SurfaceFlinger启动篇

Posted by Gityuan on February 11, 2017

基于Android 6.0源码, 分析SurfaceFlinger原理

frameworks/native/services/surfaceflinger/
  - main_surfaceflinger.cpp
  - SurfaceFlinger.cpp
  - DispSync.cpp
  - MessageQueue.cpp
  - DisplayHardware/HWComposer.cpp

frameworks/native/libs/gui/
  - DisplayEventReceiver.cpp
  - BitTube.cpp

一. 概述

Android系统的图形处理相关的模块,就不得不提surfaceflinger,这是由init进程所启动的 守护进程,在init.rc中该服务如下:

service surfaceflinger /system/bin/surfaceflinger
    class core
    user system
    group graphics drmrpc
    onrestart restart zygote
    writepid /dev/cpuset/system-background/tasks

surfaceflinger服务属于核心类(core class),另外,当surfaceflinger重启时会触发zygote的重启。 surfaceflinger服务启动的起点便是如下的main()函数。

二. 启动过程

2.1 main

[-> main_surfaceflinger.cpp]

int main(int, char**) {
    ProcessState::self()->setThreadPoolMaxThreadCount(4);

    sp<ProcessState> ps(ProcessState::self());
    ps->startThreadPool();

    //实例化surfaceflinger【见小节2.2】
    sp<SurfaceFlinger> flinger =  new SurfaceFlinger();

    setpriority(PRIO_PROCESS, 0, PRIORITY_URGENT_DISPLAY);
    set_sched_policy(0, SP_FOREGROUND);

    //初始化【见小节2.3】
    flinger->init();

    //发布surface flinger,注册到Service Manager
    sp<IServiceManager> sm(defaultServiceManager());
    sm->addService(String16(SurfaceFlinger::getServiceName()), flinger, false);

    // 运行在当前线程【见小节2.11】
    flinger->run();

    return 0;
}

该方法的主要功能:

  • 设定surfaceflinger进程的binder线程池个数上限为4,并启动binder线程池;
  • 创建SurfaceFlinger对象;【见小节2.1】
  • 设置surfaceflinger进程为高优先级以及前台调度策略;
  • 初始化SurfaceFlinger;【见小节2.3】
  • 将”SurfaceFlinger”服务注册到Service Manager;
  • 在当前主线程执行SurfaceFlinger的run方法。【见小节2.11】

2.2 创建SurfaceFlinger

[-> SurfaceFlinger.cpp]

SurfaceFlinger::SurfaceFlinger()
    :   BnSurfaceComposer(),
        mTransactionFlags(0),
        mTransactionPending(false),
        mAnimTransactionPending(false),
        mLayersRemoved(false),
        mRepaintEverything(0),
        mRenderEngine(NULL),
        mBootTime(systemTime()),
        mVisibleRegionsDirty(false),
        mHwWorkListDirty(false),
        mAnimCompositionPending(false),
        mDebugRegion(0),
        mDebugDDMS(0),
        mDebugDisableHWC(0),
        mDebugDisableTransformHint(0),
        mDebugInSwapBuffers(0),
        mLastSwapBufferTime(0),
        mDebugInTransaction(0),
        mLastTransactionTime(0),
        mBootFinished(false),
        mForceFullDamage(false),
        mPrimaryHWVsyncEnabled(false),
        mHWVsyncAvailable(false),
        mDaltonize(false),
        mHasColorMatrix(false),
        mHasPoweredOff(false),
        mFrameBuckets(),
        mTotalTime(0),
        mLastSwapTime(0)
{
    ALOGI("SurfaceFlinger is starting");
    char value[PROPERTY_VALUE_MAX];

    property_get("ro.bq.gpu_to_cpu_unsupported", value, "0");
    mGpuToCpuSupported = !atoi(value);

    property_get("debug.sf.showupdates", value, "0");
    mDebugRegion = atoi(value);

    property_get("debug.sf.ddms", value, "0");
    mDebugDDMS = atoi(value);       
}

SurfaceFlinger继承于BnSurfaceComposer,IBinder::DeathRecipient,HWComposer::EventHandler

flinger的数据类型为sp强指针类型,当首次被强指针引用时则执行OnFirstRef()

2.2.1 SF.onFirstRef

void SurfaceFlinger::onFirstRef()
{
    mEventQueue.init(this);
}

2.2.2 MQ.init

[-> MessageQueue.cpp]

void MessageQueue::init(const sp<SurfaceFlinger>& flinger)
{
    mFlinger = flinger;
    mLooper = new Looper(true);
    mHandler = new Handler(*this); //【见小节2.2.3】
}

