client端的bindService()
我们在Application或者Activity里调用bindService()的时候，其实调用的是Context中的抽象方法：
public abstract boolean bindService(@RequiresPermission Intent service,
@NonNull ServiceConnection conn, @BindServiceFlags int flags);
复制代码真正的实现在ContextImpl中：
@Override
public boolean bindService(Intent service, ServiceConnection conn,
int flags) {
warnIfCallingFromSystemProcess();
return bindServiceCommon(service, conn, flags, mMainThread.getHandler(), Process.myUserHandle());
}
复制代码而bindServiceCommon()方法如下:
private boolean bindServiceCommon(Intent service, ServiceConnection conn, int flags, Handler handler, UserHandle user) {
IServiceConnection sd;
if (conn == null) {
throw new IllegalArgumentException("connection is null");
}
if (mPackageInfo != null) {
//获取的是内部静态类LoadedApk.ServiceDispatcher.InnerConnection
sd = mPackageInfo.getServiceDispatcher(conn, getOuterContext(), handler, flags);
} else {
throw new RuntimeException("Not supported in system context");
}
validateServiceIntent(service);
try {
IBinder token = getActivityToken();
if (token == null && (flags&BIND_AUTO_CREATE) == 0 && mPackageInfo != null
&& mPackageInfo.getApplicationInfo().targetSdkVersion
< android.os.Build.VERSION_CODES.ICE_CREAM_SANDWICH) {
flags |= BIND_WAIVE_PRIORITY;
}
service.prepareToLeaveProcess(this);
//通过IActivityManager接口,向AMS发送bind请求.
int res = ActivityManagerNative.getDefault().bindService(
mMainThread.getApplicationThread(),//ApplicationThread对象, 该对象继承于ApplicationThreadNative(Binder服务端)
getActivityToken(), service,
service.resolveTypeIfNeeded(getContentResolver()),
sd, flags, getOpPackageName(), user.getIdentifier());
if (res < 0) {
throw new SecurityException(
"Not allowed to bind to service " + service);
}
return res != 0;
} catch (RemoteException e) {
throw e.rethrowFromSystemServer();
}
}
复制代码该方法主要做了几件事：
1.调用LoadedAPK对象的getServiceDispatcher()方法。
2.如果getActivityToken()为null, 或者flags不为BIND_AUTO_CREATE,并且targetSdkVersion小于14的话，flags会添加BIND_WAIVE_PRIORITY标志，这个标志的含义是弃权，即不影响提供Service的进程的优先级。
3.调用AMS.bindService()时，会传递getActivityToken()的值和一些其他参数,然后通过IPC调用到AMS的bindService()方法。
LoadedApk.getServiceDispatcher()
public final IServiceConnection getServiceDispatcher(ServiceConnection c,
Context context, Handler handler, int flags) {
synchronized (mServices) {
LoadedApk.ServiceDispatcher sd = null;
ArrayMap<ServiceConnection, LoadedApk.ServiceDispatcher> map = mServices.get(context);
if (map != null) {
sd = map.get(c);
}
if (sd == null) {
//创建服务分发对象
sd = new ServiceDispatcher(c, context, handler, flags);
if (map == null) {
map = new ArrayMap<ServiceConnection, LoadedApk.ServiceDispatcher>();
mServices.put(context, map);
}
//以ServiceConnection为key, ServiceDispatcher为value保存到map
map.put(c, sd);
} else {
sd.validate(context, handler);
}
//返回的内部类的对象LoadedApk.ServiceDispatcher.InnerConnection，
//该对象继承于IServiceConnection.Stub, 该类是由IServiceConnection.aidl自动生成的 作为binder服务端。
return sd.getIServiceConnection();
}
}
复制代码mServices是一个Map，里面缓存了ServiceDispatcher对象。
private final ArrayMap<Context, ArrayMap<ServiceConnection, LoadedApk.ServiceDispatcher>> mServices = new ArrayMap<Context, ArrayMap<ServiceConnection, LoadedApk.ServiceDispatcher>>();
如果没有取到就新建一个ServiceDispatcher对象，而ServiceConnection就是此时保存到ServiceDispatcher对象中的。
最后返回的是IServiceConnection对象，它是一个IPC接口, 而IServiceConnection对象的初始化就是在ServiceDispatcher的构造方法中。这个IServiceConnection的binder会在IPC时传递到AMS中。
IServiceConnection的aidl定义如下:
import android.content.ComponentName;
/** @hide */
oneway interface IServiceConnection {
void connected(in ComponentName name, IBinder service);
}
复制代码connected方法没有返回值，ComponentName参数也是in类型的。对于server端是单向的。
AMS.bindService()
public int bindService(IApplicationThread caller, IBinder token, Intent service,
String resolvedType, IServiceConnection connection, int flags, String callingPackage,
int userId) throws TransactionTooLargeException {
enforceNotIsolatedCaller("bindService");
// Refuse possible leaked file descriptors
if (service != null && service.hasFileDescriptors() == true) {
throw new IllegalArgumentException("File descriptors passed in Intent");
}
if (callingPackage == null) {
throw new IllegalArgumentException("callingPackage cannot be null");
}
synchronized(this) {
//锁定绑定服务操作
return mServices.bindServiceLocked(caller, token, service,
resolvedType, connection, flags, callingPackage, userId);
}
}
复制代码mServices是ActiveServices对象，其bindServiceLocked()方法如下:
int bindServiceLocked(IApplicationThread caller, IBinder token, Intent service,
String resolvedType, final IServiceConnection connection, int flags,
String callingPackage, final int userId) throws TransactionTooLargeException {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "bindService: " + service
+ " type=" + resolvedType + " conn=" + connection.asBinder()
+ " flags=0x" + Integer.toHexString(flags));
//查询发起端所对应的进程记录结构
final ProcessRecord callerApp = mAm.getRecordForAppLocked(caller);
if (callerApp == null) {
throw new SecurityException(
"Unable to find app for caller " + caller
+ " (pid=" + Binder.getCallingPid()
+ ") when binding service " + service);
}
ActivityRecord activity = null;
if (token != null) {
activity = ActivityRecord.isInStackLocked(token);
if (activity == null) {
Slog.w(TAG, "Binding with unknown activity: " + token);
//发起方的activity为空，则直接返回
return 0;
}
}
int clientLabel = 0;
PendingIntent clientIntent = null;
final boolean isCallerSystem = callerApp.info.uid == Process.SYSTEM_UID;
if (isCallerSystem) {
// Hacky kind of thing -- allow system stuff to tell us
// what they are, so we can report this elsewhere for
// others to know why certain services are running.
//发起方是system进程的情况
service.setDefusable(true);
clientIntent = service.getParcelableExtra(Intent.EXTRA_CLIENT_INTENT);
if (clientIntent != null) {
clientLabel = service.getIntExtra(Intent.EXTRA_CLIENT_LABEL, 0);
if (clientLabel != 0) {
// There are no useful extras in the intent, trash them.
// System code calling with this stuff just needs to know
// this will happen.
service = service.cloneFilter();
}
}
}
if ((flags&Context.BIND_TREAT_LIKE_ACTIVITY) != 0) {
mAm.enforceCallingPermission(android.Manifest.permission.MANAGE_ACTIVITY_STACKS,
"BIND_TREAT_LIKE_ACTIVITY");
}
if ((flags & Context.BIND_ALLOW_WHITELIST_MANAGEMENT) != 0 && !isCallerSystem) {
throw new SecurityException(
"Non-system caller " + caller + " (pid=" + Binder.getCallingPid()
+ ") set BIND_ALLOW_WHITELIST_MANAGEMENT when binding service " + service);
}
//根据发送端所在进程的SchedGroup来决定是否为前台service.
final boolean callerFg = callerApp.setSchedGroup != ProcessList.SCHED_GROUP_BACKGROUND;
final boolean isBindExternal = (flags & Context.BIND_EXTERNAL_SERVICE) != 0;
//根据用户传递进来Intent来检索相对应的服务
ServiceLookupResult res =
retrieveServiceLocked(service, resolvedType, callingPackage, Binder.getCallingPid(),
Binder.getCallingUid(), userId, true, callerFg, isBindExternal);
if (res == null) {
return 0;
}
if (res.record == null) {
return -1;
}
//查询到相应的Service
ServiceRecord s = res.record;
boolean permissionsReviewRequired = false;
// If permissions need a review before any of the app components can run,
// we schedule binding to the service but do not start its process, then
// we launch a review activity to which is passed a callback to invoke
// when done to start the bound service's process to completing the binding.
if (Build.PERMISSIONS_REVIEW_REQUIRED) {
if (mAm.getPackageManagerInternalLocked().isPermissionsReviewRequired(
s.packageName, s.userId)) {
permissionsReviewRequired = true;
// Show a permission review UI only for binding from a foreground app
if (!callerFg) {
Slog.w(TAG, "u" + s.userId + " Binding to a service in package"
+ s.packageName + " requires a permissions review");
return 0;
}
final ServiceRecord serviceRecord = s;
final Intent serviceIntent = service;
RemoteCallback callback = new RemoteCallback(
new RemoteCallback.OnResultListener() {
@Override
public void onResult(Bundle result) {
synchronized(mAm) {
final long identity = Binder.clearCallingIdentity();
try {
if (!mPendingServices.contains(serviceRecord)) {
return;
}
// If there is still a pending record, then the service
// binding request is still valid, so hook them up. We
// proceed only if the caller cleared the review requirement
// otherwise we unbind because the user didn't approve.
if (!mAm.getPackageManagerInternalLocked()
.isPermissionsReviewRequired(
serviceRecord.packageName,
serviceRecord.userId)) {
try {
bringUpServiceLocked(serviceRecord,
serviceIntent.getFlags(),
callerFg, false, false);
} catch (RemoteException e) {
/* ignore - local call */
}
} else {
unbindServiceLocked(connection);
}
} finally {
Binder.restoreCallingIdentity(identity);
}
}
}
});
final Intent intent = new Intent(Intent.ACTION_REVIEW_PERMISSIONS);
intent.addFlags(Intent.FLAG_ACTIVITY_NEW_TASK
| Intent.FLAG_ACTIVITY_EXCLUDE_FROM_RECENTS);
intent.putExtra(Intent.EXTRA_PACKAGE_NAME, s.packageName);
intent.putExtra(Intent.EXTRA_REMOTE_CALLBACK, callback);
if (DEBUG_PERMISSIONS_REVIEW) {
Slog.i(TAG, "u" + s.userId + " Launching permission review for package "
+ s.packageName);
}
mAm.mHandler.post(new Runnable() {
@Override
public void run() {
mAm.mContext.startActivityAsUser(intent, new UserHandle(userId));
}
});
}
}
final long origId = Binder.clearCallingIdentity();
try {
//取消服务的重启调度
if (unscheduleServiceRestartLocked(s, callerApp.info.uid, false)) {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "BIND SERVICE WHILE RESTART PENDING: " + s);
}
if ((flags&Context.BIND_AUTO_CREATE) != 0) {
//更新当前service活动时间
s.lastActivity = SystemClock.uptimeMillis();
if (!s.hasAutoCreateConnections()) {
// This is the first binding, let the tracker know.
