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.
This enables optimizing / dead-stripping of witness methods across modules at
LTO time.
- Under -internalize-at-link, restrict visibility of wtables to linkage unit.
- Emit thunks for cross-module wcalls when WME is enabled.
- Use thunks for wcalls across modules when WME is enabled.
- Adjust TBDGen to account for witness method thunks when WME is enabled.
- Add an IR test to check that thunks are used when doing cross-module calls.
- Add an end-to-end test case for cross-module WME.
When we deploy to a minimum target that is not known to support extended
frame information the function prolog of Swift async functions will
contain a reference to swift_async_extendedFramePointerFlags. This
reference needs to be weak like any async symbols.
rdar://83412550
- Under -internalize-at-link, stop unconditionally marking all globals as used.
- Under -internalize-at-link, restrict visibility of vtables to linkage unit.
- Emit virtual method thunks for cross-module vcalls when VFE is enabled.
- Use thunks for vcalls across modules when VFE is enabled.
- Adjust TBDGen to account for virtual method thunks when VFE is enabled.
- Add an end-to-end test case for cross-module VFE.
Support for addresses with arbitrary alignment as opposed to their
element type's natural in-memory alignment.
Required for bytestream encoding/decoding without resorting to memcpy.
SIL instruction flag, documentation, printing, parsing, serialization,
and IRGen.
This is a new instruction that can be used by SILGen to perform a semantic move
in between two entities that are considered separate variables at the AST
level. I am going to use it to implement an experimental borrow checker.
This PR contains the following:
1. I define move_value, setup parsing, printing, serializing, deserializing,
cloning, and filled in all of the visitors as appropriate.
2. I added createMoveValue and emitMoveValueOperation SILBuilder
APIs. createMoveValue always creates a move and asserts is passed a trivial
type. emitMoveValueOperation in contrast, will short circuit if passed a
trivial value and just return the trivial value.
3. I added IRGen tests to show that we can push this through the entire system.
This is all just scaffolding for the instruction to live in SIL land and as of
this PR doesn't actually do anything.
Later stages use the name to disambiguate variables and this amgiguity
can lead to incorrect debug info that crashes LLVM. This also makes
the artificial variable names visible in textual SIL output.
rdar://82313550
This patch replace all in-memory objects of DebugValueAddrInst with
DebugValueInst + op_deref, and duplicates logics that handles
DebugValueAddrInst with the latter. All related check in the tests
have been updated as well.
Note that this patch neither remove the DebugValueAddrInst class nor
remove `debug_value_addr` syntax in the test inputs.
This new SIL di-expression represents the dereference on the SSA value.
Similar to DW_OP_deref in DWARF. It is also going to replace the
existing `debug_value_addr`. Namely, replacing the following
instruction:
```
debug_value_addr %a : $*T, name "my_var"
```
with this one:
```
debug_value %a : $*T, name "my_var", expr op_deref
```
Just for convenicence.
* Replace `llvm::isa_and_nonnull` with imported `isa_and_nonnull`
* Repalce some `EXPR && isa<T>(EXPR)` with `isa_and_nonnull<T>(EXPR)`
This commit is similar to cbb89c78d, but for explosions. emitShadowCopyIfNeeded()
has some extra code that was added when function arguments were moved out of the
async context to ensure that they are being lifetime-extended, and there is also
code that generates an incorrect load from the shadow copy. However,
emitShadowCopyIfNeeded is supposed return either an alloca or the value, and
IRGenDebugInfo knows to describe the value in the alloca already. The load is
counterproductive it's only valid until whatever register it ends up in is
clobbered, whereas the alloca is valid throughout the function.
This patch removes the load and updates the tests accordingly.
rdar://81565869
emitShadowCopyIfNeeded has some extra code that was added when function
arguments were moved out of the async context to ensure that they are being
lifetime-extended, and there is also code that generates an incorrect load from
the shadow copy. However, emitShadowCopyIfNeeded is supposed return either an
alloca or the value, and IRGenDebugInfo knows to describe the value in the
alloca already. The load is counterproductive it's only valid until whatever
register it ends up in is clobbered, whereas the alloca is valid throughout the
function.
This patch removes the load and updates the tests accordingly.
rdar://81805727
SROA and Mem2Reg now can leverage DIExpression -- op_fragment, more
specifically -- to generate correct debug info for optimized SIL. Some
important highlights:
- The new swift::salvageDebugInfo, similar to llvm::salvageDebugInfo,
tries to restore / transfer debug info from a deleted instruction.
Currently I only implemented this for store instruction whose
destination is an alloc_stack value.
- Since we now have source-variable-specific SIL location inside a
`debug_value` instruction (and its friends), this patch teaches
SILCloner and SILInliner to remap the debug scope there in addition
to debug scope of the instruction.
- DCE now does not remove `debug_value` instruction whose associating
with a function argument SSA value that is not used elsewhere. Since
that SSA value will not disappear so we should keep the debug info.
This is to deal with the fact that LLVM's coroutine can't handle instructions
with side-effects well that are inserted before the coro.begin.
rdar://81113950
Change the code generation patterns for `async let` bindings to use an ABI based on the following
functions:
- `swift_asyncLet_begin`, which starts an `async let` child task, but which additionally
now associates the `async let` with a caller-owned buffer to receive the result of the task.