这个Handler是MessageQueue的内部类Handler,此处传递的*this便是MessageQueue本身。

2.2.3 MQ.Handler

[-> MessageQueue.cpp]

class MessageQueue {
    class Handler : public MessageHandler {
        enum {
            eventMaskInvalidate     = 0x1,
            eventMaskRefresh        = 0x2,
            eventMaskTransaction    = 0x4
        };
        MessageQueue& mQueue;
        int32_t mEventMask;
    public:
        Handler(MessageQueue& queue) : mQueue(queue), mEventMask(0) { }
        virtual void handleMessage(const Message& message);
        void dispatchRefresh();
        void dispatchInvalidate();
        void dispatchTransaction();
    };
    ...
}

2.3 SF.init

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::init() {
    Mutex::Autolock _l(mStateLock);

    //初始化EGL,作为默认的显示
    mEGLDisplay = eglGetDisplay(EGL_DEFAULT_DISPLAY);
    eglInitialize(mEGLDisplay, NULL, NULL);

    // 初始化硬件composer对象【见小节2.4】
    mHwc = new HWComposer(this, *static_cast<HWComposer::EventHandler *>(this));

    //获取RenderEngine引擎
    mRenderEngine = RenderEngine::create(mEGLDisplay, mHwc->getVisualID());

    //检索创建的EGL上下文
    mEGLContext = mRenderEngine->getEGLContext();

    //初始化非虚拟显示屏【见小节2.5】
    for (size_t i=0 ; i<DisplayDevice::NUM_BUILTIN_DISPLAY_TYPES ; i++) {
        DisplayDevice::DisplayType type((DisplayDevice::DisplayType)i);
        //建立已连接的显示设备
        if (mHwc->isConnected(i) || type==DisplayDevice::DISPLAY_PRIMARY) {
            bool isSecure = true;
            createBuiltinDisplayLocked(type);
            wp<IBinder> token = mBuiltinDisplays[i];

            sp<IGraphicBufferProducer> producer;
            sp<IGraphicBufferConsumer> consumer;
            //创建BufferQueue的生产者和消费者
            BufferQueue::createBufferQueue(&producer, &consumer,
                    new GraphicBufferAlloc());

            sp<FramebufferSurface> fbs = new FramebufferSurface(*mHwc, i, consumer);
            int32_t hwcId = allocateHwcDisplayId(type);
            //创建显示设备
            sp<DisplayDevice> hw = new DisplayDevice(this,
                    type, hwcId, mHwc->getFormat(hwcId), isSecure, token,
                    fbs, producer,
                    mRenderEngine->getEGLConfig());
            if (i > DisplayDevice::DISPLAY_PRIMARY) {
                hw->setPowerMode(HWC_POWER_MODE_NORMAL);
            }
            mDisplays.add(token, hw);
        }
    }

    getDefaultDisplayDevice()->makeCurrent(mEGLDisplay, mEGLContext);

    //当应用和sf的vsync偏移量一致时,则只创建一个EventThread线程
    if (vsyncPhaseOffsetNs != sfVsyncPhaseOffsetNs) {
        sp<VSyncSource> vsyncSrc = new DispSyncSource(&mPrimaryDispSync,
                vsyncPhaseOffsetNs, true, "app");
        mEventThread = new EventThread(vsyncSrc);
        sp<VSyncSource> sfVsyncSrc = new DispSyncSource(&mPrimaryDispSync,
                sfVsyncPhaseOffsetNs, true, "sf");
        mSFEventThread = new EventThread(sfVsyncSrc);
        mEventQueue.setEventThread(mSFEventThread);
    } else {
        //创建DispSyncSource对象【2.6】
        sp<VSyncSource> vsyncSrc = new DispSyncSource(&mPrimaryDispSync,
                vsyncPhaseOffsetNs, true, "sf-app");
        //创建线程EventThread 【见小节2.7】
        mEventThread = new EventThread(vsyncSrc);
        //设置EventThread 【见小节2.8】
        mEventQueue.setEventThread(mEventThread);
    }

    //【见小节2.9】
    mEventControlThread = new EventControlThread(this);
    mEventControlThread->run("EventControl", PRIORITY_URGENT_DISPLAY);

    //当不存在HWComposer时,则设置软件vsync
    if (mHwc->initCheck() != NO_ERROR) {
        mPrimaryDispSync.setPeriod(16666667);
    }

