Files
swift-mirror/stdlib/public/Concurrency/Task.cpp
John McCall 1177cde4e3 Use current public Dispatch API to schedule global work.
We expect to iterate on this quite a bit, both publicly
and internally, but this is a fine starting-point.

I've renamed runAsync to runAsyncAndBlock to underline
very clearly what it does and why it's not long for this
world.  I've also had to give it a radically different
implementation in an effort to make it continue to work
given an actor implementation that is no longer just
running all work synchronously.

The major remaining bit of actor-scheduling work is to
make swift_task_enqueue actually do something sensible
based on the executor it's been given; currently it's
expecting a flag that IRGen simply doesn't know to set.
2020-12-10 19:18:53 -05:00

529 lines
17 KiB
C++

//===--- Task.cpp - Task object and management ----------------------------===//
//
// This source file is part of the Swift.org open source project
//
// Copyright (c) 2014 - 2020 Apple Inc. and the Swift project authors
// Licensed under Apache License v2.0 with Runtime Library Exception
//
// See https://swift.org/LICENSE.txt for license information
// See https://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
//
//===----------------------------------------------------------------------===//
//
// Object management routines for asynchronous task objects.
//
//===----------------------------------------------------------------------===//
#include "swift/Runtime/Concurrency.h"
#include "swift/ABI/Task.h"
#include "swift/ABI/Metadata.h"
#include "swift/Runtime/Mutex.h"
#include "swift/Runtime/HeapObject.h"
#include "TaskPrivate.h"
#include "AsyncCall.h"
#if defined(__APPLE__)
// TODO: We shouldn't need this
#include <dispatch/dispatch.h>
#endif
using namespace swift;
using FutureFragment = AsyncTask::FutureFragment;
void FutureFragment::destroy() {
auto queueHead = waitQueue.load(std::memory_order_acquire);
switch (queueHead.getStatus()) {
case Status::Executing:
assert(false && "destroying a task that never completed");
case Status::Success:
resultType->vw_destroy(getStoragePtr());
break;
case Status::Error:
swift_unknownObjectRelease(reinterpret_cast<OpaqueValue *>(getError()));
break;
}
}
FutureFragment::Status AsyncTask::waitFuture(AsyncTask *waitingTask) {
using Status = FutureFragment::Status;
using WaitQueueItem = FutureFragment::WaitQueueItem;
assert(isFuture());
auto fragment = futureFragment();
auto queueHead = fragment->waitQueue.load(std::memory_order_acquire);
while (true) {
switch (queueHead.getStatus()) {
case Status::Error:
case Status::Success:
// The task is done; we don't need to wait.
return queueHead.getStatus();
case Status::Executing:
// Task is now complete. We'll need to add ourselves to the queue.
break;
}
// Put the waiting task at the beginning of the wait queue.
waitingTask->getNextWaitingTask() = queueHead.getTask();
auto newQueueHead = WaitQueueItem::get(Status::Executing, waitingTask);
if (fragment->waitQueue.compare_exchange_weak(
queueHead, newQueueHead, std::memory_order_release,
std::memory_order_acquire)) {
// Escalate the priority of this task based on the priority
// of the waiting task.
swift_task_escalate(this, waitingTask->Flags.getPriority());
return FutureFragment::Status::Executing;
}
}
}
namespace {
/// An asynchronous context within a task that describes a general "Future".
/// task.
///
/// This type matches the ABI of a function `<T> () async throws -> T`, which
/// is the type used by `Task.runDetached` and `Task.group.add` to create
/// futures.
class TaskFutureWaitAsyncContext : public AsyncContext {
public:
// Error result is always present.
SwiftError *errorResult = nullptr;
// No indirect results.
TaskFutureWaitResult result;
// FIXME: Currently, this is always here, but it isn't technically
// necessary.
void* Self;
// Arguments.