ServiceState stracker = s.getTracker();
if (stracker != null) {
stracker.setBound(true, mAm.mProcessStats.getMemFactorLocked(),
s.lastActivity);
}
}
}
mAm.startAssociationLocked(callerApp.uid, callerApp.processName, callerApp.curProcState,
s.appInfo.uid, s.name, s.processName);
AppBindRecord b = s.retrieveAppBindingLocked(service, callerApp);
//创建对象ConnectionRecord,此处connection来自发起方
ConnectionRecord c = new ConnectionRecord(b, activity,
connection, flags, clientLabel, clientIntent);
IBinder binder = connection.asBinder();
ArrayList<ConnectionRecord> clist = s.connections.get(binder);
if (clist == null) {
clist = new ArrayList<ConnectionRecord>();
s.connections.put(binder, clist);
}
// clist是ServiceRecord.connections的成员变量
clist.add(c);
//b是指AppBindRecord
b.connections.add(c);
if (activity != null) {
if (activity.connections == null) {
activity.connections = new HashSet<ConnectionRecord>();
}
activity.connections.add(c);
}
b.client.connections.add(c);
if ((c.flags&Context.BIND_ABOVE_CLIENT) != 0) {
b.client.hasAboveClient = true;
}
if ((c.flags&Context.BIND_ALLOW_WHITELIST_MANAGEMENT) != 0) {
s.whitelistManager = true;
}
if (s.app != null) {
updateServiceClientActivitiesLocked(s.app, c, true);
}
clist = mServiceConnections.get(binder);
if (clist == null) {
clist = new ArrayList<ConnectionRecord>();
mServiceConnections.put(binder, clist);
}
clist.add(c);
if ((flags&Context.BIND_AUTO_CREATE) != 0) {
s.lastActivity = SystemClock.uptimeMillis();
//启动service，这个过程跟startService过程一致
if (bringUpServiceLocked(s, service.getFlags(), callerFg, false,
permissionsReviewRequired) != null) {
return 0;
}
}
if (s.app != null) {
if ((flags&Context.BIND_TREAT_LIKE_ACTIVITY) != 0) {
s.app.treatLikeActivity = true;
}
if (s.whitelistManager) {
s.app.whitelistManager = true;
}
// This could have made the service more important.
//更新service所在进程的优先级
mAm.updateLruProcessLocked(s.app, s.app.hasClientActivities
|| s.app.treatLikeActivity, b.client);
mAm.updateOomAdjLocked(s.app);
}
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Bind " + s + " with " + b
+ ": received=" + b.intent.received
+ " apps=" + b.intent.apps.size()
+ " doRebind=" + b.intent.doRebind);
if (s.app != null && b.intent.received) {
// Service is already running, so we can immediately
// publish the connection.
try {
//Service已经正在运行，则调用InnerConnection的代理对象
c.conn.connected(s.name, b.intent.binder);
} catch (Exception e) {
Slog.w(TAG, "Failure sending service " + s.shortName
+ " to connection " + c.conn.asBinder()
+ " (in " + c.binding.client.processName + ")", e);
}
// If this is the first app connected back to this binding,
// and the service had previously asked to be told when
// rebound, then do so.
if (b.intent.apps.size() == 1 && b.intent.doRebind) {
//rebind过程
requestServiceBindingLocked(s, b.intent, callerFg, true);
}
} else if (!b.intent.requested) {
requestServiceBindingLocked(s, b.intent, callerFg, false);
}
getServiceMap(s.userId).ensureNotStartingBackground(s);
} finally {
Binder.restoreCallingIdentity(origId);
}
return 1;
}
复制代码方法调用过程主要包括：
- 根据传递的IApplicationThread，通过AMS.getRecordForAppLocked()方法，从LRU缓存中获取ProcessRecord对象;这里可以保证一定能从缓存中取到，因为在App调用bindService()之前，所在的进程一定已经启动，只要启动了就会在AMS中留下进程记录。
- 前面提到的Activity的token在这里用上了，如果token不为null,就会从缓存中取出ActivityRecord, 类似地，如果是从Activity中调用bindService()，由于Activity已经启动，所以这里也一定可以取到ActivityRecord对象;
- 如果是系统uid,那么会从Intent中提取出clientIntent;
- 调用retrieveServiceLocked()方法来获取ServiceLookupResult这个查询结果，这个方法会先尝试从缓存中取出ServiceRecord对象，如果没有则新建ServiceRecord并且存入缓存中，最后返回的ServiceLookupResult是ServiceRecord的包装类;
- 调用unscheduleServiceRestartLocked(),即如果要绑定的Service在重启名单中，那么就将它从AMS的mHandler中移除重启的Callback;
- 调用ServiceRecord的retrieveAppBindingLocked()从缓存中获取或者新建AppBindRecord对象；
- 然后新建ConnectionRecord对象,这个对象包含的信息非常多，包括AppBindRecord对象，ActivityRecord对象，IServiceConnection代理，flags, clientLabel和clientIntent;
- 之后将ConnectionRecord对象放入缓存中，这个缓存的key是IServiceConnection的binder, value是ArrayList，这说明一个IServiceConnection可能对应多个ConnectionRecord,其实这也很好理解，因为一个client进程只有一个用于处理服务连接状态的IServiceConnection,但是一个client进程中却可能有多个连接;
- 之后，如果ServiceRecord中的ProcessRecord不为空，则调用updateServiceClientActivitiesLocked()方法，这个方法主要作用就是查找client进程中是否存在activity,如果存在则更新AMS中ProcessRecord的LRU记录;
- 如果flags中含有Context.BIND_AUTO_CREATE标志，那么调用bringUpServiceLocked()将启动Service，这个过程还可能伴随启动Service所在进程(如果进程没有启动的话);
service端进程启动以及service启动
ActiveServices.bringUpServiceLocked()
private String bringUpServiceLocked(ServiceRecord r, int intentFlags, boolean execInFg,
boolean whileRestarting, boolean permissionsReviewRequired)
throws TransactionTooLargeException {
//Slog.i(TAG, "Bring up service:");
//r.dump(" ");
// 进程已存在的情况
if (r.app != null && r.app.thread != null) {
//调用service.onStartCommand()过程
sendServiceArgsLocked(r, execInFg, false);
return null;
}
if (!whileRestarting && r.restartDelay > 0) {
// If waiting for a restart, then do nothing.
//等待延迟重启的过程，则直接返回
return null;
}
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Bringing up " + r + " " + r.intent);
// We are now bringing the service up, so no longer in the
// restarting state.
// 启动service前，把service从重启服务队列中移除
if (mRestartingServices.remove(r)) {
r.resetRestartCounter();
clearRestartingIfNeededLocked(r);
}
// Make sure this service is no longer considered delayed, we are starting it now.
if (r.delayed) {
//service正在启动，将delayed设置为false
if (DEBUG_DELAYED_STARTS) Slog.v(TAG_SERVICE, "REM FR DELAY LIST (bring up): " + r);
getServiceMap(r.userId).mDelayedStartList.remove(r);
r.delayed = false;
}
// Make sure that the user who owns this service is started. If not,
// we don't want to allow it to run.
if (!mAm.mUserController.hasStartedUserState(r.userId)) {
//确保拥有该服务的user已经启动，否则停止；
String msg = "Unable to launch app "
+ r.appInfo.packageName + "/"
+ r.appInfo.uid + " for service "
+ r.intent.getIntent() + ": user " + r.userId + " is stopped";
Slog.w(TAG, msg);
bringDownServiceLocked(r);
return msg;
}
// Service is now being launched, its package can't be stopped.
try {
//服务正在启动，设置package停止状态为false
AppGlobals.getPackageManager().setPackageStoppedState(r.packageName, false, r.userId);
} catch (RemoteException e) {
} catch (IllegalArgumentException e) {
Slog.w(TAG, "Failed trying to unstop package "
+ r.packageName + ": " + e);
}
final boolean isolated = (r.serviceInfo.flags&ServiceInfo.FLAG_ISOLATED_PROCESS) != 0;
final String procName = r.processName;
ProcessRecord app;
if (!isolated) {
//根据进程名和uid，查询ProcessRecord
app = mAm.getProcessRecordLocked(procName, r.appInfo.uid, false);
if (DEBUG_MU) Slog.v(TAG_MU, "bringUpServiceLocked: appInfo.uid=" + r.appInfo.uid
+ " app=" + app);
if (app != null && app.thread != null) {
try {
app.addPackage(r.appInfo.packageName, r.appInfo.versionCode, mAm.mProcessStats);
// 启动服务
realStartServiceLocked(r, app, execInFg);
return null;
} catch (TransactionTooLargeException e) {
throw e;
} catch (RemoteException e) {
Slog.w(TAG, "Exception when starting service " + r.shortName, e);
}
// If a dead object exception was thrown -- fall through to
// restart the application.
}
} else {
// If this service runs in an isolated process, then each time
// we call startProcessLocked() we will get a new isolated
// process, starting another process if we are currently waiting
// for a previous process to come up. To deal with this, we store
// in the service any current isolated process it is running in or
// waiting to have come up.
app = r.isolatedProc;
}
// Not running -- get it started, and enqueue this service record
// to be executed when the app comes up.
if (app == null && !permissionsReviewRequired) {//对于进程没有启动的情况
//启动service所要运行的进程
if ((app=mAm.startProcessLocked(procName, r.appInfo, true, intentFlags,
"service", r.name, false, isolated, false)) == null) {
String msg = "Unable to launch app "
+ r.appInfo.packageName + "/"
+ r.appInfo.uid + " for service "
+ r.intent.getIntent() + ": process is bad";
Slog.w(TAG, msg);
// 进程启动失败
bringDownServiceLocked(r);
return msg;
}
if (isolated) {
r.isolatedProc = app;
}
}
if (!mPendingServices.contains(r)) {
mPendingServices.add(r);
}
if (r.delayedStop) {
// Oh and hey we've already been asked to stop!
r.delayedStop = false;
if (r.startRequested) {
if (DEBUG_DELAYED_STARTS) Slog.v(TAG_SERVICE,
"Applying delayed stop (in bring up): " + r);
//停止服务
stopServiceLocked(r);
}
}
return null;
}
复制代码方法主要分为两部分，进程已启动和未启动
进程已启动
首先尝试从AMS.getProcessRecordLocked()方法中获取ProcessRecord,如果获取到了，说明进程已经启动，就可以调用realStartServiceLocked()方法启动Service了:
private final void realStartServiceLocked(ServiceRecord r,
ProcessRecord app, boolean execInFg) throws RemoteException {
if (app.thread == null) {
throw new RemoteException();
}
if (DEBUG_MU)
Slog.v(TAG_MU, "realStartServiceLocked, ServiceRecord.uid = " + r.appInfo.uid
+ ", ProcessRecord.uid = " + app.uid);
r.app = app;
r.restartTime = r.lastActivity = SystemClock.uptimeMillis();
final boolean newService = app.services.add(r);
//发送delay消息
bumpServiceExecutingLocked(r, execInFg, "create");
mAm.updateLruProcessLocked(app, false, null);
mAm.updateOomAdjLocked();
boolean created = false;
try {
if (LOG_SERVICE_START_STOP) {
String nameTerm;
int lastPeriod = r.shortName.lastIndexOf('.');
nameTerm = lastPeriod >= 0 ? r.shortName.substring(lastPeriod) : r.shortName;
EventLogTags.writeAmCreateService(
r.userId, System.identityHashCode(r), nameTerm, r.app.uid, r.app.pid);
}
synchronized (r.stats.getBatteryStats()) {
r.stats.startLaunchedLocked();
}
mAm.notifyPackageUse(r.serviceInfo.packageName,
PackageManager.NOTIFY_PACKAGE_USE_SERVICE);
app.forceProcessStateUpTo(ActivityManager.PROCESS_STATE_SERVICE);
//服务进入 onCreate()
app.thread.scheduleCreateService(r, r.serviceInfo,
mAm.compatibilityInfoForPackageLocked(r.serviceInfo.applicationInfo),
app.repProcState);
r.postNotification();
created = true;
} catch (DeadObjectException e) {
Slog.w(TAG, "Application dead when creating service " + r);
//应用死亡处理
mAm.appDiedLocked(app);
throw e;
} finally {
if (!created) {
// Keep the executeNesting count accurate.
final boolean inDestroying = mDestroyingServices.contains(r);
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
// Cleanup.
if (newService) {
app.services.remove(r);
r.app = null;
}
// Retry.