    //初始化绘图状态
    mDrawingState = mCurrentState;

    //初始化显示设备
    initializeDisplays();

    //启动开机动画【2.10】
    startBootAnim();
}

主要功能:

  • 初始化EGL相关;
  • 创建HWComposer对象;
  • 初始化非虚拟显示屏;
  • 启动app和sf两个EventThread线程;
  • 启动开机动画;

另外,当应用和sf的vsync偏移量一致时,则只创建一个EventThread线程;否则会创建两个DispSyncSource对象,分别是用于绘制(app)和合成(SurfaceFlinger)。

2.4 创建HWComposer

[-> HWComposer.cpp]

HWComposer::HWComposer(
        const sp<SurfaceFlinger>& flinger,
        EventHandler& handler)
    : mFlinger(flinger),
      mFbDev(0), mHwc(0), mNumDisplays(1),
      mCBContext(new cb_context),
      mEventHandler(handler),
      mDebugForceFakeVSync(false)
{
    ...
    bool needVSyncThread = true;
    int fberr = loadFbHalModule(); //加载framebuffer的HAL层模块
    loadHwcModule(); //加载HWComposer模块

    //标记已分配的display ID
    for (size_t i=0 ; i<NUM_BUILTIN_DISPLAYS ; i++) {
        mAllocatedDisplayIDs.markBit(i);
    }

    if (mHwc) {
        if (mHwc->registerProcs) {
            mCBContext->hwc = this;
            mCBContext->procs.invalidate = &hook_invalidate;
            //VSYNC信号的回调方法
            mCBContext->procs.vsync = &hook_vsync;
            if (hwcHasApiVersion(mHwc, HWC_DEVICE_API_VERSION_1_1))
                mCBContext->procs.hotplug = &hook_hotplug;
            else
                mCBContext->procs.hotplug = NULL;
            memset(mCBContext->procs.zero, 0, sizeof(mCBContext->procs.zero));
            //注册回调函数
            mHwc->registerProcs(mHwc, &mCBContext->procs);
        }

        //进入此处,说明已成功打开硬件composer设备,则不再需要vsync线程
        needVSyncThread = false;
        eventControl(HWC_DISPLAY_PRIMARY, HWC_EVENT_VSYNC, 0);
        ...
    }
    ...

    if (needVSyncThread) {
        //不支持硬件的VSYNC,则会创建线程来模拟定时VSYNC信号
        mVSyncThread = new VSyncThread(*this);
    }
}

HWComposer代表着硬件显示设备,注册了VSYNC信号的回调。VSYNC信号本身是由显示驱动产生的, 在不支持硬件的VSYNC,则会创建“VSyncThread”线程来模拟定时VSYNC信号。

2.5 显示设备

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::init() {
    ...
    for (size_t i=0 ; i<DisplayDevice::NUM_BUILTIN_DISPLAY_TYPES ; i++) {
        DisplayDevice::DisplayType type((DisplayDevice::DisplayType)i);
        //建立已连接的显示设备
        if (mHwc->isConnected(i) || type==DisplayDevice::DISPLAY_PRIMARY) {
            bool isSecure = true;
            createBuiltinDisplayLocked(type);
            wp<IBinder> token = mBuiltinDisplays[i];

            sp<IGraphicBufferProducer> producer;
            sp<IGraphicBufferConsumer> consumer;
            //创建BufferQueue的生产者和消费者
            BufferQueue::createBufferQueue(&producer, &consumer,
                    new GraphicBufferAlloc());

            sp<FramebufferSurface> fbs = new FramebufferSurface(*mHwc, i, consumer);
            int32_t hwcId = allocateHwcDisplayId(type);
            //创建显示设备
            sp<DisplayDevice> hw = new DisplayDevice(this,
                    type, hwcId, mHwc->getFormat(hwcId), isSecure, token,
                    fbs, producer,
                    mRenderEngine->getEGLConfig());
            if (i > DisplayDevice::DISPLAY_PRIMARY) {
                hw->setPowerMode(HWC_POWER_MODE_NORMAL);
            }
            mDisplays.add(token, hw);
        }
    }
    ...
}

创建IGraphicBufferProducer和IGraphicBufferConsumer,以及FramebufferSurface,DisplayDevice对象。另外, 显示设备有3类:主设备,扩展设备,虚拟设备。其中前两个都是内置显示设备,故NUM_BUILTIN_DISPLAY_TYPES=2,