AsyncTask *task;
using AsyncContext::AsyncContext;
};
}
/// Run the given task, privoding it with the result of the future.
static void runTaskWithFutureResult(
AsyncTask *waitingTask, ExecutorRef executor,
FutureFragment *futureFragment, bool hadErrorResult) {
auto waitingTaskContext =
static_cast<TaskFutureWaitAsyncContext *>(waitingTask->ResumeContext);
waitingTaskContext->result.hadErrorResult = hadErrorResult;
if (hadErrorResult) {
waitingTaskContext->result.storage =
reinterpret_cast<OpaqueValue *>(futureFragment->getError());
} else {
waitingTaskContext->result.storage = futureFragment->getStoragePtr();
}
// TODO: schedule this task on the executor rather than running it
// directly.
waitingTask->run(executor);
}
void AsyncTask::completeFuture(AsyncContext *context, ExecutorRef executor) {
using Status = FutureFragment::Status;
using WaitQueueItem = FutureFragment::WaitQueueItem;
assert(isFuture());
auto fragment = futureFragment();
// If an error was thrown, save it in the future fragment.
auto futureContext = static_cast<FutureAsyncContext *>(context);
bool hadErrorResult = false;
if (auto errorObject = futureContext->errorResult) {
fragment->getError() = errorObject;
hadErrorResult = true;
}
// Update the status to signal completion.
auto newQueueHead = WaitQueueItem::get(
hadErrorResult ? Status::Error : Status::Success,
nullptr
);
auto queueHead = fragment->waitQueue.exchange(
newQueueHead, std::memory_order_acquire);
assert(queueHead.getStatus() == Status::Executing);
// Schedule every waiting task on the executor.
auto waitingTask = queueHead.getTask();
while (waitingTask) {
// Find the next waiting task.
auto nextWaitingTask = waitingTask->getNextWaitingTask();
// Run the task.
runTaskWithFutureResult(waitingTask, executor, fragment, hadErrorResult);
// Move to the next task.
waitingTask = nextWaitingTask;
}
}
SWIFT_CC(swift)
static void destroyTask(SWIFT_CONTEXT HeapObject *obj) {
auto task = static_cast<AsyncTask*>(obj);
// For a future, destroy the result.
if (task->isFuture()) {
task->futureFragment()->destroy();
}
// The task execution itself should always hold a reference to it, so
// if we get here, we know the task has finished running, which means
// swift_task_complete should have been run, which will have torn down
// the task-local allocator. There's actually nothing else to clean up
// here.
free(task);
}
/// Heap metadata for an asynchronous task.
static FullMetadata<HeapMetadata> taskHeapMetadata = {
{
{
&destroyTask
},
{
/*value witness table*/ nullptr
}
},
{
MetadataKind::Task
}
};
/// The function that we put in the context of a simple task
/// to handle the final return.
SWIFT_CC(swift)
static void completeTask(AsyncTask *task, ExecutorRef executor,
AsyncContext *context) {
// Tear down the task-local allocator immediately; there's no need
// to wait for the object to be destroyed.
_swift_task_alloc_destroy(task);
// Complete the future.
if (task->isFuture()) {
task->completeFuture(context, executor);
}
// TODO: set something in the status?
// TODO: notify the parent somehow?
// TODO: remove this task from the child-task chain?
// Release the task, balancing the retain that a running task
// has on itself.
swift_release(task);
}
AsyncTaskAndContext
swift::swift_task_create(JobFlags flags, AsyncTask *parent,
const ThinNullaryAsyncSignature::FunctionPointer *function) {
return swift_task_create_f(flags, parent, function->Function.get(),
function->ExpectedContextSize);
}
AsyncTaskAndContext
swift::swift_task_create_f(JobFlags flags, AsyncTask *parent,
ThinNullaryAsyncSignature::FunctionType *function,
size_t initialContextSize) {
return swift_task_create_future_f(
flags, parent, nullptr, function, initialContextSize);
}
AsyncTaskAndContext swift::swift_task_create_future(
JobFlags flags, AsyncTask *parent, const Metadata *futureResultType,
const FutureAsyncSignature::FunctionPointer *function) {
return swift_task_create_future_f(
flags, parent, futureResultType, function->Function.get(),
function->ExpectedContextSize);
}
AsyncTaskAndContext swift::swift_task_create_future_f(
JobFlags flags, AsyncTask *parent, const Metadata *futureResultType,
FutureAsyncSignature::FunctionType *function, size_t initialContextSize) {
assert((futureResultType != nullptr) == flags.task_isFuture());
assert(!flags.task_isFuture() ||
initialContextSize >= sizeof(FutureAsyncContext));
assert((parent != nullptr) == flags.task_isChildTask());
// Figure out the size of the header.
size_t headerSize = sizeof(AsyncTask);
if (parent) headerSize += sizeof(AsyncTask::ChildFragment);
if (futureResultType) {
headerSize += FutureFragment::fragmentSize(futureResultType);
}
headerSize = llvm::alignTo(headerSize, llvm::Align(alignof(AsyncContext)));
// Allocate the initial context together with the job.