//尝试重新启动服务
if (!inDestroying) {
scheduleServiceRestartLocked(r, false);
}
}
}
if (r.whitelistManager) {
app.whitelistManager = true;
}
requestServiceBindingsLocked(r, execInFg);
updateServiceClientActivitiesLocked(app, null, true);
// If the service is in the started state, and there are no
// pending arguments, then fake up one so its onStartCommand() will
// be called.
if (r.startRequested && r.callStart && r.pendingStarts.size() == 0) {
r.pendingStarts.add(new ServiceRecord.StartItem(r, false, r.makeNextStartId(),
null, null));
}
//服务 进入onStartCommand()
sendServiceArgsLocked(r, execInFg, true);
if (r.delayed) {
if (DEBUG_DELAYED_STARTS) Slog.v(TAG_SERVICE, "REM FR DELAY LIST (new proc): " + r);
getServiceMap(r.userId).mDelayedStartList.remove(r);
r.delayed = false;
}
if (r.delayedStop) {
// Oh and hey we've already been asked to stop!
r.delayedStop = false;
if (r.startRequested) {
if (DEBUG_DELAYED_STARTS) Slog.v(TAG_SERVICE,
"Applying delayed stop (from start): " + r);
stopServiceLocked(r); //停止服务
}
}
}
复制代码方法主要包括了：
- 调用AMS.updateLruProcessLocked()方法更新ProcessRecord的LRU缓存;
- 调用AMS.updateOomAdjLocked()方法更新Service所在进程的oom_adj值;
- 通过ProcessRecord中的IApplicationThread的代理，通过IPC调用scheduleCreateService()方法，进行Service的创建，这个过程发生在ApplicationThread中;
- 如果发现不需要重新创建(比如已经创建了),那么就进行一些状态维护即可;
- 之后调用requestServiceBindingsLocked()方法，这个方法非常重要，就是在这里，把Service的onBind()与要提供的服务关联起来了。
private final void requestServiceBindingsLocked(ServiceRecord r, boolean execInFg)
throws TransactionTooLargeException {
for (int i=r.bindings.size()-1; i>=0; i--) {
IntentBindRecord ibr = r.bindings.valueAt(i);
if (!requestServiceBindingLocked(r, ibr, execInFg, false)) {
break;
}
}
}
复制代码这里是遍历ServiceRecord中的IntentBindRecord,即所有的绑定记录，然后调用requestServiceBindingLocked()方法，只要有一个执行成功即中断。而requestServiceBindingLocked()方法如下:
private final boolean requestServiceBindingLocked(ServiceRecord r, IntentBindRecord i,
boolean execInFg, boolean rebind) throws TransactionTooLargeException {
if (r.app == null || r.app.thread == null) {
// If service is not currently running, can't yet bind.
return false;
}
if ((!i.requested || rebind) && i.apps.size() > 0) {
try {
//发送bind开始的消息
bumpServiceExecutingLocked(r, execInFg, "bind");
r.app.forceProcessStateUpTo(ActivityManager.PROCESS_STATE_SERVICE);
//服务进入 onBind()
r.app.thread.scheduleBindService(r, i.intent.getIntent(), rebind, r.app.repProcState);
if (!rebind) {
i.requested = true;
}
i.hasBound = true;
i.doRebind = false;
} catch (TransactionTooLargeException e) {
// Keep the executeNesting count accurate.
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Crashed while binding " + r, e);
final boolean inDestroying = mDestroyingServices.contains(r);
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
throw e;
} catch (RemoteException e) {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Crashed while binding " + r);
// Keep the executeNesting count accurate.
final boolean inDestroying = mDestroyingServices.contains(r);
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
return false;
}
}
return true;
}
复制代码这个方法的重点是调用ApplicationThread的代理，通过IPC调用scheduleBindService(), 注意这里会将ServiceRecord(继承自Binder)传递过去(后面要给它赋值),而ApplicationThread中的scheduleBindService()方法如下:
public final void scheduleBindService(IBinder token, Intent intent,
boolean rebind, int processState) {
updateProcessState(processState, false);
BindServiceData s = new BindServiceData();
s.token = token;
s.intent = intent;
s.rebind = rebind;
if (DEBUG_SERVICE)
Slog.v(TAG, "scheduleBindService token=" + token + " intent=" + intent + " uid="
+ Binder.getCallingUid() + " pid=" + Binder.getCallingPid());
sendMessage(H.BIND_SERVICE, s);
}
复制代码这个方法很简单，就是新建了一个包装类BindServiceData对象，然后通过H传递消息到主线程，之后调用handleBindService()方法:
private void handleBindService(BindServiceData data) {
Service s = mServices.get(data.token);
if (DEBUG_SERVICE)
Slog.v(TAG, "handleBindService s=" + s + " rebind=" + data.rebind);
if (s != null) {
try {
data.intent.setExtrasClassLoader(s.getClassLoader());
data.intent.prepareToEnterProcess();
try {
if (!data.rebind) {
// 执行Service.onBind()回调方法
IBinder binder = s.onBind(data.intent);
//将onBind返回值传递回去
ActivityManagerNative.getDefault().publishService(data.token, data.intent, binder);
} else {
s.onRebind(data.intent);
ActivityManagerNative.getDefault().serviceDoneExecuting(
data.token, SERVICE_DONE_EXECUTING_ANON, 0, 0);
}
ensureJitEnabled();
} catch (RemoteException ex) {
throw ex.rethrowFromSystemServer();
}
} catch (Exception e) {
if (!mInstrumentation.onException(s, e)) {
throw new RuntimeException(
"Unable to bind to service " + s
+ " with " + data.intent + ": " + e.toString(), e);
}
}
}
}
复制代码这里的data.token就是IPC调用传递过来的ServiceRecord对象，由于在前面已经通过IPC调用启动了Service对象，所以这里可以通过data.token从缓存中取出Service对象，然后分两种情况:
- 如果不是重连，那么就调用Service的onBind()方法获取IBinder,而获取的这个binder会在pushlishService()时传递到AMS中，之后又会通过AMS到client进程中，这也就是client中获取到Service中onBind()所返回的binder的原因。之后就进行IPC调用，最终会调用到AMS.publishService()方法;
- 否则回调Service的onRebind()方法,这个方法经常被我们忽视，实际上也很重要。之后通过IPC调用到AMS的serviceDoneExecuting()方法。
Service侧的调用就基本结束了，主要是AMS对其的两次IPC调用，分别是IApplicationThread的scheduleCreateService()和scheduleBindService(), 前者是为了新建Service, 而后者是获取Service的onBind()所返回的IBinder对象。
进程未启动
对于还未启动或者进程已经回收的进程的情况，需要先调用AMS.startProcessLocked()方法来启动进程，并且等到进程启动完成之后再进行绑定。
// Not running -- get it started, and enqueue this service record
// to be executed when the app comes up.
if (app == null && !permissionsReviewRequired) {//对于进程没有启动的情况
//启动service所要运行的进程
if ((app=mAm.startProcessLocked(procName, r.appInfo, true, intentFlags,
"service", r.name, false, isolated, false)) == null) {
String msg = "Unable to launch app "
+ r.appInfo.packageName + "/"
+ r.appInfo.uid + " for service "
+ r.intent.getIntent() + ": process is bad";
Slog.w(TAG, msg);
// 进程启动失败
bringDownServiceLocked(r);
return msg;
}
if (isolated) {
r.isolatedProc = app;
}
}
if (!mPendingServices.contains(r)) {
mPendingServices.add(r);
}
复制代码将ServiceRecord先放到mPendingServices中,这个mPendingServices顾名思义，就是暂时挂起的服务。在启动Service所在进程后，会通过IPC调用到AMS的attachApplication()—>attachApplicationLocked()方法:
// Find any services that should be running in this process...
if (!badApp) {
try {
//寻找所有需要在该进程中运行的服务
didSomething |= mServices.attachApplicationLocked(app, processName);
} catch (Exception e) {
Slog.wtf(TAG, "Exception thrown starting services in " + app, e);
badApp = true;
}
}
复制代码注释已经写得很清楚了，就是执行到这里时，查找那些本来应该运行在这个进程的Service(由于进程没启动而暂时挂起的那些), 然后调用到ActiveServices的attachApplicationLocked()方法:
boolean attachApplicationLocked(ProcessRecord proc, String processName)
throws RemoteException {
boolean didSomething = false;
// Collect any services that are waiting for this process to come up.
//启动mPendingServices队列中，等待在该进程启动的服务
if (mPendingServices.size() > 0) {
ServiceRecord sr = null;
try {
for (int i=0; i<mPendingServices.size(); i++) {
sr = mPendingServices.get(i);
if (proc != sr.isolatedProc && (proc.uid != sr.appInfo.uid
|| !processName.equals(sr.processName))) {
continue;
}
mPendingServices.remove(i);
i--;
// 将当前服务的包信息加入到proc
proc.addPackage(sr.appInfo.packageName, sr.appInfo.versionCode,
mAm.mProcessStats);
// 启动服务，即将进入服务的生命周期
realStartServiceLocked(sr, proc, sr.createdFromFg);
didSomething = true;
if (!isServiceNeeded(sr, false, false)) {
// We were waiting for this service to start, but it is actually no
// longer needed. This could happen because bringDownServiceIfNeeded
// won't bring down a service that is pending... so now the pending
// is done, so let's drop it.
bringDownServiceLocked(sr);
}
}
} catch (RemoteException e) {
Slog.w(TAG, "Exception in new application when starting service "
+ sr.shortName, e);
throw e;
}
}
// Also, if there are any services that are waiting to restart and
// would run in this process, now is a good time to start them. It would
// be weird to bring up the process but arbitrarily not let the services
// run at this point just because their restart time hasn't come up.
// 对于正在等待重启并需要运行在该进程的服务，现在是启动它们的大好时机
if (mRestartingServices.size() > 0) {
ServiceRecord sr;
for (int i=0; i<mRestartingServices.size(); i++) {
sr = mRestartingServices.get(i);
if (proc != sr.isolatedProc && (proc.uid != sr.appInfo.uid
|| !processName.equals(sr.processName))) {
continue;
}
mAm.mHandler.removeCallbacks(sr.restarter);
mAm.mHandler.post(sr.restarter);
}
}
return didSomething;
}
复制代码在这个方法里面，会调用realStarServiceLocked()方法完成Service的创建和绑定工作,如前所述。
AMS中更新oom_adj值
由于在上一步可能进行启动Service所在进程的操作，所以在这里再次对s.app是否为null进行判断，然后同样也更新了ProcessRecord的LRU记录；并且，有一个很重要的操作是调用了AMS的updateOomAdjLocked()方法更新client进程的oom_adj值，该方法如下:
final boolean updateOomAdjLocked(ProcessRecord app) {
final ActivityRecord TOP_ACT = resumedAppLocked();
final ProcessRecord TOP_APP = TOP_ACT != null ? TOP_ACT.app : null;
final boolean wasCached = app.cached;
mAdjSeq++;
// This is the desired cached adjusment we want to tell it to use.
