gecko-dev/widget/android/AndroidUiThread.cpp

369 строки
9.2 KiB
C++
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "base/message_loop.h"
#include "GeneratedJNIWrappers.h"
#include "mozilla/Atomics.h"
Bug 1382922 - Refactor event queue to allow multiple implementations (r=erahm) This patch refactors the nsThread event queue to clean it up and to make it easier to restructure. The fundamental concepts are as follows: Each nsThread will have a pointer to a refcounted SynchronizedEventQueue. A SynchronizedEQ takes care of doing the locking and condition variable work when posting and popping events. For the actual storage of events, it delegates to an AbstractEventQueue data structure. It keeps a UniquePtr to the AbstractEventQueue that it uses for storage. Both SynchronizedEQ and AbstractEventQueue are abstract classes. There is only one concrete implementation of SynchronizedEQ in this patch, which is called ThreadEventQueue. ThreadEventQueue uses locks and condition variables to post and pop events the same way nsThread does. It also encapsulates the functionality that DOM workers need to implement their special event loops (PushEventQueue and PopEventQueue). In later Quantum DOM work, I plan to have another SynchronizedEQ implementation for the main thread, called SchedulerEventQueue. It will have special code for the cooperatively scheduling threads in Quantum DOM. There are two concrete implementations of AbstractEventQueue in this patch: EventQueue and PrioritizedEventQueue. EventQueue replaces the old nsEventQueue. The other AbstractEventQueue implementation is PrioritizedEventQueue, which uses multiple queues for different event priorities. The final major piece here is ThreadEventTarget, which splits some of the code for posting events out of nsThread. Eventually, my plan is for multiple cooperatively scheduled nsThreads to be able to share a ThreadEventTarget. In this patch, though, each nsThread has its own ThreadEventTarget. The class's purpose is just to collect some related code together. One final note: I tried to avoid virtual dispatch overhead as much as possible. Calls to SynchronizedEQ methods do use virtual dispatch, since I plan to use different implementations for different threads with Quantum DOM. But all the calls to EventQueue methods should be non-virtual. Although the methods are declared virtual, all the classes used are final and the concrete classes involved should all be known through templatization. MozReview-Commit-ID: 9Evtr9oIJvx
2017-06-21 05:42:13 +03:00
#include "mozilla/EventQueue.h"
#include "mozilla/LinkedList.h"
#include "mozilla/Monitor.h"
#include "mozilla/Mutex.h"
#include "mozilla/RefPtr.h"
#include "mozilla/StaticPtr.h"
Bug 1382922 - Refactor event queue to allow multiple implementations (r=erahm) This patch refactors the nsThread event queue to clean it up and to make it easier to restructure. The fundamental concepts are as follows: Each nsThread will have a pointer to a refcounted SynchronizedEventQueue. A SynchronizedEQ takes care of doing the locking and condition variable work when posting and popping events. For the actual storage of events, it delegates to an AbstractEventQueue data structure. It keeps a UniquePtr to the AbstractEventQueue that it uses for storage. Both SynchronizedEQ and AbstractEventQueue are abstract classes. There is only one concrete implementation of SynchronizedEQ in this patch, which is called ThreadEventQueue. ThreadEventQueue uses locks and condition variables to post and pop events the same way nsThread does. It also encapsulates the functionality that DOM workers need to implement their special event loops (PushEventQueue and PopEventQueue). In later Quantum DOM work, I plan to have another SynchronizedEQ implementation for the main thread, called SchedulerEventQueue. It will have special code for the cooperatively scheduling threads in Quantum DOM. There are two concrete implementations of AbstractEventQueue in this patch: EventQueue and PrioritizedEventQueue. EventQueue replaces the old nsEventQueue. The other AbstractEventQueue implementation is PrioritizedEventQueue, which