gecko-dev/xpcom/threads/TaskQueue.h

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

226 строки
7.6 KiB
C
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* 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/. */
#ifndef TaskQueue_h_
#define TaskQueue_h_
#include <queue>
#include "mozilla/AbstractThread.h"
#include "mozilla/Maybe.h"
#include "mozilla/Monitor.h"
#include "mozilla/MozPromise.h"
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
#include "mozilla/RefPtr.h"
#include "mozilla/TaskDispatcher.h"
#include "mozilla/Unused.h"
#include "nsIDirectTaskDispatcher.h"
#include "nsThreadUtils.h"
namespace mozilla {
typedef MozPromise<bool, bool, false> ShutdownPromise;
// Abstracts executing runnables in order on an arbitrary event target. The
// runnables dispatched to the TaskQueue will be executed in the order in which
// they're received, and are guaranteed to not be executed concurrently.
// They may be executed on different threads, and a memory barrier is used
// to make this threadsafe for objects that aren't already threadsafe.
//
// Note, since a TaskQueue can also be converted to an nsIEventTarget using
// WrapAsEventTarget() its possible to construct a hierarchy of TaskQueues.
// Consider these three TaskQueues:
//
// TQ1 dispatches to the main thread
// TQ2 dispatches to TQ1
// TQ3 dispatches to TQ1
//
// This ensures there is only ever a single runnable from the entire chain on
// the main thread. It also ensures that TQ2 and TQ3 only have a single
// runnable in TQ1 at any time.
//
// This arrangement lets you prioritize work by dispatching runnables directly
// to TQ1. You can issue many runnables for important work. Meanwhile the TQ2
// and TQ3 work will always execute at most one runnable and then yield.
//
// A TaskQueue does not require explicit shutdown, however it provides a
// BeginShutdown() method that places TaskQueue in a shut down state and returns
// a promise that gets resolved once all pending tasks have completed
class TaskQueue : public AbstractThread, public nsIDirectTaskDispatcher {
class EventTargetWrapper;
public:
explicit TaskQueue(already_AddRefed<nsIEventTarget> aTarget,
bool aSupportsTailDispatch = false);
TaskQueue(already_AddRefed<nsIEventTarget> aTarget, const char* aName,
bool aSupportsTailDispatch = false);
NS_DECL_ISUPPORTS_INHERITED
NS_DECL_NSIDIRECTTASKDISPATCHER
TaskDispatcher& TailDispatcher() override;
NS_IMETHOD Dispatch(already_AddRefed<nsIRunnable> aEvent,
uint32_t aFlags) override {
nsCOMPtr<nsIRunnable> runnable = aEvent;
{
MonitorAutoLock mon(mQueueMonitor);
return DispatchLocked(/* passed by ref */ runnable, aFlags,
NormalDispatch);
}
// If the ownership of |r| is not transferred in DispatchLocked() due to
// dispatch failure, it will be deleted here outside the lock. We do so
// since the destructor of the runnable might access TaskQueue and result
// in deadlocks.
}
[[nodiscard]] nsresult Dispatch(
already_AddRefed<nsIRunnable> aRunnable,
DispatchReason aReason = NormalDispatch) override {
nsCOMPtr<nsIRunnable> r = aRunnable;
{
MonitorAutoLock mon(mQueueMonitor);
Bug 1593802 - don't drop dispatch flags in TaskQueue's EventTargetWrapper; r=erahm `TaskQueue` wraps an `nsIEventTarget` to provide "one runnable at a time" execution policies regardless of the underlying implementation of the wrapped `nsIEventTarget` (e.g. a thread pool). `TaskQueue` also provides a `nsISerialEventTarget` wrapper, `EventTargetWrapper`, around itself (!) for consumers who want to continue to provide a more XPCOM-ish API. One would think that dispatching tasks to `EventTargetWrapper` with a given set of flags would pass that set of flags through, unchanged, to the underlying event target of the wrapped `TaskQueue`. This pass-through is not the case. `TaskQueue` supports a "tail dispatch" mode of operation that is somewhat underdocumented. Roughly, tail dispatch to a `TaskQueue` says (with several other conditions) that dispatched tasks are held separately and not passed through to the underlying event target. If a given `TaskQueue` supports tail dispatch and the dispatcher also supports tail dispatch, any tasks dispatched to said `TaskQueue` are silently converted to tail dispatched tasks. Since tail dispatched tasks can't meaningfully have flags associated with them, the current implementation simply drops any passed flags on the floor. These flags, however, might be meaningful, and we should attempt to honor them in the cases we're not doing tail dispatch. (And when we are doing tail dispatch, we can verify that the requested flags are not asking for peculiar things.) Differential Revision: https://phabricator.services.mozilla.com/D51702 --HG-- extra : moz-landing-system : lando
2019-11-05 19:59:30 +03:00
return DispatchLocked(/* passed by ref */ r, NS_DISPATCH_NORMAL, aReason);
}
// If the ownership of |r| is not transferred in DispatchLocked() due to
// dispatch failure, it will be deleted here outside the lock. We do so
// since the destructor of the runnable might access TaskQueue and result
// in deadlocks.
}
// So we can access nsIEventTarget::Dispatch(nsIRunnable*, uint32_t aFlags)
using nsIEventTarget::Dispatch;
// Puts the queue in a shutdown state and returns immediately. The queue will
// remain alive at least until all the events are drained, because the Runners
// hold a strong reference to the task queue, and one of them is always held
// by the target event queue when the task queue is non-empty.