// If our app is currently cached, we know it, and that is it. Otherwise,
// we don't know it yet, and it needs to now be cached we will then
// need to do a complete oom adj.
final int cachedAdj = app.curRawAdj >= ProcessList.CACHED_APP_MIN_ADJ
? app.curRawAdj : ProcessList.UNKNOWN_ADJ;
boolean success = updateOomAdjLocked(app, cachedAdj, TOP_APP, false,
SystemClock.uptimeMillis());
if (wasCached != app.cached || app.curRawAdj == ProcessList.UNKNOWN_ADJ) {
// Changed to/from cached state, so apps after it in the LRU
// list may also be changed.
updateOomAdjLocked();
}
return success;
}
复制代码就是如果app进程(即发起bindService()的那个进程)是缓存的，那么取app进程当前的oom_adj值和ProcessList.CACHED_APP_MIN_ADJ中的较大值，然后调用updateOomAdjLocked()方法更新app进程的oom_adj值即可。该方法如下:
private final boolean updateOomAdjLocked(ProcessRecord app, int cachedAdj,
ProcessRecord TOP_APP, boolean doingAll, long now) {
if (app.thread == null) {
return false;
}
computeOomAdjLocked(app, cachedAdj, TOP_APP, doingAll, now);
return applyOomAdjLocked(app, doingAll, now, SystemClock.elapsedRealtime());
}
复制代码可见这里主要就是调用了computeOomAdjLocked()方法对于app进程的oom_adj值进行了一次计算，然后调用applyOomAdjLocked()方法将计算结果运用上去。
其中computeOomAdjLocked()方法非常重要，现在市面上所谓的进程保活措施的依据有很多就是来自于这里。
AMS.computeOomAdjLocked()
private final int computeOomAdjLocked(ProcessRecord app, int cachedAdj, ProcessRecord TOP_APP,
boolean doingAll, long now) {
if (mAdjSeq == app.adjSeq) {
// This adjustment has already been computed.
return app.curRawAdj;
}
if (app.thread == null) {
app.adjSeq = mAdjSeq;
app.curSchedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
app.curProcState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
return (app.curAdj=app.curRawAdj=ProcessList.CACHED_APP_MAX_ADJ);
}
app.adjTypeCode = ActivityManager.RunningAppProcessInfo.REASON_UNKNOWN;
app.adjSource = null;
app.adjTarget = null;
app.empty = false;
app.cached = false;
final int activitiesSize = app.activities.size();
if (app.maxAdj <= ProcessList.FOREGROUND_APP_ADJ) {
// The max adjustment doesn't allow this app to be anything
// below foreground, so it is not worth doing work for it.
app.adjType = "fixed";
app.adjSeq = mAdjSeq;
app.curRawAdj = app.maxAdj;
app.foregroundActivities = false;
app.curSchedGroup = ProcessList.SCHED_GROUP_DEFAULT;
app.curProcState = ActivityManager.PROCESS_STATE_PERSISTENT;
// System processes can do UI, and when they do we want to have
// them trim their memory after the user leaves the UI. To
// facilitate this, here we need to determine whether or not it
// is currently showing UI.
app.systemNoUi = true;
if (app == TOP_APP) {
app.systemNoUi = false;
app.curSchedGroup = ProcessList.SCHED_GROUP_TOP_APP;
app.adjType = "pers-top-activity";
} else if (activitiesSize > 0) {
for (int j = 0; j < activitiesSize; j++) {
final ActivityRecord r = app.activities.get(j);
if (r.visible) {
app.systemNoUi = false;
}
}
}
if (!app.systemNoUi) {
app.curProcState = ActivityManager.PROCESS_STATE_PERSISTENT_UI;
}
return (app.curAdj=app.maxAdj);
}
app.systemNoUi = false;
final int PROCESS_STATE_CUR_TOP = mTopProcessState;
// Determine the importance of the process, starting with most
// important to least, and assign an appropriate OOM adjustment.
int adj;
int schedGroup;
int procState;
boolean foregroundActivities = false;
BroadcastQueue queue;
if (app == TOP_APP) {
// The last app on the list is the foreground app.
adj = ProcessList.FOREGROUND_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_TOP_APP;
app.adjType = "top-activity";
foregroundActivities = true;
procState = PROCESS_STATE_CUR_TOP;
} else if (app.instrumentationClass != null) {
// Don't want to kill running instrumentation.
adj = ProcessList.FOREGROUND_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
app.adjType = "instrumentation";
procState = ActivityManager.PROCESS_STATE_FOREGROUND_SERVICE;
} else if ((queue = isReceivingBroadcast(app)) != null) {
// An app that is currently receiving a broadcast also
// counts as being in the foreground for OOM killer purposes.
// It's placed in a sched group based on the nature of the
// broadcast as reflected by which queue it's active in.
adj = ProcessList.FOREGROUND_APP_ADJ;
schedGroup = (queue == mFgBroadcastQueue)
? ProcessList.SCHED_GROUP_DEFAULT : ProcessList.SCHED_GROUP_BACKGROUND;
app.adjType = "broadcast";
procState = ActivityManager.PROCESS_STATE_RECEIVER;
} else if (app.executingServices.size() > 0) {
// An app that is currently executing a service callback also
// counts as being in the foreground.
adj = ProcessList.FOREGROUND_APP_ADJ;
schedGroup = app.execServicesFg ?
ProcessList.SCHED_GROUP_DEFAULT : ProcessList.SCHED_GROUP_BACKGROUND;
app.adjType = "exec-service";
procState = ActivityManager.PROCESS_STATE_SERVICE;
//Slog.i(TAG, "EXEC " + (app.execServicesFg ? "FG" : "BG") + ": " + app);
} else {
// As far as we know the process is empty. We may change our mind later.
schedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
// At this point we don't actually know the adjustment. Use the cached adj
// value that the caller wants us to.
adj = cachedAdj;
procState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
app.cached = true;
app.empty = true;
app.adjType = "cch-empty";
}
// Examine all activities if not already foreground.
if (!foregroundActivities && activitiesSize > 0) {
int minLayer = ProcessList.VISIBLE_APP_LAYER_MAX;
for (int j = 0; j < activitiesSize; j++) {
final ActivityRecord r = app.activities.get(j);
if (r.app != app) {
Log.e(TAG, "Found activity " + r + " in proc activity list using " + r.app
+ " instead of expected " + app);
if (r.app == null || (r.app.uid == app.uid)) {
// Only fix things up when they look sane
r.app = app;
} else {
continue;
}
}
if (r.visible) {
// App has a visible activity; only upgrade adjustment.
if (adj > ProcessList.VISIBLE_APP_ADJ) {
adj = ProcessList.VISIBLE_APP_ADJ;
app.adjType = "visible";
}
if (procState > PROCESS_STATE_CUR_TOP) {
procState = PROCESS_STATE_CUR_TOP;
}
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
app.cached = false;
app.empty = false;
foregroundActivities = true;
if (r.task != null && minLayer > 0) {
final int layer = r.task.mLayerRank;
if (layer >= 0 && minLayer > layer) {
minLayer = layer;
}
}
break;
} else if (r.state == ActivityState.PAUSING || r.state == ActivityState.PAUSED) {
if (adj > ProcessList.PERCEPTIBLE_APP_ADJ) {
adj = ProcessList.PERCEPTIBLE_APP_ADJ;
app.adjType = "pausing";
}
if (procState > PROCESS_STATE_CUR_TOP) {
procState = PROCESS_STATE_CUR_TOP;
}
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
app.cached = false;
app.empty = false;
foregroundActivities = true;
} else if (r.state == ActivityState.STOPPING) {
if (adj > ProcessList.PERCEPTIBLE_APP_ADJ) {
adj = ProcessList.PERCEPTIBLE_APP_ADJ;
app.adjType = "stopping";
}
// For the process state, we will at this point consider the
// process to be cached. It will be cached either as an activity
// or empty depending on whether the activity is finishing. We do
// this so that we can treat the process as cached for purposes of
// memory trimming (determing current memory level, trim command to
// send to process) since there can be an arbitrary number of stopping
// processes and they should soon all go into the cached state.
if (!r.finishing) {
if (procState > ActivityManager.PROCESS_STATE_LAST_ACTIVITY) {
procState = ActivityManager.PROCESS_STATE_LAST_ACTIVITY;
}
}
app.cached = false;
app.empty = false;
foregroundActivities = true;
} else {
if (procState > ActivityManager.PROCESS_STATE_CACHED_ACTIVITY) {
procState = ActivityManager.PROCESS_STATE_CACHED_ACTIVITY;
app.adjType = "cch-act";
}
}
}
if (adj == ProcessList.VISIBLE_APP_ADJ) {
adj += minLayer;
}
}
if (adj > ProcessList.PERCEPTIBLE_APP_ADJ
|| procState > ActivityManager.PROCESS_STATE_FOREGROUND_SERVICE) {
if (app.foregroundServices) {
// The user is aware of this app, so make it visible.
adj = ProcessList.PERCEPTIBLE_APP_ADJ;
procState = ActivityManager.PROCESS_STATE_FOREGROUND_SERVICE;
app.cached = false;
app.adjType = "fg-service";
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
} else if (app.forcingToForeground != null) {
// The user is aware of this app, so make it visible.
adj = ProcessList.PERCEPTIBLE_APP_ADJ;
procState = ActivityManager.PROCESS_STATE_IMPORTANT_FOREGROUND;
app.cached = false;
app.adjType = "force-fg";
app.adjSource = app.forcingToForeground;
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
}
}
if (app == mHeavyWeightProcess) {
if (adj > ProcessList.HEAVY_WEIGHT_APP_ADJ) {
// We don't want to kill the current heavy-weight process.
adj = ProcessList.HEAVY_WEIGHT_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
app.cached = false;
app.adjType = "heavy";
}
if (procState > ActivityManager.PROCESS_STATE_HEAVY_WEIGHT) {
procState = ActivityManager.PROCESS_STATE_HEAVY_WEIGHT;
}
}
if (app == mHomeProcess) {
if (adj > ProcessList.HOME_APP_ADJ) {
// This process is hosting what we currently consider to be the
// home app, so we don't want to let it go into the background.
adj = ProcessList.HOME_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
app.cached = false;
app.adjType = "home";
}
if (procState > ActivityManager.PROCESS_STATE_HOME) {
procState = ActivityManager.PROCESS_STATE_HOME;
}
}
if (app == mPreviousProcess && app.activities.size() > 0) {
if (adj > ProcessList.PREVIOUS_APP_ADJ) {
// This was the previous process that showed UI to the user.
// We want to try to keep it around more aggressively, to give
// a good experience around switching between two apps.
adj = ProcessList.PREVIOUS_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
app.cached = false;
app.adjType = "previous";
}
if (procState > ActivityManager.PROCESS_STATE_LAST_ACTIVITY) {
procState = ActivityManager.PROCESS_STATE_LAST_ACTIVITY;
}
}
if (false) Slog.i(TAG, "OOM " + app + ": initial adj=" + adj
+ " reason=" + app.adjType);
// By default, we use the computed adjustment. It may be changed if
// there are applications dependent on our services or providers, but
// this gives us a baseline and makes sure we don't get into an
// infinite recursion.
app.adjSeq = mAdjSeq;
app.curRawAdj = adj;
app.hasStartedServices = false;
if (mBackupTarget != null && app == mBackupTarget.app) {
// If possible we want to avoid killing apps while they're being backed up
if (adj > ProcessList.BACKUP_APP_ADJ) {
if (DEBUG_BACKUP) Slog.v(TAG_BACKUP, "oom BACKUP_APP_ADJ for " + app);
adj = ProcessList.BACKUP_APP_ADJ;
if (procState > ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND) {
procState = ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND;
}
app.adjType = "backup";
app.cached = false;
}
if (procState > ActivityManager.PROCESS_STATE_BACKUP) {
procState = ActivityManager.PROCESS_STATE_BACKUP;
}
}
boolean mayBeTop = false;
for (int is = app.services.size()-1;
is >= 0 && (adj > ProcessList.FOREGROUND_APP_ADJ
|| schedGroup == ProcessList.SCHED_GROUP_BACKGROUND
|| procState > ActivityManager.PROCESS_STATE_TOP);
is--) {
ServiceRecord s = app.services.valueAt(is);
if (s.startRequested) {
app.hasStartedServices = true;
if (procState > ActivityManager.PROCESS_STATE_SERVICE) {
procState = ActivityManager.PROCESS_STATE_SERVICE;
}
if (app.hasShownUi && app != mHomeProcess) {
// If this process has shown some UI, let it immediately
// go to the LRU list because it may be pretty heavy with
// UI stuff. We'll tag it with a label just to help
// debug and understand what is going on.
if (adj > ProcessList.SERVICE_ADJ) {
app.adjType = "cch-started-ui-services";
}
} else {
if (now < (s.lastActivity + ActiveServices.MAX_SERVICE_INACTIVITY)) {
// This service has seen some activity within
// recent memory, so we will keep its process ahead
// of the background processes.
if (adj > ProcessList.SERVICE_ADJ) {
adj = ProcessList.SERVICE_ADJ;
app.adjType = "started-services";
app.cached = false;
}
}
// If we have let the service slide into the background
// state, still have some text describing what it is doing
// even though the service no longer has an impact.
if (adj > ProcessList.SERVICE_ADJ) {
app.adjType = "cch-started-services";
}
}
}
for (int conni = s.connections.size()-1;
conni >= 0 && (adj > ProcessList.FOREGROUND_APP_ADJ
|| schedGroup == ProcessList.SCHED_GROUP_BACKGROUND
|| procState > ActivityManager.PROCESS_STATE_TOP);
conni--) {
ArrayList<ConnectionRecord> clist = s.connections.valueAt(conni);
for (int i = 0;
i < clist.size() && (adj > ProcessList.FOREGROUND_APP_ADJ
|| schedGroup == ProcessList.SCHED_GROUP_BACKGROUND
|| procState > ActivityManager.PROCESS_STATE_TOP);
i++) {
// XXX should compute this based on the max of
// all connected clients.