uses multiple queues for different event priorities. The final major piece here is ThreadEventTarget, which splits some of the code for posting events out of nsThread. Eventually, my plan is for multiple cooperatively scheduled nsThreads to be able to share a ThreadEventTarget. In this patch, though, each nsThread has its own ThreadEventTarget. The class's purpose is just to collect some related code together. One final note: I tried to avoid virtual dispatch overhead as much as possible. Calls to SynchronizedEQ methods do use virtual dispatch, since I plan to use different implementations for different threads with Quantum DOM. But all the calls to EventQueue methods should be non-virtual. Although the methods are declared virtual, all the classes used are final and the concrete classes involved should all be known through templatization. MozReview-Commit-ID: 9Evtr9oIJvx
2017-06-21 05:42:13 +03:00
#include "mozilla/ThreadEventQueue.h"
#include "mozilla/TimeStamp.h"
Bug 1382922 - Refactor event queue to allow multiple implementations (r=erahm) This patch refactors the nsThread event queue to clean it up and to make it easier to restructure. The fundamental concepts are as follows: Each nsThread will have a pointer to a refcounted SynchronizedEventQueue. A SynchronizedEQ takes care of doing the locking and condition variable work when posting and popping events. For the actual storage of events, it delegates to an AbstractEventQueue data structure. It keeps a UniquePtr to the AbstractEventQueue that it uses for storage. Both SynchronizedEQ and AbstractEventQueue are abstract classes. There is only one concrete implementation of SynchronizedEQ in this patch, which is called ThreadEventQueue. ThreadEventQueue uses locks and condition variables to post and pop events the same way nsThread does. It also encapsulates the functionality that DOM workers need to implement their special event loops (PushEventQueue and PopEventQueue). In later Quantum DOM work, I plan to have another SynchronizedEQ implementation for the main thread, called SchedulerEventQueue. It will have special code for the cooperatively scheduling threads in Quantum DOM. There are two concrete implementations of AbstractEventQueue in this patch: EventQueue and PrioritizedEventQueue. EventQueue replaces the old nsEventQueue. The other AbstractEventQueue implementation is PrioritizedEventQueue, which uses multiple queues for different event priorities. The final major piece here is ThreadEventTarget, which splits some of the code for posting events out of nsThread. Eventually, my plan is for multiple cooperatively scheduled nsThreads to be able to share a ThreadEventTarget. In this patch, though, each nsThread has its own ThreadEventTarget. The class's purpose is just to collect some related code together. One final note: I tried to avoid virtual dispatch overhead as much as possible. Calls to SynchronizedEQ methods do use virtual dispatch, since I plan to use different implementations for different threads with Quantum DOM. But all the calls to EventQueue methods should be non-virtual. Although the methods are declared virtual, all the classes used are final and the concrete classes involved should all be known through templatization. MozReview-Commit-ID: 9Evtr9oIJvx
2017-06-21 05:42:13 +03:00
#include "mozilla/UniquePtr.h"
#include "nsThread.h"
#include "nsThreadManager.h"
#include "nsThreadUtils.h"
using namespace mozilla;
namespace {
class AndroidUiThread;
class AndroidUiTask;
StaticAutoPtr<LinkedList<AndroidUiTask> > sTaskQueue;
StaticAutoPtr<mozilla::Mutex> sTaskQueueLock;
StaticRefPtr<AndroidUiThread> sThread;
static bool sThreadDestroyed;
static MessageLoop* sMessageLoop;
static Atomic<Monitor*> sMessageLoopAccessMonitor;
void EnqueueTask(already_AddRefed<nsIRunnable> aTask, int aDelayMs);
/*
* The AndroidUiThread is derived from nsThread so that nsIRunnable objects that get
* dispatched may be intercepted. Only nsIRunnable objects that need to be synchronously
* executed are passed into the nsThread to be queued. All other nsIRunnable object
* are immediately dispatched to the Android UI thread.