// This means that we never get rid of this allocation.
size_t amountToAllocate = headerSize + initialContextSize;
assert(amountToAllocate % MaximumAlignment == 0);
void *allocation = malloc(amountToAllocate);
AsyncContext *initialContext =
reinterpret_cast<AsyncContext*>(
reinterpret_cast<char*>(allocation) + headerSize);
// Initialize the task so that resuming it will run the given
// function on the initial context.
AsyncTask *task =
new(allocation) AsyncTask(&taskHeapMetadata, flags,
function, initialContext);
// Initialize the child fragment if applicable.
// TODO: propagate information from the parent?
if (parent) {
auto childFragment = task->childFragment();
new (childFragment) AsyncTask::ChildFragment(parent);
}
// Initialize the future fragment if applicable.
if (futureResultType) {
auto futureFragment = task->futureFragment();
new (futureFragment) FutureFragment(futureResultType);
// Set up the context for the future so there is no error, and a successful
// result will be written into the future fragment's storage.
auto futureContext = static_cast<FutureAsyncContext *>(initialContext);
futureContext->errorResult = nullptr;
futureContext->indirectResult = futureFragment->getStoragePtr();
}
// Configure the initial context.
//
// FIXME: if we store a null pointer here using the standard ABI for
// signed null pointers, then we'll have to authenticate context pointers
// as if they might be null, even though the only time they ever might
// be is the final hop. Store a signed null instead.
initialContext->Parent = nullptr;
initialContext->ResumeParent = &completeTask;
initialContext->ResumeParentExecutor = ExecutorRef::generic();
initialContext->Flags = AsyncContextKind::Ordinary;
initialContext->Flags.setShouldNotDeallocateInCallee(true);
// Initialize the task-local allocator.
// TODO: consider providing an initial pre-allocated first slab to the
// allocator.
_swift_task_alloc_initialize(task);
return {task, initialContext};
}
void swift::swift_task_future_wait(
AsyncTask *waitingTask, ExecutorRef executor,
AsyncContext *rawContext) {
// Suspend the waiting task.
waitingTask->ResumeTask = rawContext->ResumeParent;
waitingTask->ResumeContext = rawContext;
// Wait on the future.
auto context = static_cast<TaskFutureWaitAsyncContext *>(rawContext);
auto task = context->task;
assert(task->isFuture());
switch (task->waitFuture(waitingTask)) {
case FutureFragment::Status::Executing:
// The waiting task has been queued on the future.
return;
case FutureFragment::Status::Success:
// Run the task with a successful result.
// FIXME: Want to guarantee a tail call here
runTaskWithFutureResult(
waitingTask, executor, task->futureFragment(),
/*hadErrorResult=*/false);
return;
case FutureFragment::Status::Error:
// Run the task with an error result.
// FIXME: Want to guarantee a tail call here
runTaskWithFutureResult(
waitingTask, executor, task->futureFragment(),
/*hadErrorResult=*/true);
return;
}
}
namespace {
/// The header of a function context (closure captures) of
/// a thick async function with a non-null context.
struct ThickAsyncFunctionContext: HeapObject {
uint32_t ExpectedContextSize;
};
struct RunAndBlockSemaphore {
ConditionVariable Queue;
ConditionVariable::Mutex Lock;
bool Finished = false;
};
using RunAndBlockSignature =
AsyncSignature<void(HeapObject*), /*throws*/ false>;
struct RunAndBlockContext: AsyncContext {
const void *Function;
HeapObject *FunctionContext;
RunAndBlockSemaphore *Semaphore;
};
using RunAndBlockCalleeContext =
AsyncCalleeContext<RunAndBlockContext, RunAndBlockSignature>;
} // end anonymous namespace
/// Second half of the runAndBlock async function.
SWIFT_CC(swiftasync)
static void runAndBlock_finish(AsyncTask *task, ExecutorRef executor,
AsyncContext *_context) {
auto calleeContext = static_cast<RunAndBlockCalleeContext*>(_context);
auto context = popAsyncContext(task, calleeContext);
auto semaphore = context->Semaphore;
semaphore->Lock.withLockThenNotifyAll(semaphore->Queue, [&]{
semaphore->Finished = true;
});
return context->ResumeParent(task, executor, context);
}
/// First half of the runAndBlock async function.