ConnectionRecord cr = clist.get(i);
if (cr.binding.client == app) {
// Binding to ourself is not interesting.
continue;
}
if ((cr.flags&Context.BIND_WAIVE_PRIORITY) == 0) {
ProcessRecord client = cr.binding.client;
int clientAdj = computeOomAdjLocked(client, cachedAdj,
TOP_APP, doingAll, now);
int clientProcState = client.curProcState;
if (clientProcState >= ActivityManager.PROCESS_STATE_CACHED_ACTIVITY) {
// If the other app is cached for any reason, for purposes here
// we are going to consider it empty. The specific cached state
// doesn't propagate except under certain conditions.
clientProcState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
}
String adjType = null;
if ((cr.flags&Context.BIND_ALLOW_OOM_MANAGEMENT) != 0) {
// Not doing bind OOM management, so treat
// this guy more like a started service.
if (app.hasShownUi && app != mHomeProcess) {
// If this process has shown some UI, let it immediately
// go to the LRU list because it may be pretty heavy with
// UI stuff. We'll tag it with a label just to help
// debug and understand what is going on.
if (adj > clientAdj) {
adjType = "cch-bound-ui-services";
}
app.cached = false;
clientAdj = adj;
clientProcState = procState;
} else {
if (now >= (s.lastActivity
+ ActiveServices.MAX_SERVICE_INACTIVITY)) {
// This service has not seen activity within
// recent memory, so allow it to drop to the
// LRU list if there is no other reason to keep
// it around. We'll also tag it with a label just
// to help debug and undertand what is going on.
if (adj > clientAdj) {
adjType = "cch-bound-services";
}
clientAdj = adj;
}
}
}
if (adj > clientAdj) {
// If this process has recently shown UI, and
// the process that is binding to it is less
// important than being visible, then we don't
// care about the binding as much as we care
// about letting this process get into the LRU
// list to be killed and restarted if needed for
// memory.
if (app.hasShownUi && app != mHomeProcess
&& clientAdj > ProcessList.PERCEPTIBLE_APP_ADJ) {
adjType = "cch-bound-ui-services";
} else {
if ((cr.flags&(Context.BIND_ABOVE_CLIENT
|Context.BIND_IMPORTANT)) != 0) {
adj = clientAdj >= ProcessList.PERSISTENT_SERVICE_ADJ
? clientAdj : ProcessList.PERSISTENT_SERVICE_ADJ;
} else if ((cr.flags&Context.BIND_NOT_VISIBLE) != 0
&& clientAdj < ProcessList.PERCEPTIBLE_APP_ADJ
&& adj > ProcessList.PERCEPTIBLE_APP_ADJ) {
adj = ProcessList.PERCEPTIBLE_APP_ADJ;
} else if (clientAdj >= ProcessList.PERCEPTIBLE_APP_ADJ) {
adj = clientAdj;
} else {
if (adj > ProcessList.VISIBLE_APP_ADJ) {
adj = Math.max(clientAdj, ProcessList.VISIBLE_APP_ADJ);
}
}
if (!client.cached) {
app.cached = false;
}
adjType = "service";
}
}
if ((cr.flags&Context.BIND_NOT_FOREGROUND) == 0) {
// This will treat important bound services identically to
// the top app, which may behave differently than generic
// foreground work.
if (client.curSchedGroup > schedGroup) {
if ((cr.flags&Context.BIND_IMPORTANT) != 0) {
schedGroup = client.curSchedGroup;
} else {
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
}
}
if (clientProcState <= ActivityManager.PROCESS_STATE_TOP) {
if (clientProcState == ActivityManager.PROCESS_STATE_TOP) {
// Special handling of clients who are in the top state.
// We *may* want to consider this process to be in the
// top state as well, but only if there is not another
// reason for it to be running. Being on the top is a
// special state, meaning you are specifically running
// for the current top app. If the process is already
// running in the background for some other reason, it
// is more important to continue considering it to be
// in the background state.
mayBeTop = true;
clientProcState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
} else {
// Special handling for above-top states (persistent
// processes). These should not bring the current process
// into the top state, since they are not on top. Instead
// give them the best state after that.
if ((cr.flags&Context.BIND_FOREGROUND_SERVICE) != 0) {
clientProcState =
ActivityManager.PROCESS_STATE_BOUND_FOREGROUND_SERVICE;
} else if (mWakefulness
== PowerManagerInternal.WAKEFULNESS_AWAKE &&
(cr.flags&Context.BIND_FOREGROUND_SERVICE_WHILE_AWAKE)
!= 0) {
clientProcState =
ActivityManager.PROCESS_STATE_BOUND_FOREGROUND_SERVICE;
} else {
clientProcState =
ActivityManager.PROCESS_STATE_IMPORTANT_FOREGROUND;
}
}
}
} else {
if (clientProcState <
ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND) {
clientProcState =
ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND;
}
}
if (procState > clientProcState) {
procState = clientProcState;
}
if (procState < ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND
&& (cr.flags&Context.BIND_SHOWING_UI) != 0) {
app.pendingUiClean = true;
}
if (adjType != null) {
app.adjType = adjType;
app.adjTypeCode = ActivityManager.RunningAppProcessInfo
.REASON_SERVICE_IN_USE;
app.adjSource = cr.binding.client;
app.adjSourceProcState = clientProcState;
app.adjTarget = s.name;
}
}
if ((cr.flags&Context.BIND_TREAT_LIKE_ACTIVITY) != 0) {
app.treatLikeActivity = true;
}
final ActivityRecord a = cr.activity;
if ((cr.flags&Context.BIND_ADJUST_WITH_ACTIVITY) != 0) {
if (a != null && adj > ProcessList.FOREGROUND_APP_ADJ &&
(a.visible || a.state == ActivityState.RESUMED ||
a.state == ActivityState.PAUSING)) {
adj = ProcessList.FOREGROUND_APP_ADJ;
if ((cr.flags&Context.BIND_NOT_FOREGROUND) == 0) {
if ((cr.flags&Context.BIND_IMPORTANT) != 0) {
schedGroup = ProcessList.SCHED_GROUP_TOP_APP;
} else {
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
}
}
app.cached = false;
app.adjType = "service";
app.adjTypeCode = ActivityManager.RunningAppProcessInfo
.REASON_SERVICE_IN_USE;
app.adjSource = a;
app.adjSourceProcState = procState;
app.adjTarget = s.name;
}
}
}
}
}
for (int provi = app.pubProviders.size()-1;
provi >= 0 && (adj > ProcessList.FOREGROUND_APP_ADJ
|| schedGroup == ProcessList.SCHED_GROUP_BACKGROUND
|| procState > ActivityManager.PROCESS_STATE_TOP);
provi--) {
ContentProviderRecord cpr = app.pubProviders.valueAt(provi);
for (int i = cpr.connections.size()-1;
i >= 0 && (adj > ProcessList.FOREGROUND_APP_ADJ
|| schedGroup == ProcessList.SCHED_GROUP_BACKGROUND
|| procState > ActivityManager.PROCESS_STATE_TOP);
i--) {
ContentProviderConnection conn = cpr.connections.get(i);
ProcessRecord client = conn.client;
if (client == app) {
// Being our own client is not interesting.
continue;
}
int clientAdj = computeOomAdjLocked(client, cachedAdj, TOP_APP, doingAll, now);
int clientProcState = client.curProcState;
if (clientProcState >= ActivityManager.PROCESS_STATE_CACHED_ACTIVITY) {
// If the other app is cached for any reason, for purposes here
// we are going to consider it empty.
clientProcState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
}
if (adj > clientAdj) {
if (app.hasShownUi && app != mHomeProcess
&& clientAdj > ProcessList.PERCEPTIBLE_APP_ADJ) {
app.adjType = "cch-ui-provider";
} else {
adj = clientAdj > ProcessList.FOREGROUND_APP_ADJ
? clientAdj : ProcessList.FOREGROUND_APP_ADJ;
app.adjType = "provider";
}
app.cached &= client.cached;
app.adjTypeCode = ActivityManager.RunningAppProcessInfo
.REASON_PROVIDER_IN_USE;
app.adjSource = client;
app.adjSourceProcState = clientProcState;
app.adjTarget = cpr.name;
}
if (clientProcState <= ActivityManager.PROCESS_STATE_TOP) {
if (clientProcState == ActivityManager.PROCESS_STATE_TOP) {
// Special handling of clients who are in the top state.
// We *may* want to consider this process to be in the
// top state as well, but only if there is not another
// reason for it to be running. Being on the top is a
// special state, meaning you are specifically running
// for the current top app. If the process is already
// running in the background for some other reason, it
// is more important to continue considering it to be
// in the background state.
mayBeTop = true;
clientProcState = ActivityManager.PROCESS_STATE_CACHED_EMPTY;
} else {
// Special handling for above-top states (persistent
// processes). These should not bring the current process
// into the top state, since they are not on top. Instead
// give them the best state after that.
clientProcState =
ActivityManager.PROCESS_STATE_BOUND_FOREGROUND_SERVICE;
}
}
if (procState > clientProcState) {
procState = clientProcState;
}
if (client.curSchedGroup > schedGroup) {
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
}
}
// If the provider has external (non-framework) process
// dependencies, ensure that its adjustment is at least
// FOREGROUND_APP_ADJ.
if (cpr.hasExternalProcessHandles()) {
if (adj > ProcessList.FOREGROUND_APP_ADJ) {
adj = ProcessList.FOREGROUND_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
app.cached = false;
app.adjType = "provider";
app.adjTarget = cpr.name;
}
if (procState > ActivityManager.PROCESS_STATE_IMPORTANT_FOREGROUND) {
procState = ActivityManager.PROCESS_STATE_IMPORTANT_FOREGROUND;
}
}
}
if (app.lastProviderTime > 0 && (app.lastProviderTime+CONTENT_PROVIDER_RETAIN_TIME) > now) {
if (adj > ProcessList.PREVIOUS_APP_ADJ) {
adj = ProcessList.PREVIOUS_APP_ADJ;
schedGroup = ProcessList.SCHED_GROUP_BACKGROUND;
app.cached = false;
app.adjType = "provider";
}
if (procState > ActivityManager.PROCESS_STATE_LAST_ACTIVITY) {
procState = ActivityManager.PROCESS_STATE_LAST_ACTIVITY;
}
}
if (mayBeTop && procState > ActivityManager.PROCESS_STATE_TOP) {
// A client of one of our services or providers is in the top state. We
// *may* want to be in the top state, but not if we are already running in
// the background for some other reason. For the decision here, we are going
// to pick out a few specific states that we want to remain in when a client
// is top (states that tend to be longer-term) and otherwise allow it to go
// to the top state.
switch (procState) {
case ActivityManager.PROCESS_STATE_IMPORTANT_FOREGROUND:
case ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND:
case ActivityManager.PROCESS_STATE_SERVICE:
// These all are longer-term states, so pull them up to the top
// of the background states, but not all the way to the top state.
procState = ActivityManager.PROCESS_STATE_BOUND_FOREGROUND_SERVICE;
break;
default:
// Otherwise, top is a better choice, so take it.
procState = ActivityManager.PROCESS_STATE_TOP;
break;
}
}
if (procState >= ActivityManager.PROCESS_STATE_CACHED_EMPTY) {
if (app.hasClientActivities) {
// This is a cached process, but with client activities. Mark it so.
procState = ActivityManager.PROCESS_STATE_CACHED_ACTIVITY_CLIENT;
app.adjType = "cch-client-act";
} else if (app.treatLikeActivity) {
// This is a cached process, but somebody wants us to treat it like it has
// an activity, okay!
procState = ActivityManager.PROCESS_STATE_CACHED_ACTIVITY;
app.adjType = "cch-as-act";
}
}
if (adj == ProcessList.SERVICE_ADJ) {
if (doingAll) {
app.serviceb = mNewNumAServiceProcs > (mNumServiceProcs/3);
mNewNumServiceProcs++;
//Slog.i(TAG, "ADJ " + app + " serviceb=" + app.serviceb);
if (!app.serviceb) {
// This service isn't far enough down on the LRU list to
// normally be a B service, but if we are low on RAM and it
// is large we want to force it down since we would prefer to
// keep launcher over it.
