
* Unsafe performance of opaque/raw pointer
* Dump doesn't need to be fast
* Error paths shouldn't require inlining
* Consistency with surrounding code
* Lazy performance needs specialization
Now that bridging is enabled, unify the ObjC and non-ObjC paths. Expand
the comment with how to efficiently grab the object reference and when
it can be enabled.