//
// The returned promise is resolved when the queue goes empty.
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<ShutdownPromise> BeginShutdown();
// Blocks until all task finish executing.
void AwaitIdle();
// Blocks until the queue is flagged for shutdown and all tasks have finished
// executing.
void AwaitShutdownAndIdle();
bool IsEmpty();
// Returns true if the current thread is currently running a Runnable in
// the task queue.
bool IsCurrentThreadIn() const override;
using nsISerialEventTarget::IsOnCurrentThread;
protected:
virtual ~TaskQueue();
// Blocks until all task finish executing. Called internally by methods
// that need to wait until the task queue is idle.
// mQueueMonitor must be held.
void AwaitIdleLocked();
Bug 1593802 - don't drop dispatch flags in TaskQueue's EventTargetWrapper; r=erahm `TaskQueue` wraps an `nsIEventTarget` to provide "one runnable at a time" execution policies regardless of the underlying implementation of the wrapped `nsIEventTarget` (e.g. a thread pool). `TaskQueue` also provides a `nsISerialEventTarget` wrapper, `EventTargetWrapper`, around itself (!) for consumers who want to continue to provide a more XPCOM-ish API. One would think that dispatching tasks to `EventTargetWrapper` with a given set of flags would pass that set of flags through, unchanged, to the underlying event target of the wrapped `TaskQueue`. This pass-through is not the case. `TaskQueue` supports a "tail dispatch" mode of operation that is somewhat underdocumented. Roughly, tail dispatch to a `TaskQueue` says (with several other conditions) that dispatched tasks are held separately and not passed through to the underlying event target. If a given `TaskQueue` supports tail dispatch and the dispatcher also supports tail dispatch, any tasks dispatched to said `TaskQueue` are silently converted to tail dispatched tasks. Since tail dispatched tasks can't meaningfully have flags associated with them, the current implementation simply drops any passed flags on the floor. These flags, however, might be meaningful, and we should attempt to honor them in the cases we're not doing tail dispatch. (And when we are doing tail dispatch, we can verify that the requested flags are not asking for peculiar things.) Differential Revision: https://phabricator.services.mozilla.com/D51702 --HG-- extra : moz-landing-system : lando
2019-11-05 19:59:30 +03:00
nsresult DispatchLocked(nsCOMPtr<nsIRunnable>& aRunnable, uint32_t aFlags,
DispatchReason aReason = NormalDispatch);
void MaybeResolveShutdown() {
mQueueMonitor.AssertCurrentThreadOwns();
if (mIsShutdown && !mIsRunning) {
mShutdownPromise.ResolveIfExists(true, __func__);
mTarget = nullptr;
}
}
nsCOMPtr<nsIEventTarget> mTarget;
// Monitor that protects the queue and mIsRunning;
Monitor mQueueMonitor;
typedef struct TaskStruct {
nsCOMPtr<nsIRunnable> event;
uint32_t flags;
} TaskStruct;
// Queue of tasks to run.
std::queue<TaskStruct> mTasks;
// The thread currently running the task queue. We store a reference
// to this so that IsCurrentThreadIn() can tell if the current thread
// is the thread currently running in the task queue.
//
// This may be read on any thread, but may only be written on mRunningThread.
// The thread can't die while we're running in it, and we only use it for
// pointer-comparison with the current thread anyway - so we make it atomic
// and don't refcount it.
Atomic<PRThread*> mRunningThread;
// RAII class that gets instantiated for each dispatched task.
class AutoTaskGuard {
public:
explicit AutoTaskGuard(TaskQueue* aQueue)
: mQueue(aQueue), mLastCurrentThread(nullptr) {
// NB: We don't hold the lock to aQueue here. Don't do anything that
// might require it.
MOZ_ASSERT(!mQueue->mTailDispatcher);
mTaskDispatcher.emplace(aQueue,
/* aIsTailDispatcher = */ true);
mQueue->mTailDispatcher = mTaskDispatcher.ptr();
mLastCurrentThread = sCurrentThreadTLS.get();
sCurrentThreadTLS.set(aQueue);
MOZ_ASSERT(mQueue->mRunningThread == nullptr);
mQueue->mRunningThread = PR_GetCurrentThread();
}
~AutoTaskGuard() {
mTaskDispatcher->DrainDirectTasks();
mTaskDispatcher.reset();
MOZ_ASSERT(mQueue->mRunningThread == PR_GetCurrentThread());
mQueue->mRunningThread = nullptr;
sCurrentThreadTLS.set(mLastCurrentThread);
mQueue->mTailDispatcher = nullptr;
}
private:
Maybe<AutoTaskDispatcher> mTaskDispatcher;
TaskQueue* mQueue;
AbstractThread* mLastCurrentThread;
};
TaskDispatcher* mTailDispatcher;
// True if we've dispatched an event to the target to execute events from
// the queue.
bool mIsRunning;
// True if we've started our shutdown process.
bool mIsShutdown;
MozPromiseHolder<ShutdownPromise> mShutdownPromise;
// The name of this TaskQueue. Useful when debugging dispatch failures.
const char* const mName;
SimpleTaskQueue mDirectTasks;
class Runner : public Runnable {
public:
explicit Runner(TaskQueue* aQueue)
: Runnable("TaskQueue::Runner"), mQueue(aQueue) {}
NS_IMETHOD Run() override;
private:
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<TaskQueue> mQueue;
};
};
} // namespace mozilla
#endif // TaskQueue_h_