if (mLastMemoryLevel > ProcessStats.ADJ_MEM_FACTOR_NORMAL
&& app.lastPss >= mProcessList.getCachedRestoreThresholdKb()) {
app.serviceHighRam = true;
app.serviceb = true;
//Slog.i(TAG, "ADJ " + app + " high ram!");
} else {
mNewNumAServiceProcs++;
//Slog.i(TAG, "ADJ " + app + " not high ram!");
}
} else {
app.serviceHighRam = false;
}
}
if (app.serviceb) {
adj = ProcessList.SERVICE_B_ADJ;
}
}
app.curRawAdj = adj;
//Slog.i(TAG, "OOM ADJ " + app + ": pid=" + app.pid +
// " adj=" + adj + " curAdj=" + app.curAdj + " maxAdj=" + app.maxAdj);
if (adj > app.maxAdj) {
adj = app.maxAdj;
if (app.maxAdj <= ProcessList.PERCEPTIBLE_APP_ADJ) {
schedGroup = ProcessList.SCHED_GROUP_DEFAULT;
}
}
// Do final modification to adj. Everything we do between here and applying
// the final setAdj must be done in this function, because we will also use
// it when computing the final cached adj later. Note that we don't need to
// worry about this for max adj above, since max adj will always be used to
// keep it out of the cached vaues.
app.curAdj = app.modifyRawOomAdj(adj);
app.curSchedGroup = schedGroup;
app.curProcState = procState;
app.foregroundActivities = foregroundActivities;
return app.curRawAdj;
}
复制代码方法主要包括几个部分：
1)如果当前service所在进程的oom_adj值已经被计算过了，那就无需再次计算，直接使用当前的值即可;
2)如果app.thread==null,表明这个进程在缓存中，那么返回ProcessList.CACHED_APP_MAX_ADJ值即可,而这个值的优先级并不高，因为这个数值为15;
3)如果app.maxAdj值小于等于ProcessList.FOREGROUND_APP_ADJ(数值为0),表明最大的调整都不允许Service进程达到ProcessList.FOREGROUND_APP_ADJ,那么就不值得为这个客户端进程工作，所以此时要进行调整，以使app.maxAdj的值能够达到ProcessList.FOREGROUND_APP_ADJ值;
4)根据service进程的情况来为其设置oom_adj值，分别考虑前台app进程，运行instrumentation的进程，正在接收广播的进程，正在执行service callback的进程以及空进程这5种情况。
5)如果不是前台进程，并且service所在进程的Activity数量大于0，就需要依次检查所有activity以更新ProcessRecord和foregroundActivities的状态;
6)如果adj值比ProcessList.PERCEPTIBLE_APP_ADJ(值为2)还大，则要对结果进行调整，如果是ProcessRecord是前台service进程，或者forcingToForeground不为null,则会将adj值调整到ProcessList.PERCEPTIBLE_APP_ADJ值;
7)如果service所在进程对应的ProcessRecord正在执行重量级任务，如果adj>ProcessList.HEAVY_WEIGHT_APP_ADJ值，那么会调整到ProcessList.HEAVY_WEIGHT_APP_ADJ值，这显然是一种均衡策略，不让某个太繁重的任务影响到其他进程的流畅运行;
8)如果服务进程就是mHomeProcess,那么也会依据adj值进行调整;
9)如果服务进程是前一个进程，并且它有activity的话，那么会调整到ProcessList.PREVIOUS_APP_ADJ值;
10)下面根据ProcessRecord中所有ServiceRecord的状态来更新adj和ProcessRecord的值，分以下情况:
如果Service是显式启动的,并且进程还拥有activity,则将其adj值设置为ProcessList.SERVICE_ADJ; 之后遍历每个ServiceRecord中包含的ConnectionRecord,根据ConnectionRecord的值进行adj和ProcessRecord的状态更新，此时会涉及到client端的oom_adj值; 11)最后根据ProcessRecord中所有ContentProviderRecord的状态，逐个分析每个ContentProviderRecord的状态，根据其状态进行adj值和ProcessRecord状态的调整;
12)最后将计算的adj值存放到ProcessRecord中的curRawAdj值中。并且调用ProcessRecord的modifyRawOomAdj()方法来对计算得到的原始adj值进行可能的调整，最后将结果赋值结ProcessRecord的curAdj值中。
AMS.updateOomAdjLocked()
final void updateOomAdjLocked() {
final ActivityRecord TOP_ACT = resumedAppLocked();
final ProcessRecord TOP_APP = TOP_ACT != null ? TOP_ACT.app : null;
final long now = SystemClock.uptimeMillis();
final long nowElapsed = SystemClock.elapsedRealtime();
final long oldTime = now - ProcessList.MAX_EMPTY_TIME;
final int N = mLruProcesses.size();
if (false) {
RuntimeException e = new RuntimeException();
e.fillInStackTrace();
Slog.i(TAG, "updateOomAdj: top=" + TOP_ACT, e);
}
// Reset state in all uid records.
for (int i=mActiveUids.size()-1; i>=0; i--) {
final UidRecord uidRec = mActiveUids.valueAt(i);
if (false && DEBUG_UID_OBSERVERS) Slog.i(TAG_UID_OBSERVERS,
"Starting update of " + uidRec);
uidRec.reset();
}
mStackSupervisor.rankTaskLayersIfNeeded();
mAdjSeq++;
mNewNumServiceProcs = 0;
mNewNumAServiceProcs = 0;
final int emptyProcessLimit;
final int cachedProcessLimit;
if (mProcessLimit <= 0) {
emptyProcessLimit = cachedProcessLimit = 0;
} else if (mProcessLimit == 1) {
emptyProcessLimit = 1;
cachedProcessLimit = 0;
} else {
emptyProcessLimit = ProcessList.computeEmptyProcessLimit(mProcessLimit);
cachedProcessLimit = mProcessLimit - emptyProcessLimit;
}
// Let's determine how many processes we have running vs.
// how many slots we have for background processes; we may want
// to put multiple processes in a slot of there are enough of
// them.
int numSlots = (ProcessList.CACHED_APP_MAX_ADJ
- ProcessList.CACHED_APP_MIN_ADJ + 1) / 2;
int numEmptyProcs = N - mNumNonCachedProcs - mNumCachedHiddenProcs;
if (numEmptyProcs > cachedProcessLimit) {
// If there are more empty processes than our limit on cached
// processes, then use the cached process limit for the factor.
// This ensures that the really old empty processes get pushed
// down to the bottom, so if we are running low on memory we will
// have a better chance at keeping around more cached processes
// instead of a gazillion empty processes.
numEmptyProcs = cachedProcessLimit;
}
int emptyFactor = numEmptyProcs/numSlots;
if (emptyFactor < 1) emptyFactor = 1;
int cachedFactor = (mNumCachedHiddenProcs > 0 ? mNumCachedHiddenProcs : 1)/numSlots;
if (cachedFactor < 1) cachedFactor = 1;
int stepCached = 0;
int stepEmpty = 0;
int numCached = 0;
int numEmpty = 0;
int numTrimming = 0;
mNumNonCachedProcs = 0;
mNumCachedHiddenProcs = 0;
// First update the OOM adjustment for each of the
// application processes based on their current state.
int curCachedAdj = ProcessList.CACHED_APP_MIN_ADJ;
int nextCachedAdj = curCachedAdj+1;
int curEmptyAdj = ProcessList.CACHED_APP_MIN_ADJ;
int nextEmptyAdj = curEmptyAdj+2;
for (int i=N-1; i>=0; i--) {
ProcessRecord app = mLruProcesses.get(i);
if (!app.killedByAm && app.thread != null) {
app.procStateChanged = false;
computeOomAdjLocked(app, ProcessList.UNKNOWN_ADJ, TOP_APP, true, now);
// If we haven't yet assigned the final cached adj
// to the process, do that now.
if (app.curAdj >= ProcessList.UNKNOWN_ADJ) {
switch (app.curProcState) {
case ActivityManager.PROCESS_STATE_CACHED_ACTIVITY:
case ActivityManager.PROCESS_STATE_CACHED_ACTIVITY_CLIENT:
// This process is a cached process holding activities...
// assign it the next cached value for that type, and then
// step that cached level.
app.curRawAdj = curCachedAdj;
app.curAdj = app.modifyRawOomAdj(curCachedAdj);
if (DEBUG_LRU && false) Slog.d(TAG_LRU, "Assigning activity LRU #" + i
+ " adj: " + app.curAdj + " (curCachedAdj=" + curCachedAdj
+ ")");
if (curCachedAdj != nextCachedAdj) {
stepCached++;
if (stepCached >= cachedFactor) {
stepCached = 0;
curCachedAdj = nextCachedAdj;
nextCachedAdj += 2;
if (nextCachedAdj > ProcessList.CACHED_APP_MAX_ADJ) {
nextCachedAdj = ProcessList.CACHED_APP_MAX_ADJ;
}
}
}
break;
default:
// For everything else, assign next empty cached process
// level and bump that up. Note that this means that
// long-running services that have dropped down to the
// cached level will be treated as empty (since their process
// state is still as a service), which is what we want.
app.curRawAdj = curEmptyAdj;
app.curAdj = app.modifyRawOomAdj(curEmptyAdj);
if (DEBUG_LRU && false) Slog.d(TAG_LRU, "Assigning empty LRU #" + i
+ " adj: " + app.curAdj + " (curEmptyAdj=" + curEmptyAdj
+ ")");
if (curEmptyAdj != nextEmptyAdj) {
stepEmpty++;
if (stepEmpty >= emptyFactor) {
stepEmpty = 0;
curEmptyAdj = nextEmptyAdj;
nextEmptyAdj += 2;
if (nextEmptyAdj > ProcessList.CACHED_APP_MAX_ADJ) {
nextEmptyAdj = ProcessList.CACHED_APP_MAX_ADJ;
}
}
}
break;
}
}
applyOomAdjLocked(app, true, now, nowElapsed);
// Count the number of process types.
switch (app.curProcState) {
case ActivityManager.PROCESS_STATE_CACHED_ACTIVITY:
case ActivityManager.PROCESS_STATE_CACHED_ACTIVITY_CLIENT:
mNumCachedHiddenProcs++;
numCached++;
if (numCached > cachedProcessLimit) {
app.kill("cached #" + numCached, true);
}
break;
case ActivityManager.PROCESS_STATE_CACHED_EMPTY:
if (numEmpty > ProcessList.TRIM_EMPTY_APPS
&& app.lastActivityTime < oldTime) {
app.kill("empty for "
+ ((oldTime + ProcessList.MAX_EMPTY_TIME - app.lastActivityTime)
/ 1000) + "s", true);
} else {
numEmpty++;
if (numEmpty > emptyProcessLimit) {
app.kill("empty #" + numEmpty, true);
}
}
break;
default:
mNumNonCachedProcs++;
break;
}
if (app.isolated && app.services.size() <= 0) {
// If this is an isolated process, and there are no
// services running in it, then the process is no longer
// needed. We agressively kill these because we can by
// definition not re-use the same process again, and it is
// good to avoid having whatever code was running in them
// left sitting around after no longer needed.
app.kill("isolated not needed", true);
} else {
// Keeping this process, update its uid.
final UidRecord uidRec = app.uidRecord;
if (uidRec != null && uidRec.curProcState > app.curProcState) {
uidRec.curProcState = app.curProcState;
}
}
if (app.curProcState >= ActivityManager.PROCESS_STATE_HOME
&& !app.killedByAm) {
numTrimming++;
}
}
}
mNumServiceProcs = mNewNumServiceProcs;
// Now determine the memory trimming level of background processes.
