We would run into an assertion during building the partial apply forwarding
thunk because we passed down a parameter convention of an empty argument
although we ignore it for the purpose of adding to the closure context.
However, the code path for the non-allocated closure context (0 or one capture)
would assert that there is only one convention.
Reconcile this by not adding the empty parameter to the convention array.
rdar://31487947
This was an unnecessary complication and didn't make a lot of
logical sense, because we can recover the witness table from
substitutions when we call a @convention(witness_method) anyway.
Also, to fix materializeForSet for generic subscripts, I want the
materializeForSet *callback* of a protocol witness to have
@convention(witness_method), which requires representing such
functions as a single function pointer in IRGen.
The hasPolymorphicParameters() method always returns true for a
SILFunctionType with witness_method calling convention, because
we have to pass the type metadata for 'Self'.
However, it is possible for a witness_method to have no generic
signature, if we're directly calling a concrete witness method
thunk, as can happen after devirtualization.
In this case, we would crash if the witness_method was partially
applied. Handle this case.
This fixes a regression from the recent cleanup to have IRGen use
SubstitutionMap: <https://github.com/apple/swift/pull/7211>
Use the generic type lowering algorithm described in
"docs/CallingConvention.rst#physical-lowering" to map from IRGen's explosion
type to the type expected by the ABI.
Change IRGen to use the swift calling convention (swiftcc) for native swift
functions.
Use the 'swiftself' attribute on self parameters and for closures contexts.
Use the 'swifterror' parameter for swift error parameters.
Change functions in the runtime that are called as native swift functions to use
the swift calling convention.
rdar://19978563
SubstitutionList is going to be a more compact representation of
a SubstitutionMap, suitable for inline allocation inside another
object.
For now, it's just a typedef for ArrayRef<Substitution>.
Separate formal lowered types from SIL types.
The SIL type of an argument will depend on the SIL module's conventions.
The module conventions are determined by the SIL stage and LangOpts.
Almost NFC, but specialized manglings are broken incidentally as a result of
fixes to the way passes handle book-keeping of aruments. The mangler is fixed in
the subsequent commit.
Otherwise, NFC is intended, but quite possible do to rewriting the logic in many
places.
Fixed for the difference of Cygwin with other Windows variants (MSVC,
Itanium, MinGW).
- The platform name is renamed to "cygwin" from "windows" which is used
for searching the standard libraries.
- The consideration for DLL storage class (DllExport/DllImport) is not
required for Cygwin and MinGW. There is no problem when linking in
these environment.
- Cygwin should use large memory model as default.(This may be changed
if someone ports to 32bit)
- Cygwin and MinGW should use the autolink feature in the sameway of
Linux due to the linker's limit.
This is dead code and can be re-added if it is needed. Right now though there
really isnt a ValueOwnershipKind that corresponds to deallocating and I do not
want to add a new ValueOwnershipKind for dead code.
TSan does not observe the guaranteed syncronization between the ref
count drop to zero and object destruction. This can lead to false positive
reports.
This patch adds an attribute to deinitializers to ignore memory accesses
at run time. It also moves the logic to add sanitizer attributes from
IRGenFunction to IRGenSILFunction, which means that the automatically
generated code such as _Block_release handler will not be instrumented
and the accesses made in them will be invisible to TSan.
Solves a problem similar to what's addressed in clang commit:
https://reviews.llvm.org/D25857
Allocate buffers for local generic/resilient values on the stack. alloc_stack
instructions in the entry block are translated using a dynamic alloca
instruction with variables size. All other alloc_stack instructions in addition
use llvm's stacksave/restore instrinsics to reset the stack (they could be
executed multiple times and with varying sizes).
Changes:
* Terminate all namespaces with the correct closing comment.
* Make sure argument names in comments match the corresponding parameter name.
* Remove redundant get() calls on smart pointers.
* Prefer using "override" or "final" instead of "virtual". Remove "virtual" where appropriate.
We don't want the machine calling conventions for closure invocation functions to necessarily be tied to the convention for normal thin functions or methods. NFC yet; for now, 'closure' follows the same behavior as the 'method' convention, but as part of partial_apply simplification it will be a requirement that partial_apply takes a @convention(closure) function and a box and produces a @convention(thick) function from them.
We need to splice self inbetween the polymorphic arguments and the error
argument but before the self metadata and witness table arguments.
rdar://28302820
In the case we are forwarding from partial apply that captures a concrete ref
counted value to a generic function we need to use the swift retainable type
from the captured value rather than the type of the generic function we are
calling. The type of the generic argument is opaque and we can't allocate
storage for an opaque type.
rdar://24306823
Add initial support for modelling DLL Storage semantics for global values. This
is needed to support the indirect addressing mechanism used on Windows.
It is not valid LLVM IR to have a function call without a location to an
inlinable function inside a function with debug info — this makes it impossible
to construct inline information.
This patch adds an assertion and fixes up several places across IRGen where
such a situation could happen.
rdar://problem/26955467
@convention(witness_method) values were changed to carry a pointer to their source witness table, but the type info wasn't changed to match. Fixing this fixes rdar://problem/26268544.
When emitting capture descriptors for functions with a smaller number of parameters
than SIL parameters, the compiler can crash indexing into the heap layout's element
types, because the capture index underflows to UINT_MAX.
rdar://problem/26404583
- Fix caller/callee confusion, and use the right SIL function type
for obtaining the generic signature.
- Correctly interpret the NecessaryBindings structure and the
substitutions therein.
- Fix alignment for capture and builtin type descriptors
- Put capture descriptor typerefs in the correct section
Add new SIL-level tests to precisely trigger various scenarios.
Now we can discern the types of values in heap boxes at runtime!
Closure reference captures are a common way of creating reference
cycles, so this provides some basic infrastructure for detecting those
someday.
A closure capture descriptor has the following:
- The number of captures.
- The number of sources of metadata reachable from the closure.
This is important for substituting generics at runtime since we
can't know precisely what will get captured until we observe a
closure.
- The number of types in the NecessaryBindings structure.
This is a holding tank in a closure for sources of metadata that
can't be gotten from the captured values themselves.
- The metadata source map, a list of pairs, for each
source of metadata for every generic argument needed to perform
substitution at runtime.
Key: The typeref for the generic parameter visible from the closure
in the Swift source.
Value: The metadata source, which describes how to crawl the heap from
the closure to get to the metadata for that generic argument.
- A list of typerefs for the captured values themselves.
Follow-up: IRGen tests for various capture scenarios, which will include
MetadataSource encoding tests.
rdar://problem/24989531
Properly lower reference counting SIL instructions with nonatomic attribute as invocations of corresponding non-atomic reference counting runtime functions.
It is a hint to the optimizer that the code, where this builtin is called, is on the fast path.
Specifically, the inliner takes it into account and increases the assumed benefit for code where the builtin is located.
Compared to the fastPath/slowPath builtins, this builtin can be placed into plain linear code and doesn't need to be used in conditions.
Compared to the @inline(__always) attribute, this builtin has also an effect on the caller function. Let's assume
foo() calls bar() contains onFastPath
and both foo and bar are small functions. Then if bar gets inlined into foo, the builtin also increases the chances that foo gets inlined.
This would not be the case if @inline(__always) is used just for bar.
Fix a crash in emitBuiltinCall() which occurs because we drop function
linkage information when creating SILCoverageMaps.
This re-applies 45c7e4e86 with the MachO-specific checks in the test
case removed.