* AndroidUiThread is derived from nsThread instead of being an nsIEventTarget
* wrapper that contains an nsThread object because if nsIRunnable objects with a
* delay were dispatch directly to an nsThread object, such as obtained from
* nsThreadManager::GetCurrentThread(), the nsIRunnable could get stuck in the
* nsThread nsIRunnable queue. This is due to the fact that Android controls the
* event loop in the Android UI thread and has no knowledge of when the nsThread
* needs to be drained.
*/
class AndroidUiThread : public nsThread
{
public:
NS_INLINE_DECL_REFCOUNTING_INHERITED(AndroidUiThread, nsThread)
Bug 1382922 - Refactor event queue to allow multiple implementations (r=erahm) This patch refactors the nsThread event queue to clean it up and to make it easier to restructure. The fundamental concepts are as follows: Each nsThread will have a pointer to a refcounted SynchronizedEventQueue. A SynchronizedEQ takes care of doing the locking and condition variable work when posting and popping events. For the actual storage of events, it delegates to an AbstractEventQueue data structure. It keeps a UniquePtr to the AbstractEventQueue that it uses for storage. Both SynchronizedEQ and AbstractEventQueue are abstract classes. There is only one concrete implementation of SynchronizedEQ in this patch, which is called ThreadEventQueue. ThreadEventQueue uses locks and condition variables to post and pop events the same way nsThread does. It also encapsulates the functionality that DOM workers need to implement their special event loops (PushEventQueue and PopEventQueue). In later Quantum DOM work, I plan to have another SynchronizedEQ implementation for the main thread, called SchedulerEventQueue. It will have special code for the cooperatively scheduling threads in Quantum DOM. There are two concrete implementations of AbstractEventQueue in this patch: EventQueue and PrioritizedEventQueue. EventQueue replaces the old nsEventQueue. The other AbstractEventQueue implementation is PrioritizedEventQueue, which uses multiple queues for different event priorities. The final major piece here is ThreadEventTarget, which splits some of the code for posting events out of nsThread. Eventually, my plan is for multiple cooperatively scheduled nsThreads to be able to share a ThreadEventTarget. In this patch, though, each nsThread has its own ThreadEventTarget. The class's purpose is just to collect some related code together. One final note: I tried to avoid virtual dispatch overhead as much as possible. Calls to SynchronizedEQ methods do use virtual dispatch, since I plan to use different implementations for different threads with Quantum DOM. But all the calls to EventQueue methods should be non-virtual. Although the methods are declared virtual, all the classes used are final and the concrete classes involved should all be known through templatization. MozReview-Commit-ID: 9Evtr9oIJvx
2017-06-21 05:42:13 +03:00
AndroidUiThread()
: nsThread(MakeNotNull<ThreadEventQueue<mozilla::EventQueue>*>(
MakeUnique<mozilla::EventQueue>()),
nsThread::NOT_MAIN_THREAD,
0)
{}
nsresult Dispatch(already_AddRefed<nsIRunnable> aEvent, uint32_t aFlags) override;