// Unfortunately we need to start at the back of the list to do this
// properly. We only do this if the number of background apps we
// are managing to keep around is less than half the maximum we desire;
// if we are keeping a good number around, we'll let them use whatever
// memory they want.
final int numCachedAndEmpty = numCached + numEmpty;
int memFactor;
if (numCached <= ProcessList.TRIM_CACHED_APPS
&& numEmpty <= ProcessList.TRIM_EMPTY_APPS) {
if (numCachedAndEmpty <= ProcessList.TRIM_CRITICAL_THRESHOLD) {
memFactor = ProcessStats.ADJ_MEM_FACTOR_CRITICAL;
} else if (numCachedAndEmpty <= ProcessList.TRIM_LOW_THRESHOLD) {
memFactor = ProcessStats.ADJ_MEM_FACTOR_LOW;
} else {
memFactor = ProcessStats.ADJ_MEM_FACTOR_MODERATE;
}
} else {
memFactor = ProcessStats.ADJ_MEM_FACTOR_NORMAL;
}
// We always allow the memory level to go up (better). We only allow it to go
// down if we are in a state where that is allowed, *and* the total number of processes
// has gone down since last time.
if (DEBUG_OOM_ADJ) Slog.d(TAG_OOM_ADJ, "oom: memFactor=" + memFactor
+ " last=" + mLastMemoryLevel + " allowLow=" + mAllowLowerMemLevel
+ " numProcs=" + mLruProcesses.size() + " last=" + mLastNumProcesses);
if (memFactor > mLastMemoryLevel) {
if (!mAllowLowerMemLevel || mLruProcesses.size() >= mLastNumProcesses) {
memFactor = mLastMemoryLevel;
if (DEBUG_OOM_ADJ) Slog.d(TAG_OOM_ADJ, "Keeping last mem factor!");
}
}
if (memFactor != mLastMemoryLevel) {
EventLogTags.writeAmMemFactor(memFactor, mLastMemoryLevel);
}
mLastMemoryLevel = memFactor;
mLastNumProcesses = mLruProcesses.size();
boolean allChanged = mProcessStats.setMemFactorLocked(memFactor, !isSleepingLocked(), now);
final int trackerMemFactor = mProcessStats.getMemFactorLocked();
if (memFactor != ProcessStats.ADJ_MEM_FACTOR_NORMAL) {
if (mLowRamStartTime == 0) {
mLowRamStartTime = now;
}
int step = 0;
int fgTrimLevel;
switch (memFactor) {
case ProcessStats.ADJ_MEM_FACTOR_CRITICAL:
fgTrimLevel = ComponentCallbacks2.TRIM_MEMORY_RUNNING_CRITICAL;
break;
case ProcessStats.ADJ_MEM_FACTOR_LOW:
fgTrimLevel = ComponentCallbacks2.TRIM_MEMORY_RUNNING_LOW;
break;
default:
fgTrimLevel = ComponentCallbacks2.TRIM_MEMORY_RUNNING_MODERATE;
break;
}
int factor = numTrimming/3;
int minFactor = 2;
if (mHomeProcess != null) minFactor++;
if (mPreviousProcess != null) minFactor++;
if (factor < minFactor) factor = minFactor;
int curLevel = ComponentCallbacks2.TRIM_MEMORY_COMPLETE;
for (int i=N-1; i>=0; i--) {
ProcessRecord app = mLruProcesses.get(i);
if (allChanged || app.procStateChanged) {
setProcessTrackerStateLocked(app, trackerMemFactor, now);
app.procStateChanged = false;
}
if (app.curProcState >= ActivityManager.PROCESS_STATE_HOME
&& !app.killedByAm) {
if (app.trimMemoryLevel < curLevel && app.thread != null) {
try {
if (DEBUG_SWITCH || DEBUG_OOM_ADJ) Slog.v(TAG_OOM_ADJ,
"Trimming memory of " + app.processName + " to " + curLevel);
app.thread.scheduleTrimMemory(curLevel);
} catch (RemoteException e) {
}
if (false) {
// For now we won't do this; our memory trimming seems
// to be good enough at this point that destroying
// activities causes more harm than good.
if (curLevel >= ComponentCallbacks2.TRIM_MEMORY_COMPLETE
&& app != mHomeProcess && app != mPreviousProcess) {
// Need to do this on its own message because the stack may not
// be in a consistent state at this point.
// For these apps we will also finish their activities
// to help them free memory.
mStackSupervisor.scheduleDestroyAllActivities(app, "trim");
}
}
}
app.trimMemoryLevel = curLevel;
step++;
if (step >= factor) {
step = 0;
switch (curLevel) {
case ComponentCallbacks2.TRIM_MEMORY_COMPLETE:
curLevel = ComponentCallbacks2.TRIM_MEMORY_MODERATE;
break;
case ComponentCallbacks2.TRIM_MEMORY_MODERATE:
curLevel = ComponentCallbacks2.TRIM_MEMORY_BACKGROUND;
break;
}
}
} else if (app.curProcState == ActivityManager.PROCESS_STATE_HEAVY_WEIGHT) {
if (app.trimMemoryLevel < ComponentCallbacks2.TRIM_MEMORY_BACKGROUND
&& app.thread != null) {
try {
if (DEBUG_SWITCH || DEBUG_OOM_ADJ) Slog.v(TAG_OOM_ADJ,
"Trimming memory of heavy-weight " + app.processName
+ " to " + ComponentCallbacks2.TRIM_MEMORY_BACKGROUND);
app.thread.scheduleTrimMemory(
ComponentCallbacks2.TRIM_MEMORY_BACKGROUND);
} catch (RemoteException e) {
}
}
app.trimMemoryLevel = ComponentCallbacks2.TRIM_MEMORY_BACKGROUND;
} else {
if ((app.curProcState >= ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND
|| app.systemNoUi) && app.pendingUiClean) {
// If this application is now in the background and it
// had done UI, then give it the special trim level to
// have it free UI resources.
final int level = ComponentCallbacks2.TRIM_MEMORY_UI_HIDDEN;
if (app.trimMemoryLevel < level && app.thread != null) {
try {
if (DEBUG_SWITCH || DEBUG_OOM_ADJ) Slog.v(TAG_OOM_ADJ,
"Trimming memory of bg-ui " + app.processName
+ " to " + level);
app.thread.scheduleTrimMemory(level);
} catch (RemoteException e) {
}
}
app.pendingUiClean = false;
}
if (app.trimMemoryLevel < fgTrimLevel && app.thread != null) {
try {
if (DEBUG_SWITCH || DEBUG_OOM_ADJ) Slog.v(TAG_OOM_ADJ,
"Trimming memory of fg " + app.processName
+ " to " + fgTrimLevel);
app.thread.scheduleTrimMemory(fgTrimLevel);
} catch (RemoteException e) {
}
}
app.trimMemoryLevel = fgTrimLevel;
}
}
} else {
if (mLowRamStartTime != 0) {
mLowRamTimeSinceLastIdle += now - mLowRamStartTime;
mLowRamStartTime = 0;
}
for (int i=N-1; i>=0; i--) {
ProcessRecord app = mLruProcesses.get(i);
if (allChanged || app.procStateChanged) {
setProcessTrackerStateLocked(app, trackerMemFactor, now);
app.procStateChanged = false;
}
if ((app.curProcState >= ActivityManager.PROCESS_STATE_IMPORTANT_BACKGROUND
|| app.systemNoUi) && app.pendingUiClean) {
if (app.trimMemoryLevel < ComponentCallbacks2.TRIM_MEMORY_UI_HIDDEN
&& app.thread != null) {
try {
if (DEBUG_SWITCH || DEBUG_OOM_ADJ) Slog.v(TAG_OOM_ADJ,
"Trimming memory of ui hidden " + app.processName
+ " to " + ComponentCallbacks2.TRIM_MEMORY_UI_HIDDEN);
app.thread.scheduleTrimMemory(
ComponentCallbacks2.TRIM_MEMORY_UI_HIDDEN);
} catch (RemoteException e) {
}
}
app.pendingUiClean = false;
}
app.trimMemoryLevel = 0;
}
}
if (mAlwaysFinishActivities) {
// Need to do this on its own message because the stack may not
// be in a consistent state at this point.
mStackSupervisor.scheduleDestroyAllActivities(null, "always-finish");
}
if (allChanged) {
requestPssAllProcsLocked(now, false, mProcessStats.isMemFactorLowered());
}
// Update from any uid changes.
for (int i=mActiveUids.size()-1; i>=0; i--) {
final UidRecord uidRec = mActiveUids.valueAt(i);
int uidChange = UidRecord.CHANGE_PROCSTATE;
if (uidRec.setProcState != uidRec.curProcState) {
if (DEBUG_UID_OBSERVERS) Slog.i(TAG_UID_OBSERVERS,
"Changes in " + uidRec + ": proc state from " + uidRec.setProcState
+ " to " + uidRec.curProcState);
if (ActivityManager.isProcStateBackground(uidRec.curProcState)) {
if (!ActivityManager.isProcStateBackground(uidRec.setProcState)) {
uidRec.lastBackgroundTime = nowElapsed;
if (!mHandler.hasMessages(IDLE_UIDS_MSG)) {
// Note: the background settle time is in elapsed realtime, while
// the handler time base is uptime. All this means is that we may
// stop background uids later than we had intended, but that only
// happens because the device was sleeping so we are okay anyway.
mHandler.sendEmptyMessageDelayed(IDLE_UIDS_MSG, BACKGROUND_SETTLE_TIME);
}
}
} else {
if (uidRec.idle) {
uidChange = UidRecord.CHANGE_ACTIVE;
uidRec.idle = false;
}
uidRec.lastBackgroundTime = 0;
}
uidRec.setProcState = uidRec.curProcState;
enqueueUidChangeLocked(uidRec, -1, uidChange);
noteUidProcessState(uidRec.uid, uidRec.curProcState);
}
}
if (mProcessStats.shouldWriteNowLocked(now)) {
mHandler.post(new Runnable() {
@Override public void run() {
synchronized (ActivityManagerService.this) {
mProcessStats.writeStateAsyncLocked();
}
}
});
}
if (DEBUG_OOM_ADJ) {
final long duration = SystemClock.uptimeMillis() - now;
if (false) {
Slog.d(TAG_OOM_ADJ, "Did OOM ADJ in " + duration + "ms",
new RuntimeException("here").fillInStackTrace());
} else {
Slog.d(TAG_OOM_ADJ, "Did OOM ADJ in " + duration + "ms");
}
}
}
复制代码- 首先基于app当前状态更新每个app进程的oom_adj值，在这里面也可以看到，对于孤立的进程(即无service在其中运行的进程)，AMS会有很大的概率将其杀死(原话是agressively kill these)，所以如果想保活的话，要尽可能避免这种情况的出现
- 然后决定后台进程的memory trimming level
- 最后会根据uid的改变来更新oom_adj值
client端的连接回调
再回到ActiveServices中的bindServiceLocked()方法中，在更新完Service所在进程的oom_adj值之后，会通过IServiceConnection进行IPC调用，对应的代码片段为:
if (s.app != null && b.intent.received) {
// Service is already running, so we can immediately
// publish the connection.
try {
//Service已经正在运行，则调用InnerConnection的代理对象
c.conn.connected(s.name, b.intent.binder);
} catch (Exception e) {
Slog.w(TAG, "Failure sending service " + s.shortName
+ " to connection " + c.conn.asBinder()
+ " (in " + c.binding.client.processName + ")", e);
}
// If this is the first app connected back to this binding,
// and the service had previously asked to be told when
// rebound, then do so.