2.6 DispSyncSource

[-> SurfaceFlinger.cpp]

class DispSyncSource : public VSyncSource, private DispSync::Callback {
  DispSyncSource(DispSync* dispSync, nsecs_t phaseOffset, bool traceVsync,
      const char* label) :
          mValue(0),
          mTraceVsync(traceVsync),
          mVsyncOnLabel(String8::format("VsyncOn-%s", label)),
          mVsyncEventLabel(String8::format("VSYNC-%s", label)),
          mDispSync(dispSync),
          mCallbackMutex(),
          mCallback(),
          mVsyncMutex(),
          mPhaseOffset(phaseOffset),
          mEnabled(false) {}
  ...   }

2.7 EventThread线程

[-> EventThread.cpp]

EventThread::EventThread(const sp<VSyncSource>& src)
    : mVSyncSource(src),
      mUseSoftwareVSync(false),
      mVsyncEnabled(false),
      mDebugVsyncEnabled(false),
      mVsyncHintSent(false) {

    for (int32_t i=0 ; i<DisplayDevice::NUM_BUILTIN_DISPLAY_TYPES ; i++) {
        mVSyncEvent[i].header.type = DisplayEventReceiver::DISPLAY_EVENT_VSYNC;
        mVSyncEvent[i].header.id = 0;
        mVSyncEvent[i].header.timestamp = 0;
        mVSyncEvent[i].vsync.count =  0;
    }
    struct sigevent se;
    se.sigev_notify = SIGEV_THREAD;
    se.sigev_value.sival_ptr = this;
    se.sigev_notify_function = vsyncOffCallback;
    se.sigev_notify_attributes = NULL;
    timer_create(CLOCK_MONOTONIC, &se, &mTimerId);
}

EventThread继承于Thread和VSyncSource::Callback两个类。

2.7.1 ET.onFirstRef

[-> EventThread.cpp]

void EventThread::onFirstRef() {
    //运行EventThread线程【见小节2.7.2】
    run("EventThread", PRIORITY_URGENT_DISPLAY + PRIORITY_MORE_FAVORABLE);
}

2.7.2 ET.threadLoop

[-> EventThread.cpp]

bool EventThread::threadLoop() {
    DisplayEventReceiver::Event event;
    Vector< sp<EventThread::Connection> > signalConnections;
    // 等待事件【见小节2.7.3】
    signalConnections = waitForEvent(&event);

    //分发事件给所有的监听者
    const size_t count = signalConnections.size();
    for (size_t i=0 ; i<count ; i++) {
        const sp<Connection>& conn(signalConnections[i]);
        //传递事件【见小节3.10】
        status_t err = conn->postEvent(event);
        if (err == -EAGAIN || err == -EWOULDBLOCK) {
            //可能此时connection已满,则直接抛弃事件
            ALOGW("EventThread: dropping event (%08x) for connection %p",
                    event.header.type, conn.get());
        } else if (err < 0) {
            //发生致命错误,则清理该连接
            removeDisplayEventConnection(signalConnections[i]);
        }
    }
    return true;
}

2.7.3 waitForEvent

[-> EventThread.cpp]

Vector< sp<EventThread::Connection> > EventThread::waitForEvent(
        DisplayEventReceiver::Event* event)
{
    Mutex::Autolock _l(mLock);
    Vector< sp<EventThread::Connection> > signalConnections;

    do {
        bool eventPending = false;
        bool waitForVSync = false;

        size_t vsyncCount = 0;
        nsecs_t timestamp = 0;
        for (int32_t i=0 ; i<DisplayDevice::NUM_BUILTIN_DISPLAY_TYPES ; i++) {
            timestamp = mVSyncEvent[i].header.timestamp;
            if (timestamp) {
                *event = mVSyncEvent[i];
                mVSyncEvent[i].header.timestamp = 0;
                vsyncCount = mVSyncEvent[i].vsync.count;
                break;
            }
        }

        if (!timestamp) {
            //没有vsync事件,则查看其它事件
            eventPending = !mPendingEvents.isEmpty();
            if (eventPending) {
                //存在其它事件可用于分发
                *event = mPendingEvents[0];
                mPendingEvents.removeAt(0);
            }
        }