nsresult DelayedDispatch(already_AddRefed<nsIRunnable> aEvent, uint32_t aDelayMs) override;
private:
~AndroidUiThread()
{}
};
NS_IMETHODIMP
AndroidUiThread::Dispatch(already_AddRefed<nsIRunnable> aEvent, uint32_t aFlags)
{
if (aFlags & NS_DISPATCH_SYNC) {
return nsThread::Dispatch(std::move(aEvent), aFlags);
} else {
EnqueueTask(std::move(aEvent), 0);
return NS_OK;
}
}
NS_IMETHODIMP
AndroidUiThread::DelayedDispatch(already_AddRefed<nsIRunnable> aEvent, uint32_t aDelayMs)
{
EnqueueTask(std::move(aEvent), aDelayMs);
return NS_OK;
}
static void
PumpEvents() {
NS_ProcessPendingEvents(sThread.get());
}
class ThreadObserver : public nsIThreadObserver
{
public:
NS_DECL_THREADSAFE_ISUPPORTS
NS_DECL_NSITHREADOBSERVER
ThreadObserver()
{}
private:
virtual ~ThreadObserver()
{}
};
NS_IMPL_ISUPPORTS(ThreadObserver, nsIThreadObserver)
NS_IMETHODIMP
ThreadObserver::OnDispatchedEvent()
{
EnqueueTask(NS_NewRunnableFunction("PumpEvents", &PumpEvents), 0);
return NS_OK;
}
NS_IMETHODIMP
ThreadObserver::OnProcessNextEvent(nsIThreadInternal *thread, bool mayWait)
{
return NS_OK;
}
NS_IMETHODIMP
ThreadObserver::AfterProcessNextEvent(nsIThreadInternal *thread, bool eventWasProcessed)
{
return NS_OK;
}
class AndroidUiTask : public LinkedListElement<AndroidUiTask> {
using TimeStamp = mozilla::TimeStamp;
using TimeDuration = mozilla::TimeDuration;
public:
explicit AndroidUiTask(already_AddRefed<nsIRunnable> aTask)
: mTask(aTask)
, mRunTime() // Null timestamp representing no delay.
{}
AndroidUiTask(already_AddRefed<nsIRunnable> aTask, int aDelayMs)
: mTask(aTask)
, mRunTime(TimeStamp::Now() + TimeDuration::FromMilliseconds(aDelayMs))
{}
bool IsEarlierThan(const AndroidUiTask& aOther) const
{
if (mRunTime) {
return aOther.mRunTime ? mRunTime < aOther.mRunTime : false;
}
// In the case of no delay, we're earlier if aOther has a delay.
// Otherwise, we're not earlier, to maintain task order.
return !!aOther.mRunTime;
}
int64_t MillisecondsToRunTime() const
{
if (mRunTime) {
return int64_t((mRunTime - TimeStamp::Now()).ToMilliseconds());
}
return 0;
}
already_AddRefed<nsIRunnable> TakeTask()
{
return mTask.forget();
}
private:
nsCOMPtr<nsIRunnable> mTask;
const TimeStamp mRunTime;
};
class CreateOnUiThread : public Runnable {
public:
CreateOnUiThread() : Runnable("CreateOnUiThread")
{}
NS_IMETHOD Run() override {
MOZ_ASSERT(!sThreadDestroyed);
MOZ_ASSERT(sMessageLoopAccessMonitor);
MonitorAutoLock lock(*sMessageLoopAccessMonitor);
sThread = new AndroidUiThread();
sThread->InitCurrentThread();
sThread->SetObserver(new ThreadObserver());
sMessageLoop = new MessageLoop(MessageLoop::TYPE_MOZILLA_ANDROID_UI, sThread.get());
lock.NotifyAll();
return NS_OK;
}
};
class DestroyOnUiThread : public Runnable {
public:
DestroyOnUiThread() : Runnable("DestroyOnUiThread"), mDestroyed(false)
{}
NS_IMETHOD Run() override {
MOZ_ASSERT(!sThreadDestroyed);
MOZ_ASSERT(sMessageLoopAccessMonitor);
MOZ_ASSERT(sTaskQueue);
MonitorAutoLock lock(*sMessageLoopAccessMonitor);
sThreadDestroyed = true;
{
// Flush the queue
MutexAutoLock lock (*sTaskQueueLock);
while (AndroidUiTask* task = sTaskQueue->getFirst()) {
delete task;
}
}
delete sMessageLoop;