if (b.intent.apps.size() == 1 && b.intent.doRebind) {
//rebind过程
requestServiceBindingLocked(s, b.intent, callerFg, true);
}
} else if (!b.intent.requested) {
requestServiceBindingLocked(s, b.intent, callerFg, false);
}
复制代码显然是调用IServiceConnection的connected()方法，再回到ContextImpl中，IServiceConnection的真正实现在ServiceDispatcher中的InnerConnection中:
private static class InnerConnection extends IServiceConnection.Stub {
final WeakReference<LoadedApk.ServiceDispatcher> mDispatcher;
InnerConnection(LoadedApk.ServiceDispatcher sd) {
mDispatcher = new WeakReference<LoadedApk.ServiceDispatcher>(sd);
}
public void connected(ComponentName name, IBinder service) throws RemoteException {
LoadedApk.ServiceDispatcher sd = mDispatcher.get();
if (sd != null) {
sd.connected(name, service);
}
}
}
复制代码通过IPC调用之后会调用ServiceDispatcher的connected()方法，该方法如下:
public void connected(ComponentName name, IBinder service) {
if (mActivityThread != null) {
//这是主线程的Handler
mActivityThread.post(new RunConnection(name, service, 0));
} else {
doConnected(name, service);
}
}
复制代码由于回调是在binder线程中，这里通过ActivityThread中的Handler将事件发布到主线程中，其中RunConnection如下:
private final class RunConnection implements Runnable {
RunConnection(ComponentName name, IBinder service, int command) {
mName = name;
mService = service;
mCommand = command;
}
public void run() {
if (mCommand == 0) {
doConnected(mName, mService);
} else if (mCommand == 1) {
doDeath(mName, mService);
}
}
final ComponentName mName;
final IBinder mService;
final int mCommand;
}
复制代码而doConnected()方法如下:
public void doConnected(ComponentName name, IBinder service) {
ServiceDispatcher.ConnectionInfo old;
ServiceDispatcher.ConnectionInfo info;
synchronized (this) {
if (mForgotten) {
// We unbound before receiving the connection; ignore
// any connection received.
return;
}
old = mActiveConnections.get(name);
if (old != null && old.binder == service) {
// Huh, already have this one. Oh well!
return;
}
if (service != null) {
// A new service is being connected... set it all up.
info = new ConnectionInfo();
info.binder = service;
//创建死亡监听对象
info.deathMonitor = new DeathMonitor(name, service);
try {
//建立死亡通知
service.linkToDeath(info.deathMonitor, 0);
mActiveConnections.put(name, info);
} catch (RemoteException e) {
// This service was dead before we got it... just
// don't do anything with it.
mActiveConnections.remove(name);
return;
}
} else {
// The named service is being disconnected... clean up.
mActiveConnections.remove(name);
}
if (old != null) {
old.binder.unlinkToDeath(old.deathMonitor, 0);
}
}
// If there was an old service, it is not disconnected.
if (old != null) {
mConnection.onServiceDisconnected(name);
}
// If there is a new service, it is now connected.
if (service != null) {
//回调用户定义的ServiceConnection()
mConnection.onServiceConnected(name, service);
}
}
复制代码实际上主要就是做了三件事:
- 处理之前的ServiceConnection
- 为IBinder添加DeathMonitor,一旦发现binderDied()回调，就进行连接记录的修改，以及调用ServiceConnection的onDisconnected()方法
- 调用我们创建的ServiceConnection的onServiceConnected()方法。
AMS.bindServiceLocked()末尾
if (s.app != null && b.intent.received) {
// Service is already running, so we can immediately
// publish the connection.
try {
//Service已经正在运行，则调用InnerConnection的代理对象
c.conn.connected(s.name, b.intent.binder);
} catch (Exception e) {
Slog.w(TAG, "Failure sending service " + s.shortName
+ " to connection " + c.conn.asBinder()
+ " (in " + c.binding.client.processName + ")", e);
}
// If this is the first app connected back to this binding,
// and the service had previously asked to be told when
// rebound, then do so.
if (b.intent.apps.size() == 1 && b.intent.doRebind) {
//rebind过程
requestServiceBindingLocked(s, b.intent, callerFg, true);
}
} else if (!b.intent.requested) {
requestServiceBindingLocked(s, b.intent, callerFg, false);
}
复制代码有两种情形，一种是某个进程重连到远程Service时，会调用requestServiceBindingLocked()方法:
private final boolean requestServiceBindingLocked(ServiceRecord r, IntentBindRecord i,
boolean execInFg, boolean rebind) throws TransactionTooLargeException {
if (r.app == null || r.app.thread == null) {
// If service is not currently running, can't yet bind.
return false;
}
if ((!i.requested || rebind) && i.apps.size() > 0) {
try {
//发送bind开始的消息
bumpServiceExecutingLocked(r, execInFg, "bind");
r.app.forceProcessStateUpTo(ActivityManager.PROCESS_STATE_SERVICE);
//服务进入 onBind()
r.app.thread.scheduleBindService(r, i.intent.getIntent(), rebind, r.app.repProcState);
if (!rebind) {
i.requested = true;
}
i.hasBound = true;
i.doRebind = false;
} catch (TransactionTooLargeException e) {
// Keep the executeNesting count accurate.
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Crashed while binding " + r, e);
final boolean inDestroying = mDestroyingServices.contains(r);
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
throw e;
} catch (RemoteException e) {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Crashed while binding " + r);
// Keep the executeNesting count accurate.
final boolean inDestroying = mDestroyingServices.contains(r);
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
return false;
}
}
return true;
}
复制代码注意r.app.thread.scheduleBindService()这个调用，这其实是会IPC调用到Service所在进程的ApplicationThread的scheduleBindService()方法,之后通过Handler在主线程调用handleBindService()方法：
private void handleBindService(BindServiceData data) {
Service s = mServices.get(data.token);
if (DEBUG_SERVICE)
Slog.v(TAG, "handleBindService s=" + s + " rebind=" + data.rebind);
if (s != null) {
try {
data.intent.setExtrasClassLoader(s.getClassLoader());
data.intent.prepareToEnterProcess();
try {
if (!data.rebind) {
// 执行Service.onBind()回调方法
IBinder binder = s.onBind(data.intent);
//将onBind返回值传递回去
ActivityManagerNative.getDefault().publishService(data.token, data.intent, binder);
} else {
s.onRebind(data.intent);
ActivityManagerNative.getDefault().serviceDoneExecuting(
data.token, SERVICE_DONE_EXECUTING_ANON, 0, 0);
}
ensureJitEnabled();
} catch (RemoteException ex) {
throw ex.rethrowFromSystemServer();
}
} catch (Exception e) {
if (!mInstrumentation.onException(s, e)) {
throw new RuntimeException(
"Unable to bind to service " + s
+ " with " + data.intent + ": " + e.toString(), e);
}
}
}
}
复制代码分两种情况，如果不是重新连接，调用AMP的publishService()方法;否则调用AMP的serviceDoneExecuting()方法，这个方法其实是一个IPC调用，告诉AMS重新连接上了，然后AMS就会更新它的ServiceRecord和相应进程的oom_adj值。
AMS.publishService()
public void publishService(IBinder token, Intent intent, IBinder service) {
// Refuse possible leaked file descriptors
if (intent != null && intent.hasFileDescriptors() == true) {
throw new IllegalArgumentException("File descriptors passed in Intent");
}
synchronized(this) {
if (!(token instanceof ServiceRecord)) {
throw new IllegalArgumentException("Invalid service token");
}
mServices.publishServiceLocked((ServiceRecord)token, intent, service);
}
}
复制代码调用到ActiveServices的publishServiceLocked()方法:
void publishServiceLocked(ServiceRecord r, Intent intent, IBinder service) {
final long origId = Binder.clearCallingIdentity();
try {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "PUBLISHING " + r + " " + intent + ": " + service);
if (r != null) {
Intent.FilterComparison filter = new Intent.FilterComparison(intent);
IntentBindRecord b = r.bindings.get(filter);
if (b != null && !b.received) {
b.binder = service;
b.requested = true;
b.received = true;
for (int conni=r.connections.size()-1; conni>=0; conni--) {
ArrayList<ConnectionRecord> clist = r.connections.valueAt(conni);
for (int i=0; i<clist.size(); i++) {
ConnectionRecord c = clist.get(i);
if (!filter.equals(c.binding.intent.intent)) {
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Not publishing to: " + c);
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Bound intent: " + c.binding.intent.intent);
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Published intent: " + intent);
continue;
}
if (DEBUG_SERVICE) Slog.v(TAG_SERVICE, "Publishing to: " + c);
try {
//LoadedApk.ServiceDispatcher.InnerConnection.connected()
c.conn.connected(r.name, service);
} catch (Exception e) {
Slog.w(TAG, "Failure sending service " + r.name +
" to connection " + c.conn.asBinder() +
" (in " + c.binding.client.processName + ")", e);
}
}
}
}
serviceDoneExecutingLocked(r, mDestroyingServices.contains(r), false);
}
} finally {
Binder.restoreCallingIdentity(origId);
}
}
复制代码这个方法主要做了两件事:
- 给IntentBindRecord赋值，其中的binder就是Service中onBind()所返回的binder;
- 这里会先判断b.received是否为false,如果为false说明发起绑定流程的进程还没收到绑定回调，从而此时会再次通过IPC调用到InnerConnection的connected()方法。
serviceDoneExecutingLocked()方法:
void serviceDoneExecutingLocked(ServiceRecord r, int type, int startId, int res) {
boolean inDestroying = mDestroyingServices.contains(r);
if (r != null) {
if (type == ActivityThread.SERVICE_DONE_EXECUTING_START) {
// This is a call from a service start... take care of
// book-keeping.
r.callStart = true;
switch (res) {
case Service.START_STICKY_COMPATIBILITY:
case Service.START_STICKY: {
// We are done with the associated start arguments.
r.findDeliveredStart(startId, true);
// Don't stop if killed.
r.stopIfKilled = false;
break;
}
case Service.START_NOT_STICKY: {
// We are done with the associated start arguments.
r.findDeliveredStart(startId, true);
if (r.getLastStartId() == startId) {
// There is no more work, and this service
// doesn't want to hang around if killed.
r.stopIfKilled = true;
}
break;
}
case Service.START_REDELIVER_INTENT: {
// We'll keep this item until they explicitly
// call stop for it, but keep track of the fact
// that it was delivered.
ServiceRecord.StartItem si = r.findDeliveredStart(startId, false);
if (si != null) {
si.deliveryCount = 0;
si.doneExecutingCount++;
// Don't stop if killed.
r.stopIfKilled = true;
}
break;
}
case Service.START_TASK_REMOVED_COMPLETE: {
// Special processing for onTaskRemoved(). Don't
// impact normal onStartCommand() processing.
r.findDeliveredStart(startId, true);
break;
}
default:
throw new IllegalArgumentException(
"Unknown service start result: " + res);
}
if (res == Service.START_STICKY_COMPATIBILITY) {
r.callStart = false;
}
} else if (type == ActivityThread.SERVICE_DONE_EXECUTING_STOP) {
// This is the final call from destroying the service... we should
// actually be getting rid of the service at this point. Do some
// validation of its state, and ensure it will be fully removed.
if (!inDestroying) {
// Not sure what else to do with this... if it is not actually in the
// destroying list, we don't need to make sure to remove it from it.
// If the app is null, then it was probably removed because the process died,
// otherwise wtf
if (r.app != null) {
Slog.w(TAG, "Service done with onDestroy, but not inDestroying: "
+ r + ", app=" + r.app);
}
} else if (r.executeNesting != 1) {
Slog.w(TAG, "Service done with onDestroy, but executeNesting="
+ r.executeNesting + ": " + r);
// Fake it to keep from ANR due to orphaned entry.
r.executeNesting = 1;
}
}
final long origId = Binder.clearCallingIdentity();
//
serviceDoneExecutingLocked(r, inDestroying, inDestroying);
Binder.restoreCallingIdentity(origId);
} else {
Slog.w(TAG, "Done executing unknown service from pid "
+ Binder.getCallingPid());
}
}
复制代码整个方法中，对我们来说最重要的就是mAm.updateOomAdjLocked(r.app);这个方法了,因为它涉及到更新Service所在进程的oom_adj值，而oom_adj值直接关系到内存回收。
总结
通过前面的分析，可以知道bindService()其实主要做了以下事情:
- 在client侧，ServiceConnection对象会保存在ServiceDispatcher对象中，并且利用IServiceConnection进行IPC
- AMS会根据bindService中的Intent信息，会对Service侧进行两次IPC调用，分别是IApplicationThread的scheduleCreateService()和scheduleBindService(), 前者是为了新建Service, 而后者是获取Service的onBind()所返回的IBinder对象
- 之后AMS会根据client端和Service本身的的情况来调整Service所在进程的oom_adj值
- 最后通过IServiceConnection的proxy进行IPC,最终调用到client端的ServiceConnection的onServiceConnected()方法
扫一扫
4304
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