SWIFT_CC(swiftasync)
static void runAndBlock_start(AsyncTask *task, ExecutorRef executor,
AsyncContext *_context) {
auto callerContext = static_cast<RunAndBlockContext*>(_context);
size_t calleeContextSize;
RunAndBlockSignature::FunctionType *function;
// If the function context is non-null, then the function pointer is
// an ordinary function pointer.
auto functionContext = callerContext->FunctionContext;
if (functionContext) {
function = reinterpret_cast<RunAndBlockSignature::FunctionType*>(
const_cast<void*>(callerContext->Function));
calleeContextSize =
static_cast<ThickAsyncFunctionContext*>(functionContext)
->ExpectedContextSize;
// Otherwise, the function pointer is an async function pointer.
} else {
auto fnPtr = reinterpret_cast<const RunAndBlockSignature::FunctionPointer*>(
callerContext->Function);
function = fnPtr->Function;
calleeContextSize = fnPtr->ExpectedContextSize;
}
auto calleeContext =
pushAsyncContext<RunAndBlockSignature>(task, executor, callerContext,
calleeContextSize,
&runAndBlock_finish,
functionContext);
return function(task, executor, calleeContext);
}
// TODO: Remove this hack.
void swift::swift_task_runAndBlockThread(const void *function,
HeapObject *functionContext) {
RunAndBlockSemaphore semaphore;
// Set up a task that runs the runAndBlock async function above.
auto pair = swift_task_create_f(JobFlags(JobKind::Task,
JobPriority::Default),
/*parent*/ nullptr,
&runAndBlock_start,
sizeof(RunAndBlockContext));
auto context = static_cast<RunAndBlockContext*>(pair.InitialContext);
context->Function = function;
context->FunctionContext = functionContext;
context->Semaphore = &semaphore;
// Enqueue the task.
swift_task_enqueueGlobal(pair.Task);
// Wait for the task to finish.
semaphore.Lock.withLockOrWait(semaphore.Queue, [&] {
return semaphore.Finished;
});
}
size_t swift::swift_task_getJobFlags(AsyncTask *task) {
return task->Flags.getOpaqueValue();
}
namespace {
/// Structure that gets filled in when a task is suspended by `withUnsafeContinuation`.
struct AsyncContinuationContext {
// These fields are unnecessary for resuming a continuation.
void *Unused1;
void *Unused2;
// Storage slot for the error result, if any.
SwiftError *ErrorResult;
// Pointer to where to store a normal result.
OpaqueValue *NormalResult;
// Executor on which to resume execution.
ExecutorRef ResumeExecutor;
};
static void resumeTaskAfterContinuation(AsyncTask *task,
AsyncContinuationContext *context) {
#if __APPLE__
// TODO: Enqueue the task on the specific executor in the continuation
// context.
//
// For now, just enqueue the task resumption on the global concurrent queue
// so that we're able to return back to the caller of resume.
dispatch_async(dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_DEFAULT, 0),
^{
task->run(context->ResumeExecutor);
});
#else
swift_unreachable("not implemented");
#endif
}
}
SWIFT_CC(swift)
void swift::swift_continuation_resume(/* +1 */ OpaqueValue *result,
void *continuation,
const Metadata *resumeType) {
auto task = reinterpret_cast<AsyncTask*>(continuation);
auto context = reinterpret_cast<AsyncContinuationContext*>(task->ResumeContext);
resumeType->vw_initializeWithTake(context->NormalResult, result);
resumeTaskAfterContinuation(task, context);
}
SWIFT_CC(swift)
void swift::swift_continuation_throwingResume(/* +1 */ OpaqueValue *result,
void *continuation,
const Metadata *resumeType) {
return swift_continuation_resume(result, continuation, resumeType);
}
SWIFT_CC(swift)
void swift::swift_continuation_throwingResumeWithError(/* +1 */ SwiftError *error,
void *continuation,
const Metadata *resumeType) {
auto task = reinterpret_cast<AsyncTask*>(continuation);
auto context = reinterpret_cast<AsyncContinuationContext*>(task->ResumeContext);
context->ErrorResult = error;
resumeTaskAfterContinuation(task, context);
}