sMessageLoop = nullptr;
MOZ_ASSERT(sThread);
nsThreadManager::get().UnregisterCurrentThread(*sThread);
sThread = nullptr;
mDestroyed = true;
lock.NotifyAll();
return NS_OK;
}
void WaitForDestruction()
{
MOZ_ASSERT(sMessageLoopAccessMonitor);
MonitorAutoLock lock(*sMessageLoopAccessMonitor);
while (!mDestroyed) {
lock.Wait();
}
}
private:
bool mDestroyed;
};
void
EnqueueTask(already_AddRefed<nsIRunnable> aTask, int aDelayMs)
{
if (sThreadDestroyed) {
return;
}
// add the new task into the sTaskQueue, sorted with
// the earliest task first in the queue
AndroidUiTask* newTask = (aDelayMs ? new AndroidUiTask(std::move(aTask), aDelayMs)
: new AndroidUiTask(std::move(aTask)));
bool headOfList = false;
{
MOZ_ASSERT(sTaskQueue);
MOZ_ASSERT(sTaskQueueLock);
MutexAutoLock lock(*sTaskQueueLock);
AndroidUiTask* task = sTaskQueue->getFirst();
while (task) {
if (newTask->IsEarlierThan(*task)) {
task->setPrevious(newTask);
break;
}
task = task->getNext();
}
if (!newTask->isInList()) {
sTaskQueue->insertBack(newTask);
}
headOfList = !newTask->getPrevious();
}
if (headOfList) {
// if we're inserting it at the head of the queue, notify Java because
// we need to get a callback at an earlier time than the last scheduled
// callback
GeckoThread::RequestUiThreadCallback(int64_t(aDelayMs));
}
}
} // namespace
namespace mozilla {
void
CreateAndroidUiThread()
{
MOZ_ASSERT(!sThread);
MOZ_ASSERT(!sMessageLoopAccessMonitor);
sTaskQueue = new LinkedList<AndroidUiTask>();
sTaskQueueLock = new Mutex("AndroidUiThreadTaskQueueLock");
sMessageLoopAccessMonitor = new Monitor("AndroidUiThreadMessageLoopAccessMonitor");
sThreadDestroyed = false;
RefPtr<CreateOnUiThread> runnable = new CreateOnUiThread;
EnqueueTask(do_AddRef(runnable), 0);
}
void
DestroyAndroidUiThread()
{
MOZ_ASSERT(sThread);
RefPtr<DestroyOnUiThread> runnable = new DestroyOnUiThread;
EnqueueTask(do_AddRef(runnable), 0);
runnable->WaitForDestruction();
delete sMessageLoopAccessMonitor;
sMessageLoopAccessMonitor = nullptr;
}
MessageLoop*
GetAndroidUiThreadMessageLoop()
{
if (!sMessageLoopAccessMonitor) {
return nullptr;
}
MonitorAutoLock lock(*sMessageLoopAccessMonitor);
while (!sMessageLoop) {
lock.Wait();
}
return sMessageLoop;
}
RefPtr<nsThread>
GetAndroidUiThread()
{
if (!sMessageLoopAccessMonitor) {
return nullptr;
}
MonitorAutoLock lock(*sMessageLoopAccessMonitor);
while (!sThread) {
lock.Wait();
}
return sThread;
}
int64_t
RunAndroidUiTasks()
{
MutexAutoLock lock(*sTaskQueueLock);
if (sThreadDestroyed) {
return -1;
}
while (!sTaskQueue->isEmpty()) {
AndroidUiTask* task = sTaskQueue->getFirst();
const int64_t timeLeft = task->MillisecondsToRunTime();
if (timeLeft > 0) {
// this task (and therefore all remaining tasks)
// have not yet reached their runtime. return the
// time left until we should be called again
return timeLeft;
}
// Retrieve task before unlocking/running.
nsCOMPtr<nsIRunnable> runnable(task->TakeTask());
// LinkedListElements auto remove from list upon destruction
delete task;
// Unlock to allow posting new tasks reentrantly.
MutexAutoUnlock unlock(*sTaskQueueLock);
runnable->Run();
if (sThreadDestroyed) {
return -1;
}
}
return -1;
}
} // namespace mozilla