        //查找正在等待事件的连接
        size_t count = mDisplayEventConnections.size();
        for (size_t i=0 ; i<count ; i++) {
            sp<Connection> connection(mDisplayEventConnections[i].promote());
            if (connection != NULL) {
                bool added = false;
                if (connection->count >= 0) {
                    //需要vsync事件,由于至少存在一个连接正在等待vsync
                    waitForVSync = true;
                    if (timestamp) {
                        if (connection->count == 0) {
                            connection->count = -1;
                            signalConnections.add(connection);
                            added = true;
                        } else if (connection->count == 1 ||
                                (vsyncCount % connection->count) == 0) {
                            signalConnections.add(connection);
                            added = true;
                        }
                    }
                }

                if (eventPending && !timestamp && !added) {
                    //没有vsync事件需要处理(timestamp==0),但存在pending消息
                    signalConnections.add(connection);
                }
            } else {
                //该连接已死亡,则直接清理
                mDisplayEventConnections.removeAt(i);
                --i; --count;
            }
        }

        if (timestamp && !waitForVSync) {
            //接收到VSYNC,但没有client需要它,则直接关闭VSYNC
            disableVSyncLocked();
        } else if (!timestamp && waitForVSync) {
            //至少存在一个client,则需要使能VSYNC
            enableVSyncLocked();
        }

        if (!timestamp && !eventPending) {
            if (waitForVSync) {
                bool softwareSync = mUseSoftwareVSync;
                nsecs_t timeout = softwareSync ? ms2ns(16) : ms2ns(1000);
                if (mCondition.waitRelative(mLock, timeout) == TIMED_OUT) {
                    mVSyncEvent[0].header.type = DisplayEventReceiver::DISPLAY_EVENT_VSYNC;
                    mVSyncEvent[0].header.id = DisplayDevice::DISPLAY_PRIMARY;
                    mVSyncEvent[0].header.timestamp = systemTime(SYSTEM_TIME_MONOTONIC);
                    mVSyncEvent[0].vsync.count++;
                }
            } else {
                //不存在对vsync感兴趣的连接,即将要进入休眠
                mCondition.wait(mLock);
            }
        }
    } while (signalConnections.isEmpty());

    //到此处,则保证存在timestamp以及连接
    return signalConnections;
}

EventThread线程,进入mCondition的wait()方法,等待唤醒。

2.8 setEventThread

[-> MessageQueue.cpp]

void MessageQueue::setEventThread(const sp<EventThread>& eventThread)
{
    mEventThread = eventThread;
    //创建连接
    mEvents = eventThread->createEventConnection();
    //获取BitTube对象
    mEventTube = mEvents->getDataChannel();
    //监听BitTube,一旦有数据到来则调用cb_eventReceiver()
    mLooper->addFd(mEventTube->getFd(), 0, Looper::EVENT_INPUT,
            MessageQueue::cb_eventReceiver, this);
}

此处mEvents的数据类型为sp,mEventTube的数据类型为sp

2.9 EventControlThread线程

[-> EventControlThread.cpp]

EventControlThread::EventControlThread(const sp<SurfaceFlinger>& flinger):
        mFlinger(flinger),
        mVsyncEnabled(false) {
}

bool EventControlThread::threadLoop() {
    Mutex::Autolock lock(mMutex);
    bool vsyncEnabled = mVsyncEnabled;

    mFlinger->eventControl(HWC_DISPLAY_PRIMARY, SurfaceFlinger::EVENT_VSYNC,
            mVsyncEnabled);

    while (true) {
        status_t err = mCond.wait(mMutex);
        ...

        if (vsyncEnabled != mVsyncEnabled) {
            mFlinger->eventControl(HWC_DISPLAY_PRIMARY,
                    SurfaceFlinger::EVENT_VSYNC, mVsyncEnabled);
            vsyncEnabled = mVsyncEnabled;
        }
    }

    return false;
}

EventControlThread也是继承于Thread。

2.10 startBootAnim

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::startBootAnim() {
    property_set("service.bootanim.exit", "0");
    property_set("ctl.start", "bootanim");
}

通过控制ctl.start属性,设置成bootanim值,则触发init进程来创建开机动画进程bootanim, 到此,则开始显示开机过程的动画。 从小节[2.4 ~2.9]都是介绍SurfaceFlinger的init()过程, 紧接着便执行其run()方法。

2.11 SF.run

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::run() {
    do {
        //不断循环地等待事件【见小节2.12】
        waitForEvent();
    } while (true);
}