This is intended to allow the task to emplace its result in caller-owned memory, allowing the
child task to be deallocated after completion without invalidating the result buffer.
- `swift_asyncLet_get[_throwing]`, which replaces `swift_asyncLet_wait[_throwing]`. Instead of
returning a copy of the value, this entry point concerns itself with populating the local buffer.
If the buffer hasn't been populated, then it awaits completion of the task and emplaces the
result in the buffer; otherwise, it simply returns. The caller can then read the result out of
its owned memory. These entry points are intended to be used before every read from the
`async let` binding, after which point the local buffer is guaranteed to contain an initialized
value.
- `swift_asyncLet_finish`, which replaces `swift_asyncLet_end`. Unlike `_end`, this variant
is async and will suspend the parent task after cancelling the child to ensure it finishes
before cleaning up. The local buffer will also be deinitialized if necessary. This is intended
to be used on exit from an `async let` scope, to handle cleaning up the local buffer if necessary
as well as cancelling, awaiting, and deallocating the child task.
- `swift_asyncLet_consume[_throwing]`, which combines `get` and `finish`. This will await completion
of the task, leaving the result value in the result buffer (or propagating the error, if it
throws), while destroying and deallocating the child task. This is intended as an optimization
for reading `async let` variables that are read exactly once by their parent task.
To avoid an epoch break with existing swiftinterfaces and ABI clients, the old builtins and entry
points are kept intact for now, but SILGen now only generates code using the new interface.
This new interface fixes several issues with the old async let codegen, including use-after-free
crashes if the `async let` was never awaited, and the inability to read from an `async let` variable
more than once.
rdar://77855176
This patch is primarily doing two things:
- Teach IRGen for some of the instructions to generate debug info based
on SILDebugVariable rather than AST node, which is not always
available when reading from SIL.
- Generate corresponding `llvm.dbg.*` intrinsics and `!DIExpression`
from the newly added SIL DIExpression.
Currently IRGen requires the AST node of a variable declaration to
generate debug location. However, this will fail if the input is SIL due
to the lack of AST reconstruction. Plus, it's unnecessary since we can
just use the `SILLocation` attached on `debug_value` (and its friends) SIL
instruction to generate the correct LLVM debug metadata.
Resolves SR-14868
This patch removes a heuristic to promote all debug intrinsics pointing into
allocas to llvm.dbg.declare() intrinsics and instead more accurate classifies
variables in async contexts by adding the missing cases alloc_box and
alloc_stack cases.
rdar://78977132
Changes the task, taskGroup, asyncLet wait funtion call ABIs.
To reduce code size pass the context parameters and resumption function
as arguments to the wait function.
This means that the suspend point does not need to store parent context
and resumption to the suspend point's context.
```
void swift_task_future_wait_throwing(
OpaqueValue * result,
SWIFT_ASYNC_CONTEXT AsyncContext *callerContext,
AsyncTask *task,
ThrowingTaskFutureWaitContinuationFunction *resume,
AsyncContext *callContext);
```
The runtime passes the caller context to the resume entry point saving
the load of the parent context in the resumption function.
This patch adds a `Metadata *` field to `GroupImpl`. The await entry
pointer no longer pass the metadata pointer and there is a path through
the runtime where the task future is no longer available.
Through various means, it is possible for a synchronous actor-isolated
function to escape to another concurrency domain and be called from
outside the actor. The problem existed previously, but has become far
easier to trigger now that `@escaping` closures and local functions
can be actor-isolated.
Introduce runtime detection of such data races, where a synchronous
actor-isolated function ends up being called from the wrong executor.
Do this by emitting an executor check in actor-isolated synchronous
functions, where we query the executor in thread-local storage and
ensure that it is what we expect. If it isn't, the runtime complains.
The runtime's complaints can be controlled with the environment
variable `SWIFT_UNEXPECTED_EXECUTOR_LOG_LEVEL`:
0 - disable checking
1 - warn when a data race is detected
2 - error and abort when a data race is detected
At an implementation level, this introduces a new concurrency runtime
entry point `_checkExpectedExecutor` that checks the given executor
(on which the function should always have been called) against the
executor on which is called (which is in thread-local storage). There
is a special carve-out here for `@MainActor` code, where we check
against the OS's notion of "main thread" as well, so that `@MainActor`
code can be called via (e.g.) the Dispatch library's
`DispatchQueue.main.async`.
The new SIL instruction `extract_executor` performs the lowering of an
actor down to its executor, which is implicit in the `hop_to_executor`
instruction. Extend the LowerHopToExecutor pass to perform said
lowering.
Previously, the partial application of a generic method at the receiver
would result in the emission of a "simple" partial apply where the
receiver was treated as the context and the original function was
treated as the function. That was problematic because such partial
applications must capture the polymorphic arguments supplied at
partial_apply time. Here, a full partial application forwarder is
emitted for this case.
rdar://76479222
As part of bringup, specifically in order to support storing the size of
the async context as the first entry in the thick context, the
optimization that allows the partial application of a single refcounted
object to avoid the allocation of a thick context was disabled.
Now that we have async function pointers for partial application
forwarders, that rationale is moot, so, here, the optimization is
restored.
rdar://76372871