2.12 SF.waitForEvent

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::waitForEvent() {
    mEventQueue.waitMessage(); //【2.13】
}

mEventQueue的数据类型为MessageQueue。

2.13 MQ.waitMessage

[-> MessageQueue.cpp]

void MessageQueue::waitMessage() {
    do {
        IPCThreadState::self()->flushCommands();
        int32_t ret = mLooper->pollOnce(-1);
        ...
    } while (true);
}

可见SurfaceFlinger主线程进入waitMessage来等待消息的到来。

三. Vsync信号

HWComposer对象创建过程,会注册一些回调方法,当硬件产生VSYNC信号时,则会回调hook_vsync()方法。

3.1 HWC.hook_vsync

[-> HWComposer.cpp]

void HWComposer::hook_vsync(const struct hwc_procs* procs, int disp,
        int64_t timestamp) {
    cb_context* ctx = reinterpret_cast<cb_context*>(
            const_cast<hwc_procs_t*>(procs));
    ctx->hwc->vsync(disp, timestamp); //【见小节3.2】
}

3.2 HWC.vsync

[-> HWComposer.cpp]

void HWComposer::vsync(int disp, int64_t timestamp) {
    if (uint32_t(disp) < HWC_NUM_PHYSICAL_DISPLAY_TYPES) {
        {
            Mutex::Autolock _l(mLock);
            if (timestamp == mLastHwVSync[disp]) {
                return; //忽略重复的VSYNC信号
            }
            mLastHwVSync[disp] = timestamp;
        }
        //【见小节3.3】
        mEventHandler.onVSyncReceived(disp, timestamp);
    }
}

当收到VSYNC信号则会回调EventHandler的onVSyncReceived()方法,此处mEventHandler是指SurfaceFlinger对象。

3.3 SF.onVSyncReceived

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::onVSyncReceived(int type, nsecs_t timestamp) {
    bool needsHwVsync = false;

    {
        Mutex::Autolock _l(mHWVsyncLock);
        if (type == 0 && mPrimaryHWVsyncEnabled) {
            // 此处mPrimaryDispSync为DispSync类【见小节3.4】
            needsHwVsync = mPrimaryDispSync.addResyncSample(timestamp);
        }
    }

    if (needsHwVsync) {
        enableHardwareVsync();
    } else {
        disableHardwareVsync(false);
    }
}

3.4 DS.addResyncSample

此处调用addResyncSample对象的addResyncSample方法,那么先来看看DispSync对象的初始化过程

3.4.1 创建DispSync

[-> DispSync.cpp]

DispSync::DispSync() :
        mRefreshSkipCount(0),
        mThread(new DispSyncThread()) {
    //【见小节3.4.2】
    mThread->run("DispSync", PRIORITY_URGENT_DISPLAY + PRIORITY_MORE_FAVORABLE);

    reset();
    beginResync();
    ...
}

3.4.2 DispSyncThread线程

[-> DispSync.cpp]

virtual bool threadLoop() {
     status_t err;
     nsecs_t now = systemTime(SYSTEM_TIME_MONOTONIC);
     nsecs_t nextEventTime = 0;

     while (true) {
         Vector<CallbackInvocation> callbackInvocations;

         nsecs_t targetTime = 0;
         { // Scope for lock
             Mutex::Autolock lock(mMutex);
             if (mStop) {
                 return false;
             }

             if (mPeriod == 0) {
                 err = mCond.wait(mMutex);
                 continue;
             }

             nextEventTime = computeNextEventTimeLocked(now);
             targetTime = nextEventTime;
             bool isWakeup = false;

             if (now < targetTime) {
                 err = mCond.waitRelative(mMutex, targetTime - now);
                 if (err == TIMED_OUT) {
                     isWakeup = true;
                 } else if (err != NO_ERROR) {
                     return false;
                 }
             }

             now = systemTime(SYSTEM_TIME_MONOTONIC);

             if (isWakeup) {
                 mWakeupLatency = ((mWakeupLatency * 63) +
                         (now - targetTime)) / 64;
                 if (mWakeupLatency > 500000) {
                     mWakeupLatency = 500000;
                 }
             }
             //收集vsync信号的所有回调方法
             callbackInvocations = gatherCallbackInvocationsLocked(now);
         }

         if (callbackInvocations.size() > 0) {
             //回调所有对象的onDispSyncEvent方法
             fireCallbackInvocations(callbackInvocations);
         }
     }

     return false;
 }   

线程”DispSync”停留在mCond的wait()过程,等待被唤醒。

3.4.3 addResyncSample

[-> DispSync.cpp]

bool DispSync::addResyncSample(nsecs_t timestamp) {
    Mutex::Autolock lock(mMutex);

    size_t idx = (mFirstResyncSample + mNumResyncSamples) % MAX_RESYNC_SAMPLES;
    mResyncSamples[idx] = timestamp;

    if (mNumResyncSamples < MAX_RESYNC_SAMPLES) {
        mNumResyncSamples++;
    } else {
        mFirstResyncSample = (mFirstResyncSample + 1) % MAX_RESYNC_SAMPLES;
    }

    updateModelLocked(); //【见小节3.5】

    if (mNumResyncSamplesSincePresent++ > MAX_RESYNC_SAMPLES_WITHOUT_PRESENT) {
        resetErrorLocked();
    }

    if (kIgnorePresentFences) {
        return mThread->hasAnyEventListeners();
    }

    return mPeriod == 0 || mError > kErrorThreshold;
}

3.5 DS.updateModelLocked

[-> DispSync.cpp]

void DispSync::updateModelLocked() {
    ...
    //【见小节3.6】
    mThread->updateModel(mPeriod, mPhase);
}

3.6 DST.updateModel

[-> DispSyncThread.cpp]

class DispSyncThread: public Thread {
    void updateModel(nsecs_t period, nsecs_t phase) {
        Mutex::Autolock lock(mMutex);
        mPeriod = period;
        mPhase = phase;
        mCond.signal(); //唤醒目标线程
    }
}

唤醒DispSyncThread线程,接下里进入DispSyncThread线程。

3.7 DispSyncThread线程

[-> DispSync.cpp]

virtual bool threadLoop() {
     ...
     while (true) {
         Vector<CallbackInvocation> callbackInvocations;

         nsecs_t targetTime = 0;
         { // Scope for lock
             Mutex::Autolock lock(mMutex);
            ...
             if (now < targetTime) {
                 err = mCond.waitRelative(mMutex, targetTime - now);
                 ...
             }
             ...
             //收集vsync信号的所有回调方法
             callbackInvocations = gatherCallbackInvocationsLocked(now);
         }

         if (callbackInvocations.size() > 0) {
             //回调所有对象的onDispSyncEvent方法 【见小节3.7.1】
             fireCallbackInvocations(callbackInvocations);
         }
     }

     return false;
 }

3.7.1 fireCallbackInvocations

void fireCallbackInvocations(const Vector<CallbackInvocation>& callbacks) {
    for (size_t i = 0; i < callbacks.size(); i++) {
        //【见小节3.8】
        callbacks[i].mCallback->onDispSyncEvent(callbacks[i].mEventTime);
    }
}

在前面小节【2.3】SurfaceFlinger调用init()的过程,创建过DispSyncSource对象。接下里便是回调该对象的 onDispSyncEvent。

3.8 DSS.onDispSyncEvent

[-> SurfaceFlinger.cpp ::DispSyncSource]

virtual void onDispSyncEvent(nsecs_t when) {
    sp<VSyncSource::Callback> callback;
    {
       Mutex::Autolock lock(mCallbackMutex);
       callback = mCallback;
    }

    if (callback != NULL) {
      callback->onVSyncEvent(when); //【见小节3.9】
    }
}

3.9 ET.onVSyncEvent

[-> EventThread.java]

void EventThread::onVSyncEvent(nsecs_t timestamp) {
    Mutex::Autolock _l(mLock);
    mVSyncEvent[0].header.type = DisplayEventReceiver::DISPLAY_EVENT_VSYNC;
    mVSyncEvent[0].header.id = 0;
    mVSyncEvent[0].header.timestamp = timestamp;
    mVSyncEvent[0].vsync.count++;
    mCondition.broadcast(); //唤醒EventThread线程
}

mCondition.broadcast能够唤醒处理waitForEvent()过程的EventThread【见小节2.7.2】,并往下执行conn的postEvent().

3.10 ET.postEvent

[-> EventThread.java]

status_t EventThread::Connection::postEvent(
        const DisplayEventReceiver::Event& event) {
    ssize_t size = DisplayEventReceiver::sendEvents(mChannel, &event, 1);
    return size < 0 ? status_t(size) : status_t(NO_ERROR);
}

3.11 DER.sendEvents

[-> DisplayEventReceiver.cpp]

ssize_t DisplayEventReceiver::sendEvents(const sp<BitTube>& dataChannel,
        Event const* events, size_t count)
{
    return BitTube::sendObjects(dataChannel, events, count);
}

根据小节【2.8】可知监听BitTube,此处调用BitTube来sendObjects。一旦收到数据,则调用MQ.cb_eventReceiver()方法。

3.11.1 MQ.cb_eventReceiver

[-> MessageQueue.cpp]

int MessageQueue::cb_eventReceiver(int fd, int events, void* data) {
    MessageQueue* queue = reinterpret_cast<MessageQueue *>(data);
    return queue->eventReceiver(fd, events);
}

3.11.2 MQ.eventReceiver

[-> MessageQueue.cpp]

int MessageQueue::eventReceiver(int /*fd*/, int /*events*/) {
    ssize_t n;
    DisplayEventReceiver::Event buffer[8];
    while ((n = DisplayEventReceiver::getEvents(mEventTube, buffer, 8)) > 0) {
        for (int i=0 ; i<n ; i++) {
            if (buffer[i].header.type == DisplayEventReceiver::DISPLAY_EVENT_VSYNC) {
#if INVALIDATE_ON_VSYNC
                mHandler->dispatchInvalidate();
#else
                mHandler->dispatchRefresh(); //【见小节3.12】
#endif
                break;
            }
        }
    }
    return 1;
}

3.12 MQ.dispatchRefresh

void MessageQueue::Handler::dispatchRefresh() {
    if ((android_atomic_or(eventMaskRefresh, &mEventMask) & eventMaskRefresh) == 0) {
        //发送消息,则进入handleMessage过程【见小节3.13】
        mQueue.mLooper->sendMessage(this, Message(MessageQueue::REFRESH));
    }
}

3.13 MQ.handleMessage

void MessageQueue::Handler::handleMessage(const Message& message) {
    switch (message.what) {
        case INVALIDATE:
            android_atomic_and(~eventMaskInvalidate, &mEventMask);
            mQueue.mFlinger->onMessageReceived(message.what);
            break;
        case REFRESH:
            android_atomic_and(~eventMaskRefresh, &mEventMask);
            mQueue.mFlinger->onMessageReceived(message.what);//【见小节3.14】
            break;
        case TRANSACTION:
            android_atomic_and(~eventMaskTransaction, &mEventMask);
            mQueue.mFlinger->onMessageReceived(message.what);
            break;
    }
}

对于REFRESH操作,则进入onMessageReceived().

3.14 SF.onMessageReceived

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::onMessageReceived(int32_t what) {
    ATRACE_CALL();
    switch (what) {
        case MessageQueue::TRANSACTION: {
            handleMessageTransaction();
            break;
        }
        case MessageQueue::INVALIDATE: {
            bool refreshNeeded = handleMessageTransaction();
            refreshNeeded |= handleMessageInvalidate();
            refreshNeeded |= mRepaintEverything;
            if (refreshNeeded) {
                signalRefresh();
            }
            break;
        }
        case MessageQueue::REFRESH: {
            handleMessageRefresh();
            break;
        }
    }
}

3.15 SF.handleMessageRefresh

[-> SurfaceFlinger.cpp]

void SurfaceFlinger::handleMessageRefresh() {
    ATRACE_CALL();
    preComposition();
    rebuildLayerStacks();
    setUpHWComposer();
    doDebugFlashRegions();
    doComposition();
    postComposition();
}

下一篇文章,再来介绍图形输出过程。

四 总结

前面讲述过程中所涉及到的线程情况:

  • 主线程“/system/bin/surfaceflinger”: 主线程
  • 线程”EventThread”:EventThread
  • 线程”EventControl”: EventControlThread
  • 线程”DispSync”:DispSyncThread

Vsync处理流程图:点击查看大图

vsync

  1. 底层vsync信号发送过来,一路执行到【小节3.6】DispSyncThread.updateModel()方法中调用mCond.signal() 来唤醒DispSyncThread线程;
  2. DispSyncThread线程:执行到【小节3.9】EventThread::onVSyncEvent()方法中调用mCondition.broadcast() 唤醒EventThread线程;
  3. EventThread线程:执行到【小节3.11】DisplayEventReceiver::sendEvents()方法中调用BitTube::sendObjects(); 由【小节2.8】可知当收到数据则调用MQ.cb_eventReceiver(),然后再经过handler消息机制,进入SurfaceFlinger主线程;
  4. SurfaceFlinger主线程:【小节3.13】进入到MesageQueue的handleMessage(),最终调用SurfaceFlinger的handleMessageRefresh()。

微信公众号 Gityuan | 微博 weibo.com/gityuan | 博客 留言区交流